Powdery Mildew of Grape on Leaves – Is It a Problem?

IOBC / WPRS

Working Group “Integrated Protection in Viticulture”

OILB / SROP

Groupe de Travail “Lutte Intégrée en Viticulture”

Proceedings of the meeting

Marsala, Sicily (Italy)

25-27 October, 2007

Editors: Giuseppe Carlo Lozzia, Andrea Lucchi, Salvatore Ragusa Di Chiara & Haralabos Tsolakis

IOBC wprs Bulletin Bulletin OILB srop Vol. 36, 2008 2

The content of the contributions is in the responsibility of the authors

The IOBC/WPRS Bulletin is published by the International Organization for Biological and Integrated Control of Noxious Animals and Plants, West Palearctic Regional Section (IOBC/WPRS)

Le Bulletin OILB/SROP est publié par l‘Organisation Internationale de Lutte Biologique et Intégrée contre les Animaux et les Plantes Nuisibles, section Regionale Ouest Paléarctique (OILB/SROP)

The Publication Commission of the IOBC/WPRS:

Horst Bathon Luc Tirry Julius Kühn-Institute (JKI) University of Gent Federal Research Center for Cultivated Plants Laboratory of Agrozoology Institute for Biological Control Department of Crop Protection Heinrichstr. 243 Coupure Links 653 D-64287 Darmstadt (Germany) B-9000 Gent (Belgium) Tel +49 6151 407-225, Fax +49 6151 407-290 Tel +32-9-2646152, Fax +32-9-2646239 e-mail: [email protected] e-mail: [email protected]

Address General Secretariat:

Dr. Philippe C. Nicot INRA – Unité de Pathologie Végétale Domaine St Maurice - B.P. 94 F-84143 Montfavet Cedex (France)

ISBN 978-92-9067-210-4 www.iobc-wprs.org Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. i-iii

Editorial

In leaving the position of convenor of this group, that has given me great satisfaction in the last 6 years, I would like to thank all the people, reported in Table 1, who have helped in organizing the 5 Meetings that have taken place under my auspices in Florence, Ponte de Lima, Volos, Boario Terme and, lastly, this one in Marsala. These occasions have been very important moments of confrontation that have led to very important results in the protection of the vine, so much so that most of our results have been included in local, regional and national directives and legislations. The agro-ecosystem of the vineyard with its interactions, phytoseids as controllers of mite populations, the finalization of forecasting models for diseases, the determination of threshold levels and the practical and widespread use of mating disruption methods are some of the undeniable results that have turned viticulture into a practice that can be defined highly integrated, fully respecting the environment. All this has been obtained thanks to the collaboration of researchers from different countries, to the constant exchange of information. As a matter of fact the international group has worked in the interest of the European Viticulture sector, and remains an example for the researchers because is one of the most long-standing working group of the IOBC/wprs, and its scientific production is relevant and regularly published by the IOBC/wprs bulletins.

Table 1. WG Meetings Local Organizers from 1999 to 2007

FIRENZE: Dott. BRUNO BAGNOLI, ISZA Florence; ARSIA, Regione Toscana

PONTE DE LIMA: Dr. JOSÈ RIBEIRO, UTAD, Vila Real; Prof. PEDRO AMARO, Instituto Superior de Agromonia, Lisboa; Agricultural High School, Ponte de Lima; Polytechnic Institute, Viana de Castelo

VOLOS: Dr. C. SOULIOTIS, Benaki Phytopatological Institute

DARFO – BOARIO TERME: Prof. IVO RIGAMONTI, Milan University; Università della Montagna, Edolo; Centro Vitivinicolo Bresciano

MARSALA: Prof. SALVATORE RAGUSA, SENFIMIZO, Palermo University; Prof. HARALABOS TSOLAKIS, SENFIMIZO, Palermo University; Regione Autonoma Sicilia

In Fig. 2 is reported the detail of the papers presented in the vatious Meetings in comparison with those presented in 1981 at Gargnano. It is noteworthy that new topics as biodiversity and IPM applications have become more and more important, as well as the influence of the cultivation activities. It would be very difficult to remember all the people who have worked and coordinated this group, from MARC BAJLOD to AUGUSTIN SCHMIDT and ERNST BOLLER. Many of them

i ii

have left, retired or passed away. I would like to remember my friend JACQUES STOCKEL and professor MINOS MARTELLI and MARIO BAGGIOLINI, who died last summer; they have been, together with others, two of the founders of the group.

Fig. 2. Number of papers presented in the different Meetings from 1999 to 2007 in comparison with the Meeting at Gargnano.

Fig. 3. Number of researchers attending to the Meetings at: Firenze (122), Ponte de Lima (138), Volos (217) and Darfo-Boario Terme (82). iii

In the Meeting room I see many new faces…..of young people. They are the future, and this continuous change can only be positive for the future. I also see new additions from other countries such as Israel, the USA and Hungary, which give us the possibility to exchange viticultural knowledge and practices from other countries. In Fig. 3 are represented, for each Country, the participants to the different Meetings. Now, I don't think I can say what subjects will be faced in the future, but surely biodiversity, the climatic changes and the new defense perspectives are all aspects that will be studied, as well as how new diseases or insects must be faced and eradicated in the next years. I'm sure this group will be able to face the new problems and to propose innovative and suitable solutions maintaining the role of guide for plant/vine protection. I would like to thank the local organizers of this meeting. I wish you all a good and profitable work. I also want to offer my congratulations and wishes for a good job on Agnes Calonnec, recently elected as new group convenor, hoping that all the colleagues will continue to help her as they helped me in the last years in the group management. Finally I’d like to offer my gratitude to CESARE GESSLER, a friend that worked with me to keep in touch with the General IOBC Committee.

As far as I'm concerned, I will remain a member of this group and I will be very happy to take part to the next IOBC meeting on IPM of grapevine, with my usual working enthusiasm, not forgetting to have a drink with old friends. Maybe the wine will help us to produce new good ideas! Who knows??!?

Prof. Carlo Lozzia Istituto di Entomologia Agraria Università di Milano

List of participants

ADDANTE Rocco BAUS-REICHEL Ottmar Università dagli Studi di Bari Research Center Geisenheim Biologia e Chimica Agro-Forestale ed Ambientale Phytomedicine Via Amendola, 165/A Von-Lade-Strasse 1 79126 Bari – Italy 65366 Geisenheim – Germany [email protected] [email protected]

AGRÒ Alfonso BLEYER Gottfried Università degli Studi di Palermo Staatliches Weinbauinstitut Freiburg Dipartimento S.En.Fi.Mi.Zo. Abteilung Biologie Viale delle Scienze, 13 Dept. of Biology / Ecology 90128 Palermo – Italy Merzhauserstr. 119 [email protected] 79100 Freiburg – Germany [email protected] ALFONZO Antonio Università degli Studi di Palermo BLÜMEL Sylvia Dipartimento S.En.Fi.Mi.Zo. Austrian Agency of Health and Food Safety Viale delle Scienze, 13 Institute for Plant Health 90128 Palermo – Italy Spargelfeldstraße 191 [email protected] 1226 Vienna – Austria [email protected] ANFORA Gianfranco IASMA Research Center BOTTON Marcos SafeCrop Centre Embrapa, Entomology Via E. Mach 1 Rua Livramento 515 38010 San Michele all'Adige – Italy 95700000 Bento Gonçalves – RS – Brazil [email protected] BRETH Karl ANGELI Dario DLR-Oppenheim IASMA Research Center Department of viticulture SafeCrop Centre Bachstrasse 15 Via E. Mach 1 67577 Alsheim – Germany 38010 San Michele all'Adige – Italy [email protected] [email protected] BREUER Michael ANGELI Gino State Institute for Viticulture and Enology, Biology IASMA Research Centre Merzhauser Str. 119 Plant Protection dept. 79100 Freiburg – Germany via E. Mach, 1 [email protected] 38010 San Michele all'Adige – Italy [email protected] BUONOCORE Emanuele Regione Siciliana ASCIUTO Antonio Servizio Fitosanitario Dipartimento di Economia dei Sistemi Agro- C.da Fanello Forestali 97019 Vittoria (RG) – Italy Università degli Studi di Palermo [email protected] Viale delle Scienze, 13 90128 Palermo – Italy BURRUANO Santella [email protected] Università degli Studi di Palermo Dipartimento S.En.Fi.Mi.Zo. Viale delle Scienze, 13 90128 Palermo – Italy [email protected] vi

CALONNEC Agnès DELRIO Gavino INRA Università degli studi di Sassari Department: UMR1065 Santé végétale Dipartimento di protezione delle Piante 71 avenue Edouard Bourlaux Via E. De Nicola 33183 Villenave d'Ornon – France 07100 Sassari – Italy [email protected] [email protected]

CANGELOSI Benedetta DUSO Carlo Università degli Studi di Palermo Università degli studi di Padova Dipartimento S.En.Fi.Mi.Zo. Dipartimento di Agronomia Ambientale e Viale delle Scienze, 13 Produzioni Vegetali 90128 Palermo – Italy Viale dell'Università, 16 [email protected] 35020 Legnaro (Padova) – Italy [email protected] CARTOLARO Philippe INRA, Department: Santé des Plantes FERRARO Valeria 71 avenue E. Bourlaux Università degli Studi di Palermo 33140 Villenave d'Ornon – France Dipartimento S.En.Fi.Mi.Zo. [email protected] Viale delle Scienze, 13 90128 Palermo – Italy CHEBIL Samir [email protected] Laboratoire de Physiologie Moléc. de la Vigne Centre de Biotechnologie Borj Cédria FISCHER Michael BP. 901 – Route Touristique Weinbauinstitut Freiburg 2050 Hammam-lif – Tunisia Department: Plant protection [email protected] Merzhauser Str. 119 79100 Freiburg – Germany CONIGLIARO Gaetano [email protected] Università degli Studi di Palermo Dipartimento S.En.Fi.Mi.Zo. FORNASIERO Diego Viale delle Scienze, 13 Università degli studi di Padova 90128 Palermo – Italy Dipartimento di Agronomia Ambientale e Produzioni Vegetali COULON Thierry Viale dell'Università, 16 ENTAV/ITV FRANCE 35020 Legnaro (Padova) – Italy Department: Viticulture [email protected] 39, rue Michel Montaigne 33290 Blanquefort – France FORTE Vally [email protected] Università degli studi di Padova Dipartimento di Agronomia Ambientale e DAGOSTIN Silvia Produzioni Vegetali IASMA Research Center Viale dell'Università, 16 SafeCrop Centre 35020 Legnaro (Padova) – Italy Via E. Mach 1 [email protected] 38010 San Michele all'Adige – Italy [email protected] FUCARINO Alessandro Università degli Studi di Palermo DAL CORTIVO Gianluca Dipartimento S.En.Fi.Mi.Zo. Universita' di Padova Viale delle Scienze, 13, Dipartimento di Agronomia Ambientale e 90128 Palermo – Italy Produzioni Vegetali [email protected] Viale dell'uniersita' 16 35020 Legnaro (Padova) – Italy FULCHIN Emma Association in sustainable viticulture DE CRISTOFARO Antonio 1 cours du Général de Gaulle Department of Animal, Plant and Environmental 33170 Gradignan – France Sciences, University of Molise, Via De Sanctis, [email protected] 86100 Campobasso – Italy

vii

GESSLER Cesare HRNCIC Snjezana ETH Zürich University of Monenegro – Biotechnical Department of Agronomy Department of Plant Protection Universitätstrasse 2 Kralja Nikole bb, 8092 Zürich – Switzerland 81 000 Podgorica – Montenegro [email protected] [email protected]

GOBBIN Davide IORIATTI Claudio ETH Zürich IASMA Research Center Department of Plant Pathology Department of Plant Protection Universitätstrasse 2 via E. Mach 1 8092 Zürich – Switzerland 38010 S.Michele all'Adige – Italy [email protected] [email protected]

GUARINO Salvatore JERMINI Mauro Università degli Studi di Palermo Agroscope Changins-Wädenswil ACW Dipartimento S.En.Fi.Mi.Zo. Centro di Cadenazzo Viale delle Scienze, 13 6594 Contone – Switzerland 90128 Palermo – Italy [email protected] [email protected] JORDA PALOMERO Raul HARARI Ally Università degli Studi di Palermo (The Volcani Center, Israel) Dipartimento S.En.Fi.Mi.Zo. Department of Entomology Viale delle Scienze, 13 Lund University, Chemical Ecology 90128 Palermo – Italy SE-223 62 Lund – Sweden [email protected] KAST Walter Klaus Staatliche Lehr- und Versuchsanstalt Viticulture & Horticulture HEIBERTSHAUSEN Dagmar Geisenheim Research Center Traubenplatz 5 Department: Section of Phytomedicine 74189 Weinsberg – Germany Von-Lade-Str. 1 [email protected]

65366 Geisenheim – Germany KEHRLI Patrik [email protected] Agroscope Changins-Wädenswil ACW

Entomology HERRMANN Josef Valentin Route de Duillier CP 1012 Bayerische Landesanstalt für 1261 Nyon – Switzerland Weinbau und Gartenbau [email protected] Herrnstraße 8

97209 Veitshöchheim – Germany KUNTZMANN Philippe [email protected] Entav-Itv France Department of Viticulture HILL Georg 28, rue Herrlisheim DLR Rheinhessen-Nahe 68000 Colmar – France Department of viticulture [email protected] DLR RNH 55276 Oppenheim – Germany LASRAM Salma [email protected] Centre de Biotechnologie Borj Cédria Laboratoire de Physiologie Moléculaire de la Vigne HOFFMANN Christoph BP. 901 – Route Touristique JKI Institute for plant protection in viticulture 2050 Hammam-lif – Tunisia Department of Entomology [email protected] Brüningstr. 84 54470 Bernkastel-Kues – Germany LATINOVIC Nedeljko [email protected] University of Montenegro, Biotechnical Department of Plant Protection Kralja Nikole bb 81 000 Podgorica – Montenegro [email protected] viii

LÉGER Bertrand 90128 Palermo – Italy INRA UMR Sante vegetal [email protected] Domaine de la Grande Ferrade 71, Avenue Edouard Bourlaux BP81 LOZZIA Giuseppe Carlo 33883 Villenave d'Ornon – France Università degli studi di Milano [email protected] Istituto di Entomologia Agraria Via Caloria, 2 LENTINI Andrea 20133 Milano – Italy Università degli studi di Sassari [email protected] Dipartimento di Protezione delle Piante Via E. De Nicola LUCAS ESPADAS Alfonso 07100 Sassari – Italy Servicio de Sanitad Vegetal [email protected] Consejeria de Agricultura y Agua de Murcia C/ Mayor, S/N LINDER Christian 30150 Murcia – Spain Agroscope Changins-Wädenswil ACW [email protected] Department of Entomology Route de Duillier CP 1012 LUCCHI Andrea 1261 Nyon – Switzerland Università degli studi di Pisa [email protected] Dipartimento Coltivazione e Difesa Specie Legnose Sez. Entomologia Agraria LO BUE Mauro Via S. Michele 2 Università degli Studi di Palermo 56124 Pisa – Italy Dipartimento S.En.Fi.Mi.Zo. [email protected] Viale delle Scienze, 13 90128 Palermo – Italy LUCIDO Paolo [email protected] Università degli Studi di Palermo Dipartimento S.En.Fi.Mi.Zo. LO BUE Paolo Viale delle Scienze, 13 Università degli Studi di Palermo 90128 Palermo – Italy Dipartimento S.En.Fi.Mi.Zo. [email protected] Viale delle Scienze, 13 90128 Palermo – Italy MAcDONALD Al Chemeketa Community College LO GENCO Alessandro Department of Viticulture Università degli Studi di Palermo 4000 Lancaster Drive NE Dipartimento S.En.Fi.Mi.Zo. 97308 Salem, Oregon – USA Viale delle Scienze, 13 [email protected] 90128 Palermo – Italy [email protected] MAIXNER Michael Julius-Kühn-Institute (JKI) LO PICCOLO Sandra Institute for Plant Protection in Viticulture Università degli Studi di Palermo Brueningstrasse 84 Dipartimento S.En.Fi.Mi.Zo. 54470 Bernkastel-Kues – Germany Viale delle Scienze, 13 [email protected] 90128 Palermo – Italy [email protected] MARTORANA Alessandra Università degli Studi di Palermo LO PINTO Mirella Dipartimento S.En.Fi.Mi.Zo. Università degli Studi di Palermo Viale delle Scienze, 13 Dipartimento S.En.Fi.Mi.Zo. 90128 Palermo – Italy Viale delle Scienze, 13 [email protected] 90128 Palermo – Italy [email protected] MATASCI Caterina ETH Zürich LO VERDE Gabriella Department of Plant Pathology Università degli Studi di Palermo Universitätsstr. 2 Dipartimento S.En.Fi.Mi.Zo. 8092 Zürich – Switzerland Viale delle Scienze, 13 [email protected] ix

MAZZOCCHETTI Angelo POZZEBON Alberto A.R.S.S.A. – Agenzia Regionale Università degli Studi di Padova per i Servizi di Sviluppo Agricolo Dipartimento DAAPV Integrated Pest Management viale dell'Università, 16 Via Nazionale, 38 35020 Legnaro – Italy 65010 Villanova – Italy [email protected] [email protected] POZZOLINI Elena MAZZONI Valerio Università degli studi di Pisa IASMA Research Center Dipartimento Coltivazione e Difesa Specie Legnose SafeCrop Centre Sez. Entomologia Agraria Via E. Mach 1 Via S. Michele 2 38010 San Michele all'Adige – Italy 56124 Pisa – Italy [email protected] [email protected]

MICHELON Lorenza RAGUSA Ernesto IASMA Research Center Università degli Studi di Palermo SafeCrop Centre Dipartimento S.En.Fi.Mi.Zo. Via E. Mach 1 Viale delle Scienze, 13 38010 San Michele all'Adige – Italy 90128 Palermo – Italy [email protected] [email protected]

MOLITOR Daniel RAGUSA Salvatore Geisenheim Research Centre Università degli Studi di Palermo Phytomedicine Dipartimento S.En.Fi.Mi.Zo. Von-Lade-Str. 1 Viale delle Scienze, 13 65366 Geisenheim – Germany 90128 Palermo – Italy [email protected] [email protected]

MONDELLO Vincenzo RAYNAL Marc Università degli Studi di Palermo ENTAV-ITV France Dipartimento S.En.Fi.Mi.Zo. Department of viticulture Viale delle Scienze, 13 39 Rue Michel Montaigne 90128 Palermo – Italy 33290 Blanquefort – France [email protected] [email protected]

MOSCHETTI Giancarlo REINEKE Annette Università degli Studi di Palermo Geisenheim Research Center Dipartimento S.En.Fi.Mi.Zo. Phytomedicine Viale delle Scienze, 13 Von-Lade-Str. 1 90128 Palermo – Italy 65366 Geisenheim – Germany [email protected] [email protected]

PAPAIOANNOU-SOULIOTIS Peggy REISENZEIN Helga Benaki Phytopath. Institute AGES Entomology and Agric. Zoology Department: Inst. for Plant Health Ekalis 2 Spargelfeldstrasse 191 14561 Kiphissia – Greece 1230 Vienna – Austria [email protected] [email protected]

PELLEGRINI Alberto REYNAUD Catherine PERTOT Ilaria La Tapy IASMA Research Center Domaine experimental SafeCrop Centre 1881 Chemin des Galères Via E. Mach 1 84200 Carpentras-Serres – France 38010 San Michele all'Adige – Italy [email protected] [email protected]

RIGAMONTI Ivo SHARON Rakefet Università degli studi di Milano MIGAL- Galilee Technology center Istituto di Entomologia Agraria South industrial area, P.O.B 831 Via Caloria, 2 11016 Kiryat Shmona – Israel 20133 Milano – Italy [email protected] [email protected] SOULIOTIS Costantino SAMBADO Paolo Benaki Phytopath. Institute CBC (Europe) Ltd Entomology and Agric. Zoology Department of Bicontrol Ekalis 2 Via E. Majorana, 2 14561 Kiphissia - Athens – Greece 20054 Nova Milanese – Italy [email protected] [email protected] TASIN Marco SAVINO Francesco IASMA Research Center CBC (Europe) Ltd SafeCrop Centre Department of Biocontrol Via E. Mach 1 Via E. Majorana, 2 38010 San Michele all'Adige – Italy 20054 Nova Milanese – Italy [email protected] [email protected] TIRELLO Paola SCHIRRA Karl-Josef Università degli Studi di Padova DLR Rheinpfalz DAAPV Phytomedicine Viale dell'Università, 16 Breitenweg 71 35020 Legnaro – Italy 67435 Neustadt/Weinstraße – Germany [email protected] [email protected] TORTA Livio Università degli Studi di Palermo SCHWAPPACH Peter Bayerische Landesanstalt für Dipartimento S.En.Fi.Mi.Zo. Weinbau und Gartenbau, Viale delle Scienze, 13 Rebschutz und Physiologie 90128 Palermo – Italy Herrnstrasse 8 [email protected]

97209 Veitshoechheim – Germany TROPEA GARZIA Giovanna [email protected] Università degli studi di Catania

Dipartimento DISTEF SCIARRETTA Andrea Via Santa Sofia Università degli studi del Molise 95123 Catania – Italy Department: SAVA [email protected] via De Sanctis

86100 Campobasso – Italy TSOLAKIS Haralabos [email protected] Università degli Studi di Palermo

Dipartimento S.En.Fi.Mi.Zo. SENTENAC Gilles Viale delle Scienze, 13 ENTAV-ITV France 90128 Palermo – Italy Department of Viticulture [email protected] 6, rue du 16e chasseur

21200 Beaune – France TUTONE Livia [email protected] Università degli Studi di Palermo

Dipartimento S.En.Fi.Mi.Zo. SERRA Giuseppe Viale delle Scienze, 13 CNR 90128 Palermo – Italy Istituto per lo Studio degli Ecosistemi

Via E. De Nicola VAN HELDEN Maarten 07100 Sassari – Italy ENITA Bordeaux [email protected] Plant protection

1 cours général de Gaulle

33175 Gradignan – France

[email protected] xi

VERONELLI Vittorio ZEISNER Norbert CBC (Europe) Ltd Austrian Agency of Health and Food Safety Department of Biocontrol Institute for Plant Health Via E. Majorana, 2 Spargelfeldstraße 191 20054 Nova Milanese – Italy 1226 Vienna – Austria [email protected] [email protected]

WALTER Ruth ZOIDA Giuseppe DLR-Rheinpfalz Università degli Studi di Palermo Phytomedizin Dipartimento S.En.Fi.Mi.Zo. Breitenweg 71 Viale delle Scienze, 13 67435 Neustadt an der Weinstraße – Germany 90128 Palermo – Italy [email protected] [email protected]

ZAHAVI Tirtza ZULINI Luca Ministry of Agriculture IASMA Research Center Plant Protection Department of Agricultural Resources Kibbutz Gshur Via E. Mach 1 12942 Ramat Hagolan – Israel 38010 San Michele all'Adige – Italy [email protected] [email protected]

xii

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. xiii-xvi

Contents

Editorial...... i List of participants...... v

Economic and structural aspects of vitiviniculture in Sicily...... 1-18 A. Asciuto, S. Bacarella “In Vitro” antagonism of a grapevine endophytic Bacillus subtilis strain towards “esca” fungi ...... 19-24 A. Alfonzo, V. Ventorino, L. Torta, S. Burruano, G. Moschetti Efficacy of microorganisms and natural products against grapevine powdery mildew .... 25-30 D. Angeli, L. Maines, C. Sicher, H.-A. Assaf, C. Longa, Y. Elad, V. Simeone, I. Pertot Natural occurrence of Ampelomyces spp. as E. necator mycoparasite in the vineyards of Trentino Province (Northern Italy) and efficacy evaluation of A. quisqualis in integrated powdery mildew management ...... 31-34 D. Angeli, L. Maines, E. di Marino, Enzo Mescalchin, I. Pertot VitiMeteoPlasmopara – a modern tool for integrated fungicide strategies...... 35-36 G. Bleyer, H.-H. Kassemeyer; O. Viret, W. Siegfried; R. Krause Grapevine infectious diseases in Sicily...... 37-44 S. Burruano, G. Granata Evaluation of grapevine resistance to downy and powdery mildew in a population segregating for run1 and rpv1 resistance genes...... 45-52 A.Calonnec, L. Delière, P. Cartolaro, F. Delmotte, D. Forget, S. Wiedemann- Merdinoglu, D. Merdinoglu, C. Schneider Detection of 16SrXII-A phytoplasma in insects collected in vineyards of South Italy ..... 53-60 V. Cavalieri, V. D’Urso, L. Ferretti, C. Rapisarda Tolerance of Tunisian Grapevine to Uncinula necator...... 61-68 S. Chebil, N. Zeghonda, N. Jallouli, S. Lasram, H. Zemni, A. Ghorbel and A. Mliki In semi-vivo antagonism of Acremoinum byssoides towards Plasmopara viticola ...... 69-72 G. Conigliaro, S. Lo Piccolo, L. Torta and S. Burruano Approach of the sociological factors influencing the vine growers commitment while apprehending the integrated production in the Aquitaine and Charentes area...... 73-79 T. Coulon, F. Hugueniot, C. Compagnone Technical and economical validation of the integrated vine growing production approach on a network of reference farms in the Bordeaux and Cognac vineyards Evaluation 2000-2006 ...... 81-86 T. Coulon, F. Hugueniot Replacement of copper in organic viticulture: efficacy evaluation of new natural fungicides against downy mildew...... 87-90 S. Dagostin, T. Formolo, I. Pertot A multivariate analysis of combined effects of (micro)climate, vegetative and reproductive growth on grey mould incidence in grapevine...... 91-94 M. Fermaud, H. Valdés-Gómez, A. Calonnec, J. Roudet, C. Gary Development of monitoring schedules for grape diseases at regional scale...... 95-99 E. Fulchin, M. Van Helden Control of Blackrot (Guignardia bidwellii) on the hybrid vitis cultivar Isabella...... 101-105 C. Gessler, F. Foiada, M. Jermini, I. Pertot

xiv

Sloping ice from the wings of airplanes – A potential cause for locally limited Plasmopara viticola primary infections? ...... 106 G. K. Hill Flavescence dorée and Scaphoideus titanus: Distribution and control in Switzerland.. 107-111 M. Jermini, M. Gusberti, L. Schaub, C. Linder, P. Gugerli, S. Schärer, P. Kehrli, L. Colombi, S. Bellion, S. Emery Effect of late season sprays against Botrytis on quality of the wines ...... 113-116 W. K. Kast, O. Schmidt, K. Bleyer Water-glass (SiO2), an excellent, mostly overseen tool in fruit-rot control...... 117-120 W. K. Kast, K. Bleyer, R. Fox Bois noir, a severe outbreak of stolbur type A in Southern Germany - disease abundance and treatments against disease-causing agents and vectors ...... 121-125 W. K. Kast, M. Stark-Urnau, K.Bleyer Bois noir disease of the grapevine in Alsace: field transmission, observations made on symptomatology and reduction of transmission risk by the vector Hyalesthes obsoletus Signoret ...... 127-136 P. Kuntzmann, E. Bogen, C. Renel Time of treatment and selection of fungicides importance in control of Phomopsis cane and leaf spot disease of grapevine...... 137-143 N. Latinovic, P. Vuksa, Z. Vucinic, J. Latinovic An expert-based crop protection decision strategy against grapevine’s powdery and downy mildews epidemics: Part 1) formalization...... 145-153 B. Leger, P. Cartolaro, L. Delière, L. Delbac, M. Clerjeau, O. Naud Detection of endophytic bacteria in leaves of Vitis vinifera by using Fluorescence in Situ Hybridization (FISH) ...... 155-159 S. Lo Piccolo, G. Conigliaro, D. Ercolini, L. Torta, S. Burruano and G. Moschetti Biodiversity in the grapevine: rhizosphere arthropods and mycorrhizal fungi...... 161-165 A. Martorana, L. Torta, G. Lo Verde, E. Ragusa, S. Burruano and S. Ragusa Early detection of selection for resistance in Plasmopara viticola populations treated with organically based fungicides ...... 167-174 C. L. Matasci, D. Gobbin, H.-J. Schärer, C. Stutz, L. Tamm, C. Gessler Incidence and development of Esca disease in Trentino Province (Northern Italy)...... 175-180 L. Michelon, C. Pellegrini, I. Pertot Preliminary studies on “esca” disease in Sicilian vineyards...... 181-187 V. Mondello, G. Conigliaro, A. Alfonzo, V. Ferraro, L. Torta , S. Burruano Coptimizer”: a decision support system to reduce copper in organic viticulture...... 189-192 D. Prodorutti, A. Pellegrini, S. Simon, Y. Gafni, T. Kuflik, A. Frizzi, I. Pertot Impact de l'estimation des pluies par radarsur la representation spatiale de la modelisation du risque d'epidemie de mildiou sur le vignoble de Bordeaux...... 193-201 M. Raynal, C. Debord, K. Griaud, S. Strizyk, D. Boisgontier, J. Congnard, D. Grimal Powdery mildew of grape on leaves – is it a problem?...... 203-206 Z. Tirtza, R. Moshe Competitive colonisation of Penicillium expansum and Botrytis cinerea on grapes..... 207-214 R. Walter, M. Harms, H. Buchenauer Characteristics of wine and table grapevine hybrids tested for cultivation in Trentino (northern Italy) ...... 215-219 L. Zulini, A. Vecchione, L. Antonelli, M. Stefanini

The impact of spiders (Araneae) on Lobesia botrana (Denis & Schiffermüller) population density ...... 221-231 R. Addante, S. Di Gioia, C. Calculli, A. Pollice Olfactory responses of Eupoecilia ambiguella (Hübner) (Lepidoptera Tortricidae) females to volatiles from grapevine* ...... 233-236 G. Anfora, M. Tasin, A.-C. Bäckmann, E. Leonardelli, A. De Cristofaro, A. Lucchi, C. Ioriatti Comparison of two methods for the agrochemicals side effect evaluation on Phytoseiid mites in vineyards...... 237-243 M. Baldessari, R. Maines, G. Angeli Olfactory cells responding to the main pheromone component and plant volatiles in Lobesia botrana (Den. & Schiff.): possible effects on monitoring systems ...... 245-249 A. De Cristofaro, S. Vitagliano, G. Anfora, G.S. Germinara, M. Tasin, A. Lucchi, C. Ioriatti, G. Rotundo. Notes on the distribution and the phenology of Erasmoneura vulnerata (Fitch) (Homoptera: Cicadellidae) in North-Eastern Italy...... 251-254 C. Duso, R. Moret, G. Marchegiani, A. Pozzebon Effects of irrigation on Empoasca vitis populations ...... 255-258 D. Fornasiero, F. M. Buzzetti, A. Pozzebon, C. Duso Simulation of Lobesia-botrana-egg-laying for autecological and insecticide studies... 259-265 C. Hoffmann Seasonal abundance and distribution of Planococcus ficus on grape vine in Sardinia.. 267-272 A. Lentini, G. Serra, S. Ortu, G. Delrio Impact of the erineum mite Colomerus vitis on Muscat ...... 273-277 C. Linder, M. Jermini, V. Zufferey Ecological infrastructures in a vineyard of Western Sicily...... 279-281 Lo Genco A., Fucarino A. and Lo Pinto M. Arthropods as biological soil quality indicators in a vineyard under different soil management ...... 283-288 G. Lo Verde, V. Palermo, A. Santoro, L. Gristina Organizational traits for the grapevine moth control in Tuscany (Italy)...... 289-293 A. Lucchi, E. Pozzolini, L. Santini & B. Bagnoli Mating disruption technique vs Lobesia botrana (Denis & Schiffermüller) (Tortricidae): 3 years of experience in vineyards of Abruzzo (Italy) ...... 295-300 A. Mazzocchetti, A. Zinni Daily “chores” of the Nearctic leafhopper Scaphoideus titanus Ball (Hemiptera Cicadellidae)...... 301-303 V. Mazzoni, M. Virant Doberlet, L. Santini and A. Lucchi Olfactory stimuli involved in host plant detection of Scaphoideus titanus Ball nymphs 305-307 V. Mazzoni, G. Anfora, F. Trona, A. Lucchi, C. Ioriatti Comparison between downy mildew fungicides in a vineyard of the Avellino province (Campania South Italy) and their influence on the population of Phytoseid mites (Parasitiformes, Phytoseiidae)...... 309-319 M. Nicòtina, G. C. Capone Population dynamics of grapevine moths reflect weather conditions over the past decade...... 321-323 D. Pasquier, P.-J. Charmillot, P. Kehrli Effects of grape downy mildew on interactions between fungicides and predatory mites on grapevines...... 325-329 A. Pozzebon, C. Duso and P. Tirello

xvi

Soil Pest Management with Herbicides?...... 331-336 Peter Schwappach Spatio-temporal distribution of Lobesia botrana (Denis & Schiffermüller) male population in a central Italy agro-ecosystem...... 337-342 A. Sciarretta, A. Zinni, A. Mazzocchetti, P. Trematerra Lutte biologique contre les cochenilles farineuses Heliococcus bohemicus Sulc et Phenacoccus aceris (Signoret) au moyen de lâchers de Chrysoperla lucasina (Lacroix)...... 343-349 G. Sentenac, T. Pham, A. Salaun, J. Souvignet Volatiles from grape drive the oviposition of Lobesia botrana at short distance...... 351-353 M. Tasin, G. Anfora, A.-C. Bäckman, C. Ioriatti, A. De Cristofaro, E. Pozzolini, E. Leonardelli and A. Lucchi. Grapevine Pests in Sicily...... 355-361 H. Tsolakis, E. Ragusa Incidence of grapevine moth Lobesia botrana (Den. & Schiff.) on occurrence of ochratoxin A in grapes ...... 363-368 H. Tsolakis, O. Corona, A. S. Pulizzi, F. Grippi, V. Mondello Landscape characteristics influencing pest populations in viticulture...... 369-373 M. Van Helden, G. Pain, J. Pithon Occurrence and spread of Scaphoideus titanus in Austria...... 375-377 N. Zeisner

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 1-18

Economic and structural aspects of vitiviniculture in Sicily

Antonio Asciuto, Simona Bacarella1 University of Palermo, Dipartimento di Economia dei Sistemi AgroForestali, Viale delle Scienze, 90128 Palermo

Abstract: World wine market is among the most dynamic in the agro-food sector and in the last few years it has undergone deep changes concerning production, consumption and trade. The growing presence of new wine-producing countries and the reform of the vitivinicultural Common Market Organization (CMO) have strongly contributed to the changes occurred in this sector. Accounting for 49% of world wine supply, Italy, France and Spain are the three main wine producers in the world. Italian situation is characterised by few wine-producing regions which altogether represent the major part of cultivated areas and wine production. In this scenario Sicily, according to 2004 ISTAT data, is the Italian region with the largest area destined to vineyards (over 135,000 hectares, correspondent to 17.6% of national grape area), followed by Puglia (14.3%) and Veneto (10%). In terms of wine production, Sicilian supply represents around 13% of national total, after Veneto (16.6%) and Puglia (14.3%). Vitiviniculture is mainly concentrated in western part of Sicily with a prevalence of white on red wines, both of them mainly originated from autochthonous vines, although international varieties are also quite common. At present, notwithstanding a general reduction of grapes yields in favour of quality, wine production in Sicily is still mainly represented by ordinary table wines and grape musts, so that this region is predominantly considered a raw material supplier to Italian and foreign industries. Both DCO and DOCG wines have a small incidence (around 3%) in terms of wine regional production, even though the former are quite well represented (21). As to productive structure there are lots of “wine growers’ cooperatives”, which sell their production mainly for distillation or as loose wine in Northern Italy or abroad, with a consequent loss of value added. There is also a group of small and medium farms which, although they have started a process of wine quality improvement, run into difficulties when trying to find new markets for their product. In the end, there is a restricted number of leading farms, strongly oriented to product and process innovation and to marketing, and successful in placing their wines in a medium-high segment of national and international markets.

Key words: wine market, Sicily, vitiviniculture

1. International scenario

World wine market is among the most dynamic in the agro-food sector and in the last few years it has undergone deep changes concerning production, consumption and trade. The growing presence of new wine-producing countries and the reform of the Common Market Organization (CMO) for wine have strongly contributed to the changes occurred in this sector. All in all in 2006 world vineyard surface area (wine vines and table grapes) was over 7.92 million hectares, mainly concentrated in the EU countries where 45% of total area is

1 Antonio Asciuto has drawn up section 1; Simona Bacarella has written sections 2, 3 and the final considerations.

1 2

found, correspondent to 3.6 million hectares. Spain, France and Italy are the countries with the widest vineyard surface area, representing an overall 36.4% of world areas under vines (respectively 14.8%, 11.1% and 10.5%). Among other wine-producing countries, the main international competitors for Europe are USA (5% of overall total), Argentina (2.8%), Chile (2.4%), Australia (2.1%) and South Africa (1.7%). Although other countries account for 41.0% of total vineyard area, they are not important wine-producers since these areas are mainly destined to table grape, as it can be seen in figure 1.

Spain Other countries 14.8% 41.0% Italy E.U. (2 5 ) 10.5% 44.9% France 11.1%

Other EU countries South Africa 8.5% 1.7% Australia Chile 2.1% 2.4% US A Argentina 5.0% 2.8%

Fig. 1. Distribution of world areas under vines (%). Source: Our elaborations on FAO and OIV data

Other countries 16.5% France 17.8% South Africa 3.1% Italy E.U. (2 5 ) Australia 16.6% 59.2% 4.4% Spain Chile 14.4% 2.8% Other EU countries Argentina 10.6% 7.3% US A 6.6%

Fig. 2. Wine production in the world by country (%). Source: Our elaborations on FAO data.

2. Italian scenario

According to the most recent data from the national “Vitivinicultural Inventory”, in 2005 productive wine-grape area in Italy was 730,439 hectares, and 796,700 hectares when considering also national replanting or new planting rights. Going into details, Sicily is characterised by the largest area devoted to wine-grapes, equal to 128,144 hectares (17.5% of national total); in Puglia there are 105,601 hectares (14.5%) and in Veneto 72,460 hectares (9.9%). In the other regions vitivinicultural areas are far below 65,000 hectares (see figure 3).

Other regions 58.1%

Sicily 17.5%

Puglia Veneto 14.5% 9.9%

Fig. 3. Distribution of vineyard area in Italy by region (%).Source: Data taken from Agea Inventory, on 1/9/2005.

100.0 97.7 100 45.000 86.8 90 40.000 79.7 73.9 75.3 80 35.000 70 30.000 60 56.3 49.9 25.000 44.8 50 46.5 46.4 44.0 44.0 20.000 40 36.3 31.8 15.000 30 23.4 21.1 20.6 (CDO-CGDO areas,%) 10.000 20 13.8 14.2 (CDO-CGDO areas, hectares) 10 6.7 5.000 0 0 Lazio Sicilia Puglia Liguria Molise Veneto Umbria Marche Calabria Toscana Sardegna Abruzzo Piemonte Basilicata Campania Lombardia Valled'Aosta Emilia Romagna Prov. Aut. Trento Prov. Aut. Bolzano Friuli Venezia Giulia

Fig. 4. CDO-CGDO wine areas on regional wine-grape total (2005) (%).Source: Our elaborations on AGEA data.

From these data we can notice differentiated situations within the country, especially with reference to CDO and CGDO wine-producing areas (in all they represent 37.6% of total viticultural areas), mainly found in northern and central regions. In particular, Piedmont is the most represented region, since these areas extend on more than 40,000 hectares of CDO and CGDO vineyards (14.8% of national total). Other important regions are Toscana (12.3%), Veneto (12.3%), Emilia-Romagna (10.4%) and Lombardia (7%). The remaining regions in all account on the national CDO and CGDO areas for just 42.8%. When analysing the relative importance of areas destined to quality viticulture compared to the total regional areas for viticulture, it emerges that regions from North and Centre of Italy play a remarkable role. In Piedmont, Friuli-Venezia Giulia, Lombardia and Valle D’Aosta quite high percentages of areas under vines are destined to CDO-CGDO wines, in a range comprised between 73% and nearly 100% (Valle D’Aosta). With a further exception (Toscana, 56.3%), in the other Italian regions CDO-CGDO areas never exceed 50% of the regional vine areas. Sicily in particular has allocated less than 7% of total vine areas to a high-quality viticulture (see figure 4). As to Italian wine production, in 2006 it was around 49.6 million hectolitres, showing a 1.8% decrease rate in comparison with 2005 (ISTAT, see table 1).

Table 1. Wine and Must Production in Italy (mhl, thousand of hectolitres). Source: Our elaborations on ISTAT data.

Regions 2002 2003 2004 2005 2006 Piedmont 2,329 2,282 3,263 3,054 3,229 Valle d'Aosta 16 18 22 20 20 Lombardia 1,123 856 1,168 1,100 1,081 Trentino Alto Adige 1,063 1,076 1,269 1,057 1,159 Veneto 6,847 7,369 8,843 7,093 7,208 Friuli Venezia Giulia 1,006 1,113 1,344 1,159 1,014 Liguria 93 106 91 84 77 Emilia Romagna 5,682 5,305 7,155 6,608 6,768 Toscana 2,319 2,274 3,166 2,780 2,978 Umbria 776 812 1,078 998 1,103 Marche 1,258 940 1,248 1,206 1,090 Lazio 2,859 2,441 2,492 2,362 2,316 Abruzzo 3,808 3,319 3,585 3,469 3,233 Molise 307 274 328 390 376 Campania 1,761 1,655 1,878 1,826 2,020 Puglia 5,580 6,089 7,610 8,348 7,397 Basilicata 309 284 201 267 246 Calabria 531 476 485 539 484 Sicily 6,209 6,553 6,964 7,283 6,974 Sardegna 729 856 943 924 859 Italy total 44,604 44,096 53,135 50,566 49,631 Northern - Central Italy 25,370 24,592 31,140 27,520 28,43 Southern Italy 19,234 19,505 21,995 23,046 21,588

During the 2002-2006 period, Italian wine production has slightly increased (11.3% in the five-year period), showing a trend inversion after years of decline, due to the changes occurred in the international wine market, and particularly in the EU market.

Other regions 42.9%

Sicily 14.1%

Emilia Romagna 13.6% Puglia Veneto 14.9% 14.5%

Fig. 5. Wine production in Italy. Source: Our elaborations on ISTAT data.

In detail, four regions account for 57% of national wine production: Puglia is the leading region in quantitative terms, with 7.3 million hectolitres equal to 14.9% of national total, Veneto with 7.2 million hectolitres follows Puglia closely (14.5%), Sicily with 6.9 million hectolitres, equal to 14.1%, and Emilia-Romagna with 6.7 million hectolitres, equal to 13.6% of national wine supply (see fig. 5).

Table 2. CDO, CGDO and TGI wines acknowledged in Italy. Source: Ministry of Agricultural, Food and Forest Policies.

CGDO CDO TGI Total Piedmont 9 45 54 Valle d'Aosta 1 1 Lombardia 4 15 14 33 Trentino Alto Adige 8 4 12 Veneto 3 25 10 38 Friuli Venezia Giulia 2 9 3 14 Liguria 8 3 11 Emilia Romagna 1 20 10 31 Toscana 7 36 6 49 Umbria 2 11 6 19 Marche 2 15 1 18 Lazio 27 4 31 Abruzzo 1 3 9 13 Molise 3 2 5 Campania 3 17 9 29 Puglia 25 6 31 Basilicata 3 2 5 Calabria 12 13 25 Sicily 1 22 6 29 Sardegna 1 19 15 35 Italy* 36 316 118 470

* The total is not equal to the summation of addends as there are 8 interregional CDO wines and 4 TGI wines (two over three regions and two over two regions). Updated to September 2007.

Veneto, Puglia and Sicily have been for years the Italian leading regions in wine production showing different trends: Sicily was the leading region for wine production up to 1998; Veneto took Sicily’s place for the 1999-2004 period, and since 2005 Puglia has become the first Italian region in terms of wine production, closely followed by Veneto. The reasons for this recent trend inversion are to be found in the modern evolution of Italian vitiviniculture, the generalised strategy of which has concerned wine quality improvement. One of the main consequences has been a sensible yield reduction of grapes in order to produce wines more appreciated by national and international markets. According to these aims, all the regions above mentioned have modified the structure of their wine production both in quantitative and qualitative terms, even though to a different extent. With reference to the pyramid of wine quality, in Italy there are 470 types of qualitative acknowledgements for wines, shared out among CGDO, CDO and TGI (see table 2). The highest number of acknowledgements refers to Central and Northern regions, and in particular to Piedmont and Tuscany, represented respectively with 54 and 49 quality denominations. Italian wine production is therefore mainly characterised by wines of medium-high quality, as showed by the share of table wines (43.6%) compared to the share of quality wines, just over 56%, distributed between CDO and CGDO (30.8%) and TGI (25.7%) (see figure 6).

CDO-CGDO wines Table wines 30.8% 43.6%

TGI wines 25.7%

Fig. 6. Italian wine production by denomination. Source: Our elaborations on ISTAT data

From the analysis of regional viticultural sector, differences between North and South are once again significant. The major contributes to the national wine production with typical geographical indication belong to the four leading regions for wine production: Veneto (33.5% of national production of TGI wines), Emilia-Romagna (20.5%), Sicily (11.4%) and Puglia (9.0%). Quite a different situation comes out if we analyse the importance of TGI segment in the regional context; in this case some regions (Veneto, Emilia-Romagna, Marche and Umbria) have a share of TGI segment far higher than the national average, few regions (Sicily, Sardegna and Friuli-Venezia Giulia) have a share close to the national average and all the other regions are below the average. CDO-CGDO wine productions are similarly represented along Italian territory, being mainly concentrated in Piedmont (17.7% of Italian CDO-CGDO production), Veneto (14.9%) and Toscana (11.2%), whereas the remaining regions record just a 7.5% share of this wine denomination.

The analysis of the contribution of Origin Denominations within regional wine productions has revealed its significant importance, especially in some areas where the incidence of quality wine is remarkable. This is the case of Piedmont (83.4% of regional wine production), Trentino Alto Adige (81.6%), Friuli Venezia Giulia (64.6%) and Toscana (57.4%). In the other Italian regions, the wine productions under Denominations of Origin represent less than 50% of total production, and Sicily is last with a 3.6% share out of regional wine production (see figure 7).

90 83.9 81.6 80 70 64.6 57.5 59.2 57.4 58.8 60 47.0 50 45.5 40 31.6 32.2 34.8 33.9 27.6 30 22.2 17.7 14.9 20 9.8 11.2 12.5 11.7 11.8 11.0 4.2 6.2 4.3 6.9 7.2 5.6 3.6 10 0.1 0.2 2.3 2.5 1.4 1.7 0.2 0.3 1.6 1.6 0 Lazio Puglia Sicilia Molise Veneto Liguria Umbria Marche Toscana Abruzzo Calabria Sardegna Piemonte Basilicata Campania Lombardia Valle d'Aosta Valle Emilia Romagna Emilia Trentino Alto Adige Alto Trentino Friuli Venezia Giulia Venezia Friuli

% of total CDO-CGDO prod. Italy % of regional wine production

Fig. 7. Incidence of CDO and GCDO on the production of national and regional wine (%). Source: Elaborations on ISTAT data

3. Vitiviniculture sector in Sicily

3.1 Vineyard surface area The leading role played by Sicilian vitiviniculture at a national level can easily be noted by the analysis of official statistical data. In recent years the long-term trends noticed in the national and European wine market have nevertheless driven to a gradual reduction of both vineyard surface area and wine production (see figure 8). According to the latest ISTAT data (2005), Sicily is the Italian region with the largest vineyard area (128,144 hectares) of which 9,282 hectares is represented by replanting and new planting quotas (rights). Analysing the evolution of vineyard area in Italy and Sicily in the 1983-2005 period, it can be observed that trends and reduction rates are closely, respectively equal to 30% and 28%. Sicilian viticulture is mainly localized in the western part of the Island, which holds – in just three provinces – 87% of regional vineyards; in particular, the province of Trapani records the largest area under vines (65,497 hectares) among Italian provinces and represents over 55% of Sicilian vineyard surface area, followed by the provinces of Agrigento (17.7%) and Palermo (14.1%). The other six provinces in all account for 13% of regional vineyard areas (see figure 9).

1.200 250 1.000 ) 200 800 150 600 100 400 Italy (000 ha (000 Italy Sicily (000 ha) (000 Sicily 50 200 0 - 2003 2005 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

SICILY ITALY

Fig. 8. Evolution of vineyard surface area in Italy and Sicily (thousand hectares). Source: Our elaborations on ISTAT data.

Other provinces Agrigento 13.1% 17.7%

Palermo 14.1%

Trapani 55.1%

Fig. 9. Distribution of vineyard surface area in Sicily (%).Source: Our elaborations on ISTAT data.

Red 2000 Red 2007 grapes grapes 23.2% 35,4%

White White grapes grapes 76.8% 64,4%

Fig. 10. Evolution of vineyards surface area in Sicily by colour of grapes (%). Source: Our elaborations on data taken from Vitiviniculture Inventory 2000/Sicilian Region – Regional Board for agriculture and forest/SRRFV, 2007.

Sicilian vineyards have always been characterised by a continuous predominance of white grapes, nevertheless in the last years their presence has been greatly reduced in favour

of red grapes, in consequence of the growing demand for red wines. Red grapes varieties in 2000 accounted approximately for 23% of Sicilian vineyards, while in 2007 their share has grown to over 35%, showing an increase rate of about 12% (see figure 10). In the figure it can be noted a 0.2% share (2007) referred to areas allocated for other vine uses, such as vines for grafts or rootstocks, experimental vineyards or other uses. Distribution of the above typologies of grapes is differentiated in spatial terms; white grapes are actually widespread in western Sicily in the provinces of Trapani, Agrigento and Palermo, whereas red grapes are mainly cultivated in the eastern part of the Island. “Catarratto Bianco Comune”, an autochthonous white grape variety, is cultivated on an area of over 39,000 hectares, accounting for more than 33% of regional wine grape vine area and 52% of white grapes varieties. “Nero d’Avola”, an autochthonous red grape variety, is the second cultivar in terms of cultivated areas with 18,812 hectares, correspondent to 15.8% and to 45.0% of respectively regional vineyard total and red grapes area (see figure 11 and table 3).

33.2 35 30 25 20 15.8

(%) 15 10 7.2 6.7 4.5 4.0 4.4 4.1 2.9 2.1 3.3 3.3 5 1.4 0.5 0.2 1.6 0.7 0.7 0.3 0.2 0 Perricone Merlot Viogner Chardonnay Ansonica o Insolia Ansonica Nerello Cappuccio Nerello Cabernet Sauvignon Cabernet Catarratto Bianco Lucido Catarratto Bianco Comune Bianco Catarratto Calabrese o Nero d'Avola o Nero Calabrese

Fig. 11. Distribution of vineyard surface area in Sicily by variety on 10/01/2007 (%).Source: Our elaborations on data by Sicilian Region - Regional Board for Agriculture and Forests / SRRFV.

Other three autochthonous white grape varieties are also significant in terms of vineyard areas, Trebbiano Toscano, Insolia and Grecanico, which respectively account for 7.1%, 6.6% and 4.5% of wine vine surface area in Sicily. Allochthonous varieties, on a whole, are well adapted to Sicilian environment and have had a good spatial growth: among these ones, Syrah, Merlot, Chardonnay and Cabernet Sauvignon, all international varieties, have to be mentioned as their share in terms of area is at least equal to 3% of the regional area under vines (their areas range between 3.3% and 4.4% of the regional total). In the context of a modern viticulture which is more and more quality-oriented, also the training system is a factor that can heavily affect the achievement of this objective. In the Sicilian viticulture sector, espalier is the most common training system, being present in 77% of total surface area, followed by arbour (12%) and by alberello (head-trained) (10.6%) (see figure 12).

Table 3. Wine grape varieties in Sicily by grape colour (2007). Source: Sicilian Region –Regional Board for Agriculture and Forests / SRRFV.

% regional area Surface area % regional area CULTIVAR under vines (ha) under vines by grape colour WHITE GRAPES Catarratto Bianco Comune 39,495 33.21 51.58 Trebbiano Toscano 8,520 7.16 11.13 Ansonica o Insolia 7,925 6.66 10.35 Grecanico 5,388 4.53 7.04 Chardonnay 4,785 4.02 6.25 Grillo 3,385 2.85 4.42 Catarratto Bianco Lucido 2,488 2.09 3.25 Zibibbo 1,602 1.35 2.09 Viogner 548 0.46 0.72 Other white grapes 2,434 0.18 3.18 White grapes 76,571 64.43 100.00 RED GRAPES Calabrese or Nero d'Avola 18,812 15.82 44.77 Syrah 5,179 4.35 12.32 Merlot 4,814 4.05 11.45 Nerello Mascalese 3,968 3.34 9.44 Cabernet Sauvignon 3,908 3.29 9.30 Sangiovese 1,846 1.55 4.39 Nerello Cappuccio 857 0.72 2.04 Frappato di Vittoria 792 0.67 188 Perricone 320 0.27 0.76 Other red grapes 1,528 0.24 3.64 Red grapes 42,023 35.37 100.00 Other 268 0.20 Total Sicily (hectares) 118,862 100.00

Espalier 77.0%

Arbour Other Alberello 12.0% 0.3% 10.6%

Fig. 12. Distribution of Sicilian vineyard surface area by training system. Source: our elaborations on IRVV data, 2005.

In the last years both arbour-trained and head-trained vineyard areas have decreased: the former neglects the qualitative aspects of production, thus being excluded by EU incentives, the latter has heavy production costs because it is not compatible with crop mechanisation.

3.2 Grape production In the last twenty years, Sicilian wine grape production has significantly decreased, in line with national trends, due to the combined reduction of both vineyard surface areas and average yields, passed from over 100 quintals per hectare in 1983 to 78 quintals per hectare in 2006. The decreasing rate of Sicilian grape production in the 1983-2006 period – from 16 million quintals to 9.2 million – has been greater than the national one, from 112 million quintals to 68.2 million quintals, that is -42.5% versus -39.1% (see figure 13).

25.000 120.000

20.000 100.000 80.000 15.000 60.000 10.000 40.000 5.000 20.000 - - 2003 2005 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

Sicily Italy

Fig. 13. Evolution of wine grape production in Sicily and Italy (mqx). Source: ISTAT data.

Such differences are likely to be associated with the recent guidelines of farm strategies and, more generally, with Common Agricultural Policy, which, focusing more and more on the improvement of wine production quality, have driven the whole sector to modernize the production system. The province of Trapani records the largest wine grape production (4.6 Mqx, equal to 50.0% of regional total), followed by the provinces of Palermo and Agrigento (respectively 1.7 Mqx, equal to 18.9%, and 1.5 Mqx, equal to 17.2%) (see figure 14). The province of Enna records the highest yields (133.4 quintals per hectare) which do not allow a significant contribution to the regional wine grape production as a consequence of a very small vineyard area, followed by the provinces of Ragusa and Palermo (respectively 124.5 and 104.5 quintals/hectare). The lowest yields are recorded in the province of Catania (41.3 quintals/hectare), followed at a distance by the province of Trapani (71.3 quintals/hectare). As to wine and must production, in the 1983-2006 period the weight of Sicily at national level has been ranging between 13% and 18%, to settle at 14.1% in 2006, showing a downward trend in absolute terms (see table 4). According to ISTAT data, Italy has produced on average 48.2 Mhl of wine on a five-year basis (2002-2006), showing a 34.8% decrease when compared to the 1983-1987 period.

In the 1983-2006 period, the Sicilian wine production has followed the above trend, but in the comparison between the averages of the last (2002-2006) and the first (1983-1987) five-year periods it has showed a higher rate of reduction (-43% decrease with respect to the average 1983-1987).

5.000 160 4.500 140 4.000 133,4 124,5 120 3.500 100 3.000 93,3 104,5 87,4 2.500 75,5 78,8 80

2.000 71,3 60 (q.li/ha) 1.500 (000 quintals)(000 41,3 40 1.000 500 20 - 0 Enna Ragusa Catania Trapani Messina Palermo Siracusa Agrigento Caltanissetta Grape Prod. Yields

Fig. 14. Distribution of wine grape production and yields in Sicily (2006). Source: Our elaborations on ISTAT data.

Table 4. Wine and must production in Sicily and Italy in the 1983-2006 period (mhl). Source: ISTAT data.

SICILY ITALY % SICILY ITALY % 1983 13,060 83,280 15.7 1995 10,391 56,201 18.5 1984 10,893 70,900 15.4 1996 9,009 58,772 15.3 1985 10,488 62,340 16.8 1997 8,073 50,563 16.0 1986 12,271 77,093 15.9 1998 9,200 56,912 16.2 1987 11,899 75,875 15.7 1999 8,160 58,072 14.1 1988 8,975 61,010 14.7 2000 7,106 54,088 13.1 1989 9,394 60,651 15.5 2001 7,149 52,293 13.7 1990 7,715 54,866 14.1 2002 5,719 43,241 13.2 1991 10,137 59,788 17.0 2003 6,553 44,086 14.9 1992 11,677 68,686 17.0 2004 6,640 53,275 13.1 1993 10,192 62,672 16.3 2005 7,283 50,566 14.4 1994 9,300 59,290 15.7 2006 6,974 49,631 14.1

Wine in Sicily has always been jointly produced with must, in different proportions depending on pedoclimatic conditions and commercial factors, ranging between 10.8% and 23.9% in the 1999-2006 period (see figure 15), with an average weight equal to 20%. Sicily is the leading region in must production, partly sold as fresh product and the greater part as sulphited must, concentrated must or concentrated rectified must. Concentrated musts are used in must enrichment by wine-makers both in Sicily and outside the region,

where they are mainly exported by Northern Italy firms with factories in Sicily. Another share of must production is directly transformed by Sicilian processors.

100 89.2 83.8 82.7 80.2 76.1 80.6 79.0 79.7 80

40 19.8 23.9 19.4 16.3 17.3 21.0 20.3 20 10.8

0 1999 2000 2001 2002 2003 2004 2005 2006

Wine Must

Fig. 15. Distribution of wine and must production in Sicily (%) (2006). Source: Our elaborations on ISTAT data.

As to the composition of Italian wine production, in the last years it has deeply changed showing a gradual shift from white to red wines, following the consumers’ preferences, until 2001 when red and rosé wines have recorded a higher share than white ones. In Sicily, despite a trend inversion that has produced a growth in the rate of red wines (from 25.7% in 1999 to 37.3% in 2006), white wines still prevail on them with a 63.7% share (in 1999 the share for white wines was 74.3%) (see figure 16). According to IRVV estimates2, just 15-17% of Sicilian wine production is packaged, the remaining part is sold as loose wine, mainly as table wine, but also as TGI (around 1 million hectolitres).

1999 Red & Red & 2006 Rosé Rosé 25.7% 37.3%

White White 74.3% 62.7%

Fig. 16. Distribution of wine production by colour. Source: elaborations on ISTAT data.

2 Estimates from IRVV are quite close to those from the Observatory of Sicilian vitiviniculture bottling firms (ISI), according to which packaged wine in 2005 was around 1,152 hectolitres.

Loose wine is largely sold in Northern Italy, where it is packaged (in bottles or in brick) gaining a value added which otherwise would be earned by Sicilian firms, or sold abroad, especially in France. With regard to wine composition by typology, in spite of the recent developments implemented by regional viticulture with the aim at improving the quality of final product, table wine still represents a significant portion (70%) of Sicilian production and only a small part of it is represented by high-quality wines. Although Sicily is characterised by a high number of Denominations of Origin (1 CGDO, 22 CDO, 6 TGI), it has a restricted production of quality wines, among which CDO and CGDO wines account for 3.6% of total regional wine production and TGI wines account for 20.8% (see table 5 and figure 17). As to CDO production, the last data can be traced back to 2004 (Chambers of Com- merce, Industry, Artisanship, Agriculture of the Sicilian provinces) and from their analysis it turns out that most Sicilian CDO wines belong to Marsala CDO (about 62%), followed by Alcamo (13 %), Etna (6%) and Moscato of Pantelleria (5%). The quota of TGI wine produced in Sicily, despite the 6 TGI present in the region, belongs almost entirely to Sicily TGI. The productive structure in Sicily is characterized by a variety of productive systems and processing structures: firms making wine with their own grapes and packaging their own products; cooperative farms and medium-small wine firms; medium and big-medium firms making wine from grapes or musts bought from a third party and packaging the wine so obtained; firms processing their partners’ grapes and mainly producing loose wine to be retailed or sold to extra-regional markets and for distillation.

Table 5. List of CGDO, CDO and IGT wines recorded in Sicily. Source: Ministry of Agricultural, Food and Forest Policies. Updated September 2007.

CGDO Cerasuolo di Vittoria CDO Alcamo o Bianco Alcamo, Contea di Sclafani, Contessa Entellina, Delia Nivolelli, Eloro, Erice, Etna, Faro, Malvasia delle Lipari, Mamertino di Milazzo o Mamertino, Marsala, Menfi, Monreale, Moscato di Noto, Moscato di Pantelleria - Passito di Pantelleria e Pantelleria, Moscato di Siracusa, Salaparuta, Sambuca di Sicily, Santa Margherita Belice, Sciacca, Riesi, Vittoria TGI Camarro, Fontanarossa di Cerda, Salemi, Salina, Sicily, Valle Belice

A survey by Istituto Regionale Vite Vino in 2005 has counted 461 firms in Sicily, 52 of which appear to be cooperative wineries, marketing packaged Sicilian wine (bottles, bricks, bag-in-boxes, etc.). Trapani has turned out to be the province with the highest number of firms (167), followed by Agrigento (56), Palermo (52), Catania (52), Ragusa (32), Messina (30), Caltanissetta (16) and Enna (4). Despite the high number of “wine bottling” firms, the major part of Sicilian packaged wine trade (73.4%) is supported by just 30 firms, two of which - a co-operative winery and a limited company- account for 21.3%. There is a remarkable number of cooperative wineries which mainly distillate or sell loose wine in Northern Italy and abroad, with the consequent loss of a large part of the value- added connected with production.

Besides these ones, there’s a group of small and medium firms which, even though are trying to improve the quality of their production, have some difficulty in putting their products successfully on the markets. Finally, there are few leading firms oriented to product, process and marketing innovation, which succeed in placing their products in a medium-high segment of national and international markets.

100

80 72.1 74.9 75.9 75.6

22.3 20.2 20.1 20.8 20 5.6 4.9 4.0 3.6

0 2003 2004 2005 2006

Table TGI CDO-CGDO

Fig. 17. Distribution of wine production by typology (%).Source: Elaborations on ISTAT data.

3.3 Sicilian wine export Since Sicilian wine production is mainly oriented to a mass product, there it follows that only a little part of it is exported. According to ISTAT data, Sicilian wine exports in 2006 were equal to 5.9%, remarkably less than the previous years, when they were equal to 25% (1996) or even to 30% (2000). In the last ten-year period at world level a heavy reduction of exports has occurred in quantitative terms, offset in value terms by the export increase of packaged and quality wine. When analysing the trend of Sicilian exports between 1995 and 2006, it appears that if on one hand the incidence of the exported production over the total has strongly decreased, together with the volumes of the exported product (-66.7%), on the other hand the overall value of the exported product has increased (+68.2%), as well as the average value (from 0.25 €/l in 1993 to 2.04 €/l in 2006). This indicates that the foreign trade of loose wine has gradually diminished in favour of packaged wine (see table 6). Packaged wine accounts for, in fact, 87 % on the value of regional exports, and almost 71 % on their volumes. Loose wine, even though representing about 28% of the exported volumes, only accounts for 10.7% on the product value to be sold abroad. It is followed by “spumante” wine and must, with a very low weight on regional exports, characterised by different trends: must exports, though always not much represented, are nearly close to zero, whereas “spumante” wines have shown a marked growth rate, from just some hundreds hectolitres in the first years of 21st century to over 6,000 hectolitres and 1.8 million euro in 2006 (see figure 18).

Table 6. Evolution of Sicilian wine exports from 1995 to 2006. Source: Elaborations on ISTAT data.

Total wine of which packaged wine Year (000) (000) € €/l (000) (000) € €/l mhl mhl 1995 2,247 100,186 0.45 n.d. n.d. n.d. 1996 1,367 74,278 0.54 n.d. n.d. n.d. 1997 1,357 65,689 0.48 112 24,403 2.18 1998 1,220 74,408 0.61 163 35,020 2.14 1999 2,141 104,437 0.49 154 34,731 2.25 2000 1,729 91,328 0.53 192 42,658 2.22 2001 1,317 84,059 0.64 215 50,105 2.32 2002 650 79,188 1.22 226 59,595 2.64 2003 421 72,319 1.72 193 56,116 2.90 2004 441 81,435 1.85 260 66,538 2.56 2005 356 78,790 2.21 267 67,775 2.54 2006 413 84,174 2.04 292 73,279 2.51

100 87.1

80 70.7

40 27.6

20 10.7 2.2 1.6 0.1 0.1 0 Spumante wine Packaged wine Loose wine Must

Value % Volume %

Fig. 18. Distribution of Sicilian wine exports by typology (%) (2006). Source: Elaborations on ISTAT data.

The main countries of destination for Sicilian packaged wine are Germany, United Kingdom, USA and Switzerland, which absorb around 70% of packaged wine exports in value and over 73% in terms of volume; exports towards other countries altogether stand respectively at 27% e del 23%; it is noteworthy to highlight the role played by Japan that has a 5.0% share in value and a 3.5% in volume on the Sicilian packaged wine exports (see figure 19). The major quantities of loose wine are mainly destined to Greece, Romania and Switzerland, even if the highest incomes are with Switzerland, Sweden, Canada and Japan. The Sicilian “spumante” wine is almost entirely destined to Canada (78% of the volume and 79% of the value of the “spumante” exported). Finally, the little quantity of must exported is largely destined to Asian countries.

40 36.9

26.6 30 23.2 19.8 19.4 20 16.4 14.7 13.6 13.4 8.2 10

- Germany United Switzerland USA Other coutries Kingdom

Value % Volume % Figure 19. Main destinations of Sicilian packaged wine exports (%) (2006). Source: Elaborations on ISTAT data.

Conclusions

As it appears from official statistics, the role of vitivinicultural sector in Sicily has undergone deep changes in the last years, following the trend in the national and international markets. Compared to other regions, Sicily has some pedoclimatic advantages, due to its geo- graphic position, which make the carrying out of an eco-compatible production possible. Besides, the region is endowed with an ampelographic platform including autochthonous vine varieties studied and selected in order to improve their productive characteristics, and widened with international vines and with vine varieties from other Italian wine-producing regions. Structural, managerial and cultural evolution pushes the modernization process both in the vine and in the cellar paying more and more attention to the improvement of product quality, to its packaging and to its development through marketing strategies. Following the market present trends, Sicilian producers have oriented their production to less alcoholic wines, also enlarging red wines supply, particularly the “Nero d’Avola”. Various firms have carried out a modernization process, paying more and more attention to quality improvement and to the productive process, but despite the remarkable efforts, there is still a high level of supply fragmentation and too much vitiviniculture oriented to quantity rather than quality, as indicated by the considerable volumes of loose wine and must produced in Sicily, the low incidence of CDO on production and the limited foreign trade. Besides, there are very few leading firms, strongly oriented to product and process innovation and to marketing, capable of placing their products in a medium-high segment of national and international markets. In conclusion, in spite of the immense potentialities of this region, the dualism between firms devoted to product, process and management innovation (marketing-oriented firms) and firms characterised by dated style, in terms of structure, management and skills (cooperative wineries in particular) is still too strong to allow the regional wine sector to make a generalised qualitative leap.

References

Bacarella S., Ciccarelli F. 2005: La vitivinicoltura nel Mezzogiorno. Master in Management della Filiera Vitivinicola. Ed. CORERAS CORERAS 2004: La filiera vitivinicola in Sicilia. Rapporto 2003. IRVV 2006: Il settore vitivinicolo Siciliano. Collana Quaderni di Ricerca e Sperimentazione. Vol. II. ISMEA: Filiera del vino. Anni 2000 - 2004. ISMEA: Filiera del vino e delle uve da tavola in Sicilia. Anni 1999, 2000, 2001. ISMEA 2001: Il marketing del vino. ISMEA 2005: Rapporto Annuale – Evoluzione del sistema agroalimentare italiano. Vol. I e II. ISMEA 2007: I vini Doc e Doc – Una mappatura della vitivinicoltura regionale a denomina- zione di origine. ISMEA 2007: Outlook dell’ agroalimentare italiano. Vol. I e II. Osservatorio Vitivinicolo Siciliano 2001: Filiera del vino e delle uve da tavola in Sicilia. Ed. CORERAS.

Internet: www.census.it www.europa.eu.int www.ismea.it www.istat.it www.politicheagricole.it

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 19-24

“In Vitro” antagonism of a grapevine endophytic Bacillus subtilis strain towards “esca” fungi

Antonio Alfonzo 1, Valeria Ventorino 2, Livio Torta 1, Santa Burruano 1, Giancarlo Moschetti 1 1 Dipartimento S.En.Fi.Mi.Zo., Sezione di Patologia vegetale e Microbiologia agraria, Università degli Studi di Palermo, Viale delle scienze 2, 90128 Palermo 2 Dipartimento di Scienza degli alimenti, Università degli Studi di Napoli “Federico II”, Via Università 100, 80055 Portici (NA)

Abstract: Preliminary investigations on grapevines (CV Catarratto) with symptoms of “esca” permitted to isolate only bacterial colonies, from black punctuations belonging to the genus Bacillus. Particularly, an isolate was Gram-positive with spore forming and, on the basis of the partial 16S rRNA sequence comparison, it showed a similarity of 99% with Bacillus subtilis subsp. subtilis. There are numerous reports on the antagonistic activity of the species towards several phytopathogenic microorganisms. For this reason the possible bacterial control against the “esca” fungi (Phaeoacre- monium aleophilum, PAL; Phaemoniella clamidospora, PCH; Fomitiporia mediterranea, FOMED), was studied. The antagonistic activity was investigated in vitro by: a) direct method by dual-culture in Petri dishes containing malt extract agar; b) indirect method by adding to fungal growth media cultural filtrate or crude extract bacterial in different concentrations. The results showed that B. subtitles isolate has a powerful antagonistic effect towards all fungal pathogens tested, reducing their growth until 80% (direct method). Moreover, the FOMED growth was reduced at least of 50% with indirect method, while the influence of this bacterial isolate on PCH and PAL growth by indirect method is in progress. As far as we know, this is the first study reporting Bacillus as “natural limiter” of “escape” fungi.

Key words: grapevine, antagonism, Bacillus subtitles, escape fungi

Introduction

Esca disease of grapevine (Vitis vinifera L.) is caused by a complex of pathogens including the ascomycetes Phaeomoniella chlamydospora (PCH) and Phaeoacremonium aleophilum (PAL) and the basidiomycete Fomitiporia mediterranea (FOMED), acting in combination or in succession. These xylem-inhabiting fungal pathogens infect the plants through large wounds and produce cancer or rot in the wood (Mugnai et al., 1999). Attempts to control esca with several chemicals have shown that complete prevention of this disease is very difficult and that the eradication of esca fungi once they have colonized a grapevine plant is impossible (Mugnai et al., 1999; Di Marco et al., 2000). Therefore, a more promising approach seems to be to apply biological control agents in the nursery and/or on young established vines (Fourie et al., 2001). The aim of this work was to isolate bacterial antagonistic agents toward esca fungi for their potential use in the biocontrol.

19 20

Material and methods

Bacterial colonies were isolated from black punctuations of grapevine cv. Catarratto affected by esca and identified at phenotypic and genotypic level on the basis of partial 16S rRNA sequence comparison. The antagonistic activity of bacteria isolate towards PAL, PCH and FOMED was investigated in vitro by: – direct method: dual-culture (fungus-bacterium) in Petri dishes containing malt extract agar (MEA; Kim and Chung, 2004; Fig. 1); – indirect method: adding to fungal growth media (MEA) cultural filtrate (25% and 50% v/v; Fig. 2) or crude extract bacterial (1 mg/ml and 0.5 mg/ml; Mc Keen et al., 1986; Ferreira et al., 1991; Fig. 3).

Fig. 1. Direct method: inhibition of FOMED, PCH and PAL by B. subtilis AG1 (C, control; Bs, dual- culture: fungus-bacterium).

Cultural filtrate production The bacteria was grown in Erlenmeyer flacks (300 ml), containing 100 ml of malt extract broth (MEB, Oxoid). The inoculated flask were incubated in a shaker at 170 rpm and 30° C for 3 days. The culture were sterile filtered under vacuum on 250 ml Stericup (0.22 µm HV Durapore Membrane, Millipore Co, Bedford, Mass).

Extraction of the secondary metabolites The culture was centrifuged for 20 min at 16,500 g to remove bacterial cells. The secondary metabolites were precipitated from the supernatant by adjusting the pH to 2.5 with concentrated HCl. This material was centrifuged for 10 min at 16,500 g. The pellet containing the active fraction was extracted three times with 96% ethanol. The ethanol extract was taken to dryness under vacuum at 45 °C in a rotary evaporator. Inactive substance were removed by sequential extraction with ethyl acetate and acetone. The resulting residue containing the active substances was dissolved in 96% ethanol, designed as crude antibiotic extract, and stored in capped bottles at 4 °C.

C CF 25% CF 50%

FOMED

PCH

PAL

Fig. 2. Indirect method: antifungal activity of B. subtilis AG1 cultural filtrate at two different concentrations 25% and 50% v/v (C, control; CF, cultural filtrate).

C SOL CE 0.5 mg/ml CE 1 mg/ml

FOMED

PCH

PAL

Fig. 3. Indirect method: antifungal activity of B. subtilis AG1 crude extract at two different concentrations: 1 mg/ml and 0.5 mg/ml (C, control; Sol, control with solvent; CE, crude extract).

Results and discussion

The bacterial strain AG1, a spore forming Gram-positive, was referred to Bacillus subtilis subsp. subtilis (B. s.) on the basis of partial 16S rRNA sequence comparison (99% similarity; Fig. 4).

Fig. 4. Phylogenetic tree based on 16S rRNA sequences. The bacterial strain AG1 was identified as B. subtilis subsp. subtilis (99% similarity). The term “unknown” indicates the bacteria strain AG1.

The results of these investigations allowed to put in evidence for the first time the B. s. antagonistic activity towards phytopathogens (FOMED, PAL, PCH) causing “esca” disease of grapevine. Particularly, direct essays showed a considerable biocontrol by the bacterium towards the three tested fungal species. Indirect essays by means of either cultural filtrates and crude extracts (Fig. 5, 6), both at two different doses, confirmed the bacterial antagonistic activity; this latter was statistically significant at both the 5% and 1% level towards all the above mentioned phytopathogens.

Particularly, the crude extract was able to inhibit PCH and PAL growth already at the lowest dose (0.5 mg/ml). Further studies are required in order to determine the chemical composition of the crude extracts and then to identify the molecules involved in the antibiosis process, in the perspective of an eventual application in defensive strategies.

PAL PCH FOMED 8 7

m 6 5 K 4

Growth in c CF Bs25% 3 **

2 ** ** CF Bs50% ** ** ** ** * P ≤ 0.05 1 ** ** ** ** ** ** ** * ** ** ** ** ** P ≤ 0.01 ** ** 0 ** ** ** 7 14 21 28 7 14 21 28 7 14 21 28 Days

Fig. 5. Growth of phytopathogens cultures (Ø in cm) at 7, 14, 21, 28 days after inoculation on MEA (control, C) and on MEA added with 50% or 25 % of B. subtilis AG1 cultural filtrate.

8 7

m 6 C PAL PCH FOMED 5 Sol 4 CE 1 mg/ml 3

Growth in c Growth CE 0.5 mg/ml 2

1 **

0 **

7 14 21 28 7 14 21 28 7 14 21 ** 28

** ** ** * P ≤ 0.05 ** ** **** ** ** ** **** **** **** Days **** **** ** P ≤ 0.01

Fig. 6. Growth of phytopathogens cultures (Ø in cm) at 7, 14, 21 and 28 days after inoculation on MEA (control, C) and on MEA added with solvent, MEA added with 1 mg/ml and 0.5 mg/ml of B. subtilis AG1 crude extract.

References

Di Marco, S., Mazzullo, A., Calzarano, F. and Cesari, A. 2000: The control of esca: status and perspectives. – Phytopath. Medit. 39(1): 232-240.

Ferreira, J.H.S, Matthee, F.N. and Thomas, A.C. 1991: Biological control of Eutypa lata on grapevine by an antagonistic strain of Bacillus subtilis. – Phytopathology 81: 283-287. Fourie, P.H., Hallen, F., van der Vyver, J. and Schreuder, W. 2001: Effect of Trichoderma treatments on the occurrence of decline pathogens in the roots and rootstocks of nursery grapevines. – Phytopath. Medit. 40: 473-478. Kim, P.I. and Chung, K.-C. 2004: Production of an antifungal protein for control of Colleto- trichum lagenarium by Bacillus amyloliquefaciens MET0908. – FEMS Microbiol. Lett. 234: 177-183. Mc Keen, C.D., Reilly, C.C. and Pusey, P.L. 1986: Production and partial characterization of antifungal substances antagonistic to Monilinia fructicola from Bacillus subtilis. – Phytopathology 76: 136-139. Mugnai, L., Graniti, A. and Surico, G. 1999: Esca (Black measles) and brown wood- streaking: two old and elusive diseases of grapevines. – Plant Dis. 83: 404-418.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 25-30

Efficacy of microorganisms and natural products against grapevine powdery mildew

Dario Angeli 1, Loris Maines 1, Carmela Sicher 1, Haya Abou Assaf 2, Claudia Longa 1, Yigal Elad 1,3, Vito Simeone 2, Ilaria Pertot 1

1SafeCrop Centre, Istituto Agrario di S. Michele all’Adige, via Mach 1, S. Michele all’Adige, 38010, Italy, e-mail: [email protected]; 2Istituto Agronomico Mediterraneo, Valenzano, Bari; 3Department of Plant Pathology and Weed Sciences, ARO, The Volcani Center, Bet Dagan 50250, Israel

Abstract: Several alternatives to synthetic fungicides such as horticultural oils, salts, plant extracts and many potential BCAs have been evaluated in order to reduce the use of pesticides against Erysiphe necator on grapevine. Until present none of them is widely used in commercial vineyards. We tested new alternatives to chemicals against grapevine powdery mildew, possibly more effective than the existing ones. Among the tested experimental microorganisms a yeast, a bacterium and Bacillus subtilis (Serenade) showed effective and consistent suppression of E. necator under greenhouse controlled conditions. Among the non synthetic fungicides tested in the greenhouse, milk and milk derivates, salts and a vegetable oil gave promising results in reducing the disease. In the vineyard, only salts had a high efficacy against powdery mildew. However, milk derivates, vegetable oils and tested microbial agents significantly reduced powdery mildew infections.

Key words: biological disease control, microbial agents, natural substances

Introduction

Powdery mildew is one of the most dangerous grapevine diseases in Italy, because environmental conditions are suitable for the development of this fungus in several areas of the country. The disease is caused by the obligate biotrofic fungus Erysiphe necator. At present grapevine protection against powdery mildew is mainly based on the use of synthetic fungicides (Hewitt, 1998) or sulphur. During the last decades, fungicide resistance, concerns regarding pesticide residues in food and revocation of some widely used fungicides pushed researchers to develop biocontrol agents (BCAs) of fungal pathogens and natural fungicides. To reduce the use of pesticides, several alternatives to chemicals against E. necator have been evaluated (Kiss, 2003). Horticultural oils, salts, plant extracts and many potential BCAs such as mycophagus mites, insects, bacteria, epiphytic yeasts and filamentous fungi have been widely studied in powdery mildews control, especially in greenhouses (Bélanger & Benyagoub, 1997; McGrath & Shishkoff, 1999; Pasini et al., 1997), but only few of them showed good disease control under field conditions. The aim of the present research was to evaluate the efficacy of microorganisms and some new compounds against grapevine powdery mildew under greenhouse controlled conditions, as a first step in the process of selecting new alternatives to chemical fungicides. The most effective ones were preliminary tested under field conditions.

25 26

Material and methods

Several fungi, bacteria and yeasts isolated from natural environment in Israel (Volcani Center, Bet Dagan) and Italy (IAMB, Bari and IASMA, S. Michele all’Adige) and some nonchemical products (Tables 1 and 2) were tested for their ability to control grapevine powdery mildew in greenhouse and field trials. Microorganisms (named here as a code) were grown in potato dextrose broth (PDB) at 25°C for 48 hours, centrifuged, suspended and sprayed in distillate water plus an adjuvant (0.1%, Tween 80). The spray suspensions contained 1×106 conidia/ml of fungi and 1×108 CFU/ml of bacteria and yeast isolates. In greenhouse trials treatments were sprayed six hours before powdery mildew inoculation by a hand sprayer. The plant resistance inducer (Milsana) and the curative compound (KBV 99-01) were applied, respectively, three days before and three days after inoculation. For each treatment, 100 ml of the control agents in water were used. Five replicates, each one consisting of one plant of the susceptible cultivar Pinot Gris with one shoot and 5-6 leaves, were used. Plants were grown under controlled conditions (25°C and 60% RH) in a pathogen free greenhouse. Powdery mildew inoculation was done by gently rubbing infected leaves with fresh symptoms (mycelium, conidia and cleistothecia) collected in the field. Seven and 14 days after inoculation the percentage of infected leaf area (severity) and the percentage of infected leaves (incidence) were assessed on all treated leaves of each replicate.

Table 1. Microorganisms tested in greenhouse during 2007. Sulphur, PDB and water represent the standard and the untreated control respectively.

Active ingredient Commercial name or code Dosage (CFU/ml) Sulphur Thiovit (Syngenta) 3* Potato dextrose broth PDB (Sigma) 24* Distillate water + 0.1 % Tween 80 Water Drip off Yeast Y1 1×108 Yeast Y2 1×108 Yeast Y13 1×108 Bacterium B1 1×108 Bacterium B2 1×108 Bacterium B3 1×108 Bacterium B4 1×108 Bacillus subtilis Serenade (Intrachem bio) 3* Filamentous fungi F1 1×106 Filamentous fungi F2 1×106 *g/l

Under field conditions the efficacy against grapevine powdery mildew of the most promising BCAs and natural fungicides was evaluated. Trials were performed in 2006 and 2007 in an experimental vineyards located in S. Michele all’Adige (Trentino) on the varieties Schiava and Chardonnay, both highly susceptible to powdery mildew. The experimental design was fully randomized blocks, with five replicates (plots with five plants each). Treatments (10 hl/ha) were weekly applied by a backpack sprayer. BCAs were applied early in the morning when temperature and relative humidity were suitable to the microorganisms. The assessments were done weekly by randomly checking 50 leaves and 25 bunches for

disease incidence and severity. Analysis of variance (ANOVA) was applied on “arcsin” transformed data, using the software Statistica 7 (Statsoft). Significant differences among treatments were separated by Duncan’s test.

Table 2. Natural compounds tested in greenhouse during 2007. Sulphur and water represent the standard and the untreated control respectively.

Active ingredient Commercial name or code Dosage (g/l) Sulphur Thiovit (Syngenta) 3 Water - Drip off Potassium bicarbonate - 10 Potassium bicarbonate (85%) Armicarb (Helena) 5 Sodium bicarbonate - 10 Cow’s milk (Mila) 250* Lactoperossidase KBV 99-01 (Koppert) 1.5 Fatty acid Tecnobiol (Tecnotrea) 7* Tee tree oil Timorex (Biomor) 10* Clove oil - 1* Reynoutria Sachalinensis Milsana (Biofa) 10* *ml/l

e 16 e e 14 de e

12 de cde 10

Severity (%) bcd bcd 6 ab ab ab ab 4 a 2

0 F1 F2 B1 B2 B3 B4 Y1 Y2 Y13 PDB Water Sulphur Serenade Untreated

Fig. 1. Effect of fungi, yeasts and bacteria sprayed 6 hours before inoculation on powdery mildew severity under greenhouse controlled conditions. Columns with same letters do not significantly differ (P≤0.05, Duncan’s test).

Results and discussion

Disease pressure in the greenhouse trials was high and severity was used to compare the efficacy of tested microbial agents.

In the greenhouse trials several bacteria and yeasts were able to reduce powdery mildew severity though to a lesser extent compared to sulphur. The best results, as effective as sulphur, were obtained with the yeast Y13 and two bacteria, B1 and B. subtilis contained in the commercial product Serenade (Figure 1). Among the natural fungicides, milk derivates (cow’s milk and the lactoperoxidase based fungicide), salts and clove oil well suppressed powdery mildew (Figure 2). Sodium bicarbonate is the most effective ingredient for powdery mildew control and as good as sulphur.

80 f f 70 60 50 e de 40

Severity (%) 30 bcd bcd bc bc bc 20 ab 10 a

0 r Milk Water Sulphur Armica Milsana Milsana Timorex Clove oil Tecnobiol KBV 99-01 K bicarbonate Na bicarbonate Na bicarbonate

Fig. 2. Effect on disease incidence of some natural compounds sprayed 6 hours before inoculation under greenhouse controlled conditions. Milsana and KBV 99-01 were applied respectively 3 days before and 3 days after inoculation. Columns with same letters do not significantly differ (P≤0.05, Duncan’s test).

30 100 25 c 80 b 20 60 b 15 10 40 a b a a a a a a a Severity (%) 5 20

0 0 Milk Milk Sulphur Sulphur Armicarb Armicarb Armicarb Armicarb Untreated Untreated K bicarbonate K Bicarbonate Na bicarbonate Na bicarbonate Na bicarbonate

Fig. 3. Disease severity on leaves (left) and incidence on bunches (right) of plants treated with milk and different salts under field conditions (2006). Columns with same letters do not significantly differ (P≤0.05, Duncan’s test).

In 2006 powdery mildew infections increased early in the season and incidence on untreated leaves and bunches at the end of the season was 100 and 70%, respectively. Good control of powdery mildew was obtained with Armicarb, sodium and potassium bicarbonate both on leaves and on grapes (Figure 3). Milk was ineffective against powdery mildew on bunches. In 2007 weather conditions were suitable to E. necator, but powdery mildew infections increased only later in the season and symptoms on bunches did not appear. Incidence on untreated leaves was conversely 100%. All tested products reduced powdery mildew severity (Figure 3). The experimental BCAs B1 and Y13 were able to reduce disease severity on leaves and were not significantly different from sulphur.

30 b 25

20 ab Severtity 15(%) a a a 10 a a a 5

0 B1 Y13 Sulphur Milsana Milsana Clove oil Serenade Untreated KBV 99-01 Fig.4. Disease severity on leaves treated with microorganisms and natural compounds under field conditions (2007). Columns with same letters do not significantly differ (P<0.05, Duncan’s test).

Conclusions

In greenhouse and field trials, several tested products had a high effectiveness against powdery mildew, especially bicarbonate solutions. Some of the microbial control agents gave promising results in reducing the disease. B1, Y13 must be now formulated to reduce their variability in disease control and increase their survival and efficacy in the vineyard. They will be further studied in greenhouse and field experiments to evaluate their potential as alternatives to sulphur against powdery mildew in organic viticulture.

Acknowledgements

This research was financially supported by SafeCrop Centre, a research project funded by Autonomous Province of Trento.

References

Bélanger, R.R. & Benyagoub, M. 1997: Challenges and prospects for integrated control of powdery mildews in the greenhouse. – Canadian Journal of Plant Pathology 19: 310-314. Hewitt, H.G. 1998: Fungicides in crop protection. – CABI Publishing, Wallingford, UK, 221 pp. Kiss, L., Russell, J.C., Szentivanyi, O., Xu, X. & Jeffries, P. 2004: Biology and biocontrol potential of Ampelomyces mycoparasites, natural antagonist of powdery mildew fungi. – Biocontrol Science and Technology 14: 635-651. McGrath, M.T. & Shishkoff, N. 1999: Evaluation of biocompatible products for managing cucurbit powdery mildew. – Crop Protection 18: 471-478. Pasini, C., D’Aquila, F., Curir, P. & Gullino, M.L. 1997: Effectiveness of antifungal compounds against rose powdery mildew in glasshouses. – Crop Protection 16: 251-256.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 31-34

Natural occurrence of Ampelomyces spp. as E. necator mycoparasite in the vineyards of Trentino Province (Northern Italy) and efficacy evaluation of A. quisqualis in integrated powdery mildew management

Dario Angeli 1, Loris Maines 1, Erica Di Marino 2, Enzo Mescalchin 2, Ilaria Pertot 1 1SafeCrop Centre, 2CAT, Istituto Agrario di S. Michele all’Adige, via Mach 1, S. Michele all’Adige, 38010, Italy, e-mail: [email protected]

Abstract: Little is known on the natural occurrence of Ampelomyces quisqualis in Trentino Province (Northern Italy), where grapevine represents the second most important crop. The natural presence of wild strains of Ampelomyces spp. in the vineyards was assessed. Three-year monitoring of Ampelomyces spp. on E. necator grapevine in autumn shows a low natural presence of Ampelomyces spp. in the vineyards. Several local strains were isolated to be hopefully developed as commercial biocontrol agents. Ampelomyces spp. were found only on leaves, as mycelium parasitizing E. necator cleistothecia and as free conidia, both in conventionally treated and organic vineyards. Among the Ampelomyces spp. strains isolated from the vineyards, strains with atypical morphology were identified, differing from the commercial A. quisqualis strain (AQ10) in the conidia shape. The efficacy of strategies (AQ10 integrated with sulphur or chemicals) in suppressing E. necator was evaluated. Promising results were obtained when AQ10 was applied at the end of the season, showing no difference compared to sulphur.

Key words: biological control, organic viticulture, mycoparasite, antagonistic fungus

Introduction

Powdery mildews are among the most important plant diseases despite extensive research on their pathogenesis, epidemiology and control. These diseases account for the greatest proportion of the fungicides applied in European agriculture. Important crops like grapevine, apple, strawberry, but also cereals and several vegetables and ornamentals, grown in the field or greenhouses, are the major targets of powdery mildew fungi. On grapevine the disease is caused by obligatory biotrofic fungus Erysiphe necator (Schwein) Burrill. At present grapevine protection against powdery mildew is mainly based on the use of synthetic fungicides (Hewitt, 1998). Fungicide resistance problems, concerns regarding pesticide residues and revocation of some widely used fungicides have led to increased research efforts to develop biocontrol agents (BCAs). Microbial biocontrol agents of plant pathogen, if carefully selected, may offer a valid alternative to chemical fungicides, being them safe and biodegradable. Many potential BCAs have been tested against powdery mildew and various mechanism of action have been described (competition for space and nutrients, antibiosis, induced resistance, hyperparasi- tism), but they are far from being fully understood (Kiss, 2003). Ampelomyces quisqualis Ces. is one of the most successfully commercialized BCAs. This fungus is an aggressive mycoparasite of powdery mildew several species, including E. necator (Kiss et al., 2004). It acts directly by invasion and destruction of host cytoplasm (Hashioka & Nakai, 1980). The fungus infects and forms pycnidia within powdery mildew hyphae and conidiophores and can parasitize E. necator immature cleistothecia, killing the parasitized cell by degradation of

31 32

cytoplasm (Falk et al., 1995; Kiss et al., 2004). In Trentino E. necator cleistothecia are the main source of inoculum for primary infections in the vineyard. Until now little is known about the natural occurrence of A. quisqualis in Trentino province (Northern Italy) where grapevine is one of the two most important crops in the area. During this experiment the natural presence of wild strains of Ampelomyces spp. in the vineyards was assessed and several local strains were isolated to be hopefully developed as commercial BCAs. Furthermore, the efficacy of the mycoparasitic fungus A. quisqualis (AQ10) was evaluated under field conditions in suppressing pathogen by cleistothecia colonization using strategies combined with conventional fungicides.

Material and methods

The natural presence of Ampelomyces spp. as E. necator hyperparasite was evaluated during three years in 80 vineyards of Trentino (northern Italy) with the method reported by Angeli et al. (2006). To obtain Ampelomyces spp. isolates, parasitized powdery mildew cleistothecia were crushed on glass slide and putative Ampelomyces spp. conidia were collected with a glass needle. Conidia were transferred to Petri dishes on Czapek dox agar medium amended with 0.5% chloramphenicol. The growing colonies were isolated by the dilution method. Isolates were maintained on potato-dextrose agar (PDA) and Czapek-Dox agar amended with 2% malt extract (MCzA) at 25°C. For long-term storage, isolates were stored in test-tubes on PDA and MCzA at 4°C. Isolates were identified as Ampelomyces spp. and the shape of their conidia compared to AQ10 conidia. The effect of the commercial product AQ10, based on A. quisqualis (Intrachem Bio, Italy), was evaluated under field conditions. The trial was performed in an experimental organic vineyard located in S. Michele all’Adige (Trentino, northern Italy). A variety highly susceptible to powdery mildew (Schiava) was used. AQ10 applications (6 g/hl) or sulphur (300 g/hl) or synthetic fungicides (standard strategy based on strobilurine 15 g/hl, quinoxifen 25 cc/hl, penconazole 25cc/hl, spiroxamine 100cc/hl) were applied with a seven days interval using an atomizer with a volume of 10 hl/ha. Five different strategies were compared (Figure 1). AQ10 was applied when temperature and relative humidity were suitable to the organism (beginning and end of the growing season) and sulphur or synthetic fungicides in the remaining periods.

Strategy May June July August UN Untreated S-S Sulphur* A-S 3 x A. quisqualis Sulphur A-C 3 x A. quisqualis Synthetic fungicides S-A Sulphur 3 x A. quisqualis * Standard strategy Fig. 1. Strategies applied in the vineyard to control powdery mildew in Trentino in 2006.

The assessments were weekly done by randomly checking 50 leaves and 25 bunches per plot. Five plots (replicates) per treatment were used in the experimental design. The percentage of leaves or bunches with symptoms (incidence) and the percentage of sympto-

matic leaf and bunch area (severity) was estimated. Area Under the Disease Progress Curve (AUDPC) was calculated. Analysis of variance (ANOVA) was applied on “arcsin” transformed data, using the software Statistica 7 (Statsoft, Italy). Significant differences among treatments were separated by Duncan’s test.

Results and discussion

The three-year monitoring of Ampelomyces spp. shows a low natural presence of Ampelomyces spp. in the Trentino’s vineyards (less than 1% of parasitized cleistothecia). Ampelomyces spp. were found only on leaves as mycelium parasitizing E. necator cleistothecia and as free conidia in all types (IPM, organic, abandoned) of vineyards (Table 1) Among the Ampelomyces spp. isolated from the vineyards, likely new strains were identified, differing in conidia shape from the commercial A. quisqualis strain (AQ10). The conidia of the wild types are more fusiform, compared to the ellipsoidal ones of AQ10 (Figure 2).

Table 1. Presence of Ampelomyces spp. in Trentino during last three years

Monitored Presence of Ampelomyces spp. in Parasitized Strains Year vineyards (No.) vineyards/total vineyards cleistothecia IPM Organic Abandoned 2004 17 0/12 0/4 1/1 wild 0.17%

2005 17 1/12 0/4 0/1 wild 0.17% wild/ 2006 44 8/32 1/6 1/6 1.32% commercial

AQ10 2004 2005 2006

Fig. 2. Shape of the commercial A. quisqualis strain (AQ10) and Ampelomyces spp conidia isolated in 2004, 2005 and 2006.

In 2005 weather conditions were suitable to E. necator and at the end of June the incidence on untreated bunches in the experimental vineyard was 100%. Data showed that if applied early in the season when powdery mildew rapidly increased, AQ10 treatments followed by sulphur or chemicals were less effective than standard (only Sulphur). Better results were obtained both on leaves and on bunches when AQ10 was applied at the end of the season, showing no difference compared to sulphur standard (Figure 3).

4500 2500 c c 4000 2000 3500 3000 1500 es (AUDPC) 2500 2000 1000 1500 1000 500 b b b b ab a a 500 a Severity on leav 0 Severity on bunches (AUDPC) 0 UN SS AS AC SA UN SS AS AC SA

Fig. 3. Powdery mildew severity (AUDPC) on leaves (left) and bunches (right) in 2005. Treatments reported on x-axis are described in Figure 1. Columns with same letter do not differ with P≤0.05 (Duncan’s test).

In conclusion in 2004, 2005 and 2006 the natural presence of Ampelomyces spp. in Trentino’s vineyards was very low, probably because of the low relative humidity and leaf wetness, which are important for Ampelomyces spp. germination or because of the low level of disease in the previous five years, which reduced the natural populations of the mycoparasite. The poor results with AQ10 applications at the beginning of the season could be related to unsuitable environmental condition for the antagonism or to an intrinsic inability of the used antagonist to fully control the disease before the exponential phase of the epidemic. A highly aggressive Ampelomyces strain or better adapted to the local environmental condition could improve powdery mildew biocontrol in the vineyard.

Acknowledgements

This research was supported by SafeCrop Centre, a research project funded by Autonomous Province of Trento.

References

Angeli, D., Di Marino, E. & Mescalchin, E. 2006: Colonization of grapevine powdery mildew cleistothecia by the mycoparasite Ampelomyces quisqualis in Trentino, Italy. IOBC/WPRS Bull. 29(11): 89-92. Falk, S.P., Gadoury, D.M., Cortesi, P., Pearson, R.C. & Seem R.C. 1995: Parasitism of Uncinula necator cleistothecia by the mycoparasite Ampelomyces quisqualis. Phytopath. 85: 794-800. Hashioka, Y. & Nakai, Y. 1980. Ultrastructure of pycnidial development and mycoparasitism of Ampelomyces quisqualis parasitic on Erysiphales. Transactions of the Mycological Society of Japan. 21(3): 329-338. Hewitt, H.G. 1998: Fungicides in crop protection. CABI Publishing, Wallingford, UK, pp.221. Kiss, L., Russell, J.C., Szentivanyi, O., Xu, X. & Jeffries, P. 2004: Biology and biocontrol potential of Ampelomyces mycoparasites, natural antagonist of Powdery mildew fungi. Biocontrol Science and Technology 14: 635-651. Kiss, L. 2003: A review of fungal antagonists of powdery mildews and their potential as biocontrol agents. Pest Management Science 59, 475-483.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 35-36

VitiMeteoPlasmopara - a modern tool for integrated fungicide strategies

Bleyer1,G.; Kassemeyer1,H.-H.; Viret2, O.; Siegfried2, W.; Krause3, R. 1Staatliches Weinbauinstitut, Freiburg (Germany), [email protected], 2Research Station ACW, CH-Changins-Wädenswil, [email protected], 3GEOsens, Messystem und Softwareentwicklung, Ebringen (Germany), [email protected]

VitiMeteoPlasmopara is an open, computerbased forecasting system for the integrated control of downy mildew. Since 2002, the Staatliches Weinbauinstitut Freiburg in Germany and the Swiss Research Station ACW, Changins-Wädenswil, in cooperation with the private company GEOsens have developed the system. Weather data are collected from the field independent of the forecasting model and are stored in a database. The structure of the database is open to different types of weather stations. The model calculates primary and secondary infections, sporulation, sporangia density and incubation time. A vine growth model completes the pure forecasting system. This allows the growers to define their spray intervals in relation to the epidemics of downy mildew and the unprotected leaf surface area. The administrator can parameterise all relevant biological parameters for example temperature etc., in the software. VitiMeteoPlasmopara was tested with a set of historical and actual climatic conditions derived from the three respective institutes. The results of the model are visualised for the plant protection service and growers on the Internet (hhtp://www.agrometeo.ch, http://www.WBI- Freiburg.de) in the form of tables and graphs twice a day. A table showing the situation of the last seven days allows a clear and short overview. From 2004 on, VitiMeteo is used as a

VitiMeteo = open system VitiMeteo – Suite

DATA VITI METEO REGISTRATION IMPORT DATABASE „Future“ of weather data Agrometeo (Expert in the field (VMImport) Software)

VITI METEO VITI METEO VITI METEO VITI METEO PLASMOPARA GROWTH INSECTS DATA GRAPH Expert Software Expert Software Expert Software Expert Software

PRESENTATION PRESENTATION PRESENTATION PRESENTATION Internet Internet, possible Internet Internet , possible • PDF, HTML, XLS • PDF, HTML, XLS • PDF, HTML, XLS • PDF, HTML, XLS

35 36

system to forecast downy mildew in Switzerland and Southern Germany (Baden- Württemberg), covering approx. 42’000 ha of grapevine, with about 100 weather stations located in the vineyards. The free Internet access is widely used by the growers and plant protection service, appreciating the simple way of presenting the information and its usefulness for disease management. In close cooperation with the users the system is steadily improved and adapted. VitiMeteoPlasmopara is a part of the so called VitiMeteoSuite. The structure of the database allows the use and the integration of weather data into any other models to forecast the development of other fungal diseases or pests. VitiMeteoDataGraph, a software to analyze and present weather data, VitiMeteoGrowth, a vine growth model and VitiMeteoBerry moth, a model for the berry moth, are already developed. VitiMeteo- Plasmopara is an expert system available for other interested institutions.

Key words: Vitis vinifera, grapevine, modeling, Plasmopara viticola, downy mildew, integrated control, disease control

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 37-44

Grapevine infectious diseases in Sicily

Santella Burruano1, Giovanni Granata2 1 Dipartimento S.En.Fi.Mi.Zo, Sez. di Patologia vegetale e Microbiologia agraria, Università degli Studi di Palermo 2 Dipartimento di Scienze e Tecnologie Fitosanitarie, Sez. di Patologia vegetale, Università degli Studi di Catania

Abstract: Erysiphe necator and Plasmopara viticola are the most recurrent fungal pathogens in the Sicilian vineyards, but appropriate schedule treatments can achieve good control. Rainy summer and hailstorms play a leading role in provoking severe Botrytis cinerea infections, lower production and reduced grape must quality. Esca disease, Nattrassia mangiferae, Phoma glomerata with Fusarium oxysporum and Cylindrocarpon destructans produce alterations of the grapevine wood. Bacterial necrosis of grapevine the causal agent identified as Xylophilus ampelinus, induces a progressive scion dieback. Many infectious agents of grapevines, including phytoplasma and viruses, occur on both vine and table cultivars in Sicily. Grapevine yellow (Bois noir) mainly affects Inzolia and Chardonnay cultivars. Grapevine fanleaf virus (GFLV), caused by a Nepovirus, presents two distinct syndromes depending on the different virus strains: infectious malformations and yellow mosaic. Grapevine leafroll GLRaV 1-3-4 Ampelovirus and GLRaV 2 Closterovirus are the most widely spread. Rugose wood complex occurs widely and includes: Rupestris stem pitting (GRSPaV); Kober stem grooving (GVA) and LN33 stem grooving (GBV). Grapevine fleck, latent in European grapevine varieties and in most American rootstocks, symptoms are expressed in V. rupestris is also widely distributed. Virus-like diseases observed in Sicily are: Enation disease and Vein necrosis.

Key words: Vine, Fungi, Bacteria, Phytoplamas, Viruses, Virus-like diseases.

Introduction

Grapevine (Vitis vinifera L.) is one of the most widespread crops in Sicily. Because of its sensitivity to pathogen infections, it is severely exposed to many infectious diseases which can be spread in vineyards by vegetative propagation of this crop. These diseases can be grouped according to their pathogenic agents (Tab. 1). The biotic alterations are caused by fungi, bacteria, phytoplasmas, viruses and virus-like organisms. Infectious diseases of grapevine are present in the major vine growing areas. They are influenced by the presence of the pathogen and climatic conditions. In Sicily, fungal diseases are generally more prevalent than bacterial ones, because the high temperatures and low humidity in our region do not favour these last diseases. Over the last thirty years, great progress has been achieved in the field of virus and virus- like disease and etiologic agents. Studies on the subject have acquired adequate information to draw a rather complete picture of the known diseases.

37 38

Table 1. Infectious diseases and pathogenic agents of grapevine in Sicily

Powdery Mildew (Erysiphe necator) Downy Mildew (Plasmopara viticola) Grey Mould (Botrytis cinerea) Fungal “Esca” complex (Phaeomoniella chlamydospora, Phaeoacremonium Diseases aleophilum, Fomitiporia punctata) Decline of young vines (Phoma glomerata; Fusarium oxysporum and Cylindrocarpon destructans; Nattrassia mangiferae) Root rot (Armillaria mellea) Bacterial Bacterial blight (Xylophilus ampelinus) Diseases Crown gall (Agrobacterium tumefaciens) Phytoplasmas Grapevine yellow (BN - Bois noir) Diseases Bois noir (Phytoplasma stolbur group 16 Sr XII) Infectious degeneration (GFV - grapevine fanleaf) Viruses Grapevine leafroll (GLRaV 1-2-3-4) Diseases Rugose wood complex (GRSPaV, GVA, GVB) Fleck disease (Grapevine fleck-GFKV) Enation disease Virus-like Vein necrosis

Fungal diseases

Powdery mildew Agent: Erysiphe necator (Schweinitz) Burrill (sin. Uncinula necator (Schw.) Burr.) development is favourite by the climate in the south of Italy. Powdery mildew is the most recurrent disease in Sicilian vineyards. In spite of early infections occurring at vine-budding, symptoms on leaves, shoots and grapes become visible after flowering. Late infections of E. necator occur until August. They are usually not hamful but increase the inoculum pathogen that influences the severity of the syndrome the following year. Control: applications of pulverulent or wettable sulphur can be repeated at 10 - 12 day intervals before flowering. Good control can also be achieved by applying inhibitors of biosynthesis of sterols (IBS), applied alone or mixed with a sulphur spray or dinocap, and repeated at the same intervals to pre/until ripening (Cannizzaro and Burruano, 1982; Granata and Pennisi, 1983; Sidoti and Zaffuto, 1998)., Axoxystrobin and quinoxyfen, applied at the same intervals, achieve good control only under a mean powdery mildew pressure of attacks. Moreover, quinoxin efficacy is significantly reduced when Ampelomyces quisqualis and Bacillus subtilis are included in scheduled treatments (Polizzi et al., 2002). Adequate management of cultural practices can considerably restrict disease severity.

Downy mildew Agent: Plasmopara viticola (Berk. et Curt.) Berl. et De Toni, biology and epidemiology can be considerably influenced by the warm, dry Sicilian climate. In particular, climatic conditions can delay or stop maturation and germination of P. viticola oospores, the only way the pathogen can overwinter. As a consequence, the first oil-spots occur in vineyards much later than indicated by the rule of the "three 10s". Primary infections can usually determine two or three secondary infective cycles until the middle of July. In coincidence with autumnal rainfall, P. viticola sporulation occurs again, causing typical mosaic spots on leaves where the oospores form. On the other hand abundant and frequent spring rains cause early onset of

primary infections. These can go on for two-three months and so can be added to the secondary ones. Downy mildew infections become epidemic and highly destructive when the host is more susceptible. Control: a rational strategy should be based on careful monitoring of both climatic and epidemiological conditions. Together with the biology of the pathogen in our region, these data can forecast the probable date of primary infections and suggest immediate treatment by protective or persistent compounds. Treatments have to be continued as long as climate favourable to downy mildew persists (Burruano et al., 2006).

Grey mould Agent: Botrytis cinerea Pers., overwinters as mycelium in shoot bark, sometimes in buds or as sclerotia on both vine shoots or self-sown plants. Usually, the grey mould infections occur in covered vineyards after late summer-autumn rains. Control: 1 or 2 chemical treatments using persistent fungicides are applied at the time of late summer rainfall. If applied together with careful and adequate agronomical techniques they can reduce damage.

Esca Agents: interaction of three fungi, Phaeomoniella chlamydospora (W. Gams, Crous. M.J. Wingfield & L. Mugnai) Crous & Gams, Phaeoacremonium aleophilum Crous, W. Gams, M.J.Wingfield &van Wyk and Fomitiporia punctata (Fr. & Karsten) Murrill, causes vine wood alterations. These have shown an increasing incidence particularly in young stressed table grapes since the 1980s (Granata 1982; Schilirò et al., 1996). Recently, xylematic fungi have been observed in cutting vines and shoots of mother plants (Sidoti, 2001), confirming that propagation material may become infected in the nursery (Sidoti et al., 2004; Surico et al., 2006). The chronic "esca", consisting in external symptoms (chlorotic turning into necrotic spots between and/or along leaf margins; Surico et al., 2000) and internal symptoms (xylematic dark brown areas, pink-brown areas, white rot) is clearly visible in late spring, with an increasing incidence during summer. The acute "esca" or apoplexy occurs as early as the beginning of July (Schilirò et al., 1996; Mondello et al. 2007). Control: preventive agronomic measures (use of healthy and vigorous cuttings, no excessive forcing of young plants, disinfection of wounds, immediate removal of infected woody parts).

Decline vines Agents: Phoma glomerata (Corda) Wr. et Ho, Fusarium oxysporum (Schlecht.) Sn. et H. with Cylindrocarpon destructans (Zins.) Scholten and Nattrassia mangiferae (H. & P. Sydon) Sutton and Dyko, are also responsible for vine decline. In particular, P. glomerata causes necrotic areas on the rootstock, its consequent scarce growth, and progressive chlorosis of leaves. The syndrome can be prevented by removing the rootstock spurs in late winter and disinfecting the consequent wounds (Refatti and Granata, 1981). F. oxysporum and C. destructans determine yellowing of leaves, cortical necrosis and xylematic browning starting from the base of the trunk (Grasso, 1984). "Grapevine blight" caused by N. mangiferae has been recorded since the end of the 1980s on cv. Moscato bianco and it is now epidemic in different cultivars of wine and table grapes. Disease symptom generally appear on the leaves as scattered chlorotic spots which then turn necrotic. A dark chromatic alteration can be observed in the cambium near the wounds. After defoliation, shoot drying starts from the apex and the infected vines die rapidly. Control: only by good cultivation practices and the elimination of infected and dead plants (Sidoti and Granata, 1998).

Root rot Agent: Armillaria mellea (Vahl ex Fr.) P. Karst determines the disease and in the Etna environment that has shown strong recurrence on old and young stressed plants after periods of high temperatures and drought. The basidiomycete can be easily identified by white mycelium mats, under decayed bark of the rot collar or the rots, or as rhizomorpha. These are the prevalent way the disease is spread from tree to tree. Control: preventive agronomic techniques (Sidoti, 2004).

Bacterial diseases

Bacterial blight Agent: Xylophilus ampelinus (Panagopoulos) Willems et al. caused vine decline on different cultivars in the Monterosso area (Catania province of Sicily) more than 40 years old (Grasso et al. 1979). Symptoms: in spring, bud-break on affected spurs is either delayed or does not take place. Young shoots are frequently weak and die in a few weeks while the more vigorous ones show longitudinal necroses, starting from the basal nodes and followed by cracking. Necrotic areas, involving all the bark tissues as the outer xylematic layers, can affect the petioles of leaves emerging from nearby portions of the cane.

Crown gall Agent: Agrobacterium vitis (Ophel and Kerr), overwintering in soil for several years, causes the most recurrent bacteriosis in Sicilian vineyards. Typical tumors, of different shape and size can occur on roots, stem, vine shoots, petioles and also on leaf veins. Control: cultural and sanitary practises.

Phytoplasmas diseases

Bois Noir (BN) Bois noir is a yellow disease presenting similar symptoms to Flavescence dorée sensu strictu but caused by different stolbur group phytoplasma. In Sicily, Insolia and Chardonnay cultivars are the most sensitive. Agent: the phytoplasma associated to these disease belong to the stolbur group 16 Sr XII. Symptoms: on white-berried varieties: rolled leaves with yellowing of the veins and following necrosis. Shoots show a lack of lignifications. Withered and dried clusters. On red-berried varieties: reddish areas follow ed by necrosis along the main veins. Control: no direct control is possible. Chemical sprays against vectors are advised to reduce vector populations. Suppression of affected branches by means of drastic prunings helps progressive recovery.

Viruses Diseases

Infectious Degeneration (Grapevine fanleaf virus) (GFLV) Fanleaf is one of the most important and widespread virus disease of the grapevine. Typical symptoms onset before the introduction of American rootstock hybrids. Now the disease is known to occur worldwide. The disease is characterized by two distinct syndromes caused by different strains of the causal agent: a) Infectious malformations b) Yellow mosaic

Agent: Grapevine fanleaf virus (GLFV) is a Nepovirus with isometric particles of about 30 nm in diameter. Some strains of GLFV have an additional satellite RNA (Martelli G.P. et al., 2003). Symptoms: a) Infectious malformations: leaves are variously malformed, asymmetrical,with open petiolar sinuses, deep lobes and acute denticulation. Shoots are also malformed, showing fasciations and double nodes or short internodes and fork divarication. Bunches are smaller and fewer in number; berries ripen irregularly, are small-sized and set poorly. b) Yellow mosaic: caused by chromogenic virus strain. The leaves appear bright chrome yellow discoloration also shoots, tendrils and influorescences. Bunches are small and few. Transmission: grapevine fanleaf is transmitted by grafting. Vector are there dagger and different nematodes: Xiphinema index, X. italiae. Control: use of virus-free propagating material of the rootstock and scionwood.

Grapevine Leafroll (GLRaV) In Sicily leafroll is more important than fanleaf for economic importance. It is probably the most widespread virus disease of grapevine. Agents: to date, nine different viruses, with filamentous particles have been found in leafroll infected grapevines. They are differentiated by a progressive number: GLRaV-1 and 3-4-5-6- 8-9 belong in the genus Ampelovirus; GLRaV-2 in the genus Closterovirus, whereas GLRaV- 7 classifies an unassigned species to the family Clostreroviridae. (Gugerli P., 2003). In Sicily GLRaV1-2-3-4 are present Symptoms: in red-berried cultivars of Vitis vinifera reddish spots develop on the leaves in summer. In autumn these spots coalesce so that the leaf is almost red-purple, except for a narrow band along the primary and secondary veins. Rolling of the leaf blade. The cluster is paler in colour due to irregular and delayed ripening; they are inferior in quantity and quality, and low in sugar. Symptoms in white-berried cultivars of Vitis vinifera are similar, but the leaves became chlorotic to yellowish. There are different types of leafroll symptoms, differing in aspect or in severity, thus suggesting that there can be several causal agents. Infection on rootstocks is symptomless. Transmission: leafroll is graft-transmissible which is largely responsible for spread. Natural field spread of leafroll disease is transmitted by pseudococcid mealybugs: Heliococcus bohemicus and Phenacoccus aceris for GLRaV1; Planococcus ficus, P. citri, Pseudococcus longispinus, P. maritimus for GLRaV3. GRLaV5 and GLRaV9 are transmitted by Pseudococcus longispinus. Control: use of clonally selected and sanitized propagation material. Control of vector populations with insecticides is recommended.

Rugose Wood Complex The rugose wood complex includes four different and several diseases latent in rootstock hybrids. In Sicily the following are present: 1) Grapevine rupestris stem pitting 2) Kober stem grooving 3) LN33 stem grooving 4) Corky bark Agent: grapevine rupestris stem pitting associated with Foveavirus (GRSPaV). Kober stem grooving associated with Vitivirus A (GVA) LN33 stem grooving associated whit Vitivirus B (GVB) Corky bark associated whit Vitivirus B (GVB)

Symptoms: grapevine rupestris stem pitting; in Vitis rupestris it presents as pitting limited to a band extending downward from the point of inoculation; LN33 and Kober 5 BB present no symptoms. Kober stem grooving: grooving on the stem of Kober 5BB; no symptoms in V. rupestris and LN33. LN33 stem grooving: grooves on the stem of LN33; no symptoms in V. rupestris and Kober 5BB. Corky bark: grooving and pitting on stem of Vitis rupetris and LN33; no symptoms on Kober 5BB. Stunting of LN33 with rolling and redding of leaves and clusters. Internodal LN33 swelling and cracking (Savino V. et al., 1985). Transmission: the vector of grapevine rupestris stem pitting (GRSPaV) is not known. Kober stem grooving (GVA) and LN33 stem grooving (GVB) are transmitted from grapevine to grapevine by pseudococcid mealybugs: GVA is transmitted by Planococcus citri, Pl. ficus, Pseudococcus longispinus, Ps. affinis, Heliococcus bohemicus; GVB by Pseudococcus longispinus, Ps. affinis and Planococcus ficus. Control: use of propagation sound material of scionwood and rootstock obtained by sanitary selection. Insect control is difficult.

Fleck Disease Grapevine fleck causes latent or semi-latent infections in Vitis vinifera and rootstock hybrids. Disease to be one of the fleck complex: Asteroid mosaic, Rupestris necrosis, Rupestris vein feathering and Grapevine red globe virus, the latter not being present in Sicily. Its presence influences the vigor of the grape infected (Refatti E. and Granata G., 1981). Agent: grapevine fleck (GFKV) is associated with Maculavirus characterized by isometric phloematic particles about 30nm in diameter. The disease is not seed transmitted. Symptoms: the symptoms are expressed in Vitis rupestris and consist of clearing of small veins on leaves. Leaf malformation with upward curling of the blade. Severe strains also stunt growth. Control: on account of the latency of symptoms, the use of healthy European grapevine cultivars and rootstocks is effective way to control the disease.

Virus-Like Diseases

The diseases below reported, are still unknown. Their agents are transmitted by grafting and persistence in propagation material.

Enation disease This grapevine disease is characterized by complex symptomatology and is latent or semi- latent. Agent: its agent is still unknown. Symptoms: delayed opening of the buds. Shoot with short internodes and zig-zag growth. The leaves are dwarfed, misshapen and present typical lamellar proliferations on the lower surface of the blade along primary and secondary veins (Graniti A. and G.P. Martelli, 1970). Control: Use of healthy propagation material.

Vein Necrosis The disease is widespread on Sicilian grapevine cultivars. It is latent in Vitis vinifera and American rootstocks but produces manifest symptoms on vines at Vitis rupestris x V. Berlandieri 110 Richter (110R) (Granata G. and E. Refatti, 1984).

Agent: its agent is still unknown. Symptoms: vein necrosis of different severity on the lower face of the leaf blade on the rootstock (110 R). Control: use of indexed planting material.

References

Burruano, S., Conigliaro, G., Lo Piccolo, S., Alfonso, A., Torta, L. 2006: Plasmopara viticola: three decades of observation in Sicily. – Proceedings of the 5° Int. Work. Grapevine Downy and Powdery Mildew. San Michele all'Adige, 18-23 June 2006, 58-59. Cannizzaro, G. & Burruano, S. 1982: Prove di lotta contro l'oidio della vite in Sicilia. – Tec. Agric. 2, 3-7. Granata, G. 1982: Degats de l'esca sur raisin cv. Italia in Sicile. – Symp. Inter. Le raisin de table et le raisin sec. 5-11 September 1982, Héraklion - ile de Crete, Grèce. Granata, G. & Refatti, E. 1981: Decline and death of young grapevines by infection of Phoma glomerata on the rootstock. – Vitis 20: 341-346. Granata, G. & Pennisi, A. 1983: Efficacia di alcuni fitofarmaci triazolici e pirimidinici nella lotta contro l'oidio della vite. – Notiz. Mal. Piante 104: 135-140. Granata, G. & Refatti, E. 1984: Indagine sulla diffusione della “Necrosi delle nervature fogliari” in alcuni vitigni ad uva da vino in Sicilia. – Informatore Fitopatol. 34: 37-39. Graniti, A. & Martelli, G.P. 1970: Enations. – In: Virus Diseases of Small Fruits and Grapevines. (A Handbook), (N.W. Frazier), University of California, Division of Agricultural Sciences, Berkley: 241-243. Grasso, S., Moller, W. J., Refatti, E., Magnano di San Lio, G., Granata, G. 1979: The bacterium Xanthomonas ampelina as casual agent of a grape decline in Sicily. – Riv. Pat. Veg. IV 15 (3/4): 91-96. Grasso, S. 1984: Infezioni di Fusarium oxysporum e di Cylindrocarpon destructans associate a una moria di giovani piante di vite in Sicilia. – Informatore Fitopatol. 34: 59-63. Gugerli, P., 2003: Grapevine leafroll and related viruses. – Extended abstracts 14th Meeting of ICVG, Locorotondo: 25-31. Martelli, G.P., Walter, B., Pinck, L. 2003: Grapevine fanleaf virus. – AAB Descriptions of Plant Viruses. Polizzi, G., Cirvilleri, G., Catara, A. 2002: Esperienze pluriennali di difesa antioidica su vite in Sicilia con formulati chimici e biologici. – Atti Gior. Fitopat. 2002(2): 381-387. Refatti, E., Granata, G. 1981. Distribution of Grapevine fleck disease in Sicily and reaction some indicators. – Proc. VII Conf. ICVG, Niagara Falls, Canada, 8-12 Settembre 1980: 35-39. Savino, V., Boscia, D., Musci, D., Martelli, G.P. 1985: Effect of Legno Riccio (Stem pitting) on “Italia” vines grafted onto rootstock of different origin. – Phytopatologia mediteranea 24: 68-72. Schilirò, E., Bonocore, E., Di Natale, A., Zaffuto G. 1996: Incidenza del mal dell'esca della vite nella Sicilia centro-orientale. – Tec. Agric. 2/3: 71-79. Sidoti, A. & Granata, G. 1998: Alterazione del legno della vite causata dal fungo Nattrassia toruloidea. – Riv. Frut. Ortic. 60(3): 61-62. Sidoti, A. & Zaffuto, G. 1998: Difesa razionale dell'uva Italia dalle principali malattie fungine. – Inf. Agr. 54(47): 59-62.

Sidoti, A., Bonocore, E., Serges, T., Mugnai, L. 2000: Decline of young grapevines associated with Phaeoacremonium chlamydospora in Sicily (Italy). – Phytopath. Medit. 39(1): 87- 91. Sidoti, A. 2004: Protezione della vite dalle malattie nell'area di produzione dell'Etna. – Inf. Agr. 24: 71-74. Surico, G., Di Marco, S., Mugnai, L., Marchi, G. 2006: Il mal dell'esca della vite: Luci ed ombre delle ricerche sulla malattia. – Incontri Fitoiatrici 2-3 marzo 2006 Torino: 21-27.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 45-52

Evaluation of grapevine resistance to downy and powdery mildew in a population segregating for run1 and rpv1 resistance genes.

Calonnec A.1, Delière L.1, Cartolaro P.1, Delmotte F. 1, Forget D.2, Wiedemann- Merdinoglu S.3, Merdinoglu D.3, Schneider C.3 1 INRA, UMR1065 Santé Végétale, 33883, Villenave d’Ornon, France; 2 INRA, UE1086 Domaine expérimental viticole de Bordeaux, 33883, Villenave d’Ornon, France; 3 INRA, UMR1131 Santé de la vigne et qualité du vin, 6802 Colmar, France.

Abstract: In this paper we present the behaviour of a population of vine segregating for several sources of resistance to powdery (PM) and downy mildews (DM). Resistance genes from Muscadinia and Vitis American species were introduced in a Vitis vinifera genetic background. Resistance to both pathogens was assessed in bioassays under controlled inoculations with two isolates for PM and three isolates for DM and in greenhouse or field experiments under natural infections. 19 out of 38 segregant genotypes were identified as totally resistant to PM on leaves and partially resistant to DM on both leaves and bunches. All these genotypes presented a level of damage satisfactory in an IPM context. There is no evidence to say that the resistance genes are expressed in bunches. However, the descriptors of resistance used in the bioassay, can be used to select with a good accuracy genotypes with an acceptable level of damage on bunches. The link between resistance to powdery mildew and downy mildew suggest that resistance in harvestable genotypes is coming from Muscadinia (Rpv1 and Run1).

Key words: mildew, screening for resistance, bioassay, field assessment, Erysiphe necator and Plasmopara viticola.

Introduction

Powdery and downy mildew are the most widespread diseases on Vitis vinifera worldwide and the main target of fungicide treatments. Different options can be used better in combination to improve the control of these diseases in the respect of IPM (Integrated Pest Management). Among these strategies we can remind i) The use of forecast models to predict the risk of disease. However, none of them are able to predict the level of primary infections. ii) Disease survey at periods known for the high susceptibility of the plant or under favourable climatic conditions iii) The use of global models leading to crop protection decision integrating knowledge on epidemiology, period and conditions of risk, crop management and treatments. iv) Finally, the growth and deployment of partially resistant or moderately susceptible varieties such as hybrids coming from American or Asian species. In this paper we would focus on the development of host resistance. Our study takes place at an advanced step of a breading program which has for objective to ensure the selection of genotypes most likely to confer durable resistance to both diseases. In this program several sources of resistance coming from Muscadinia and Vitis American species were introduced in a Vitis vinifera genetic background. The selection for resistance was based on phenotype characterisation of the resistance level. In this paper we would present results of disease assessment of resistance to both pathogens Erysiphe necator and Plasmopara viticola, in bioassays under controlled inoculations and in greenhouse or field experiments under natural infections. Finally, we would test the

45 46

ability of the bioassays to predict resistance for these varieties to downy mildew in the field on leaves and on bunches.

Material and methods

Plant material and field experimental design The population used in this paper came from a cross between 3082-1-42 (BC4 from Muscadinia rotondifolia) x Regent (Chambourcin (12.417 SV x 7053 seibel) x Diana (Sylvaner x Muller Thurgau)). Resistance to powdery mildew and downy mildew originates from two linked resistance genes in Muscadinia, Run1 (PM resistance) (Pauquet et al., 2001, Barker et al., 2005) and Rpv1 (DM resistance) (Merdinoglu et al., 2003) and also from the partially resistant cultivar Regent which inheritates its resistance to both pathogen from different Vitis American parents (1 QTL of resistance to PM and 1 QTL for endurance to DM) (Fisher et al., 2004). 200 genotypes were screened in 2001-2002 at the Geilweilerhof Institute (Germany) for their resistance to powdery mildew in greenhouse under natural infection, and for their resistance to downy mildew in a laboratory bioassay on leaf disks and on leaves after artificial inoculation, at INRA Colmar (France). 38 genotypes with different combinations of PM and DM resistance were selected and planted in Bordeaux, Colmar and Montpellier in 2004. The experimental design was made of 4 consecutive vines per genotype planted in a randomized design. Parents and other susceptible or partially resistant cultivars are included in the design.

Bioassays In 2007, bioassays were done on leaves coming from the field material to select the genotypes resistant to both pathogens. In assay 1, resistance was tested with one isolate of powdery mildew on all the 38 genotypes. The 25th of April (average grape phenology 7 leaves/shoot), 1 young leaf per vine (first to second leaf under the last expended leaf) was sampled for each genotype (4 leaves= 4 repetitions/genotype), as well as for each of the control varieties randomized within the plot (Merlot, Chardonnay, Chambourcin, Villard blanc) and for both parents Regent and 3082-1- 42. Leaves were disinfected 10 min by a hypochlorite solution (50g/l), and a sample of 1 disk of 16 mm is taken from each leaf and placed in Petri dishes (Pd) on 20g/l of agar medium (upper surface on the top). One Pd contains 5 genotypes. In each Pd one control of infection success (Cabernet-Sauvignon leaf disk coming from cuttings) was included. The day after, the 38 genotypes plus the controls were inoculated in settling towers (4 towers to infect the 4 repetitions/genotype) by blowing 600 to 800 spores/cm² of the isolate S7 (biotype B). 13 days after inoculation, disks were briefly observed at the stereomicroscope for an assessment of mycelium and / or sporulation and sporulation was assessed for each disk by a Coulter cell counter (Multisizer III) after shaking the disk in a solution of isoton with non ionic dispersant (Nacconol 90F) (particles between 17 µm et 37 µm are counted). For the totally resistant varieties an adhesive tape test was performed on one repetition to control at which stage the spores were stopped. In assay 2, the genotypes resistant to powdery mildew at bioassay 1 were tested using three isolates of downy mildew and two more isolates of powdery mildew. 14 days and 7 days before the test, leaves just expanded were marked on different shoots. The 26 of June, the marked shoots are sampled and leaves aged of 7 days are tested against powdery mildew, whereas those aged of 14 days are tested against three isolates of downy mildew. For the DM assay, 3 disks of 16 mm are taken from each leaf and placed each in one Petri dishes (Pd) (one/isolate) containing a filter paper leaks out soaks (4 ml of water/Pd),

lower side of the leaf up. One Pd contains 5 genotypes plus a control of Cabernet-Sauvignon from cuttings. The day after, each disk was inoculated with 3 droplets of 10 µl each, with a cold solution of 19000 sporangia/ml. The three isolates used were selected the week before the test in the same field either on the susceptible variety Merlot or on the two partially resistant genotypes 70 and 134, and multiplied on Cabernet Sauvignon. Isolates selected on partially resistant varieties may have acquire a higher level of aggressiveness. Two repetitions (leaves) per genotypes and pathogen are used. 7 days after inoculation, the disease was scored on a visual scale similar to that of OIV-452 (IPGRI, 1997) taken into account the sporulation intensity and the necroses. Sporulation was also assessed for each disk by the Coulter cell counter (particles between 10 µm and 25 µm are counted). For the PM assay, 2 disks of 16 mm were taken from each leaf and placed in two Pd (one /isolate) and prepared as before. The two mono-spores isolates used were coming from the two genetic groups of powdery mildew biotype B (S7) and biotype A (PVR38) (Delye et al., 1997).

Field or greenhouse assessments Whereas this year was exceptionally favorable for downy mildew, there was almost no powdery mildew observed in the field except few colonies on Chardonnay. Then for powdery mildew, results of bioassays were compared to previous greenhouse assessments performed on cuttings under natural infection and based on OIV scale 455 (IPGRI, 1997). For downy mildew, disease was assessed in the vineyard on the 30th of June. Each vine was characterised for its susceptibility to DM by a visual assessment of severity (% of area diseased) on leaves and on bunches. Severity data of every genotype on leaves corresponds to an average of the scores of four vines whereas severity on bunches results on the average of severity scores of all bunches of four vines. Genotypes were also scored for black-rot disease with the same direct severity assessment. At beginning of September, genotypes were lastly observed on bunches and classified as “harvestable” or not. The genotypes called “harvestable” could be visually classified 5 on the OIV 453 scale (20 to 30% of bunches attacked with distinct consequences for the vintage).

Results and discussion

Resistance to powdery mildew Among the 38 genotypes, 17 showed a level of resistance as strong as that of the resistant parent 3082-1-42 with no or nearly no germinated spores or sporulation. 12 genotypes had a level of sporulation not significantly different from the other parent Regent and 9 were as susceptible as the control varieties (Cabernet-Sauvignon, Merlot and Chardonnay) (Fig. 1). Regent, the partially resistant parent was surprisingly not very resistant according to the OIV score. Sporulation, was however decreased up to 40% compared to its respective Cabernet- Sauvignon control (cutting leaves), a response closed to that of Chambourcin. From the 16 most resistant genotypes all were previously scored resistant in greenhouse (R. Eibach), however 8 other genotypes scored resistant in Germany had an intermediate or fully susceptible profile in the in vitro test. The hypothesis that German isolates may have revealed resistance genes other than Run1 and Rpv1 is not likely with European isolates. The most likely hypothesis is that the in vitro test, by controlling the inoculation at a susceptible phenological stage, avoids most misclassifications of resistant genotypes. In the test 2, the 17 resistant genotypes were again totally resistant to both tested isolates, with no significant differences between isolates on the control varieties.

Height R I S

0 200000 400000 600000 800000

10 6 69 175 177 65 151 147 46 29 143 78 129 126 38 174 35 95 63 42 70 74 13 58 97 59 85 176 184 3.1 170 111 155 114 119 134 163 153 Regent 3082-1-42 Fig. 1. Result of a hierarchical clustering analysis performed on the sporulation of powdery mildew leaf disks for the 38 genotypes and parent of the population 3082-1-42 x Regent, differentiating three groups of resistant (R), intermediate (I) and susceptible (S) varieties with isolate S7 (test 1).

Resistance to downy mildew in bioassay No significant difference of sporulation was observed between the three isolates for the different genotypes and controls (Anova, F2,89=0.357, P=0.7045) (Figure 2). A good correlation is observed between the score for resistance according to the OIV scale and the sporulation amount (R²=0.79). Based on the sporulation of the 6 repetitions, 11 genotypes could be distinct for their low level of sporulation (< 14413) with an average OIV score of 5.97 and were closed to the resistant parent 3082-1-42 and to Chambourcin (Figure 3). 9 genotypes were characterised by higher level of sporulation and an average OIV of 4.18. Regent was highly susceptible with an OIV score of 3, and similar to the control varieties Cabernet-Sauvignon and Merlot.

140000 120000 100000 80000

60000 40000 20000 Number of sporangia/disk 0 Mer M70 M134 Isolates Fig. 2: Variability of sporulation (number of sporangia per disk) for the different genotypes and control varieties for the three isolates of downy mildew

Some discrepancies were observed between these results and those previously obtained at INRA- Colmar, with for example genotypes resistant at INRA-Bordeaux (OIV > 4.5, sporulation ~11500) (70-153) giving at Colmar either intermediate level of resistance (OIV = 4 for

genotypes 70 and 153) or a fully susceptibility (OIV = 3, for genotype 35). The reverse was also observed with 4 genotypes (58-74-95) expressing an intermediate level of resistance at Bordeaux (OIV < 4.5, sporulation ~ 28450) and a resistant level at Colmar (OIV > 5). These differences are probably due to variation of the expression of the resistance, however we cannot exclude the hypothesis that the diversity of isolates used allowed to reveal minor resistance genes or to reveal virulence on these genes.

A B 140000 120000 OIV >= 4.5 k 3 < OIV< 4.5 OIV <=3 100000

80000 Regent

60000

40000 20000 Height Average sporulation on leaf dis 0 0 100000 200000 300000 t r 2 4 0 3 3 u 9 8 4 e 1 2 4 0 5 7 3 5 9 8 4 3 5 6 1 6 1 L A n

. e 74 95 4 o 4 1 3 7 7 3 9 5 1 6 1 8 6 5 5 7 6 7 9 1 T 1 - B S S e 3 H i 63 1 1 1 1 b 1 1 1 M S S S 1 g C C

C 174 - V 59

C 58 85 C e C C

35 2 97 13 163 R 70 119 168 8 14 3.1 42 134

0 153 Genotypes 170 3

3082.1.42 Fig. 3. Downy mildew sporulation for the different genotypes and control varieties with their corresponding OIV score (A) and classification of the different genotypes based on the sporulation of the six repetitions (B) (3 isolates x 1 leaves) .

Resistance to downy mildew in the field. The correlation between the resistance on leaf disk (OIV score) and the leaf severity in the field is not so strong (R²=0.54). It is not as surprising as the leaf severity in the field is more complex as it results of several cycles of natural infections with probably a mixture of primary and secondary inoculum. However, the in vitro test can be used with a reasonable likelihood to select genotypes potentially resistant on leaves in the field. If we consider as satisfying (resistant) a severity in the field with less than 20% (for untreated vines), a threshold of more than 3 on the OIV scale allowed to select 92% of the resistant genotypes in the field with 30% of misclassified (individuals that would be susceptible in the field) (accuracy of a Response Operating Curve test =0.82). Our threshold of acceptable damage on bunches was quite high considering the high pressure of downy mildew. Genotypes were classified as harvestable or non harvestable. Harvestable genotypes had an average severity at the end of June of 20% with a max of 50% for some of them (Incidence of 100%). All 19 harvestable genotypes (13,168,97,119,95,3- 1,59,74,14,134,58,85,153,163,42,170,63,35,70) had a very low level of severity on leaves at the end of June (Fig. 4). Some genotypes with a moderate severity on bunch at this date but a high level of severity on leaf were not harvestable in September (175, 176, 114, 65, 38). All the harvestable genotypes fitted exactly with those highly resistant to powdery mildew (except 168 and 14 which were not tested against powdery mildew). Both parents, 3082-1-42 and Regent, as well as Villard-blanc and Chambourcin were also harvestable. Except Regent and genotype 63, all harvestable genotypes showed in the bioassay test an OIV > 3 with an average of 5.3 (Fig. 5). Therefore, the bioassay on leaf disk can be used to select genotypes resistant on bunches with an accuracy of 0.76. A threshold > 5 on the OIV scale allowed to

select 62% of the genotypes resistant on bunches among the resistant genotypes with 13% of misclassified individuals. However, in some cases harvestable genotypes can be susceptible to downy mildew on leaves in the bioassay whereas non harvestable genotypes can be resistant. This raises the question of the expression of the resistance genes in bunches. Some of the harvestable genotypes were also highly susceptible to other diseases like botrytis (63, 70, 134, 97, 3-1) or black rot (13).

A B 14 120 unharvestable unharvestable

e harvestable 12 harvestable 100 10

s 80

8 60 6

Number of genotype of Number 40 4

20 2 DownyJun mildew of on bunches (30th severity

0 0 [0;1] ]1;3] ]3:5] ]5;7] [0;1] ]1;3] ]3:5] ]5;7] OIV score on leaf disk OIV score on leaf disk

Fig. 5: Distribution of genotypes (A) and average severity on bunches at the end of June (B) function of their OIV scoring on leaf disk.

100 Resistance in the field at the end of June 90 Severity on bunches (colored histogram identified the genotypes with harvestable 80 grape at the end of September) Severity on leaves 70

Severity (%) 40 30

20 10

t L L 8 9 5 4 3 3 6 4 4 2 0 5 1 4 1 7 6 3 7 2 9 3 7 5 1 9 4 4 8 5 2 3 5 5 8 0 6 0 9 8 9 L R U U U U n - 4 1 B 6 B 9 1 9 5 7 7 1 3 5 8 5 6 7 4 1 8 7 6 6 3 3 7 4 1 5 6 5 7 7 1 4 2 2 4 7 3 2 R I I A - e 3 O O O O 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 E g 1 V V H B B B B - e

M C 2 C C C C R 8

0 Genotypes 3

Fig. 4: Average severity on bunches and on leaves in the field at the 29th of June for the 38 genotypes and control varieties. Genotypes considered as harvestable are coloured in black (bars indicate the standard deviation of the mean).

Conclusion

As expected, none of the varieties were totally resistant to DM. Even the genotypes considered as “harvestable” presented this year a high level of damage on bunches. Such levels of damage must be replaced in the context of an outstanding year in term of mildew pressure in Bordeaux and these resistant genotypes could be satisfactory in a context of IPM with fewer treatments at the right time. The low susceptibility of the leaves to DM and PM may be important to delay the epidemics when primary infection is high and to limit the multiplication of inoculum at the end of the season and consequently the damage on bunches. This was evident for some of the varieties with a moderate level of disease on bunch and a high severity on leaf at the end of June that were totally destroy in September. There is however no evidence to conclude that the resistance genes expressed in leaves are also expressed in bunches. Indeed, the field trial shows that some genotypes having a high leaf resistance in bioassay and a moderate leaf resistance in the field were not all harvestable. Final damage depends on the adequacy between the amount of inoculum and the period of bunches susceptibility. As genotypes can be very different in term of precocity, differences in resistance could also be due to delay between the epidemic on leaves and the period of susceptibility of bunches. Further analyses of the phenology and mapping resistance genes data are needed to answer this question. Even, if there is no correlation between the OIV measure of resistance on leaf disk and the disease assessment of severity on bunches in the field, the bioassay can be used to select field resistant genotypes, with a good accuracy, by using a threshold of 5 on the OIV resistance scale. The link between resistance to powdery mildew and downy mildew suggest that resistance in harvestable genotypes is coming from Muscadinia (Rpv1 and Run1). This will be confirmed by QTL analysis of the genotypes. A better understanding of the resistance of Regent would be necessary to understand its low level of resistance in bioassay compare to that in the field and to understand its contribution to genotypes having major resistance genes as Run1 and Rpv1. Field disease assessment for DM but also for the other diseases such as botrytis and black-rot may help us to identify and/or to precise the role of QTL of resistance on other diseases. Whatever the mechanism of resistance is, these resistant varieties should be tested on larger plot scale and on longer period of time to measure the ability of such strong leaves resistance to improve the control of epidemics and to be durable against new pathotypes.

Acknowledgements

We thank R. Eibach (BAZ-Institute for grapevine breeding Geilweilerhof, Germany) for its contribution in the selection program and greenhouse tests.

References

Delye, C., F. Laigret, and M.F. Corio-Costet 1997. RAPD analysis provides insight into the biology and epidemiology of Uncinula necator. – Phytopathology 87:670-677. Fischer, B.M., Salakhutdinov, I., Akkurt, M. 2004. Quantitative trait locus analysis of fungal disease resistance factors on a molecular map of grapevine. – Theoretical Applied Genetic, 108: 501-515. IPGRI, UPOV, OIV 1997. Descriptors for Grapevine (Vitis spp.). – In : International Union for the Protection of New Varieties of Plants, Geneva, Switzerland/Office International de la vigne et du Vin, Paris, France/ International Plant Genetic Resources Institute, Rome, Italy.

Pauquet, J., Bouquet, A., This, P., Adam-Blondon, A.-F. 2001. Establishment of a local map of AFLP markers around the powdery mildew resistance gene Run1 in a grapevine and assessment of their usefulness for marker assisted selection. – Theoretical Applied genetic, 103: 1201-1210. Barker, C.L., Donald, T., Pauquet, J., Ratnaparkhe, M.B., Bouquet, A., Adam-Blondon, A.-F., Thomas, M.R., Dry, I. 2005. Genetic and physical mapping of grapevine powdery mildew resistance gene, Run1, using a bacterial artificial chromosome library. Theoretical Applied genetic, 111:370-377 Merdinoglu, D., Wiedemann-Merdinoglu, S., Coste, P., Dumas, V., Haetty, S., Butterlin, G., Greif, C., Adam-Blondon, A.-F., Bouquet, A., Pauquet, J. (2003). Genetic analysis of downy mildew resistance derived from Muscadinia rotundifolia. – Acta Horticulturae 603: 451-456.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 53-60

Detection of 16SrXII-A phytoplasma in insects collected in vineyards of South Italy

V. Cavalieri1, V. D’Urso2, L. Ferretti3,4, C. Rapisarda1 1 Dipartimento di Scienze e Tecnologie Fitosanitarie, Università degli Studi di Catania, via Santa Sofia 100, I 95124 Catania 2 Dipartimento di Biologia Animale “M. La Greca”, Università degli Studi di Catania, via Androne 81, I 95124 Catania 3 C.R.A. – Centro di Ricerca per la Patologia Vegetale, via C. G. Bertero 22, I 00156 Roma 4 Dipartimento di Gestione dei Sistemi Agrari e Forestali, Università degli Studi Mediterranea di Reggio Calabria, piazza San Francesco di Sales 4, 89061 Gallina di Reggio Calabria (RC)

Abstract: Bois noir (BN) disease is a grapevine yellow associated to Stolbur phytoplasma, 16SrXII-A phylogenetic subgroup, which is responsible of severe crop losses in all grape growing areas in Italy. This phytoplasma is transmitted by the planthopper Hyalesthes obsoletus Signoret (Hemiptera Cixiidae), but the role of other Auchenorrhyncha species has been recently suspected. Studies on BN vectors have been carried out in Italy and results of surveys realised during 2004- 2007 in several vineyards located in Southern Italian regions (Calabria and Sicily) are reported in this work. Auchenorrhyncha specimens have been collected by means of entomological net, sampling on grapevine canopy and also on weeds growing both along the grape rows and on the border of vineyards. Many species of Auchenorrhyncha were detected and data on their distribution and abundance are given in this paper. In particular, Toya propinqua Fieber (Delphacidae) and Exitianus capicola Stål. (Cicadellidae) showed to be the most frequent species in the investigated vineyards. Moreover, some of their specimens resulted at molecular analysis to be positive to the16SrXII-A phytoplasma.

Key words: Bois noir, phytoplasma, vineyards, Auchenorrhyncha, vectors.

Introduction

Phytoplasmas are agents of several hundred plant diseases, including grapevine yellows (GY), which are transmitted by Hemiptera insects belonging to both Auchenorrhyncha (Cicadellidae, Cixiidae, Delphacidae, Derbidae) and Sternorrhyncha (Psyllidae). The main phytoplasma diseases affecting grapevine in Italy are Flavescence dorée (FD) and Bois noir (BN). FD is associated to “Candidatus Phytoplasma vitis” (elm yellow group, 16SrV-C and V- D), transmitted by the leafhopper Scaphoideius titanus Ball (Hemiptera Cicadellidae). The high vector specificity of this phytoplasma and the strictly ampelophagous behaviour of S. titanus makes FD the most dangerous GY disease, especially because of its rapid diffusion into the vineyards. Nevertheless, the compulsory eradication of FD infected plants and the control of S. titanus have reduced the epidemic incidence of this grapevine disease, which is mainly restricted to the northern grapevine growing areas of Italy. In contrast to FD, Bois noir is showing an increasing epidemiological and economic significance. The disease is associated to 16SrXII-A phytoplasma (Stolbur group) and is

53 54

widespread in all grapevine growing areas of Italy (Borgo et al., 2005). Outbreaks of BN have been recently reported in different northern and central Italian regions (Mori et al., 2005; Bertaccini et al., 2006; Pasquini et al., 2007; Quaglino et al., 2006; Romanazzi et al., 2007). Bois noir is usually considered to be less dangerous than FD, mainly because of the absence of a strictly ampelophagous insect vector. Nevertheless, its symptoms are identical to those ones induced by Flavescence dorée and crop losses due to the plants decline and the shrivelling of berries can be remarkable. The planthopper Hyalesthes obsoletus Signoret (Hemiptera Cixiidae) is the main vector of 16SrXII-A phytoplasma (Maixner, 1994; Sforza et al., 1998; Alma et al., 2002), yet the incidence of BN in many grapevine growing areas often appears not to be clearly linked with the occurrence and population density of this polyphagous insect vector, which is only occasionally present on grape; thus, the role of other Auchenorrhyncha species has been recently suspected. Other cixiid and cicadellid species are known to be able to acquire the 16SrXII-A phytoplasma, such as Goniagnathus guttulinervis (Garau et al., 2004), Reptalus panzeri (Palermo et al., 2004; Botti et al., 2005) and R. quinquecostatus (Trivellone et al., 2005) Thamnotettix zelleri Kirschbaum and Anoplotettix putoni Ribaut (Bagnoli et al., 2005), but their ability to transmit the phytoplasma is still not proved. In this work the results are reported of molecular investigations carried out on insects collected in infected vineyards of Calabria and Sicily, in order to clarify the role of Auchenorrhyncha species in the epidemiology of BN disease in these geographical areas.

Materials and methods

Auchenorrhyncha specimens were collected during the years 2005-2007 in 9 different vineyards where Bois noir disease has been previously detected; 6 of these vineyards have been selected in Sicily [in Caltagirone (CT), Cerda (PA), Marsala (TP), Mazara del Vallo (TP), Monreale (PA), Vicari (PA)] and 3 in Calabria [in Bianco (RC), Cirò (KR), Tarsia (CS)]. Collections have been realised on grapes of the cultivars Cabernet Sauvignon, Chardonnay, Gaglioppo, Greco Merlot, Nero d’Avola, Nerello Siciliano. On collected material, total DNA was extracted from single hoppers using protocol modified by Doyle & Doyle (1990) and adapted by Marzachì et al. (1998). The universal primers R16F2/R2 (Lee et al., 1993) were used in direct PCR assay. PCR products were diluted 1:40 and used as template in nested amplification with primers R16(I)F1/R1 specific for 16SrXII-phytoplasmas (Lee et al., 1994). One microliter of DNA was added to a reaction mixture containg: 1X PCR-buffer (Invitrogen), 1.5 mM MgCl2 (Invitrogen), 200 µM dNTPs (Promega), 0.5 µM of each primer, 0.5 U of Taq Polymerase recombinant (Invitrogen). The reaction mixture was subjected to the following temperature cycle: an initial denaturation at 94 °C for 5 min followed by 35 cicles with following conditions: 94 °C for 1 min, 50 °C for 1.5 min (1 min for R16(I)F1/R1 primers), 72 °C for 1.5 min (1.25 min for R16(I)F1/R1 primers), and a final extensions of 10 min at 72 °C terminating amplification. Reactions and cycling conditions were conducted in a authomated thermal cycler [GeneAmp® PCR System 2700 (Applied Biosystems)]. DNA of positive insects was subjected to direct amplification with primers fTufAY/rTufAY (Scheneider et al., 1997) and nested amplification with primers TufAYf2/TufAYr2 (Pasquini et al., 2006), in order to characterize the strains of Stolbur phytoplasma. Products from nested PCR were digested with restriction enzyme HpaII.

Table 1. List of Auchenorrhyncha species collected in Calabria and Sicily during the years 2005-2007.

N. Provenance (n. specimens per locality) Family Species Specime Sicily Sicily Sicily Calabri Calabria ns 2005 2006 2007 a 2006 2007 Cercopidae Neophilaenus lineatus 1 1 Philaenus spumarius 3 2 1 Philaenus sp. 8 1 7 Cicadellidae Anaceratagallia laevis 13 1 3 9 Anaceratagallia ribauti 1 1 Anaceratagallia sp. 8 5 2 1 Austroagallia sinuata 41 25 3 3 9 1 Balclutha frontalis 1 1 Balclutha gr. rhenana 3 3 Balclutha sp. 1 1 Cicadella viridis 42 2 40 Cicadula lineatopunctata 1 1 Cicadulina bipunctata 1 1 Conosanus obsoletus 1 1 Doratulia ragusai 2 2 Empoasca decipiens 6 5 1 Empoasca vitis 11 11 Empoasca sp. 75 64 4 7 Eupelix cuspidata 1 1 Euscelidius variegatus 2 2 Euscelis lineolatus 9 2 3 2 2 Euscelis sp. 2 1 1 Exitianus capicola 131 36 23 12 27 33 Fieberiella florii 1 1 Goniagnathus 1 1 guttulinervis Hauptidia provincialis 1 1 Hauptidia sp. 1 1 Hecalus glaucescens 12 2 8 1 1 Iassus sp. 1 1 Macropsis sp. 1 1 Macrosteles 2 2 quadripunctulatus Macrosteles viridigriseus 2 2 Macrosteles sp. 4 1 3 Neoaliturus fenestratus 4 2 2 Neoaliturus haematoceps 1 1 Opsius stactogalus 1 1 Psammotettix alienus 41 10 13 18 Psammotettix striatus 10 2 8

N. Provenance (n. specimens per locality) Family Species Specime Sicily Sicily Sicily Calabri Calabria ns 2005 2006 2007 a 2006 2007 Cicadellidae Psammotettix sp. 45 11 4 23 7 Recilia schmidtgeni 3 1 2 Synophropsis lauri 1 1 Zygina rhamni 17 17 Zyginidia lineata 50 50 Zyginidia serpentina 19 8 9 2 Zyginidia sp. 6 6 Cixiidae Hyalesthes luteipes 14 14 Hyalesthes obsoletus 7 2 5 Reptalus panzeri 5 4 1 Delphacidae Chloriona sicula 1 1 Laodelphax striatellus 50 45 2 3 Toya propinqua 113 7 11 5 80 10 Dictyopharidae Dictyophara europaea 12 2 3 5 2 Issidae Agalmatium flavescens 1 1 Total 791 312 78 48 221 132

Results and discussion

A total of 791 adult specimens belonging to 56 species and 6 families of Auchenorrhyncha (Cercopidae, Cicadellidae, Cixiidae, Delphacidae, Dictyopharidae, Issidae) were collected and identified (tab. 1). In particular, Exitianus capicola Stål. (Cicadellidae) and Toya propinqua Fieber (Delphacidae) resulted the most abundant species in the investigated vineyards; moreover, together with Austroagallia sinuata Mulsant & Rey (Cicadellidae), they have been the only species occurring in all investigated areas during the whole period of the study. Among Cicadellidae, significant presence has been shown, by Cicadella viridis Linnaeus (though not found in the Sicilian investigated vineyards), Empoasca spp. (especially vitis Göthe), Psammotettix spp. (especially alienus Dahlbom and striatus Linnaeus), Zygina spp. and Zyginidia spp.. Among Cixiidae, Hyalestes obsoletus Signoret and Reptalus panzeri Löw have been collected only in Calabria, while, among Dictyopharidae, Dictyophara europaea Linneus has been collected in nearly all investigated areas, though in very low numbers. Molecular analysis applied to all collected insects showed the positivity to a phytoplasma in 76 out of 791 examined specimens (tab. 2), of which 58 collected in Calabria and 18 in Sicily. The presence of 16SrXII-A phytoplasma was detected in the cixiidae R. panzeri (VK- Type II strain) as well as in C. viridis, E. capicola (fig. 1) and T. propinqua. Molecular characterization of Stolbur strain conducted with Tuf gene showed the presence of VK-Type II strain on R. panzeri and C. viridis collected in Calabrian vineyards. In addition to Stolbur phytoplasma, also 16SrI-B was found in several insects: Austro- agallia sinuata, E. capicola and Psammotettix alienus. Both phytoplasmas have not been found in mixed infections in analyzed insects. According to the literature, R. panzeri has been already found positive to stolbur phytoplasma in Hungary (Palermo et al., 2004) and in the Italian region Emilia Romagna (Botti et al., 2005); recently, Serbian researchers demonstrated that it can transmit stolbur

phytoplasma in maize (Jović et al., 2007); its presence in different Italian vineyards where BN disease occurs (Granata & Russo, 1990; Albanese et al., 1997; Mazzoni et al., 2002; Botti et al., 2005; Trivellone et al., 2005) candidates this species as possible vector of this GY disease. T. propinqua is widespread worldwide and is polyphagous on different grasses. In Italy it has been indicated as vector of Cynodon chlorotic streak virus; in various laboratory experiments it also transmitted Maize rought dwarf virus (Harpez, 1972; Raatikainen and Vasarainen 1990; Remes Lenicov et al., 1999). E. capicola [as E. taeniaticeps (Kbm)] was found as new host for the 16SrI-B, 16SrV-A, 16SrX-A and C phytoplasma groups in Sardinia (Garau et al., 2005; Prota et al., 2006). C. viridis was found positive to 16SrI-C in Tuscany (Braccini et al., 2002) and positive also to 16SrXII-A in Croatian vineyards (Mikec et al., 2006). All data from Sicily and Calabria reported in this paper seem to confirm the role of various Auchenorrhyncha species as hosts of phytoplasmas in vineyard agrosystems. In order to better understand their role in transmission and spreading of grapevine yellow diseases, it is greatly important to improve our knowledge on the biology, ecology and diffusion of these species.

12347 568910SNCM

Fig. 1. Agarose gel electrophoresis of R16F1(I)/R1 amplicons (1100 bp) from infected (3, 6, 7, 8, 9, 10) and non-infected (1, 2, 4, 5) E. capicola. S: phytoplasma reference strain (Stolbur), NC: negative control, M: 200 bp ladder (Promega).

Acknowledgements

We thank Dr Cristina Marzachì and Dr Davide Pacifico (Istituto di Virologia Vegetale, Consiglio Nazionale delle Ricerche, Torino) for molecular characterization of the Sicilan material. This study has been realized in the frame of the research project “Grapevine yellows: a limitation factor of grapevine productions – GIAVI”, financed by the Italian Ministry of Agriculture (MiPAF).

Table 2. Results of nested assay detecting stolbur phytoplasmas in hopper from different viticultural regions in Calabria (in 2006-2007) and Sicily (in 2005-2007) (n/n =number of positive insects out of total analyzed insects; n.c.=not characterized).

Results Nested Direct Family Species Locality Cultivar PCR Phytoplasma PCR R16(I) R16F2/R2 F1/R1 Cercopidae Philaenus spumarius Caltagirone Chardonnay - + (1/3) n.c. Chardonnay, Cicadellidae Anaceratagallia laevis Cirò, Vicari - + (4/13) n.c. Merlot N. siciliano, Bianco, Greco bianco, Anaceratagallia sp. Caltagirone, Cirò, - + (2/8) n.c. Merlot, Mixed Monreale,Tarsia vine Austroagallia sinuata Tarsia mixed vine - + (1/41) 16SrI-B Balclutha sp. Tarsia mixed vine - + (1/1) n.c. Cicadella viridis Tarsia mixed vine - + (4/42) 16SrXII-A Empoasca decipiens Tarsia mixed vine - + (1/6) n.c. N. siciliano, Bianco, Cirò, Greco bianco, Empoasca sp. - + (3/75) n.c. Tarsia Merlot, Mixed vine Euscelis lineolatus Tarsia mixed vine - + (1/9) n.c. Chardonnay, Caltagirone, Cirò, Gaglioppo, + 16SrXII-A; Exitianus capicola - Tarsia Merlot, Mixed (27/131) 16SrI-B vine Hauptidia provincialis Tarsia mixed vine - + (1/1) n.c. Hecalus sp. Caltagirone Chardonnay - + (1/10) n.c. Macrosteles Tarsia mixed vine - + (2/2) n.c. quadripunctulatus N. siciliano, Neoaliturus fenestratus Bianco - + (2/4) n.c. Greco bianco Opsius stactogalus Cirò Merlot - + (1/1) n.c. Gaglioppo, Psammotettix alienus Cirò, Tarsia Merlot, Mixed - + (3/41) 16SrI-B vine N. siciliano, Bianco, Cirò, Greco bianco, Psammotettix sp. - + (3/45) n.c. Tarsia Merlot, Mixed vine Gaglioppo, Cixiidae Hyalesthes obsoletus Cirò, Tarsia Merlot, Mixed - + (4/7) n.c. vine Reptalus panzeri Tarsia Mixed vine + + (1/5) 16SrXII-A Delphacidae Laodelphax striatellus Tarsia Mixed vine - + (2/5) n.c. Chardonnay, Caltagirone, Gaglioppo, Cerda, Cirò, + Toya propinqua Merlot, Mixed - 16SrXII-A Monreale, Vicari, (10/113) vine, Nero Tarsia d'Avola N. siciliano, Bianco, Cirò, Greco bianco, Dictyopharidae Dictyophara europaea - + (1/7) n.c. Tarsia Merlot, Mixed vine

References

Albanese G., D’Urso V., Granata G., Collodoro S. 1997. Individuazione di un fitoplasma in esemplari di Psammotettix striatus catturati in vigneti. – Informatore Fitopatologico 7-8: 57-60. Alma A., Soldi G., Tedeschi R., Marzachì C. 2002. Ruolo di Hyalesthes obsoletus Signoret (Homoptera, Cixiidae) nella trasmissione del legno nero della vite in Italia. – Petria 12 (3): 411-412. Bagnoli B., Pinzauti F., Trivellone V. 2005. Indagine preliminare sugli auchenorrinchi potenziali vettori di stolbur in un’area viticola del Lazio. – Petria (1/2): 55-58. Bertaccini A., Paltrinieri S., Botti S., Duduk B., Fiore N., Kolber M., Škorić D., Torres E., Conti M. 2006. Diversity of 16SrXII phytoplasmas detected in grapevine growing areas worldwide – 15th Meeting of the International Council for the Study of Virus and Virus- like Diseases of the Grapevine (ICGV), Stellenbosch, South Africa, 3-7 April 2006: 88- 89. Borgo M., Angelici E., Filippin L., Botti S., Marzachì C., Casati P., Quaglino F., Zorloni A., Albanese G., La Rosa R., Tessitori M., Pasquini G., Bertaccini A. 2005. Monitoraggio dei giallumi della vite e caratterizzazione dei fitoplasmi nell’ambito del Progetto Finalizzato ‘GIA.VI’ nel 2004. – Petria 15 (1/2): 161-164. Botti S., Paltrinieri S., Mori N., Milanesi L., Bondavalli R., Bertaccini A. 2005. Variabilità molecolare di fitoplasmi 16SrXII in vigneti delle province di Modena e Reggio Emilia. – Petria 15 (1/2): 121-124. Braccini P., Sfalanga A., Botti S., Malossi A., Bertaccini A. 2002. Prime osservazioni su indagini epidemiologiche in alcuni vigneti della Toscana centrale colpiti da legno nero. – II Incontro Nazionale sulle Malattie da Fitoplasmi, Roma, 3-4 ottobre 2006. Doyle J.J.and Doyle J.L. 1990. Isolation of plant DNA from fresh tissue. – Focus 12: 13-15. Garau R., Sechi A., Tolu G., Prota V.A., Lentini A., Prota U. 2004. Goniagnathus guttulinervis (Kirschbaum), new natural host of the Stolbur subgroup 16SrXII-A phytoplasma in Sardinia. – Journal of Plant Pathology 86 (2): 179. Garau R., Prota V.A., Paltrinieri S., Tolu G., Sechi A., Bertaccini A. 2005. Atti della Terza Giornata di studio sul Mirto. – Sassari, 23 settembre 2005: 51-58. Granata G and Russo A. 1990. Indagini su un giallume epidemico simile alla «Flavescenza dorata». – Vignevini 5: 69-71. Harpaz I. 1972. Maize Rought Dwarf. A planthopper virus disease affecting maize, rice, small grain and grasses. – Israel University Press, Jerusalem. Jović J., Cvrković T., Mitrović M., Krnjajić S., Redinbaugh M.G., Pratt R.C., Gingery R.E., Hogenhout S.A., Toševski I. 2007. Roles of stolbur phytoplasma and Reptalus panzeri (Cixiidae, Auchenorrhyncha) in the epidemiology of Maize redness in Serbia. – European Journal of Plant Pathology 118: 85-89. Lee I.M., Hammond R.W., Davis R.E., Gundersen D.E. 1993. Universal amplification and analysis of pathogen 16S rDNA for classification and identification of mycoplasmalike organisms. – Phytopathology 83: 834-842. Lee I.M., Gundersen D.E. Hammond R.W., Davis R.E. 1994. Use of mycoplasmalike organism (MLO) group-specific oligonucleotide primers for nested-PCR assays to detect mixed-MLO infections in a single host plant. – Phytopathology 84: 559-566. Maixner M. 1994. Transmission of German grapevine yellows (Vergilbungskrankheit) by the planthopper Hyalesthes obsoletus (Auchenorrhyncha: Cixiidae). – Vitis 33: 103-104. Marzachì C., Verratti F., Bosco D. 1998. Direct PCR detection of phytoplasmas in experimentally infected insects. – Annals of Applied Biology 133: 45-54.

Mazzoni V., Lucchi A. and Santini L. 2002. Indagine faunistica sugli auchenorrinchi di vigneti liguri e toscani (Rhynchota Homoptera). – Frustula Entomologica XXV (XXXVIII): 181-194. Mikec I., Križanac I., Budinščak Ž., Musić M.S., Krajačić M., Škorić D. 2006. Phytoplasmas and their potential vectors in vineyards of indigenous Croatian varities. – 15th Meeting of the International Council for the Study of Virus and Virus-like Diseases of the Grapevine (ICGV), Stellenbosch, South Africa, 3-7 April 2006: 255-257. Mori N., Milanesi L., Bondavalli R., Botti S. 2005. Prove di contenimento del Legno nero della vite. – Petria 15 (1/2): 137-138. Palermo S., Elekes M., Botti S., Ember I., Alma A., Orosz A., Bertaccini A., Kolber M. 2004. Presence of stolbur phytoplasma in Cixiidae in Hungarian vineyards. – Vitis 43 (4): 201-203. Pasquini G., Ferretti L., Albanese G., Bagnoli B., Pinzauti F., Barba M. 2006. Geographical distribution of Stolbur isolates in vineyards of Central and Southern Italy. – 15th Meeting of the International Council for the Study of Virus and Virus-like Diseases of the Grapevine (ICGV), Stellenbosch, 3-7 April 2006: 103-104. Pasquini G., Ferretti L., Barba M. 2007. Diffusione del Legno nero della vite nel Lazio e caratterizzazione molecolare dell’agente eziologico. – Informatore Fitopatologico 4: 42- 47. Prota V. A., Sechi A., Tolu A., Garau R., Botti S. and Bertaccini A. 2006. New finding on phytoplasmas-affected Auchenorrhyncha populations in Sardinia vineyards. – 2006 ISEIMA. First International Symposium on Environmental Identities and Mediterranean Area: 622-624. Quaglino F., Casati P., Zorloni A., Duranre G., Bianco P.A. 2006. Molecular characterization of phytoplasmas associated with grapevine yellows in northern Italy. – 15th Meeting of the International Council for the Study of Virus and Virus-like Diseases of the Grapevine (ICGV), Stellenbosch, South Africa, 3-7 April 2006: 90-92. Raatikainen M. and Vasarainen 1990. Metadelphax propinqua (Fieber) (Homoptera- Delphacidae). – Entomologica Fennica 1: 145-149. Remes Lenicov A.M.M. and Virla E.G. 1999. Homópteros vectores de interés fitosanitario: un problema creciente en la Argentina (Homoptera: Auchenorrhyncha). – Revista de la Sociedad Entomológica Argentina 58 (1-2): 43-47. Romanazzi G., Murolo S., Terlizzi F., Talevi S., Stimilli G., Savino V. 2007. Fitoplasmi associati ai giallumi della vite nelle Marche. – Informatore Fitopatologico 4: 48-50. Schneider B., Gibb K.S., Seemüller E. 1997. Sequence and RFLP analysis of the elongation factor Tu gene used in differentiation and classification of phytoplasmas. – Micro- biology 143: 3381-3389. Sforza R., Clair D., Dair X., Larrue J., Boudon-Padieu E. 1998. The role of Hyalesthes obsoletus (Hemiptera: Cixiidae) in the occurrence of Bois noir of grapevine in France. – Journal of Phytopathology 146: 549-556. Trivellone V., Pinzauti F., Bagnoli B. 2005. Reptalus quinquecostalus (Dufour) (Auchenor- rhyncha Cixiidae) as a possibile vector of Stolbur-phytoplasma in a vineyard in Tuscany. – Redia 88: 103-108.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 61-68

Tolerance of Tunisian Grapevine to Uncinula necator

Chebil S., Zeghonda N., Jallouli N., Lasram S., Zemni H., Ghorbel A. et A. Mliki Laboratoire Physiologie Moléculaire de la Vigne – Centre de Biotechnologie Borj Cedria – BP.901-2050 Hammam-lif – Tunisia

Abstract: Powdery mildew, caused by Uncinula necator, is the main fungus disease of grapevine in Tunisia. Chemical control, usually expensive, can be sometimes no efficient to protect fungus attack especially before veraison stage. Thus, looking for a new control method such as the cultivation of tolerant varieties, can lead to stop the epidemiological development of the disease in vineyard. In this context, autochthonous grapevines have shown different responses to the powdery mildew in the Tunisian vineyard. Indeed, an assay has been realized testing the sensitivity of more than 30 Tunisian grapevine cultivars (Vitis vinifera L.) to the fungus Uncinula necator. After inoculation of young plants from each variety, we have noticed the appearance of the symptoms on leaves. Some varieties such as “Rafraf-F” behave as tolerant to the disease and others such as “Razegui” seemed to be sensitive. In fact, with “Rafraf-F”, the symptoms took place 7 days later than with “Razegui”. The total protein composition of control and infected leaves of “Rafraf-F” and “Razegui” cultivars was analyzed by two-dimensional electrophoresis. The corresponding patterns revealed the induction of new proteins as well as the repression of certain others proteins.

Keywords: Tunisian grapevine, Uncinula necator, Powdery mildew, tolerance, two-dimensional electrophoresis.

Introduction

Powdery mildew, caused by Uncinula necator, represents the major disease in Tunisian vineyard. Every year the damage can reach more than 20% of the production even after chemical treatments. Many strategies to control the epidemiological expansion of the disease were applied by alternating fungicides during phenological grape stages. Yet, several strains of the fungus resistant to IBS fungicides were detected causing considerable reduction of treatment efficacy. In this case, the search for grapevine varieties tolerant to this disease would be a very judicious solution for this disease. In Tunisia the art of vine producing and wine making dates back to the earliest antiquity. Nowadays, Tunisia has a considerable inheritance of autochthones grapevine varieties which are very appreciated especially for their organoleptic quality (Ben Abdallah, 1999). Some of theses cultivars, generally deserted, have shown a good adaptation to the soil and climate conditions (Askri, 1997). To preserve theses varieties from extinction, our centre have collected and cultivated about fifty cultivars founded in all Tunisian territory. The collection is now used to study the agronomic characteristics of autochthones varieties such as the tolerance to the main grapevine diseases. This work contributes to understand the relationship between Uncinula necator and grapevine by studying the physiological response of two contrasting varieties after inoculation with the fungus.

61 62

Material and methods

Material About fifty Tunisian grapevines are used to study their tolerance to Uncinula necator. All theses cultivars are identified and characterised with molecular markers (Zoghlammi et al., 2001). The list of these varieties is presented in the table 1.

Table 1: Names of Tunisian varieties cultivated in the collection of the Biotechnology Centre

Number Varieties Number Varieties 1 Dalia 26 Saouadi 2 Asli 27 Bahbahi 3 Beldi Rafraf 28 Guelb Sardouk 4 Rafraf-F 29 Meguergueb Djerba 5 Ahmer bou Ahmer 30 Turky 6 Hencha 1 31 El biodh 7 Chaouch 32 Arich Djerba 8 Balta 1 33 BKB Gabes 9 Arbi Abiadh 34 Arich Dressé 10 Sakasly 35 BKB Sfax 11 Dabbouki 36 Bidh Hamem Rafraf 12 Hencha H2 37 Djerbi Dguèche 13 Blanc 3 38 Arich Ahmer 14 Khamri Tozeur 39 Beldi Baddar 15 Marsaoui 40 Bessoul Khadem Rafraf 16 Châaraoui Rafraf 41 Khamri 17 Hamri Kerkennah 42 Tounsi Djerba 18 Muscat Rafraf 43 Khedhiri 1 19 Bidh Hamem Sfax 44 Khedhiri 2 20 Khedhiri 3 45 Blanc 1 21 Akhel Meguergueb 46 Beldi Sayeb INRST 22 Sfaxi S2 Nafta 47 Medina 23 Arbia 48 Mahdaoui 24 Razegui 49 Balta 2 25 Djebbi 50 Kahli Sfax + kahli kerkennah

The Fungus The strain of Uncinula necator used to inoculate the plants is isolated from infected leaves collected in tunisian vineyard.

Plant preparation For each variety, 20 plants are placed in the glass house from which four plants are isolated to constitute the control.

Inoculum preparation Uncinula necator is conserved on grape vine leaves maintained on agar medium. The production of conidia was realised by the contamination of leaves taken from the third stage

of Carignan plants. Leaves are washed by water, disinfected with hypochlorite calcium during 10 minutes and than rinsed with sterilised distilled water. Each leave is placed in Petri dishes containing agar medium with benzimidazol added (30g/l) (Clerjeau, 1998). Contamination of the surface of the leave is realised by the deposit of conidia in dry condition with an air stream made by Pasteur pipette. Conidia were conserved on leaves about 10 -12 days before an other inoculation of new leaves.

Plants inoculation For each plant, infected leave with Uncinula necator was attached to the limb of young leaves. All inoculated plants are covered with plastic bag for about a week to maintain high level of humidity necessary for contamination process. Control plants, isolated in over compartment of the green house, are treated with fungicides (strobulurin – IBS) every week to protect against fungus attack.

Symptoms evaluation The sensitivity of Tunisian grapevine to Uncinula necator was estimated visually by the observation of the percentage of the leave’s area contaminated by the fungus. The scale of notation of the disease severity is the following: 0: now symptom 1: mycelia extension is less than 5% 2: the area contaminated is between 5% and 25% 3: the area contaminated is between 25% and 50% 4: the area contaminated is between 50% and 75% 5: the area contaminated is more than 75%

Protein analysis Proteomic study has concerned only two contrasted varieties: i) Razzegui has shown sensitivity to the disease, ii) Rafraf-F seems to be tolerant to the fungus. Also, the sampling is made in two dates: the first is one week after inoculation and the second is one week after the first. Total proteins were extracted from leaves using Hurkman & Tanaka (1986) method. After that, the measuring of proteins is realised according to Bradford et al. (1976). For protein analysis with 2D-PAGE Electrophoresis, the method used is described by O’Farrel, (1975).

Results and discussion

Sensitivity of Tunisian grapevine varieties to Uncinula necator After two week from the inoculation, the symptom of the powdery mildew was observed on six foliar stages. For each leave, the area contaminated was estimated and than the attack level per plant was using the notation scale. The results presented in table 2 show that the response of Tunisian grapevine to Uncinula necator is variable. In fact, about 66% of the cultivars were very sensitive to the fungus; in this case the symptoms began to develop after one week. At the notation day, nearly the totality of the leaves was contaminated. In the other hand, Uncinula necator has found a little difficulty to develop in tissue leaves of some varieties. In this case the symptoms appears after 10 days from inoculation and the severity of the disease is low compared to sensitive varieties (contaminated areas are less than 25%). Finally, we find from the fifty varieties, a very interesting one. It was collected from Raf- raf region in the north east of Tunisia, known from the antiquity to cultivate grapevine. This

specimen was selected by Hamadi HLEL, a grapevine grower for forty years. Even, under the best conditions to fungus development, the area of leaves contaminated by the fungus is less than 5%. This important result leads us to see the physiological modification in this cultivar after the inoculation.

Table 2: Sensitivity of Tunisian grapevine to Uncinula necator after two weeks from inoculation

Infection Infection Number varieties Number varieties degree degree 1 Dalia 3 26 Saouadi 3 2 Asli 3 27 Bahbahi 4 3 Beldi Rafraf 4 28 Guelb Sardouk 3 Meguergueb 4 Raf Raf-F 1 29 4 Djerba 5 Ahmer bou Ahmer 3 30 Turky 3 6 Hencha 1 3 31 El biodh 4 7 Chaouch 3 32 Arich Djerba 4 8 Balta 1 4 33 BKB Gabes 4 9 Arbi Abiadh 3 34 Arich Dressé 4 10 Sakasly 2 35 BKB Sfax 4 Bidh Hamem 11 Dabbouki 2 36 3 Rafraf 12 Hencha H2 2 37 Djerbi Dguech 4 13 Blanc 3 3 38 Arich Ahmer 4 14 Khamri Tozeur 3 39 Beldi Baddar 3 15 Marsaoui 3 40 Bezzoul Khadem 4 16 Châaraoui Rafraf 4 41 Khamri 4 17 Hamri Kerkennah 2 42 Tounsi Djerba 3 18 Muscat Rafraf 3 43 Khedhiri 1 3 19 Bidh Hamem Sfax 4 44 Khedhiri 2 2 20 Khedhiri 3 3 45 Blanc 1 2 Beldi Sayeb 21 Akhel Meguergueb 2 46 2 INRST 22 Sfaxi S2 Nafta 2 47 Medina 2 23 Arbia 2 48 Mahdaoui 2 24 Razzegui 4 49 Balta 2 3 25 Djebbi 3 50 Kahli kerkennah 3

Table 3: Proteins concentration in grapes leaves of the contrasted varieties

Total proteins

Sampling date (mg/g leaves) Control Rafraf-F Razzegui After one week from inoculation 1,66 2,52 3,21 After two weeks from inoculation 1,72 2,59 2,23

Protein analysis of two contrasted varieties: Razzegui and Rafraf-F The concentration of total proteins extracted from leaves of control and inoculated plants was evaluated and presented in table 3. The result shows that proteins concentration in inoculated plants is higher than the control plants. Also, Razzegui variety seems to have the highest amount of proteins after one week. This concentration decreases and becomes less than Rafraf-F variety. For electrophoresis 2D-PAGE analysis, the protein profiles of the infected leaves are presented in figure 1. The result shows, in Rafraf-F leaves, the induction of protein presented by spots 1, 2, 3, 4, 5, 13 and 14 with respectively a molecular weight: 38,5 kilo Dalton, 37,5 kilo Dalton, 34,5 kilo Dalton, 34,5 kilo Dalton, 34,5 kilo Dalton, 20,1 kilo Dalton and 19,4 kilo Dalton corresponding to isoelectric points: 5,67 ; 6,1 ; 6,53 ; 6,74 ; 6,84 ; 6,74 et 6,89. Also, spots 4 and 5 were weakly detected in control leaves but spots 1, 2,3,13 and 14 were observed only in inoculated leaves. These results show the synthesis of new proteins after fungus contamination. On the other hand, spots 6, 7 and 8 with the same molecular weight of 30 kilo Dalton corresponding to respectively isoelectric points 5,88, 5,99 and 6,1, are absent in inoculated leaves. In this case, it seems that some proteins are suppressed after the penetration of the fungus in plant tissues. In Razzegui variety, spots detected are: 9, 10, 11, 12, 13 and 14 with respectively 26 kilo Dalton, 21,5 kilo Dalton, 19,4 kilo Dalton, 19 kilo Dalton, 20,1 kilo Dalton and 19,4 kilo Dalton corresponding to isolelectric points 5,99, 6,31, 6,20, 6,42, 6,74 and 6,89. If we compare the two contrasted varieties, we notice the induction of spots 13 and 14 in the two inoculated leaves but spots 9, 10, 11 and are only observed in Razzegui leaves. Also, spots 6, 7 and 8 are inhibited in Rafraf-F leaves. These differences obtained in proteins induction and suppression between the two contrasted varieties show variable symptoms observed after fungus inoculation. Finally, the study of the electrophoretic profiles of total proteins extracted from leaves of two Tunisian varieties shows the apparition of two spots with molecular weight of 19 - 26 kilo Dalton and 30 - 38,5 kilo Dalton. These spots correspond to proteins resulting from response to Uncinula necator attack. According to Van Lorn & Van Kammen (1970) and Gianniazzi et al. (1970) proteins with weakly molecular weight seem to be Pathogenesis Related proteins (PR). In other study on the response of grapevine to abiotic stress, Ben jouira et al. (2004) have observed in Razzegui variety treated with salt, the presence of spot 13 in 2D-PAGE profile of stressed plant. This protein is identified as pathogenesis-related protéine PR-10.

Conclusion

In this preliminary work, fifty Tunisian grapes varieties were inoculated with the fungus Uncinula necator, agent of powdery mildew. The results show variability in the symptoms after two week of inoculation. More than 60% of the varieties were very sensitive and the disease begun to develop after one week. Other cultivars have shown a tolerance in the thirst week and than sporulation of fungus is observed on leaves (total area infected is less than 25% after two weeks). The best result is obtained with Rafraf-F cultivar which shows the less attack intensity after two weeks of fungus inoculation. This variety cultivated for a long time seems to be very interesting to study the relationship between Uncinula necator and grapevine. In the other hand the results from analysis of total proteins extracted from leaves of two contrasted varieties (Rafraf-F as tolerant and Razzegui as sensitive) show a differences in the

response after inoculation with fungus. Some proteins are induced and others are suppressed especially in Rafraf-F leaves. This work has to be continued to identify proteins involved in such biotic stress.

4,5 7 IEF M

SDS

97,4 66,2

31,5

21,5

14,4

Control leaves

IEF 4,5 7

SDS M

97,4 66,2 45

31,5

21,5

14,4

Inoculated leaves

Fig. 1. 2D-PAGE analysis of total proteins extracted of Rafraf-F leaves after inoculation with Uncinula necator (75 µg total proteins); M: molecular weight in kilo Dalton

IEF 7

SDS M 97,4 66,2 45

31,5

21,5

14,4

Control leaves

IEF 4,5 7

SDS M

97,4 66,2 45

31,5

21,5

14,4

Inoculated leaves

Fig. 2. 2D-PAGE analysis of total proteins extracted of Razzegui leaves after inoculation with Uncinula necator (75 µg total proteins); M: molecular weight in kilo Dalton

Acknowledgements

Authors want to thank Mr. Hamadi HLEL, grapevine grower in Rafraf region, for his cooperation.

References

Askri, F. 1997. Etude nationale sur la diversité biologique. Chapitre Arboriculture fruitière et Viticulture. – Rapport GEF/MEAT, Ministère de l’environnement, Tunis, 27 p.

Ben Abdallah, F. 1999. Les vignes autochtones; caractérisation, régénération et dépistage in vitro. – Thèse de doctoratde la faculté de science de Tunis, Tunisie. 173 p. Ben jouira , H., M. Hanana, A. Ben Salem, N. Jallouli, L. Hamrouni, S. Daldoul, I. Toumi, A. Mliki, C. Abdelly, S. Hamdi, M. Hefer, G. Reustle, A. Ghorbel 2004. Recherche des varieties de vigne adaptées à la sécheresse et à la salinité. – OIV - XXVIII th Congres 4- 9 July, Vienna, Austria. Bradford, M. 1976. A rapid and sensitive method for the quantification of micro-organism quantities of protein utilizing the principle of protein dye binding. – Anal. Biochem. 72: 248-254. Clerjeau, M. 1998. Techniques de contamination artificielles des champignons parasites de la vigne. – Stage de formation continue, ENITA-INRA Bordeaux: 55-62. Gianinazzi, S., Y. Marco, M-L. Milat, P-M. Molot, T. Rouxel, P. Boistard, M. Dron, J. Dunez, G. Bompeix, B. Fritig 1991. Les mécanismes de défense des plantes. – Les dossiers de l’INRA, les relations plantes-microorganismes. INRA Paris: 20 p. O’Farrell, P. 1975. High resolution two-dimentional electrophoresis of proteins. – J. Biol. Chem. 250: 4007-4021. Van Loon, L.C., A. Van Kammen 1970. Polyacrylamide disc electrophoresis of the soluble leaf proteins from Nicotiana tabacum var.“Samsun NN”. Changes in proteins constitution after infection with tobacco mosaic virus. – Virology 40: 199-211. Zoghlammi, N., A. Mliki, A. Ghorbel 2001. Evaluation of genetic diversity among Tunisian grapevines by RAPD markers. – Vitis 40 (1): 31-37.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 69-72

In semi-vivo antagonism of Acremoinum byssoides towards Plasmopara viticola

Gaetano Conigliaro, Sandra Lo Piccolo, Livio Torta and Santella Burruano Dipartimento S.En.Fi.Mi.Zo., Sezione di Patologia vegetale e Microbiologia agraria, Università di Palermo, Viale delle Scienze 2, 90128 Palermo. [email protected]

Abstract: In previous laboratory tests the culture filtrates and the crude extracts obtained from Acremonium byssoides were shown to completely inhibit Plasmopara viticola sporangia germination. A study was thus undertaken in order to investigate on the possible inhibiting activity of the hyphomycete towards pathogenesis of the oomycete in grapevine. A. byssoides was cultivated on malt-extract agar (MEA) to obtain pure colonies and on malt- extract broth (MEB) to obtain culture filtrate and crude extract. Healthy leaves of grapevine were put in Petri dishes containing sporangia suspended in: deionized sterile water, malt-extract broth, dimethylsulfoxide (DMSO), culture filtrate, crude extract and a suspension of A. byssoides germinated conidia. After 6 days from inoculations, the bioactivity of A. byssoides was evaluated on the basis of P. viticola evasion percentage. The results show that the higher percentage of pathogen evasion occurred in leaves inoculated with sporangia suspended in distilled water, in MEB and in DMSO; lower values of pathogen evasion were noted when leaves were treated with culture filtrate and crude extract; no evasion occurred in presence of A. byssoides germinated conidia, showing either antibiotic or hyperparasitic activity.

Key words: antagonism, Acremonium byssoides, Plasmopara viticola

Introduction

Previous studies showed Acremonium byssoides as endophyte in several asymptomatic organs and phenological stages (Burruano et al., 2002; 2004) of different grapevine cultivars (Regina, Catarratto, Insolia and Nero D’Avola). At the same time, the antagonism of the hyphomycete towards Plasmopara viticola was confirmed; in fact, A. byssoides produced secondary metabolites, such as the acremines (Assante et al., 2005), which were shown to inhibit the gamic and agamic structures of the oomycete (Burruano et al., 2004; 2006). In order to define new strategies for the downy mildew control, the biocontrol of A. byssoides towards P. viticola was tested in semi-vivo.

Material and methods

The Acremonium strain isolated from sporangia and sporangiophores of P. viticola, named A20, was cultivated on malt-extract agar (MA) and incubated for 15 days at 28 ± 1 °C to obtain pure colonies. To produce cultural filtrate (CF), the hyphomycete was inoculated in Erlenmeyer flasks (300 ml) containing 100 ml of malt extract broth (MEB, Oxoid), and then incubated at 28 ± 1 °C; after 20 days, the cultures were sterile filtered under vacuum on 250 ml Stericup (0,22 µm HV Durapore Membrane, Millipore Co, Bedford, Mass) and stored at 8 ± 1 °C. To obtain crude extracts, A20 was grown in Erlenmeyer flasks (300 ml) containing 100 ml of MEB for 15 days at 28 ± 1 °C. The crude extract extraction was made twice with

69 70

EtOAc; extract was taken to dryness by means of Na2SO4 and rotavapor under vacuum (40° C), then washed with hexane, dissolved in dimethylsulfoxide (DMSO) and tested at 0.1g/l concentration. Cultural filtrates, crude extracts and germinated conidia of Acremonium A20 were assayed as inhibitors of the P. viticola pathogenesis. Healthy leaves of grapevine (c.v. Insolia) were placed into Petri dishes (9 cm diam.) containing fresh sporangia (9x105/ml) suspended in: deionized sterile water, malt-extract broth, DMSO in water (0.3%v/v), cultural filtrate (1:1 v/v), crude extracts from MEB (1:1 v/v), and a suspension of hyphomycete germinating conidia (3x106/ml), singularly. After 6 (incubation period) and 8 days from inoculation, the A. byssoides bioactivity was evaluated on the basis of P. viticola evasion percentage. Statistical analysis was performed using ANOVA and T Tukey’s tests.

Results

The results showed the antagonistic effect of Acremonium byssoides towards P. Viticola (Tab. 1). In particular, leaves inoculated with sporangia suspended in deionized sterile water (Fig. 1a) showed abundant and homogeneous (100%) oomycete evasion on the leaf lower surface; as regards leaves inoculated with sporangial suspension in MEB (Fig. 1b) and DMSO in water (Fig. 1c), sporangiophores covered respectively about 83.3 and 91.6% of the lower surfaces.

a b c

Fig. 1. Leaves inoculated with sporangia suspended in: deionized sterile water (a), MEB (b) and DMSO in water (c).

bcb

Fig. 2. Leaves inoculated with sporangia suspended in: cultural filtrate (a), crude extract from MEB (b) and a suspension of A. byssoides germinating conidia (c).

When leaves were inoculated with sporangia suspended in cultural filtrate (Fig. 2a), pathogen evasion attained more than 2.6% at 6 and 8 days; in leaves inoculated with sporangial suspension in crude extracts from MEB (Fig. 2b), the percentage of P. viticola evasion was 16.8%. In leaves inoculated with the suspension of sporangia and A20 germinating conidia (Fig. 2c), pathogen evasion was significantly lower (0.5-1.1%); sporangiophores and sporangia were hyperparasitized by A. byssoides hyphae. This mechanism, already observed in previous studies (Burruano et al., 2004; 2005; 2006), was confirmed by microscopic observations (Figg. 3a, 3b and 3c).

5 µm 5 µm

a b c

Fig. 3. A. byssoides hyphae with phyalides and conidia bridling the oomycete structures (a); detail of a P. viticola sporangiophore enveloped and penetrated by a hypha of the antagonist (b) and sporangia with collapsed cytoplasm (c).

Conclusion

The results obtained confirmed the antibiotic and hyperparasitic activity of A20 strain towards the oomycete. Particularly, antibiotic effect towards the pathogen was proved using cultural filtrates and crude extracts. Moreover, the assays carried out with the suspension of A20 showed as the hyphomycete activity was either antibiotic and hyperparasitic. In fact, A20 inactivates the oomycete first through secondary metabolites production, then it parasitizes the few emerged sporangiophores and sporangia.

OBSERVATIONS TREATMENTS 6 gg 8 gg

Water 98.3 100 Water +DMSO 80 91.6 Table 1. P. viticola evasion (% of leaf surface covered by downy mildew) 6 and 8 MEB 73.3 83.3 d after inoculation of leaves with different Cultural filtrate 2.6○● 2.6○● sporangial suspensions. ○● ○● Crude extract from MEB 16.8 16.8 ○ = statistically significant at the 5% level ○● ○● ● Conidial suspension 0.5 1.1 = statistically significant at the 1% level

Assays are being carried out to evaluate how other media can influence the production of secondary metabolites; particularly, interesting results are being obtained by use of malt extract-peptone-glucose-glycerol-agar (MPGGIA). Further studies will be carried out in order to assess the activity of acremines towards the oomycete pathogenesis both in semi-vivo and in the field.

Acknowledgements

This research was supported by a grant from M.U.R.S.T. (60%).

References

Assante, G., Dallavalle, S., Malpezzi, L., Nasini, G., Burruano, S. & Torta L. 2005: Acremines A-F, novel secondary metabolites produced by a strain of an endophytic Acremonium, isolated from sporangiophores of Plasmopara viticola in grapevines leaves. – Tetrahedron 61: 7686-7692. Burruano, S., Conigliaro, G., Lo Piccolo, S. & Torta, L. 2002: Sull’interazione tra Acremonium sp. Vitis vinifera e Plasmopara viticola. – Atti del Convegno Nazionale: L’endofitismo di funghi e batteri patogeni in piante aboree e arbustive. Sassari-Tempio di Pausania. 19-21 Maggio 2002, 103-112. Burruano, S., Lo Piccolo, S., Alfonzo, A. & Torta, L. 2004: Evoluzione di Acremonium sp. endofita in tessuti fogliari di Vitis vinfera, durante la patogenesi di Plasmopara viticola. – Micol. It. 2: 42-45. Burruano, S., Alfonzo, A., Lo Piccolo, S., Torta, L., Moretti, M. & Assante, G. 2005: In vitro antagonism towards Plasmopara viticola from an endophyte Acremonium sp. in grapevine. – Atti Annual Scientific Meeting Exploitation of fungi, University of Manchester, 5-8 September 2005: 73. Burruano, S., Conigliaro, G., Lo Piccolo, S. & Torta L. 2007: Oospore di Plasmopara viticola: dinamica di maturazione e possibile antagonismo di Acremonium byssoides. – Micol. It. 2: 53-59.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 73-79

Approche des facteurs sociologiques influent l’engagement des viticulteurs dans la demarche de production integree en Aquitaine et Charentes

Thierry Coulon1, François Hugueniot1, Claude Compagnone2 1 Institut Français de la Vigne et du Vin, 39 rue Michel Montaigne, 33290 Blanquefort 2 Etablissement National Enseignement Supérieur Agronomique Dijon 26, boulevard Docteur Petitjean – 21000 Dijon ([email protected])

Abstract: Approach of the sociological factors influencing the vine growers commitment while apprehending the integrated production in the Aquitaine and Charentes area. – In 2000 a network of reference farm is created in Aquitaine in order to specify the feasibility and the implementation conditions of the integrated production (IP) in vine growing. At the end of 2003, we noticed that the farms started to rapidly progress towards the Integrated Production approach, but this progression had a tendency after a while to stagnate. The vine growers seem to be faced with difficulties, on certain technical points, which limited their progress to reach the goals set by the presented guidebook. In 2004, we then carried out a specific workstudy with the vine growers in order to better detect what we have called the “brakes” of the IP approach. The brake is defined as being the cause of the non achievement of an objective. The first step is to identify the "brakes", directly drawn from the results of the technical diagnoses obtained over several years on the various farms, which makes it possible to point out the objectives which are not achieved. The creation of a dialogue with the farmers on the identified critical points then allows to describe and then explain this situation. In 2005, this study on the brakes is supplemented by a sociological approach in order to see how these factors can affect and condition the vine grower’s behaviours and choices. A semi structured interview on the practices is carried out at the vine growers’ place, and sociological questions are asked to them (age, training level…). Finally, the professional insertion in the environment of each surveyed vine grower is evaluated by a sociograms graph showing the links of the vine growers with their peers and their technical support. The crosschecking of these three levels of results (interview, sociograms and sociological datas) enabled us to identify "brakes" and "motivations" specific to the implementation of the IP recommendations. This study also allowed us to draw a typology of the vine growers which shows that the relational contexts have a real influence on the implementation of the IP practices.

Key words: Integrated production, sociological factors, “brakes”, motivations, typology

Introduction

A partir de 2000 un réseau de ferme de référence est constitué en Aquitaine afin de préciser la faisabilité et les conditions de mise en œuvre de la production intégrée (PI) en viticulture. Fin 2003, nous constations que les exploitations avaient d’abord rapidement progressé dans la démarche PI, mais que cette progression tendait ensuite à "plafonner". Les exploitants semblent être confrontés, sur certains points techniques, à des difficultés qui limitent leur progrès dans la dernière phase d’atteinte des objectifs fixés par le référentiel PI, et ce pour la partie "engagements" qui correspond à la définition minimum de la déclinaison pratique de la PI. En 2004, nous avons donc mis en œuvre un travail spécifique d’investigation auprès des producteurs dans l’objectif de mieux cerner ce que nous avons appelé les "freins" à la démarche PI.

81 74

Le frein est défini comme étant la cause de non réalisation d’un objectif. La première étape est le repérage des « freins », effectué directement à partir des résultats des diagnostics techniques obtenus sur plusieurs années sur les différentes fermes, ce qui permet de repérer les objectifs qui ne sont pas atteint. L’instauration d’une conversation chez l’exploitant sur les point critiques identifiés permet ensuite de décrire puis d’expliquer cette situation. Cette étude sur les freins est complétée en 2005 par une approche sociologique afin de voir comment ces facteurs peuvent influencer et conditionner les modes de détermination et le choix des viticulteurs. Un entretien semi directif sur les pratiques est réalisé chez les viticulteurs, puis des questions d’ordre sociologique lui sont posées. Enfin l’insertion dans le milieu professionnel de chaque viticulteur enquêté est évalué par l’établissement de sociogrammes représentant les liens du viticulteurs avec ses pairs et son encadrement technique.

Identification des exploitations CHAPITRE 5 Traitement direct sur chaque face de rang / nombre total de traitements* MOMMO Traitement dirigé sur grappe réalisé face par face / nombre total CC de traitements dirigés sur grappe Arrêt de la pulvérisation lors des virages en bout de rang Grille inter- Adaptation du nombre de diffuseurs ou buses en fonction de la végétation MEC MPEC MPEC PREPARATION DE LA BOULLIE annuelle des Calcul précis de la quantité de bouillie / surface à traiter QQ CONTROLE DU PULVERISATEUR Diagnostic par un agent agrée / 3 ans FE FE FE non conformités Les autres années, autodiagnostic par le vigneron (détail notation ci contre) Réglage et étalonnage à chaque début de campagne EPQP Nombre de fois où le degré d’obstruction des buses et des filtres est vérifié avant utilisation / nombre de sortie du pulvérisateur EQUIPEMENT DU PULVERISATEUR Equipement du pulvérisateur neuf (voir détail de la notation ci-contre) QF Equipement du pulvérisateur pré existant (voir détail notation ci contre) MJEC OJEC JPFOEC Surface traitée par hélicoptère / surface non justifiée Surface traitée au canon / surface non justifiée

CHAPITRE 6 PROTECTION DE L'UTILISATEUR FMJPQECB MJPFOEQC Protection individuelle (voir détail de la notation ci-contre) BFMJECB O B

Formation de l'opérateur aux risques d'exposition FPQEC FPEQC Non OFMJPQEC Local de stockage (voir détail de la notation ci-contre) BFMJECB MJFEQC B conformités MAITRISE DU REMPLISSAGE DE LA CUVE Risque de pollution d'émissaires lors du remplissage JJPCJPC Dispositif anti retour (discontinuité hydraulique, clapet anti retour, cuve JPOEQ intermédiare)

OFMJPQEC Equipé d’un compteur d’eau programmeur BFMJECB MJPFOEQC B

Equipé d’une vanne d’arrêt type quart de tour CQCQ Si pas d’aire de récupération des effluents, nombre de débordement P / nombre total de remplissages de la cuve

Entretiens avec les viticulteurs

Par thème Analyses textuelle des discours

Comportements Raisonnements Freins/accélérateurs

Fig. 1 : Exemple d’identification des freins à la démarche PI dans les exploitations non conformes/ techniques d’application des produits phytosanitaires

Matériels et méthodes - outils supports de l’étude

Etude des « freins » et volet sociologique Deux approches sont successivement conduites:

Etude des freins en 2004 – Nous avons mis en œuvre un travail spécifique d’investigation auprès des producteurs dans l’objectif de mieux cerner ce que nous avons appelé les "freins" à la démarche PI. Le frein est défini comme étant la cause de non-réalisation d’un objectif. Afin de repérer ces « freins », des étapes et interventions successives ont été nécessaires : sur la matrice de travail (les diagnostics d’exploitation) et chez l’exploitant (enquête). Un diagnostic inter annuel et une grille des non conformités PI sont établis. Une analyse textuelle du discours de chaque exploitant enquêté est réalisée. Les comportements, raisonnements, attitudes des viticulteurs sont détaillés et permettent d’identifier les freins et d’en effectuer une typologie.

Etude sociologique en 2005 – Un entretien semi directif, organisé autour d’un questionnaire pré établi, oriente les échanges sur les pratiques des viticulteurs, les raisons des évolutions ou non évolutions, puis sur les questions d’ordre sociologique proprement dites : caractéristiques propres à l’exploitant (âge, formation, etc…) et réseau de dialogue personnel (avec ses pairs, les techniciens, autres interlocuteurs et types de contacts/échanges…).

Question: H Quel est le rôle joué par le réseau social: • sur le raisonnement? • sur le changement des pratiques ?

Hypothèse de départ: H La manière dont les viticulteurs comprennent, évaluent et mettent en œuvre leurs pratiques est construite par les relations avec leurs groupes de pairs, au sein de la commune, région ou hors région…

Fig.2. Nature de l’investigation sociologique conduite.

Résultats

Etude des freins Plusieurs types de freins sont listés. Les informations recueillies (en 2004) ne semblent pas pouvoir justifier le « plafonnement » de progression vers la PI constaté sur le réseau. Dans la plupart des cas, des possibilités techniques sont proposées aux producteurs, validées tant sur le plan expérimental que pratiquement sur d’autres exploitations viticoles. Sur le plan écono- mique, nous disposons d’indicateurs, imparfaits il est vrai, qui laissent penser que les coûts strictement spécifiques à la démarche PI influencent peu le résultat des exploitations.

13 ENTRETIENS 3 analyses

Retranscription de la partie Bilan sur les Bilan sur les sur les pratiques caractéristiques caractéristiques relationnelles sociologiques

Réorganisation par Formation des thème sociogrammes

Résultat de l’analyse Résultat de l’analyse des thématique relations professionnelles

Confrontation Résultats de l’étude des données

Fig. 3. Valorisation des entretiens réalisés dans l’étude sociologique par 3 analyses

Informat Types de freins économiques Raisonnement Techniques Humains déficien - Coûts m ajorés * investissem ent +(m atériels-installations) X X * main-d’œ uvre + X X - contexte économique défavorable X - pertes de recettes X - hiérarchisation des choix d’investissement X X X défavorable (qualité-produit/environnem ent) - PI = cadre contraignant X - contrôles + traçabilité trop stricts X X - scepticism e/certains outils ou pratiques X X X X - attentisme (conseil technique-recherche X X X X d’inform ations -aides financières… ) - im passes techniques X (viroses – FD ..) - protection individuelle (équipem ents) X X X - m anque de solution validées (effluents) X - m atériel inadapté (pulvé – entretien sols… ) X - méconnaissance technique X X X X - compétence individuelle/formation X X - diffusion références insuffisante * sur équipements demandés et leurs coûts X X * sur l’application de certaines pratiques - motivation X X - réglem entation absente ou im précise X X - valorisation insuffisante X - habitude (sécurisé par pratiques X X habituelles) - méconnaissances des risques X X X - comportements à risques X X

Fig. 4: Les freins à la démarche PI repérés sur les exploitations du réseau.

Des défauts de compétence sont repérés. Les niveaux de formation initiale et/ou continue sont parfois insuffisants. On a souvent du mal à comprendre les raisons ou plutôt la logique des choix effectués. Enfin le défaut d’information est un élément qui ressort égale- ment de l’étude, lié aux difficultés d’accès à des références technico-économiques dispersées et parfois partielles, attentisme du viticulteur, défaut de disponibilité, méconnaissance de la

réglementation…. Ces constats nous amènent à orienter notre recherche sur d’autres facteurs, d’ordre sociologique, en collaboration avec Claude Compagnone, sociologue INRA/ENESA Dijon.

Etude sociologique La question de départ de cette approche est la suivante : comment le réseau constitué de l’ensemble des relations sociales que les viticulteurs entretiennent avec d’autres joue sur la façon dont ces viticulteurs conçoivent et mettent en œuvre les pratiques préconisées ? Le traitement de cette question s’appuie sur l’hypothèse que la manière dont les viticulteurs comprennent et évaluent ces pratiques et la façon dont ils les mettent en œuvre, sont liées aux dialogues professionnels qu’ils peuvent entretenir avec d’autres. Ces dialogues par lesquels s’effectuent des échanges d’idées et d’informations, se déroulent non seulement avec des agents de l’encadrement technique mais aussi avec d’autres viticulteurs. Pour répondre à cette question, on s’est plus particulièrement intéressé aux dialogues sur les pratiques viticoles qu’entretient chaque viticulteur du réseau de protection intégrée avec des conseillers et d’autres viticulteurs. Sur treize personnes enquêtées, quatre grands types peuvent être distingués en fonction du nombre de collègues et de conseillers avec qui chacun est en relation.

Type 1 Type 2 Type 3 Type 4

Lien fort Lien faible

V iticulteur du Réseau PI

Conseiller (Chambre, ITV, agrofourniture,..) V iticulteur de la commune ou environ V iticulteur hors de la petit région viticole

Fig. 5: Sociogrammes des viticulteurs enquêtés en 2005

Lorsque l’on met en perspective ces types de réseaux avec la mise en œuvre des pratiques préconisées par les viticulteurs enquêtés, on constate que ceux du type 2 et, de manière encore plus marquée, ceux du type 1, ont les pratiques les plus conformes aux prescriptions de la production intégrée. Inversement ceux du type 3 et, de manière encore plus prononcée, ceux du type 4, ont les pratiques les moins conformes. Trois traits permettent de caractériser un gradient : le moment d’engagement dans la démarche de production intégrée, la nature des liens entretenus avec l’encadrement technique et la nature des liens établis avec des pairs. Schématiquement le type 1 est constitué des viticulteurs engagés le plus anciennement dans la démarche, ayant le nombre le plus conséquent d’appuis de la part de l’encadrement et possédant des liens conséquents avec d’autres viticulteurs en termes de nombre et de nature.

Par rapport à une stratégie d’accompagnement des viticulteurs dans la démarche vers la production intégrée, il est intéressant de bien considérer les différents positionnements professionnels que nous avons identifiés : • Certains viticulteurs, motivés et bien accompagnés techniquement, ayant pu ou su s’engager dans une dynamique de groupe stimulante avec quelques collègues motivés constitueront l’élément minoritaire, mais moteur et démonstratif de la faisabilité et de l’intérêt d’une démarche production intégrée maîtrisée, • D’autres, installés dans un certain conformisme, tendent à simplement reproduire les comportements majoritaires de la profession. Capables techniquement et économique- ment de s’engager vers une modification de leur système de production, ils ne le feront réellement et surtout complètement que sous l’incidence d’éléments extérieurs éventuellement contraignants (réglementation, contrôles, situation de marché imposant des cahiers des charges…), • D’autres enfin, insuffisamment ouverts sur leur milieu professionnel, paraissent de ce fait freinés dans les évolutions proposées.

Complétant les informations déjà obtenues sur le plan technique et partiellement sur le plan économique, l’analyse sociologique que nous avons abordée en 2005 apporte des éléments structurants permettant de mieux appréhender les comportements professionnels et ainsi, de mieux adapter les stratégies de sensibilisation à une viticulture durable. Ces éléments sont source de réflexion et d’adaptation au sein de l’équipe de conseillers viticoles des régions Aquitaine et Charentes ayant participé aux travaux.

Conclusion

«Les relations sociales constituent un capital pour le déploiement des pratiques» Les viticulteurs auxquels on s’intéresse ici dans le cadre du réseau de fermes de référence, ont la particularité d’être sensibilisés à cette question. Pour avoir été choisis dans le cadre de ce réseau, ils ont une certaine proximité avec les agents de l’encadrement technique. En position de pionniers dans leur environnement social pour la mise en œuvre des pratiques de la production intégrée, ils peuvent compter sur l’appui de l’encadrement. Du coup, les dialogues avec les pairs apparaissent, dans le cadre de l’étude réalisée, plus secondaires par rapport à cet appui. Toutefois, les travaux que nous avons pu faire en Bourgogne et en Languedoc- Roussillon sur des viticulteurs moins sensibilisés nous montrent que ces échanges entre producteurs sont déterminants dans la mise en œuvre de changements de pratiques. Et cela pour trois raisons : • Tout d’abord, parce que la capacité des viticulteurs à changer de pratiques est d’autant plus importante que ces derniers peuvent avoir accès, par la discussion, à des connaissances possédées par certains de leurs pairs. Dans ce sens, leurs relations sociales constituent réellement un capital pour le déploiement de ces pratiques, • Ensuite, parce qu’une forme d’organisation informelle peut se mettre en place entre viticulteurs d’un même réseau. Chaque viticulteur détenant un bout d’expertise sur ces pratiques apporte, au moment opportun, les éléments d’expériences et de connaissances pertinents pour pouvoir les maîtriser collectivement, • Enfin, parce que la mise en œuvre ou non de certaines pratiques par un viticulteur va dépendre des façons de faire communément admises par les membres du réseau professionnel de viticulteurs auquel il appartient. Et c’est par la discussion que ces normes peuvent évoluer localement.

Ceci amène à être particulièrement attentif à ces processus lorsque l’on veut travailler à l’élargissement de l’usage d’une démarche telle que celle de la production intégrée

Bibliographie

Coulon T., Hugueniot F. 2007: Validation technico-économique de la démarche de Production Intégrée de Raisins sur un réseau de fermes viticoles de référence dans les vignobles de Bordeaux et de Cognac. – Colloque OIV Budapest Hongrie, Juin 2007. Coulon T. et al. 2002: Diagnostic Technique de Production Viticole Intégrée. – Viticulture durable : les outils supports à une mise en pratique, publ. ITV France, Décembre 2002. Coulon T. et al. 2000: Référentiel national pour la production intégrée de raisin. – Viticulture durable : quelle mise en pratique dans le vignoble français ?, publi. ITV France, Octobre 2000. Coulon T., Hugueniot F. 2003: Technical and economical validation of the integrated production of grape in the aquitain vineyard : report after three years of study (2000 – 2002). – IOBC/wprs Bulletin 26 (8): 293-296. Coulon T., Sentenac G. 2001: Un référentiel national "Production Intégrée des raisins" pour les vignobles français. – Bulletin de l’OIV 74(845-846): 445-462. Coulon T., Sentenac G. 2001: Viticulture durable. Vers quelle mise en pratique dans le vignoble ? – Journal International des Sciences de la Vigne et du Vin, hors série, 2001: 181-187. Coulon T., Sentenac G. 2003: Proposal of review of the third IOBC Technical Guideline for the Integrated Production of Grapes. – IOBC/wprs Bulletin 26 (8): 297-299.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 81-86

Validation technico-économique de la Production Intégrée de raisins en viticulture sur un réseau de fermes viticoles de référence dans les vignobles de Bordeaux et de Cognac Bilan 2000-2006

Thierry Coulon, François Hugueniot Institut Français de la Vigne et du Vin, 39 rue Michel Montaigne, 33290 Blanquefort

Abstract: Technical and economical validation of the integrated vine growing production approach on a network of reference farms in the Bordeaux and Cognac vineyards Evaluation 2000-2006. – Two questions are asked concerning: – The Integrated Production feasibility in vine growing. Is it technically and economically possible? – The possibility to optimise a conversion of the traditional production systems into a Integrated Production system. On a network of reference farms we observed how the vine growers adapt themselves and change their methods and practices. Carried out from 2000 to 2006, the work lead to the edition of an IP guidebook and a traceability of the vine growers actions, evaluated by a technical diagnosis specifically created for the study. By their varied schemes of production (size- designation of origin…), the selected farms present a wide range of configurations which fits the reality of the regional wine farms. An estimation of the cost related to the Integrated Production practices was attempted, but methodological difficulties restrained it from the original objectives. Following the first evaluation of the vine growing methods, after realizing the problems and interacting with the technicians who followed the exploitation, all the vine growers had set objectives of improvement. Our observations made it possible to highlight well the progress achieved during one or two campaigns; all the farms progress. But this progression had a tendency after a while to stagnate. The vine growers seemed to be encounter technical difficulties on certain points which limit their progress to reach the goals set by the presented guidebook. Several types of difficulties came out. However, excepted in particular technical impossibilities clearly identified, the collected information does not seem to be able to justify the stagnation of the progression towards IP objectives noticed on the network. In most cases, technical tools (choices, alternatives…) are presented to the vine growers, validated, as well on the experimental level, that on a practical level on other vineyards. On the economical level, we have indicators, however imperfect, which allow us to think that the cost strictly specific to the IP approach, only have a small influence on the economical performances and on the attitude of the farmers. This observations lead us to aim our research on sociological factors, which could explain the behaviours and the choices of the vine growers. The results of this study constitute the subject of a specific communication.

Keywords: Integrated Production, Integrated Production guidebook, technical diagnosis, specific cost

Introduction

L’objectif est de doter la filière viticole de références permettant l'engagement maîtrisé de la démarche de production intégrée dans le cadre de l'exploitation viticole : - Définition des itinéraires techniques et évaluation des coûts, - Incidence sur l'emploi des intrants, la qualité des produits (raisins) et du produit fini (vin), - Définition des outils de pilotage de ce système de production en viticulture (procédures de suivi, modes opératoires, système d'auto contrôle des itinéraires techniques),

81 82

- Choix des outils de pilotage, et définition des critères mesurables les plus pertinents et autant que possible, les moins contraignants pour le viticulteur.

Il s’agit donc de répondre à un double questionnement: - Evaluation de la faisabilité de la production intégrée en viticulture (est ce possible tant au plan technique qu’économique) - Comment les viticulteurs s’adaptent-ils, font évoluer leurs méthodes, leurs pratiques. Il s’agit donc d’observer une population « leader » pour mieux optimiser dans un second temps la phase de transition pour un plus grand nombre de producteurs.

L'outil retenu pour cette « étude système » est le réseau de fermes de référence. La diversité des situations constitue un élément à contrôler tout en répondant à une réalité du terrain dont l'appréciation constitue bien un enjeu. La mise en place d’une telle étude est novatrice en viticulture, les méthodologies et outils supports de l’étude étaient donc à construire tant sur le plan technique qu’aux niveaux économique et sociologique.

Matériels et méthodes - outils supports de l’étude

Volet technique « Production intégrée »

⇒ Un référentiel « Production Intégrée de raisin » est rédigé (en 2000) précisant concrètement les principes et méthodes à mettre en œuvre, dans la filiation des directives établies par l’OILB (Organisation Internationale de Lutte Biologique et Intégrée). ⇒ Un diagnostic technique des pratiques du viticulteur permet de situer les exploitations par rapport aux exigences et méthodes de la production intégrée en viticulture. Répartis en sept chapitres considérés d'importance équivalente, les critères d'évaluation, quantitatifs ou qualitatifs, sont issus d'une lecture directe du référentiel. Pour chaque chapitre, l 'évaluation effectuée permet de préciser en pourcentage l'objectif rempli par rapport à l'objectif « Production Intégrée » fixé. Globalement, une représentation synthétique de cette évaluation est visualisée sous forme d'un diagramme en radar. ⇒ La Constitution d’un réseau de fermes de référence engagée en Aquitaine en 2000, a été confortée en 2001 et 2002 et compte des exploitations volontaires. De configurations technico-économiques diverses (taille, mode de faire-valoir, appellation…), elles constituent de fait un observatoire. ⇒ Conception d’une grille de traçabilité permettant à la fois de lister les informations nécessaires à l'établissement du diagnostic technique et d'évaluer le taux de traçabilité effective sur l'exploitation par rapport à la traçabilité globale nécessaire et dont la mise en œuvre concrète relève de la responsabilité du viticulteur. ⇒ Cahier de suivi Constitué des fiches de pré-saisie ou de saisie pour chaque axe technique envisagé dans le référentiel "PI".

Volet économique

La construction d'un système d'information pertinent passe par un préalable : la définition d'un modèle de fonctionnement de l'exploitation en PI. Dans notre cas, le système sera une exploitation en PI, et le processus représenté correspondra au fonctionnement évolutif permettant de passer d'une production viticole conventionnelle à une production intégrée de raisin. Le modèle induit donc les règles de fonctionnement du système d'information et la structure des données. Pour identifier toutes les données, nous proposons un schéma de 83

fonctionnement général d’une exploitation viticole (en PI ou non) à partir duquel nous pouvons faire ressortir clairement les point influencés par la PI.

Historique

Exploitant LT Salarié(s) Famille MOTIVATION PI PI

INTRANTS Achats mat 1ère COÛTS de PRODUCTION Cycle de PI production Temps de travail, CLIENT(S) sécurité, santé, Environnement naturel formation, (paysage, eau, sol) pratiques… PI PRODUITS PI (raisins-vins) PI DECHETS

Fig. 1 : Approche globale du système d'exploitation viticole

Dans la partie haute du schéma figure 1, nous considérons deux sujets d'observation, l'exploitant et l'exploitation, en interaction (cœur du système), qui constituent un couple influencé par des éléments extérieurs, locaux ou plus globaux (contexte général de la viticulture…). Dans la partie basse du schéma, les années se superposant, le cycle de production rattaché à l'ensemble statique exploitant / exploitation prend une trajectoire dynamique ; c'est pourquoi il est important de prendre en compte le facteur "temps" : notre étude doit être pluriannuelle.

Résultats après 6 années d’études

Evaluation technique. Pour cette partie technique, sur ces 6 ans d’études, quelque-soit l’exploitation étudiée, deux périodes peuvent être identifiées :

1ère période : Volonté d’adaptation des viticulteurs Après le premier bilan, une prise de conscience des problèmes s’opère chez les viticulteurs qui se fixent alors des objectifs d’amélioration. Toutes les exploitations progressent. Les adaptations se font au rythme propre de chaque viticulteur, et ce à partir de son niveau d'organisation technique de départ. Les moyens financiers de l'exploitation influencent les choix, en particulier dans les délais de réalisation de certains investissements (locaux phytos, dalle de lavage, etc…). Globalement, ces résultats sont encourageants et "rassurent" quant à la faisabilité de la Production Intégrée en viticulture.

Exploitation E - Bordeaux supérieur Exploitation B - Bordeaux supérieur Installation, plantation, Installation plantation conduite du vignoble 1 conduite du vignoble 100 Mesures écologiques 100 93 87 90 90 100 100 96 100 95 100 100 connexes 80 80 Gestion du sol, Mesures écologiques 100 Fertilisation 7 60 2 connexes 60 fertilisation 85 87 94 92 100 86 100 94 88 40 69 95 97 96 9797 95 40 20 20 Effluents et déchets 0 0 Entretien du sol 43 67 75 95 95 Entretien 95 6 3 Effluent, déchets du sol 100 100 100 100 100100 54 55 51 51 51 88 75 95 100 100

Méthodes de 5 4 100 99 96 96 87 Protection intégrée 93 96 100 100100 81 pulvérisation 100 95 100 100 100 Méthodes de pulvérisation Protection intégrée 68 67 81 81 81

Des pratiques qui évoluent Des pratiques qui plafonnent Fig. 2 : ValidationFigure 2technico : validation économique technico économique de la PI de – laRésultats PI – Résultats comparés comparés sur sur les les engagement engagements 2001, 2002, 2003, 2004 et 2005

2ème période : Vers un plafonnement de la progression Au bout de quelques années, certaines exploitations poursuivent leur progression vers une atteinte de la globalité des objectifs PI. En revanche, pour d’autres, un "plafonnement" de la progression est observé (exemples figure 2). Parmi les points les plus délicats, on retrouve toujours : la fertilisation raisonnée, l’équipement des pulvérisateurs, la protection individuelle des opérateurs lors des traitements, la maîtrise des effluents et des pollutions ponctuelles.

100

40 Premier quartile max 20 Pourcentage de réalisation de Pourcentage Min Troisième quartile 0 1 Installation, 2 Gestion du 3 Entretien du 4 Protection 5 Méthodes 6 Effluents et 7 Mesures plantation sol sol intégrée de déchets écologiques et fertilisation pulvérisation connexes

Fig. 3 : Validation technico économique de la PI. Distribution des résultats obtenus en Aquitaine en terme d’atteinte des engagements minimum PI

Situation du réseau après cinq années de suivi La figure 3 précise, pour chaque chapitre technique, les écarts constatés dans l’atteinte des objectifs PI sur le réseau. La dispersion la plus importante concerne le chapitre « effluents et déchets ». Nous pouvons remarquer que sur les sept chapitres techniques évalués, seul ce chapitre 6 n’atteint pas les 100% des objectifs atteints pour au moins une exploitation, ce qui 85

révèle les réelles difficultés d’investissement et d’adaptation des méthodes de travail aux exigences PI.

Evaluation économique

Les résultats obtenus par l’application de l’analyse financière et de la méthode des coûts complets restent très globaux et ne permettent pas d’identifier les variations de coûts liés directement à la démarche PI. Néanmoins ils nous permettent de comprendre le fonctionnement des entreprises et de confirmer certains résultats jusque là seulement pressentis : - Des écarts de coûts de production sur le centre vigne importants - Des situations financières satisfaisantes sur toutes les fermes du réseau - Pas de lien apparent entre performance technique, niveaux de coûts de production et situation financière des exploitations (figure 4).

100 14000 90 12000 80 70 10000 60 8000 50 6000 40 financière /100 financière 30 4000 20 (€/ha) vigne centre coût 2000 10 Moyenne note technique et moyenne 0 0 BCDEF J KLMOQR

Moyenne f inancière Moyenne technique Coûts du centre vigne (€/ha)

Niveau technique élevé et coût du centre vigne faible

Coût faible et moyenne financière faible

Niveau technique plus faible et un coût plus élevé

Fig. 4 : Position relative des indicateurs financiers, des coûts du centre vigne et niveaux techniques par exploitation en 2002

Conclusion

Les viticulteurs sont très réactifs par rapport aux échanges avec les techniciens basés sur l’analyse objective de leur organisation de travail et de leurs pratiques. Le diagnostic régulièrement effectué permet une prise de conscience et une appréciation qualitative et quantitative de méthodes de travail renouvelées par rapport à celles antérieures souvent 86

héritées et conservées par habitude. Des progrès restent cependant à accomplir, y compris pour les fermes appartenant au réseau depuis le début de notre étude (2000). Certaines exploitations sont très proches de remplir la totalité des objectifs minimums PI fixés, mais sont une minorité. Les marges des progrès restant à accomplir pour les autres sont plus importantes. Les résultats obtenus par les entreprises les plus avancées dans la démarche PI montrent la voie. A partir de 2004, nous complétons nos observations par un travail spécifique d’enquête auprès des producteurs dans l’objectif de mieux cerner les freins à la démarche PI, pour mieux identifier les voies de progrès dans l’application des pratiques. Enfin nous faisons le constat que les outils crées pour notre étude (Référentiel technique, Diagnostic technique…) ont rapidement été adoptés et utilisés par les agents de développement en Aquitaine mais également dans d’autres régions viticoles, preuve du réel intérêt de la démarche et de la forte demande de tous les opérateurs (techniciens, viticulteurs…).

Bibliographie

Coulon T., Hugueniot F. 2007: Validation technico-économique de la démarche de Production Intégrée de Raisins sur un réseau de fermes viticoles de référence dans les vignobles de Bordeaux et de Cognac. – Colloque OIV Budapest Hongrie, Juin 2007. Coulon T. et al. 2002: Diagnostic Technique de Production Viticole Intégrée. Viticulture durable : les outils supports à une mise en pratique. – ITV France, Décembre 2002. Coulon T. et al. 2000: Référentiel national pour la production intégrée de raisin. Viticulture durable : quelle mise en pratique dans le vignoble français ? –ITV France, Octobre 2000. Coulon T., Hugueniot F. 2003: Technical and economical validation of the integrated production of grape in the aquitain vineyard : report after three years of study (2000 – 2002). – IOBC/wprs Bulletin 26 (8): 293-296. Coulon T., Sentenac G. 2001: Un référentiel national "Production Intégrée des raisins" pour les vignobles français. – Bulletin de l’OIV 74 (845-846): 445-462. Coulon T., Sentenac G. 2001: Viticulture durable. Vers quelle mise en pratique dans le vignoble ? – Journal International des Sciences de la Vigne et du Vin, hors série, 2001: 181-187. Coulon T., Sentenac G. 2003: Proposal of review of the third IOBC Technical Guideline for the Integrated Production of Grapes. – IOBC/wprs Bulletin 26 (8): 297-299.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 87-90

Replacement of copper in organic viticulture: efficacy evaluation of new natural fungicides against downy mildew

Silvia Dagostin, Tiziano Formolo, Ilaria Pertot SafeCrop Centre, Istituto Agrario di San Michele all’Adige, via E. Mach 1, 38010 S. Michele all’Adige (TN), Italy, e-mail: [email protected]

Abstract: Downy mildew caused by the obligate biotrophic oomycete Plasmopara viticola (Berk. & Curt.) Berl. & de Toni is one of the most important and devastating grapevine diseases. So far its control in organic agriculture is mainly based on copper fungicide, but from 2006 copper use is limited by European Commission Regulation. From 2004 to 2007 our research focused on testing alternatives to replace copper in experimental field trials. We tested 29 different products divided in groups: new copper formulations, plant extracts, clay, biocontrol agents and other origins. In particular in the first three years good results were obtained with two new copper formulations, but one of them causes phytotoxicity. Among plant extracts two compounds allowed a good control of the disease both on leaves and clusters, and six compounds controlled the infection only on clusters. Biocontrol agents reduced symptoms on leaves or on clusters. Among products from other origins only three allowed a good control of disease on leaves and four of them control the disease on bunches. Clay shows an efficacy similar to copper either on leaves and or on cluster.

Key words: Plasmopara viticola, grapevine, copper, plant extract, BCAs

Introduction

Downy mildew, caused by the obligate biotrophic oomycete Plasmopara viticola (Berk. & Curt.) Berl. & de Toni, overwinters in soil as oospores, which from spring throughout the entire season can germinate (Gobbin et al., 2005) and provide the first source of inoculum. After the first infection, disease spread with rapid sequence of asexual propagation cycles during the whole growing season affecting leaves and bunches with consequent high losses. In organic viticulture control of P. viticola is currently based on copper that has a wide range of activity and does not select for resistance. However copper remains in the soil and has adverse impacts on beneficial mite species and negative consequence on the biological activity of soil microorganisms (Dumestre et al. 1999). For this incompatibility with organic farming's objective of being sustainable and environmental friendly in 2002 European commission modified the Annex II of Regulation (EEC) No 2092/91 limiting the allowed dosage of copper to 6 kg/ha/year, since 2006. The aim of our work is the evaluation of new organic compounds to replace/reduce copper in order to control P. viticola in organic viticulture.

Material and methods

Four efficacy trials were conducted from 2004 to 2007 in the experimental organic farm of the Istituto Agrario di S. Michele all’Adige, in Navicello (Rovereto, Italy) with natural infection of Plasmopara viticola. The vineyard is planted with cv. Cabernet Sauvignon grafted on Kober 5BB and pergola trentina is the trellis system. The experimental design was fully randomized blocks with four replicates with at least seven plants each. Twenty nine

87 88

different products (Table 1) divided in new copper formulations (CuB), biocontrol agents (MO), plant extracts (PLT), clay (CLA), and ingredients form other origins (OTH) were tested. Plots treated with copper hydroxide and untreated plots ware used respectively as standard and untreated control. The products were weekly applied from May to August with an experimental atomizer using 12 hl/ha of spray solution. Assessments were carried out before harvest scoring 50 randomly chosen leaves and 50 bunches for each replicate. For each replicate the percentage of infected leaf or berries surface (severity) and the percentage of infected leaves or bunches (incidence) were assessed. Data were transformed in percentage of efficacy, normalized and statistically analysed with ANOVA and Dunnett’s test (α=0.05) using SPSS software.

Table 1. Commercial name or code, active ingredient and dosage (v/v or w/v of commercial product) of the products tested in field trials for downy mildew control. REF: standard reference; CuB: copper based fungicide; MO: microorganism; PLT: plant extract or derivate, algae; CLA: clay; OTH: other origins.

Active ingredient Commercial name Dosage (%) CuB Copper gluconate Labicuper 0.3 CuB Cu peptidate Naturam 5 0.4 MO Trichoderma harzianum T39 Trichodex 0.4 MO Clonostacys rosea + T. harzianum Clonotri 0.8 MO Bacillus subtilis Serenade 0.4 MO Pseudomonas euruginea Agat-25k 1 PLT Salix alba extract Salix 2 PLT Yucca shidigera extract Saponin 1 PLT Plant based alcohol extract Elot-Vis 0.5 PLT Melaleuca alternifolia Timorex 1 PLT Plant extract BP-2 0.01 PLT Plant extract BP-3 0.04 PLT Plant extract SA 5 PLT Abies sibirica extract Novosil 1 PLT Plant extract CG 0.2 PLT Melaleuca alternifolia BM-608 0.5 PLT Quillaja saponaria Quiponin 5 PLT Plant based oil oil 0.3 PLT Camellia oleifera + Chenopodium Teawet 5 quinoa PLT Algae extract + Boron Labimar 0.8 CLA Acid clay Mycosin 0.5 OTH K phosphonate Fosfidor 0.3 OTH aminobutirric acid BABA 0.1 OTH Potassium salt of fatty acid Tecnobiol 1 OTH Chiosane Chitoplant 0.1 OTH Lactoperoxidase KBV 99-01 0.15 OTH Experimental WBI-1 0.05 OTH Experimental BPO 30 0.1 REF Cu(OH)2 Kocide 2000 0.075/0.125/0.175* *dosages varied during the growing season proportionally to the risk of infection

Results and discussion

The presence of downy mildew symptoms on leaves was very high except in 2005 and thus the ability of the tested products to control the disease was stressed. The best control of the disease on leaves was achieved with Labicuper, which contains low dosage of copper. Fosfidor and Naturam 5 gave good results too, but they are respectively not allowed in organic viticulture and phytotoxic. Other nine products suppressed downy mildew as copper. Three of them are plant extracts (Salix extract, Saponin and Elot-vis), three are microorganisms (Agat 25K, Clonotri and Trichodex), one is an acid clay (Mycosin), one is an animal derivate (Chitoplant) and last one is a product based on fatty acid (Tecnobiol). The remaining products did not control the disease as copper, even if seven of them showed some activity against downy mildew (Oil, WB-1, SA, Novosil, Serenade, Timorex, BABA) (Figure 1).

120

100

20 Efficacy on leaf severity (%)*

SO SA CG OIL BP-3 BP-2 BABA WBI-1 SALIX BPO 30 Cu(OH)2 AGAT-25 TEAWET NOVOSIL KBV-99-01 LABIMAR SAPONIN TIMOREX MYCOSIN ELOT-VIS QUIPONIN BM 608 0.75 SERENADE CLONOTRI FOSFIDOR TECNOBIOL LABICUPER TRICHODEX NATURAM 5 UNTREATED CHITOPLANT

UNTREATEDNot different from untreated DifferentSERENADE from untreated and Cu(OH)2 NotCu different(OH)2 from Cu(OH)2 *related to copper hydroxide efficacy (=100%)

Fig. 1. Efficacy of the tested products on leaf severity. Efficacy was calculated on the normalised untreated control and related to Cu(OH)2 efficacy (=100%). White bars indicate no significant difference from untreated, grey bars significant difference from Cu(OH)2 and untreated and black bars no statistically difference from Cu(OH)2 with Dunnett’s test (α=0.05).

The disease on bunches was very low in 2005, but high in 2004, 2006 and 2007. As on leaves, a very good suppression of the disease was achieved only with the formulations based on low dosage of copper (Labicuper and Naturam) or with Fosfidor. Several compounds gave a good control of symptoms on bunches. Nine plant based products, two biocontrol agents and five other compounds were as effective as copper. All the remaining products had some effect (different from the untreated), but lower than copper (Figure 2). The results suggest that at the moment a complete replacement of copper fungicide is not possible, but formulations containing low dosages of copper are available and they could be used to obtain an acceptable control of the disease with a lower copper input in the environment. Many existing commercial and experimental products, as plant extracts, BCAs and other compounds, were effective against the disease, but they need an improvement of the 90

formulation, optimization of dosages and application time to reach a commercially acceptable result.

120

100

Efficacy on bunch incidence (%)*

SA SO CG BP-2 BP-3 TEAWET BABA WBI-1 SALIX BPO 30 Cu(OH)2 AGAT-25 NOVOSIL SAPONIN TIMOREX KBV-99-01 ELOT-VIS MYCOSIN LABIMAR FOSFIDOR QUIPONIN BM 608 0.75 CLONOTRI SERENADE NATURAM 5 LABICUPER TECNOBIOL TRICHODEX UNTREATED ORANGE OIL CHITOPLANT

Not different from untreated Different from untreated and Cu(OH)2 Not different from Cu(OH)2 *related to copper hydroxide efficacy (=100%)

Fig. 2. Efficacy of tested products on bunch incidence. Efficacy was calculated on the normalised untreated control and related to Cu(OH)2 efficacy (=100%). White bars indicate no significant difference from untreated, grey bars significant difference from Cu(OH)2 and untreated and black bars no statistically difference from Cu(OH)2 with Dunnett’s test (α=0.05).

Acknowledgements

This research was supported by REPCO (Project No. 501452, 6th FP, priority 8.1), founded by the European Commission.

References

Gobbin, D., Jermini, M., Loskill, B., Pertot, I., Raynal, M. & Gessler, C. 2004: La ridefinizione del ciclo epidemiologico della peronospora della vite. – Informatore Fitopatologico 4: 12-15. Dumestre, A., Sauvé, S., McBride, M., Baveye, P. & Berthelin, J. 1999: Copper Speciation and Microbial Activity in Long-Term Contaminated Soils. – Archives of Environmental Contamination and Toxicology 36(2): 124-131.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 91-94

A multivariate analysis of combined effects of (micro)climate, vegetative and reproductive growth on grey mould incidence in grapevine

Fermaud, M.1, Valdés-Gómez, H.2, Calonnec, A.1, Roudet, J.1, Gary, C.3 1 INRA, UMR Santé Végétale, ISVV, 33883 Villenave d’Ornon, France; 2 Bioinformática- CITRA, Univ. Talca, Chile; 3 INRA, UMR SYSTEM, 34360 Montpellier, France

Abstract: Over three years (2004-06), a field experiment was carried out near Montpellier (southern France) to investigate the relationships between grey mould expression at harvest and some of the major factors affecting the disease development in vineyards, i.e. (micro)climate, fruit composition and vine vegetative and reproductive growth. By implementing irrigation and cover cropping practices, various levels of vine growth were generated and led to different levels of disease development. Disease incidence was correlated positively to key variables of vine vegetative growth: total leaf number, leaf dry matter, leaf layer number, leaf area per m of row, pruning mass and nitrogen accumulation. These relationships were established in the context of an interaction between (micro)climate and grapevine vegetative growth. In 2004, under conducive weather conditions, B. cinerea developed in all experimental plots. Favourable (micro)climatic variables were precipitation, duration of relative humidity > 90% in the canopy and low potential evapotranspiration. However, in 2005 and 2006, under dry summer conditions, disease developed only in the most vigorous vines which were both irrigated and fertilized. These vines showed a very high canopy growth associated with compact clusters and delayed fruit maturity.

Key words: Vitis vinifera, vegetative vigour, vine capacity, Botrytis cinerea, bunch compactness

Introduction

Many actors in viticulture recognize the importance of potential quantitative relationships between susceptibility to fungal pathogens and grapevine growth or vigour (Goulet et al., 2006). However, these relationships have been poorly documented in the literature. Some studies of grape quality have, incidentally, shown interactions between grey mould development and vegetative and reproductive growth patterns in grapevine (Reynolds and Wardle, 1994; Morlat and Bodin, 2006). Because of the complex relationships between B. cinerea development, canopy size, microclimate, morphological and physiological features of clusters and berries (Vail and Marois, 1991), our main objective was to find out which main variables characterizing canopy microclimate, grapevine vegetative and reproductive growth and fruit composition can be associated and best describe grey mould expression at harvest (Valdés-Gómez, 2007).

Material and methods

The 3-year field experiment was conducted from 2004 to 2006 in a commercial vineyard (cv. Aranel, white variety, grafted on Fercal rootstock) located near Montpellier (43°31' N-3°51' E) in the Mediterranean area. The vines were planted in 1997 at 2.5 m inter-row and 1.2 m intra-row spacing and were trained with a midwire bilateral cordon system and a canopy

91 92

height of 1.0 to 1.1 m (rows oriented from NW to SE). No fungicide was applied to control B. cinerea during the experiment. In order to create different vine growth levels, different types of cropping systems were used: i) chemical weed control all over the soil surface (“W”), ii) perennial cover crop sown with a mixture of tall fescue (Festuca arundinacea Shreb) and ray grass (Lolium perenne L.) in every inter-row (“C”). Weeds growing under the vine rows were controlled with glyphosate. Every treatment was divided into two blocks differing by their slope (1: slope near 0 %, 2: slope near 2 %). In 2005 a new treatment (“I”) was added in block 1, irrigated once a week from budbreak to harvest (3400 m3 ha-1 in 2005, 7400 m3 ha-1 in 2006), fertilized with nitrogen (80 kg N ha-1 in 2005, 120 kg N ha-1 in 2006) and under chemical weed control. Different variables regarding vine vegetative and reproductive growth were measured including total leaf number, leaf dry matter, leaf layer number, leaf area per m of row, pruning mass, nitrogen accumulation, vine yield, mean bunch mass and fruit composition (sugar concentration, titratable acidity). Some key canopy microclimate variables, particularly canopy air temperature (Tc) and canopy air relative humidity (RHc), were collected between veraison and harvest. At maturity stage, 180-200 bunches per sub-plot (sub-plot = cropping system x block) were scored individually and visually to assess grey mold incidence.

Results and discussion

Climatic and microclimatic conditions and grey mould development Under our experimental conditions, the primary risk factor leading to grey mould development was identified as the climatic and microclimatic conditions. From veraison to harvest, the rainfall (47 mm) was higher in 2004, increasing disease incidence, than in 2005 and 2006 (less than 20 mm). In 2004, within each block, grey mould incidence differed significantly between the two cropping systems. The perennial cover crop treatment led to, approximately, four times less disease compared with the chemical weed control treatment (Fig. 1). In 2005 and 2006 which were less conducive to grey mould, the disease developed mostly in the irrigated plot (I) with an incidence of 13.9 % and 21.5 %, respectively. A principal component analysis based on climatic and microclimatic variables (data not shown), allowed us to identify five climatic and microclimatic variables of prime importance: precipitation, potential evapotranspiration, predawn leaf water potential (PLWP), relative air humidity higher than 90% and temperature higher than 30°C in the canopy (Fig. 2).

Vine vegetative and reproductive growth and grey mould development In 2005 and 2006 under dry summer conditions, the disease developed mostly in the highly vigorous vines which were both irrigated and fertilized (Fig.1). This evidenced that unfavourable climatic conditions for the disease development can be counterbalanced by conditions of high vine growth and associated canopy and cluster features. The following key variables of vine vegetative growth were highly and positively correlated with the final disease incidence (Fig. 2): total leaf number per shoot, leaf layer number, leaf area per m. of row (m2/m), total dry matter (kg/vine), pruning mass (kg/m), nitrogen accumulation in dry matter (gN/vine). The importance of such variables linked to vine vigour, foliage density and/or cane pruning mass was already stressed under production conditions (Morlat and Bodin, 2006). As regards yield components and cluster architecture, bunch mass (g) was correlated positively and significantly (P = 0.01) with grey mould incidence (Fig. 2). This variable has been shown to make the largest contribution to cluster compactness among various cluster measurements and can be then considered as a key morphological feature increasing B. cinerea infection and mycelial colonisation within the grape cluster (Vail and Marois, 1991). Lastly, 93

concerning fruit composition, disease incidence was significantly and negatively correlated with the level of fruit maturity, as indicated for example by the SC/TA ratio (sugar concentration / titrable acidity) (Fig. 2).

0.5 2004 2005 2006 2005 ar. 0.4 (%)

0.3

0.2

0.1 Botrytis incidence Botrytis

0.0 C1 C2 W1 W2 I Cropping system

Fig. 1. Final grey mould incidence in function of the cropping system: “W”, chemical weed control treatment and “C”, perennial cover crop treatment (followed by 1 and 2 corresponding to the block number with a slope of 0 % and 2 %, respectively).

50 y = 0.1x - 5.5 y = 6.5x - 13.6 y = 3.2x - 18.4 y = 8.3x - 13.2 y = 28.1x - 15.1 40 2 2 2 R = 0.93** R = 0.90** R = 0.92** R2 = 0.90** R2 = 0.81** 30 20

Botrytis (%) Botrytis 10 0 0 100 200 300 400 0246036912150123450.0 0.5 1.0 1.5 Total leaf number Leaf layer number Leaf area/m Total dry matter Pruning mass 50 y = 28.1x - 15.1 y = 0.8x - 16.4 y = 0.1x - 32.5 y = 0.2x - 34.3 y = -0.6x + 36.2 40 2 2 2 R2 = 0.81** R = 0.95** R = 0.76** R = 0.70* R2 = 0.75* 30 20

Botrytis (%) Botrytis 10 0 0 100 200 300 400 0 25 50 75 100 -1.5 -1.0 -0.5 0.0 010203040500 125 250 375 500 PLWP N accumulation Bunch weight Berry number SC/TA Fig.2. Significant correlations between grey mould incidence (%) and key variables of vine growth, cluster architecture, yield components and fruit composition. Linear regressions include data with grey mould incidence different from 0% only. * Significant at P = 1%; ** at P = 5%. PLWP: predawn leaf water potential, SC/TA: sugar concentration / titratable acidity.

References

Goulet, E., Cady, E., Chrétien, P., Rioux, D. 2006: Grapevine sensitivity to fungal diseases: use of a combination of terroir cartography and parcel survey. – VIth International Terroir Congress: 94-100. Morlat, R., Bodin, F. 2006: Characterization of Viticultural Terroirs using a Simple Field Model Based on Soil Depth II. Validation of the Grape Yield and Berry Quality in the Anjou Vineyard (France). – Plant and Soil 281: 55-69. Reynolds, A.G. and Wardle, D.A. 1994: Impact of training system and vine spacing on vine performance and berry composition of Seyval blanc. – Am. J. Enol. Vitic. 45: 444-451. Vail M.E. and Marois J.J. 1991: Grape cluster architecture and the susceptibility of berries to Botrytis cinerea. – Phytopathology 81: 188-191. Valdés-Gómez, H. 2007. Relations entre états de croissance de la vigne et maladies cryptogamiques sous différentes modalités d’entretien du sol en région méditerranéenne. – PhD thesis, Montpellier SupAgro.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 95-99

Development of monitoring schedules for grape diseases at regional scale

Emma Fulchin, Maarten van Helden ARD-VD, Association for Research and Development in Sustainable Viticulture, 1 cours du Général de Gaulle, 33170 Gradignan, France

Abstract: The ARD-VD is an association of wine growers, and both scientific and professional organisations. The association supports the implementation of integrated pest management and, in some demonstration vineyards, called “Vignobles de Reference®”, different monitoring schedules are put into place. In order to optimise pest management, surveillance is carried out on climatic conditions, phenological growth stages and epidemiological development of insect pests and diseases. However, working at the individual farm scale clearly has its limits, such monitoring systems are time and energy consuming, and difficult to implement by the wine grower alone. So the decision was taken to evolve towards a larger regional scale aiming at the creation of local monitoring networks. This approach should lead to sharing the individual data and operating expenses, and to hiring extension workers dedicated to monitoring. These expenses will be covered by improved pest control and savings in spraying obtained trough the monitoring schedule. Usually, in a “Vignoble de Reference®”, our monitoring system combines a weather station, ‘indicator plants’ for primary contaminations of mildew, and several untreated small plots, focussing on the more sensitive sites. Transposing this protocol from farm scale to regional scale leads to various questions. In such a network the optimal density of measuring points must be adapted to the surface and topography of the region. This allows differentiating the main zones of differing sensitivity to pests and diseases and leads to adapt spraying locally. To obtain some answers and to scientifically validate a monitoring system at this regional scale, a study was carried out this year. We monitored climate (four weather stations available), phenological growth stages, epidemiological development of mildew (ten sites with untreated microplots) and pest insects (40 traps) at the scale of the Buzet wine region. This study revealed that one person is necessary to monitor approximately one thousand hectares (for both insects and diseases). Vigour and microclimate (soil, topography; landscape…) seem the main factors for sensitivity. Focussing the monitoring on ‘fairly’ to ‘very sensitive’ sites appeared the best choice, and the number of monitored sites (ten for diseases, 40 for insects in that region) seemed to be quite well adapted. On the other hand, the number of weather stations was clearly insufficient. We estimate that one weather station is needed per approximately ten square kilometres.

Key words: monitoring, regional scale, network

Introduction

The ARD-VD is an association whose aim is to support the implementation of integrated pest management as part of sustainable viticulture. It is composed of thirty-eight members. All stakeholders of the vine industry are represented: individual vine growers, groups of farmers, such as wine trade unions or cooperative wineries, distributors, agricultural suppliers (of pesticides, weather stations, spraying equipment), technical advisers, research institutes, and universities … Our purposes are to acquire, validate and spread knowledge and techniques needed to implement sustainable pest management. Demonstration vineyards, called “Vignobles de Reference®” (Bugaret, 2003), have been in operation since 2000 in France and Portugal. In these vineyards, we have validated monitoring systems and decision-making

95 96

rules for several diseases (Bugaret & Bessard, 2002; Bugaret & Burosse, 2005; Burosse, 2004; Fulchin, 2005) at farm scale. These monitoring schedules consist of a detailed surveillance of: - Phenological growth stages, particularly during the first stages in spring - Epidemiological development of diseases by means of “indicator plants” for primary contaminations of mildew (specially trimmed vines, figure 1) (Coarer et al., 2005) and of untreated microplots that are covered during pesticide applications - Climatic conditions using a weather station. This allows us to use models for risk level prediction (especially for downy mildew) (Ducros et al., 2005; Rouzet & Jacquin, 1995) and to optimise spraying schedules (first treatment and renewal).

The 3 buds on the soil break earlier than those left on the stock, which favours the exposure of the young leaves to mildew oospores.

Old shoot

Last year’s shoot, bent and fixed on the soil, with 3 buds.

“Inoculum leaves” introduced under the buds.

Fig. 1. The “indicator plant” method (Bugaret, 2001)

The sites chosen for surveillance were those identified as the most sensitive on the farm. Thanks to this monitoring system, we were able to reduce pesticide use by 20 to 30%. However, this device requires a lot of time and energy, and is very difficult to implement by the wine grower alone. At farm scale, it is not profitable for wine growers to pay a technician to carry out this monitoring: the costs are higher than savings made on sprayings. Moreover, the surroundings of the monitored plots are not taken into account. In order both to improve our methods and advice to be given, and to increase profitability for the wine grower, we decided to adopt a larger, regional scale and monitoring network. This implies working with groups of farmers (such as wine trade unions or cooperative wineries) who should collaborate in monitoring and share their individual data to compare different situations for sensitivity and risk. Even with the reduction of a single application, the savings in spraying do cover operating expenses that would be shared between all the wine growers of the network. These savings do also allow paying technicians or taking on workers dedicated to the monitoring (Table1). At the individual farm scale, we normally focus on the more sensitive sites. But transposition to regional scale raises the issue of the optimal density of measuring points: how many monitoring sites are needed, and where should they be placed? Given the scale, we assumed that it was necessary to differentiate risk levels and implement monitoring in “fairly” to “very” sensitive sites. We therefore implemented a monitoring network at regional scale as part of an engineer work experience. 97

Table 1. Calculations of yearly expenses and savings due to monitoring of a 1000 ha network

EXPENSES SAVINGS Salary (1 master level graduate in 35 000 € Average cost of a pesticide 25-30 €/ha agriculture) (diseases or insects) Operating expenses 4 000 € Average cost of 1 application 25-30 €/ha Investment 7 weather stations 6 000 € (driver + equipment) (amortized over 5 years) Average cost of 1 spraying 50-60 €/ha ARD-VD subscription (5€/ha) 5 000 € TOTAL (€) 50 000 € TOTAL (€) 50-60 000€

Material and methods

The monitoring network was implemented by two master level students in agronomy from April to August 2007 (for their end of studies project report) in the South-West of France in the Buzet wine region. This region has a vine surface area of 2000 hectares distributed over 60 square kilometres. We monitored climate, epidemiological development of diseases and insect pests and phenological growth stages, in collaboration with the Buzet wine trade union.

Climate monitoring and modelling Four weather stations were available, but only two were usable. The other two were badly situated. Four parameters were measured and recorded: air temperature, rainfall, air relative humidity and duration of humidity. The weather stations were checked twice a week by the ARD-VD. These parameters were then used for modelling the risk level of downy mildew using the POM (Prediction of the Optimum of Maturity of oospores) and EPI (Potential Infectious State) models created by the INRA (National Institute of Agronomic Research) (Tran Manh Sung et al., 1990).

Disease monitoring This year’s climatic conditions clearly favoured the development of downy mildew in France. Hence our study focussed mainly on that disease. Ten plots were chosen throughout the wine region in “fairly” and “very” sensitive sites. The risk level was estimated by the wine growers according to the history of the plots. On each site monitored, two untreated microplots, covered during sprayings, were created in order to estimate mildew pressure and the sensitivity level of the site. Two other (treated) microplots were also defined so as to compare with the untreated ones and evaluate the efficiency of sprayings. These four microplots were observed once a week and symptoms on leaves and berries were noted by vine growth stages.

Insects monitoring Forty ‘Tri ∆nglué’ traps (Van Helden et al., 2008) for grape berry moths (Lobesia botrana and Eupoecilia ambiguella) and leafhoppers (Empoasca vitis, Scaphoideus titanus) were set up and observed twice a week: once by the students (every Monday) and once by the wine growers (every Thursday). Larvae were counted 3 weeks after trapping peaks.

Results and discussion

The exceptionally strong downy mildew pressure in 2007 made it harder than in the previous years to differentiate various sensitivity levels. However we observed that the monitored plots had different sensitivity levels to mildew. This must be confirmed by future studies. The 98

choice of the two sensitivity levels “fairly” to “very” sensitive seemed appropriate but will also have to be reiterated later, especially during years with lower mildew pressure so as to confirm the observed sensitivity differences. But despite the observation of these differences, the sprayings could not be adapted locally this year: the slightest failure in protection would have had serious consequences. “Indicator plants” for primary contaminations of mildew could not be used in this study. Such plants need to be prepared in autumn/winter of the previous year (collection of mildew inoculums, pruning) and the project was launched only in the spring. Indicator plants would have been the only way to adapt spraying locally this year by delaying or bringing forward the first application. One of the main improvements for future years’ monitoring will be to include a network of “indicator plants” throughout the region, and especially to have one of these plants in each untreated microplot to monitor the first mildew cycles. 10 plots and 40 traps for diseases and insect pest monitoring seemed well adapted for the Buzet region, allowing differentiation of areas of different sensitivity. On the contrary, the number of weather stations was clearly insufficient. After meeting with experts on the subject, we estimate that one weather station is needed for approximately ten square kilometres. Field observations required two days for the two students; data analysis required half a day and data capture with a GIS (Geographical Information System) two days. This led us to estimate that approximately one person is needed to monitor 1000 hectares. When comparing disease data with detailed plot information, several factors seemed to play a major role. Soil is clearly a factor of sensitivity, possibly through its influence on bud burst, vigour and physiology. In this region, three main kinds of soil are encountered, from sandy to limey clay. Vigour appeared to be a main factor of risk. Hence, monitoring one site for each combination soil-vigour might be a good way to differentiate sensitivity levels at regional scale. In the example of the Buzet wine region, three types of soil combined with three vigour levels (not very, fairly and very vigorous) would lead to nine different situations to monitor, which corresponds to the number of sites that seems appropriate. This appears to be a worthwhile path to explore.

References

Bugaret, Y. 2001. Une méthode pour détecter les foyers primaires. – Phytoma, la défense des végétaux, n°536: 46-48. Bugaret, Y. 2003. Vignobles de référence. – VITI (supplément), n°286, 23 pages. Bugaret, Y.; Bessard, S. 2002. Une démarche de protection raisonnée contre le mildiou. – Phytoma, la défense des végétaux, n°548: 27-32. Bugaret, Y.; Burosse, L. 2005. Prise de décision en protection raisonnée. – Phytoma, la défense des végétaux, n°583, tiré à part: 2-7. Burosse, L. 2004. Formalisation et validation de règles de décision en protection raisonnée du vignoble. Etude de leur applicabilité. – Bordeaux: ENITA de Bordeaux, 57 pages. Coarer, M.; Raynal, M.; Serrano, E.; Martinez, F. 2005. Mildiou: les contaminations primaires sont-elles d’importance secondaire? – Union girondine des vins de Bordeaux, n°1012: 33-36. Ducros, Y.; Custelin, H.; Monrozies, L.; Lassort, C. 2005. L’agrométéorologie au service de la lutte raisonnée. – Phytoma, la défense des végétaux, n°578: 6-7. Fulchin, E. 2005. Formalisation et validation de règles de décision en protection raisonnée de la vigne. Identification des freins à la protection raisonnée. – Bordeaux: ENITA de Bordeaux, 71 pp. 99

Rouzet, J.; Jacquin, D. 1995. Modèles de prévision et lutte raisonnée. – Phytoma, la défense des végétaux, n°475 : 29-32. Tran Manh Sung, C.; Strizyk, S.; Clerjeau, M. 1990. Simulation of the date of maturity of Plasmopara viticola oospores to predict the severity of primary infections in grapevine. – Plant disease 74 (2): 120-124. Van Helden, M.; Pain, G.; Pithon, J. 2008. Landscape characteristics influencing pest populations in viticulture. –IOBC/wprs Bulletin 36: 369-373.

100 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 101-105

Control of Blackrot (Guignardia bidwellii) on the hybrid vitis cultivar Isabella

Cesare Gessler1,2, Flavio Foiada2, Mauro Jermini 3 and Ilaria Pertot1 1 SafeCrop c/o Istituto Agrario 38010 San Michele a/Adige TN, Italy 2 Swiss Federal Institute of Technology, 8092 ETH-Zürich, Switzerland 3 Agroscope ACW Changins, Centro di ricerca Cadenazzo, 6594 Contone, Switzerland

Abstract: Blackrot caused by Guignardia bidwelli is causing damages in recuperated vineyards in southern Switzerland planted with the cultivar Isabella. The source of inoculum is attributed to abandoned vineyards. A 5 year experiment was made to test and confirm minimal necessary fungicide application and timing to control Blackrot. Two to three well timed sprays with an appropriate fungicide are sufficient to control fully the disease. Timing should be before rain events leading to prolonged leaf wetness during the time period between flowering and six week afterwards. Heavy and prolonged rains up to 4 weeks after flowering may also favor infection by Plasmopara viticola on Isabella especially if a heavy load of primary inoculum can be expected. We recommend therefore to use fungicides or fungicide combinations with an effect on both pathogens.

Key words: Grape disease, down mildew, pathogen control

Introduction

Blackrot is a well known grape disease in North America, often it is described as the most devastating disease in the north east where cultivars such as Concord are popular (Kuo & Hoch, 1996). In Europe it is of considered of marginal importance except in a few areas with mechanical pruning and harvesting. Recently Harms et al., 2005 report on great economic losses in some vine-growing areas in Rhineland-Palatinate, Germany during 2003 and 2004 due to attacks of Blackrot. The extension of the damage came as a surprise, it was favoured by suitable climatic conditions and a large number of overgrown, recently abandoned vineyards in which the inoculum was able to build up and than spread with wind during rain periods (personal communication Harms). However disease control was possible with dithiocarbamates, strobilurins and azoles treatments. In southern Switzerland (Canton of Ticino) black rot was first discovered in 1988 (Pezet & Jermini, 1989) and its epidemiology on the dominant cultivar Merlot has been intensively studied (Jermini & Gessler, 1996). The authors concluded that if proper sanitation methods were applied, as it is the rule, the pathogen is no treat. In fact no case of loss has been known nor any grower adopted special measures for control of black rot. The disease is unknown or anecdotic among the vinegrowers. Ticino is a region with a high touristic vocation, the mountainsides of its valleys are steep and the bottoms highly anthropicised since middle age with small villages once mostly dedicated to agriculture. In Ticino in the last 100 years the grape cultivated areas dropped from over 6000 ha to about 1000, the loss was mostly concentrated in the steep side of the valleys. Since a few years efforts are made to revive the old terrace vineyards in these valleys mostly planted with hybrid cultivars among which Isabella dominates. Isabella is thought to be derived from a N. America native Vitis labrusca grape and an unknown European Vitis vinifera and has a characteristic"grapey/foxy" taste and flavour.

101 102

Even if some wine is produced for local market, the wine is mostly home consumed and the distillation product (grappa) is of priced quality. The presence of the American hybrids in the valleys has a long tradition mostly dating back to the Phylloxera and Plasmopara crisis. Isabella has a high resistance to powdery mildew, can be grown on its own roots, is vigorous sufficiently to be grown in the pergola system and finally sufficient resistant to downy mildew to allow a mostly healthy crop, whereas the Vitis vinifera cultivar Merlot, which currently occupies 90% of the grape surface, mostly in the flat areas, needs an intensive control program against Plasmopara viticola. Isabella in almost all years shows and probably always did show, starting at earliest in July, more often only in August, typical downy mildew symptoms on leaves (Gessler et al., 2006). Since a few years however owners report also black dried berries from early August on, in some years with relevant losses up to complete loss if untreated. In dry late spring – summer years such as 2003 no symptoms were reported. In the wet year 2002 untreated vineyards had almost complete loss. Some owners reported losses even if regular treatments against downy mildew were made (up to 6 “Saturday morning” sprays). As causal agent, against the general assumption which attributed all symptoms to P. viticola, we identified Guignardia bidwellii (Ellis) Viala & Ravaz (anamorph Phyllosticta ampelicida), being the symptoms typical (Kuo & Hoch 1996) without any doubt and the presence of perithecia and pycnidia easy verifiable (Gessler et al., 2006). From the literature (Ferrin & Ramsdell, 1977; Hoffman & Wilcox, 2002; Gallet, 1977), we have to deduce that Blackrot problems can be traced to particular circumstances such as i. buildup of inoculum in abandoned vineyards or in vineyards with mechanical harvest and pruning; ii. extremely favorable weather conditions for distribution and infection (Shaw et al., 1989; Spotts, 1977), usually also conducive to P. viticola and iii. the use of oomycete specific fungicides without side effect on Ascomycetes during the period of 4-6 weeks after flowering. Consequently it was necessary to develop and implement a control strategy with the minimal number, but correctly timed and targeted fungicide sprays. We used therefore a vineyard representing the average conditions, compromising with the scientific requirements of identicality of all plots and statistically sufficient repeats. Because the weather pattern in this region changes extremely from year to year during the relevant period, we did not attempt to consider experiments in different years as repeats.

Materials and methods

The various assays described were made in a terraced vineyard with ca 120 plants on ca 800 m2 of the cultivar Isabella trained as pergola of various age (Gessler et al., 2006). The fungicide used was Slick (25% Difenoconazol). Spraying was done by a portable air blower at the dosage recommended by the manufacturer. Treatments were schedule to be made from flowering on before a forecasted rain event with leaf wetness over 12 h or as soon as possible afterwards, treatments were allowed to be made only on weekends and at least at 12 days interval. Six week after full flowering berries were considered to be resistant to infection so no treatments were schedule afterwards. (see also Gessler et al., 2006) In 2006 the area was subdivived into 30 plots each of 4-6 vines, 12 plots treated twice, 6 only the first, other 6 only the second treatment and 6 without treatment. At harvest all bunches were counted and weighted. Severity was estimated as in 2005. Statistical analysis was made with SPSS-9 using ANOVA and Sheffe’s post hoc test for comparison between two treatments for all date collected in 2006.

103

140

120

100 2006 80 60 40

0 1 8 15 22 29 5 12 19 26 3 10 17 24 31 7 14 21 28

Fig. 1. Daily rain in mm (columns), flowering time of the grape cultivar Isabella and date of the fungicide treatments (arrows). Time span first of May until 30 July August for 2006.

Results and discussion

In 2006 the first rains were forecasted and occurred for the Saturday evening 24 of June, therefore the first treatment was made in the morning 24 using Slick. Even if it was a light thunderstorm, leaves remained wetted for over 12 h. The next 4 days presented evening thunderstorms. Considering the protection period of 12 days and the correctly forecasted heavy rain (over 115 mm) period starting the Wednesday 5 July evening until Friday, a second treatment was made on the following Saturday 8 of July again with Slick. No other treatments were made (Fig. 1). The year 2006 had the typical weather pattern of the region, few but intensive rains with prolonged leaf wetness. A high inoculum load was present derived from the last year control plot, probably also a high P. viticola oospore inoculum was present, as in fall 2005 most leaves were covered with the pathogen sporulation. The plots without protection showed already early July Blackrot symptoms which increased until early August, the treated plots were completely free. However end of July in all plots reddish shriveling berries appeared, ending in August with a high % of dried mummy berries. The symptoms in the twice treated plots, mostly in the only once treated plots and over half in the untreated plots were unlike Blackrot symptoms but corresponded well to late infection symptoms of P. viticola. At harvest, the dark brown mummified berries in the twice treated plots presented no pycnidia, contrary in the unsprayed control plot mummified berries appearing slightly more black violet presented pycnidias and were estimated to be about half of all damaged berries. We attribute therefore the present berry symptoms in the twice treated plot fully to down mildew (Fig. 2). The values of twice treated and the control plots are statistically different at p = 0.009. So the two slick treatments were able to fully control Blackrot which therefore can be estimated to have a severity of ca 20% in untreated plots. Average bunch weight (Fig. 2) was clearly lower as in 2005 (Gessler et al., 2006), partially due to lack of pruning out of bunches, but also due to P. viticola damage however differences, even if very evident at harvest, are not significant (p = 0154) as variability in average bunch weight per plot is very high. We conclude i- that the heavy rains in July allowed a relevant infection of the berries by P. viticola probably by primary inoculum as no lesions were visible at that time on the leaves; ii the two rain events with prolonged leaf wetness where both favorable infection by G. bidwellii, however the preventive (first treatment) as well as the curative (second treatment) were able to avoid these infections; iii- Blackrot can be fully controlled if the grapes are protected during rain/leaf wetness periods starting from flowering for about 6 weeks. 104

80 70 60 50 40 30 20 10 0 no First Second First + treatment second

Fig.2. Mean disease severity (left column) in % diseased grape berries at harvest (1 of October) and average bunch weight in gr (right column) in 2006. Standard deviation calculated from 6 plots with no treatment, 6 plots with a single early and 6 plots with a single late Slick treatment and 12 plots with both treatments

The results of this study and the previous published results aquired in the same vineyard (Gessler et al., 2006) confirm the conclusions drawn from experiments on the related cultivar Concord that berry susceptibility is maximal from bloom until 2-4 weeks after (Hoffman et al., 2002) as well as the findings that applications immediately prior to bloom plus 2 and 4 weeks later, provided virtually complete control (Hoffmann et al., 2004). This study also demonstrates that with placing the sprays just before or even in a rainy period and renouncing to treatments if leaf wetting periods are short (less than 12 h) the disease can be well controlled with a minimal number of sprays. We recommend therefore considering the application of fungicides effective against black rot during and up to four weeks after flowering if weather forecast previews rains which can lead to leaf wetness over 12 h. If such a period is not covered by an application within the last 10-12 days, an application should be made at the first possibility, e.g. the grapes bunches are dry using in any case a curative product. However to limit the problem it is important to remind that sanitation is the most effective control method. In vineyards with hand pruning and hand harvesting and consequent elimination of hanging mummified berries, the risk is represented by the inoculum deriving from abandoned vines. However to our knowledge, no report indicates minimal distances of abandoned vineyards to exclude risk for well kept vineyards, however from distribution pattern of the disease we infer that ascospores can travel at least a few hundred meters with the wind direction during rain events. So it seams imperative to eliminate abandoned vines. Moreover we infer from the results in 2006 that P. viticola can cause major damage also to Isabella if heavy rainfall with prolonged wetting periods overlap with the partial susceptibility of the berries from flowering until pea size and with the presence of a high inoculum load derived from a high leaf infection late in the past season. We recommend therefore to use a fungicide mixture with activity against both G. bidwelli and P. viticola.

105

References

Ferrin, D.M. & Ramsdell, D.C. 1977: Ascospore dispersal and infection of grapes by Guignardia bidwellii, the causal agent of grape black rot disease. – Phytopathology. 67: 1501-1505. Galet, P. 1977: Les maladies et les parasites de la vigne. Tome 1: 223-260. Gessler, C., Blaise Ph. & Jermini, M. 2006: Blackrot on the hybrid vitis cultivar Isabella. – IOBC/wprs Bulletin 29 (11): 95-102. Harms, M., Holz, B., Hoffmann, C., Lipps, H.P. & Silvanus, W. 2005: Occurrence of Guignardia bidwellii, the causal fungus of black rot on grapevine, in the vine-growing areas of Rhineland-Palatinate, Germany. – In: Abstract Plant-protection and plant-health in Europe: introduction and spread of invasive species. Humboldt-University, Berlin, Germany, 9-11 June 2005: 127-132. Hoffman, J.L.E. & Wilcox, W.F. 2002: Utilizing epidemiological investigations to optimize management of grape black rot. – Phytopathology 92: 676-680. Hoffman, J.L.E; Wilcox, W.F., Gadoury, D.A. & Seem, R.C. 2002: Influence of grape berry age on susceptibility to Guignardia bidwellii and its incubation period length. – Phytopathology. 92: 1068-1076. Hoffman, J.L.E.; Wilcox, W.F., Gadoury, D.M., Seem, R.C. & Riegel, D.G. 2004: Integrated control of grape black rot: Influence of host phenology, inoculum availability, sanitation, and spray timing. – Phytopathology. 94: 641-650. Jermini, M. & Gessler, C. 1996: Epidemiology and control of grape black rot in southern Switzerland. – Plant disease. 80: 322-325. Kuo, K.C. & Hoch, H.C. 1996: The parasitic relationship between Phyllosticta ampelicida and Vitis vinifera. – Mycologia. 88: 626-634. Pezet, R. & Jermini, M. 1989: Le Black rot de la vigne : symptômes, épidémiologie et lutte. – Revue suisse de Vitic. Arboric. Hortic. 21: 27-34. Shaw, B.D., Kuo-K.C. & Hoch, H.C. 1998: Germination and appressorium development of Phyllosticta ampelicida pycnidiospores. – Mycologia. 90: 258-268. Spotts, R.A. 1977: Effect of leaf wetness duration and temperature on the infectivity of Guignardia bidwellii on grape leaves. – Phytopathology 67: 1378-1381.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 106

Sloping ice from the wings of airplanes – A potential cause for locally limited Plasmopara viticola primary infections?

Georg K. Hill DLR R-N-H, Wormser Str. 111, D-55276 Oppenheim, Germany

Abstract: Described is a case of a very severe Plasmopara viticola primary infection on a distinct patch of 50x30 m in a large vineyard in Rheinhessen, Germany. Neither meteorological nor soil data provided any explanation for this restricted phenomenon. Windblown inoculum from more distant regions was not available due to the early date of infection just at the beginning of the growth season. Since the region is close to international airports, the hypothesis of the impact of ice layers from descending aircraft as a exceptional source of splash potential is discussed.

Key words: Plasmopara viticola, primary infection, splash

106 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 107-111

Flavescence dorée and Scaphoideus titanus: distribution and control in Switzerland

Mauro Jermini1, Michele Gusberti1, Lukas Schaub2, Christian Linder2, Paul Gugerli2, Santiago Schärer2, Patrik Kehrli2, Luigi Colombi3, Silvia Bellion4, Stefan Emery5

1 Station de recherche Agroscope Changins-Wädenswil ACW Centro di Cadenazzo, 6594 Contone, Svizzera; 2 Station de recherche Agroscope Changins-Wädenswil ACW, CP 1012, 1260 Nyon 1, Suisse; 3 Servizio fitosanitario cantonale, Viale Fransscini 17, 6501 Bellinzona, Svizzera;4 Syngenta, Suisse; 5 Service de l’agriculture, Office d’agro-écologie, CP437, 1951 Châteauneuf, Suisse.

Abstract: In 1967, Scaphoideus titanus, the vector of the flavescence dorée, was found for the first time in Switzerland. Two decades latter, this leafhopper has colonised all wine-growing areas of the Ticino. In 2004, the first case of flavescence dorée arose in the south of the Ticino. In the following two years, the disease spread to two other winegrowing areas of the Ticino, this probably due to human activities. Today, flavescence dorée is still restricted to the Ticino but its vector S. titanus is also found along the Lake of Geneva as a national monitoring programme revealed in 2006. Since flavescence dorée is a quarantine disease, control measures are mandatory. The only way to control the disease is the application of insecticides against its vectors. The mandatory insecticide control is based on a first application of buprofezin at the appearance of the third larval stadium followed by a second treatment two weeks later. Over the past three years, this control strategy was very effective. In organic viticulture, Parexan N (pyrethrum + sesame oil) is the only insecticide showing a satisfactory control of S. titanus. We recommend three applications of Parexan N at an interval of 10 days with the first application targeting hatching leafhoppers. In conclusion, the mandatory as well as the recommended organic control strategy are both effective and maintain S. titanus population at low level.

Key words: leafhopper, Scaphoideus titanus, Flavescence dorée, mandatory insecticide control, organic insecticide

Introduction

Flavescence dorée (FD) is the most important grapevine yellows disease in Europe. It is caused by a phytoplasma, which is transmitted by the leafhopper Scaphoideus titanus. In Switzerland, S. titanus has been observed for the first time in the Ticino, Swiss canton south of the Alps, in 1967 (Baggiolini et al, 1968). After the discovery of yellows disease symptoms in Ticino in 1990, a national monitoring program was launched (Cazelles et al., 1992). It revealed that the vector was present in all southern winegrowing areas of the Ticino (Jermini et al., 1992). In 1996, S. titanus was discovered for the first time North of the Alps, e.g., in the area around Geneva ; Clerc et al., 1997; Linder and Jermini, 1999). In 2004, FD has been discovered for the first time in a Ticino, respectively, Swiss vineyard (Gugerli et al., 2006). The discovery of this quarantine disease was accompanied by mandatory insecticide treatments against its vector. In 2005, 350 ha of grapevines were treated against S. titanus and in the following year 366 ha. Currently, FD is absent north of the Alps and bois noir is the only phytoplasma present.

107 108

The aim of this paper is to present the actual distribution of the FD and its vector S. titanus in Switzerland, to summarise the results of our mandatory insecticide control and to propose a control strategy in line with organic farming.

Material and methods

Distribution of flavescence dorée and S. titanus in Swiss vineyard By PCR, suspicious vines are analysed on the presence of flavescence dorée since the early nineties. To observe the presence of leafhoppers, a national monitoring program was launched in August 2006. In the north of Switzerland, a total of 124 plots were sampled. Per plot, 40 plants were screened using a beating tray method. Additional observations were carried out in 2007.

Efficacy of the mandatory insecticide control strategy against S. titanus Since 1991, we have been developing a pest control scheme of minimal ecological impact, which is in accordance with Swiss IP guidelines. After the discovery of the first FD infection in 2005, the control strategy developed has been launched. It comprises: 1. A first buprofezin application (0.075% concentration) at the peak of the first larval stadium, corresponding, generally, to the first appearance of third larval stadium. 2. A second buprofezin application 15 days later. 3. Based on a visual assessment of leafhopper abundance, a third insecticide treatment with chlorpyriphos-ethyl or chlorpyriphos-methyl can be requested. It targets nymphs of the fourth to fifth larval stadium as well as adults. The beating tray method and yellow sticky traps were used to follow the phenology of S. titanus and to calculate the efficacy of the mandatory control strategy (Jermini and Baillod, 1996). Based on this data, the official, plant protection organisation of the Ticino communicated the application date of the first buprofezin and the necessity of a third insecticide treatment.

Field trials of organic insecticide against S. titanus In 2006 and 2007, test trials were conducted in a Merlot vineyard in the Magadino. This vineyard lied outside of the area under mandatory insecticide control. The organic insecticides used in this unrepeated plot trial are summarised in Table 1.

Table 1. Product, concentration and date of application of insecticides tested in field trials.

Year Insecticide (active ingredient) Concentration Date of application used Parexan N (pyrethrum + sesame oil) 0.1% June 6, 16 and 28 Audienz (spinosad) 0.03% June 6, 16 and 28 2006 Applaud (buprofezin) 0.075% 6 and 21 June Parexan N (pyrethrum + sesame oil) 0.1% May 22 and June 8 Huile blanche (mineral oil) 2% April 6 Huile blanche (mineral oil) and 2% April 6 Parexan N (pyrethrum + sesame oil) 0.1% May 22 June 8 and 2007 July 26 Applaud (buprofezin) 0.075% May 22 June 8 and July 26 109

In 2006, application dates corresponded to the mandatory control. In the following year, we adapted the application regime. First, we increased the interval between Parexan N applications and second, we studied the impact of mineral oil on hatching eggs.

Results and discussion

Distribution of flavescence dorée and S. titanus in Swiss vineyard

Fig. 1. Distribution of S. titanus North of the Alps (black circles = confirmed habitats in 2006; white circles = confirmed habitats in 2007) and flavescence dorée in the Ticino (black squares = FD-infected areas in 2004; grey squares = FD-infected areas in 2005; white squares = FD-infected areas in 2006).

In 2006, our national monitoring program revealed that S. titanus is present North of the Alps (Fig. 1). It was found in two separated regions along the Lake of Geneva, e.g., around Geneva and in the west of Lausanne (=Lavaux). However, S. titanus was not discovered in the vineyards between these two regions. This suggests human-induced dispersal. In 2007, we observed an extension of the colonized area as a result of leafhopper migration. FD is present in Switzerland, respectively, the Ticino since 2004 (Gugerli et al., 2006). It was first detected in three vineyards of the Mendrisiotto (Figure 1). However, the high degree of infestation in one of these vineyards suggests that it is infected with FD since several years. This because 10% of vines showing symptoms were positively tested for FD. In 2005, FD spread to three adjacent municipalities as well as a previously uninfected winegrowing area more than 30 km away. In 2006, FD touched a third winegrowing area 20 km away from the second. 110

Of the 837 samples analysed between 2004 and 2006, 15% were uninfected by grapevine yellows diseases, 68% were positive to Bois noir, 11% were infected with FD and 6% were infected with both phytoplasmas. Chardonnay was the most FD sensitive variety, followed by Gamaret and Pinot noir. Merlot is the most tolerant variety displaying only few visible symptoms.

Efficacy of the mandatory insecticide control strategy against S. titanus In the first year of mandatory insecticide control, first larvae instars were found at the 25th May 2005. Exposition, altitude and grapevine variety had no impact on hatching of S. titanus. The first buprofezin application was effectuated between the 6th and 13th June followed by a second application two weeks later. Compared to an untreated vineyard outside the region of mandatory control, the two buprofezin applications showed an efficacy of 99%. Its efficacy against adults was variable and ranged from 91 to 97%. In 2006, the first buprofezin application was effectuated between the 30th May and 7th June. Once again the efficacy of the mandatory control strategy was beyond 90%. And in 2007, its efficacy was even between 98 to 100%. Theses three years of mandatory insecticide control confirmed the efficacy of the strategy proposed. Most important is the date of the first buprofezin application. It has to be applied on larvae of an age structure between the first and third larval stadium. This because buprofezin is an insect growth regulator, which inhibits the synthesis of chitin (Bosio et. al., 2001). However, a single application is certainly insufficient. As shown in Table 2, our mandatory control strategy has a long-term effect on vector population. In 2006, the number of S. titanus larvae found before insecticide application was about a twentieth of the previous year and the same trend was observed in 2007. Even though we did not catch any larvae after insecticide control in 2006, a small number of adults was nevertheless trapped. This may indicate that the population of S. titanus was only partially suppressed and adults migrate into treated vineyards from the surrounding. In 2007 neither larvae nor adults were caught after insecticide application.

Table 2. S. titanus population in the Mendrisiotto before and after the application of the mandatory insecticide controls.

Average number of larvae Average number Year per plant of adults per trap Before application After application After application 2005 6.31 0.51 0.4 2006 0.26 0.0 0.5 2007 0.06 0.0 0.0

Field trials of organic insecticide against S. titanus In 2006, Audienz showed only a low efficacy and the application of mineral oil in 2007 was also of limited successes. Latter is in opposition to previous results and might be explained by the low volume applied. Independent of the type of application, Parexan N showed an average efficacy against larvae of 95% in 2006 and 96% in 2007. This is comparable to buprofezin (=Applaud). Parexan N is therefore the only organic insecticide that can be recommended for the control of S. titanus. However, its activity and persistence is limited and its impact on the last larval instars and adults is insufficient. Consequently, our recommended strategy is: 111

• Three applications of Parexan N at an interval of ten days covering the hatching period of S. titanus. Applications must be carried out at sunset using a spraying system, which guarantees a good distribution over the entire canopy.

Conclusion

It is difficult to explain how FD arrived in the Ticino, but we can suppose that the spread of the disease within the Ticino is caused by human activities: this either by the importation of FD-infected plant material or by the transportation of FD-infected leafhoppers. The mandatory as well as the recommended organic control strategy are both effective and maintain S. titanus population at low level. However, it is difficult to evaluate insecticides targeting leafhopper adults. They are very mobile and quickly colonise treated plots from untreated vineyards. Usually, we were able to observe an efficacy beyond 95% for the first two weeks of leafhopper flight, thereafter, the effect of insecticides was diluted. Our results stress the necessity of a continuous monitoring program north of the Alps to follow the spread of S. titanus and to survey the presence of grapevine yellows diseases. In this process, good information of winegrowers is fundamental. First of all, well-trained winegrowers are a prerequisite for an early detection of FD outbreaks. And second, they are the base for a successful prevention and control of FD. In this process, we ask winegrowers in the Ticino to eradicate all plants showing symptoms of grapevine yellows diseases: first of all, to reduce confusion with bois noir disease and second to reduce any sources of inoculation.

References

Baggiolini, M., Canevascini, V., Caccia, R., Tencalla, Y. & Sobrio, G. 1968: Présence dans le vignoble du Tessin d’une cicadelle néarctique nouvelle pour la Suisse, Scaphoideus littoralis Ball (Hom., Jassidae), vecteur possible de la flavescence dorée. – Bulletin de la Société Entomologique Suisse 40 (3-4): 270-275. Bosio, G., Della valle, D., Ferrarese, D., Ferrari, D. & Occhetti, P. 2001: Evoluzione delle popolazioni di Scaphoideus titanus a seguito di interventi insetticidi. – L'informatore Agrario 21: 79-84. Clerc, L., Linder, Ch. & Günthart, H. 1997: Première observation en Suisse romande de la cicadelle Scaphoideus titanus Ball (Homoptera, Jassidae), vecteur de la flavescence dorée de la vigne. – Revue suisse Vitic. Arboric. Hortic 29 (4): 245-247. Gugerli, P., Besse, S., Colombi, L., Ramel, M.-E., Rigotti, S. & Cazelles, O. 2006: First outbreak of flavescence dorée (FD) in Swiss vineyards. – 15th meeting of the international council for the study of virus and virus-like diseases of the grapevine (ICVG). SASEV, Stellenbosch, South Africa, extended abstracts: 96-98. Jermini, M., Rossi, M. & Baillod, M. 1992: Etat actuel de la diffusion au Tessin de Scaphoideus titanus Ball, vecteur de la flavescence dorée. – Revue suisse Vitic. Arboric. Hortic 24 (3): 137-139. Jermini, M. & Baillod, M. 1996. Proposition d’une méthode de contrôle des populations de Scaphoideus titanus Ball dans le vignoble. – Revue suisse Vitic. Arboric. Hortic. 28 (3): 201-204. Linder, Ch. & Jermini, M. 1999: Le point sur la diffusion en Suisse de Scaphoideus titanus Ball, cicadelle vectrice de la flavescence dorée. – Revue suisse Vitic. Arboric. Hortic 31 (1): 53. 112

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 113-116

Effect of late season sprays against Botrytis on quality of the wines

Walter K. Kast, Oliver Schmidt and Karl Bleyer State Institute for Viticulture, Oenology and Fruit Technology, P.O. Box 1309, 74185 Weinsberg (Germany)

Abstract: Field experiments with separate vinification of 3 replicates in field were carried out to check the effect on wine quality and flavour. Wines were produced from grapes that were sprayed twice (before bunch closure, veraison) with Botrytis fungicides (Teldor: Fenhexamid and Switch: Cyprodinil and Fludioxonil) with an efficacy of 70%. They had less off-flavours than untreated but more slightly mouldy off-flavours than grapes treated only once (before bunch closure) with an efficacy of 55%. Late (veraison) treatments seam to favour the development of non-Botrytis rot diseases (Penicillium-rot). Only the treatment before bunch closure had a positive effect on the flavour of the wines.

Keywords: Botrytis cinerea, off-flavour, fungicides, Penicillium-rot, late season sprays, Fenhexamid

Introduction

Wines, produced from rotten grapes, caused by Botrytis cinerea, have typical changes in their analytical composition (Dittrich et al. 1974) and are known to have typical mouldy off- flavours (Boidron & Leveque 1979). Fungicides are used to reduce Botrytis cinerea with the intension of achieving wines with typical fruit aromas. Two applications of fungicides, one before bunch closure and another during blooming period, or at veraison, normally reduce the rotten grapes to 40-80% of untreated control (Kast 1985). In several vine estates in Southern Germany, wines produced from the fungicide treated grapes unexpectedly had more off- flavours. Off-flavours may be the result of residues of the fungicide (Lemperle 1969). Yet, fungicides registered in Germany are tested for their effect on fermentation and taste of wines and registration is rejected, if effects on taste are observed. The typical off-flavours observed in wineries in the German wine region Wuerttemberg in contrast to the expected more fruity taste were mouldy flavours. Field trials were carried out in 1999 and 2001 to evaluate the influence of Teldor (Fenhexamid) and Switch (Cyprodinil and Fludioxonil) on wine quality, especially on flavour of wines.

Material and methods

Field experiment In 1999 in a field trial with 3 replicates and 6 pesticides, 2 not registered fungicides, 4 fungicides against downy or powdery mildew: Topas (Penconazol), Ridomil Gold Combi (Metalaxyl + Folpet), Prosper (Spiroxamine) and the botryticide Teldor (Fenhexamid) and an untreated control was carried out in a Riesling vineyard at Weinsberg/Germany. All fungicides were sprayed three times (9. + 21. July, 20. August) in the period form just before bunch closure (BBCH 75) and after veraison (BBCH 87) in the recommended dose by use of a tunnel-sprayer and in addition to the standard sprays.

113 114

In 2000, a field experiment with 4 treatment plots, 2 untreated control plots and 3 replicates, was carried out. In a Riesling vineyard, two Botrytis-fungicides, Teldor (Fenhex- amid) and Switch (Fludioxonil+Cyprodinil) were both sprayed once (before bunch closure BBCH 75) and twice at BBCH 75 and after veraison (BBCH 87). Botrytis rot and Penicillium rot were evaluated one day before harvesting the grapes. The grapes of the replicates were harvested at different dates. Only grapes infected by acetic acid producing bacteria and yeasts were removed. In one untreated control of each replication only the completely healthy grapes were selected.

Vinification and tasting The grapes of the three replications in both experiments were vinified at three different dates. The wines were produced by use of steel tanks (110 l) in standard procedure (yeast added, no other treatments). The wines produced from these grapes were tasted twice, three months and nine months after bottling by sixty well experienced testers. The testers were advised to rank the seven wines of each repetition for off-flavours and typical fruit components using a 5- point scale and to rank the wines for their overall quality.

Results and discussion

In the year 1999, only very little Botrytis rot developed. The rot of grapes was mainly caused by larvae of grape berry moths with 4% of the grapes being infested by Eupoecilia ambiguella. Rot was not evaluated but should be present on a low level according to the results of Kast & Munder (1990). In 2000, Botrytis severely infected the grapes. Both fungicide treatments reduced Botrytis rot from 55 % in the untreated plots to 25 % (one treatment) and 17 % (two treatments). The grapes treated twice with one fungicide (Fenhexamid) had higher values for Penicillium rot (4 %), than the once treated and untreated grapes (<1 %).

treatment of fungicides 1999 - sensory effects

3 2,5 2 1,5

Score 0-5 1 0,5 0 l d i er r te M1 M2 pas m p do a V V o o s l re T id ro e t R P T un

overall quality (LSD5%: 0.23) apple(n.s.) mouldy (LDS5%: 0.28)

Fig. 1. Effect of different fungicides on wine quality and wine taste (3 applications: July 9, 21; August, 20) . 115

In 1999, no differences between treatments were observed for different flavour- components (Citric, peach, green apple, hydrogen sulfide). Only one treatment had significantly higher values for mouldy taste than the control (Teldor).

Fig. 2. Botrytis disease severity of Riesling grapes in 2000 (upper) and the quality (lower) of the wines [Means of rankings over 60 testers, 2 evaluations] 116

In the 2000 experiment, all once treated grapes had a significantly better quality ranking than the untreated ones but worse than the selected rot-free grapes. In both cases two applications resulted in a poorer sensory ranking than the wine of grapes sprayed once only mainly due to mouldy flavours. The difference between one and two sprays is not significant if the fungicides are proofed separately. But after pooling the values for one and two applications, the once treated grapes had significantly better quality than the twice treated. In spite of their lower Botrytis disease severity, the twice treated grapes had significantly more mouldy off-flavours. This should mainly be the result of a higher Penicillium rot incidence, which is testable in extremely low doses and could produce mouldy off-flavours. In some cases, late applications of Fenhexamid and other botryticides after veraison may favour non-Botrytis rot diseases caused by different microorganisms resulting in poorer wine quality. Sprays before bunch closure (BBCH75) don’t cause this problem and increase the quality of grapes by reducing Botrytis rot. We therefore do not recommend late season sprays because the effect seems to be 'cosmetic' only and does not reduce rotten and mouldy off- flavours.

References

Boidron, J.N. & Leveque, F. 1979: Studies on the volatile substances in grapes under the influence of Botrytis cinerea. – Rappt. Activ. Rech., Inst. Oenol., Univ. Bordeaux II; Talence : 61-62. Dittrich, H.H.; .Sponholz, W. & Kast, W.K. 1974: Comparative investigations of musts and wines from healthy and Botrytis-infected grapes. I. Oxygen metabolism, sugar metabolic products, leucoanthocyanin levels. – Vitis 13: 36-49. Kast, W.K. 1985: Botrytisabwehr im integrierten Rebschutz. – Rebe & Wein 38: 252-254 . Kast, W.K. & Munder, H. 1989: Calculation of an infestation loss relation for second brood of the grape berry moth (Eupoecilia ambiguella Hbn.) by means of regression. – Zeitschrift für Pflanzenkrankheiten und Pflanzenschutz 97: 76-83. Lemperle, E. 1969: Fermentation behaviour and measurements of active substance residues after using Botryticides in viticulture. – Deutsche Weinzeitung 105: 806.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 117-120

Water-glass (SiO2), an excellent, mostly overseen tool in fruit-rot control

W. K. Kast, K. Bleyer, R. Fox State Institute for Viticulture,Oenology and Fruit Technology, P.O. Box 1309, 74185 Weinsberg (Germany) [email protected]

Abstract: Results of field experiments and practical experiences to check the effect of silica gel (water-glass, SiO2) on Botrytis-rot were presented. Water-glass has an effect on Botrytis and sour-rot comparable to Botrytis-fungicide sprays. It is mainly effective against Botrytis infections occurring during the ripening period directly thru the berry-surface. This type of infections occurs during warm and rainy periods with very long wetness. Water-glass is less effective if the rotting is mainly caused because of pressure in compact clusters. This type of infections is a consequence of intensive berry growth forced by irrigation or rainfall. Problems may occur, if water-glass is used in mixtures with other pesticides. The application should preferably be carried out separately or by use of the double fluid application technique.

Key words: Botrytis cinerea, water-glass, SiO2, berry-splitting, fruit-rot diseases, global warming

Introduction

In Southern Germany, sour rot and Penicillium-rot of grapes was a severe and increasing problem over the last years, especially in 2000, 2002 and 2006. Global warming caused earlier ripening. Earlier ripening and higher temperatures summarize to a dramatical change in the ripening conditions. Intensity of rainfall did not change. In Southern Germany, as a result we more and more have to take a sudden outbreak of sour rot which is caused by rainfall at high temperatures into account. In cases of high amount of rainfall during ripening period, berries split or berries get micro fissures from high osmotic pressure. Control of grape-rot by fungicides is very useful, as long as the skin of grape berries is unaffected but the control is insufficient under these conditions. Wine quality in our field trials did not get better (often worse), if late season sprays were carried out, because fungicides are not active against Penicillium spec. and other rot causing fungi, yeasts and bacteria which cause the severest off-flavors. (Kast et al. 2008). Loose clusters by adapted cultural practice and hardening of cell structures are tools acting against all rot diseases. Siliceous sol (water-glass, soluble SiO2) is a very cheap chemical product used traditionally for storage of eggs and for fining of wines. Sprayed on grapes, it crystallizes on the berry surface and stabilizes the outer cell-wall (Kleber 1983). In Germany, sodium and potassium salts of SiO2 are registered as plant strengtheners and traditionally used in organic production since more than 100 years (Kleber 1983). Water-glass is used in organic viticulture against powdery mildew (Uncinula necator) (Pfleiderer; 1993; Hofmann U, 1993; Kast. et al. 1994; Reh & Schlösser; 1994), rust mites (Calepitimerus vitis, Ruehl; 1994) and showed effects on Botrytis, too (Kast & Neumann; 1997). In contrast to fungicides residues of SiO2 are not relevant and there is no need for waiting time. Fungicides resistance problems are not relevant, too. Problems with mixtures are described by Schruft et al (1993).

117 118

This paper summarizes the results of several experiments run at Weinsberg and practical experiences in the vineyards of the state wine estate at Weinsberg.

Material and methods

A commercial product (soluble potassium-SiO2 8% K, 20% SiO2) registered as plant strengthener (Kaliwasserglas Biofa Agrar GmbH Metzingen/Germany) was used in all experiments and for practical use in the state wine estate at Weinsberg/Germany. In the experients, the product was sprayed separately. Knapsack-sprayers or a tunnel sprayer were used in experiments, taking care to spray the grape zone, only. In practical use in the state estate winery mixtures with wettable sulfur and Polyram WG (Methiram) were applied, too. 3 or 4 sprays were carried out in all experiments at the growth stages BBCH75 (not if 3 sprays), BBCH77, BBCH 81 and BBCH 83. The dose of the commercial product in practical use ranged from 6kg/ha -12kg/ha dependent on the treated part of the leaf area.

Table 1: Experiments on efficacy of soluble potassium against fruit rot of grapes carried out in the years 1996 – 2007

Exp. No. of year variety dose water l/ha§ special conditions No. sprays 1 1996 Riesling 0,7% 400-800 4 Pinot 2 2005 1% 600 4 No yield reduction - compact grapes Meunier Pinot 3 2005 1% 600 4 tip of grapes (60%) cut off - loose clusters Meunier 4a Pinot early evaluation (65° Oechsle) 2006 1% 600 2 4b Meunier late evaluation (90° Oechsle) 5a very compact early evaluation 2007 Riesling 1% 400-600 4 5b grapes late evaluation 6 2007 Pinot Blanc 1% 400-600 3 compact grapes Pinot 7 2007 1% 600 4 extremely compact (1 cluster/shoot) Meunier Pinot 8 2007 1% 600 4 tip of grapes (60%) cut off - loose clusters Meunier Pinot 9 2007 1% 600 4 Gibb3 -partially loose clusters Meunier

Results and discussion

Four sprays using water-glass in a concentration of 1% in several experiments reduced Botrytis disease severity to 40 -9o% of the untreated control. Notable sour rot was observed in 2006. Same results as for Botrytis were found for sour rot disease in this year. Yield in some cases was significantly reduced, caused by smaller berries. Sugar content was unaffected if sprays were allocated on the grape berry-zone, only. No significant differences were noticed in all experiments. Actually, the trend was positive. 119

The best results of water-glass were observed in those cases, when clusters were loose and direct attack on the berry surface (Mueller-Thurgau type of Botrytis infection, according Kast and Schiefer 2005). In 2005, on grapes of Pinot Meunier that were cut in the middle to reduce the yield. These grapes became looser and the effect of water-glass was much better than on normal, compact grapes. In 2006, this type of infections was forced by high amount of rainfall and simultaneously very high temperatures during the harvest period. In contrast to 2006 in the year 2007 Riesling grapes were only attacked, if they were extremely compact. The effect of water-glass is poor, if grapes became compact and the attack of Botrytis started mainly from fissures on the inner parts of the clusters and from latent infections at the base of the berries (Pinot noir type of infection process according Kast and Schiefer (2005)). Problems may be caused by mixtures with fungicides used for control of downy and powdery mildew, insecticides or acaricides. In practical use, combinations with wettable sulfur and Polyram WG (Methiram) never caused any problems. Because only a few but not all imaginable combinations can be tested, we recommend separate sprays. The double fluid application technique (sprayers with double tanks and a second set of pump and nozzles) allows a comfortable, separate application of two different spray mixtures.

Table 2: Results of experiments on efficacy of soluble potassium against fruit rot of grapes carried out in the years 1996 – 2007

soluble SiO2 fungicide (check treatment) ex. fungicide year Variety dose Botrytis sour-rotdose Botrytis sour-rot no. Name % % efficacy % *) efficacy % 1 1996 Riesling 0.7 40 n.d.**) Ronilan (3x) 0.1 58 n.d. 2 2005 Pinot M. 1.0 23 n.d. not available 3 2005 Pinot M. 1.0 51 n.d. 4a Pinot 95 57 2006 1.0 86 Teldor (2x) 0.1 82 Meunier 4b 58 72 5a 46 n.d. 68 n.d. 2007 Riesling 1.0 Switch (2x) 0.06 5b 20 n.d. 45 n.d. 6 2007 Pinot Bl. 1.0 26 n.d. Teldor +Switch 0.1%+0.06 92 n.d. 7 2007 Pinot M. 1.0 0 n.d. 8 2007 Pinot M. 1.0 87 n.d. not available 9 2007 Pinot M. 1.0 29 n.d. *) number of sprays carried out in brackets **) no relevant disease attack

A problem after application of water-glass may be the dirtying of the equipment used for application. Water-glass produces insoluble coatings on the machines, especially on the windows of tractors. By use of special floor-cleaners used for removal of coatings on ceramic ground surface these coatings could be removed. 120

References

Kast, W.K.; Schmidt, O.; Bleyer, K. 2008: Effect of late season sprays against Botrytis on quality of the wines. IOBC/wprs Bulletin 36: 113-116. Kast, W.K.; Neumann, L., 1997: Botrytisbekämpfung – alte und neue Erfahrungen. – Rebe und Wein 50(2): 56-58. Kast, W.K.; Schiefer, H.-C. 2005: Botrytis – optimaler Bekämpfungszeitpunkt. – Der Deutsche Weinbau 1/2005: 26-29. Kast, W.K.; Ruehl, K.; Kopf, A. 1994: Silicate for mites and powdery mildew control. – Der Deutsche Weinbau, Neustadt (15): 24-25. Kleber, U. 1983: Erhöhung der Pflanzenresistenz durch Blattapplikation kieselsäurehaltiger Präparate. – Mitteilungen der Deutschen Gesellschaft für allgemeine und angewandte Entomologie 4 (1-3): 150-152. Pfleiderer, H. 1993: Wie wirksam sind Bio-Präparate. – Pflanzenschutz-Praxis 1992/2: 34-36. Reh, I.; Schloesser, E. 1994: Alternative control of powdery mildew on grapevine. – Mede- delingen van de Faculteit Landbouwkundige en Toegepaste Biologische Wetenschappen (Belgium) 89: 909-917. Ruehl, K. 1994: Kräuselmilbenbekämpfung in Württemberg 1994. – Rebe und Wein 47(3): 91-92.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 121-125

Bois noir, a severe outbreak of stolbur type A in Southern Germany – disease abundance and treatments against disease-causing agents and vectors

W. K. Kast, M. Stark-Urnau, K. Bleyer State Institute for Viticulture and Fruit Technology, P.O. Box 1309, 74185 Weinsberg (Germany)

Abstract: After 6 grapevines had been proved stolbur-positive in 2003, the Bois noir disease in grapevine caused by stolbur-phytoplasma spread over the complete region Württemberg in the years 2004 – 2006. Especially vineyards with steep slopes and vineyard walls were affected. Here, more than 70 % of the vines showed symptoms of grapevine yellows disease (Bois noir). The severe outbreak of this disease was caused by a subtype of Hyalesthes obsoletus which is specialized in feeding on stinging nettles (Urtica dioica) and which requires higher temperatures than the native race. This subtype very effectively transmits stolbur type A. Stolbur type A is more dangerous for vines than type B and causes considerable damage on the varieties Lemberger (Syn. Blaufraenkisch), Rhine-Riesling and Mueller-Thurgau. Both, the eradication of stinging nettles in the vineyards in autumn and the preservation of stinging nettles during the flight of Hyalesthes obsoletus in summer gave effective control of the disease. Infected vines of susceptible varieties should be cut 10 cm above ground immediately after observation of symptoms on leaves and grapes to keep the vines alive.

Key words: Bois noir, Hyalesthes obsoletus, Phytoplasma, stinging nettle, Urtica dioica

Introduction

Grapevine yellows disease (Bois noir) is caused by stolbur-phytoplasma type A. In Germany, so far mainly type B was found (Maixner 2007). Until 2004, the disease caused no relevant damage in the region Wuerttemberg. The only known vector was Hyalesthes obsoletus. Hyalesthes obsoletus transmits the phytoplasma during it’s search for new host plants, when it sucks by chance on vines (error-sucking). This cicada is classified as threatened by extinction in Southern Germany (Nickel and Remane 2002). Starting with only 6 PCR-proved infected vines in the region Wuerttemberg, the disease spread over the whole region in the years 2004-2006 (Kast and Stark-Urnau 2006). The disease caused severe damage on susceptible varieties (Lemberger Syn. Blaufraenkisch, Rhine-Riesling, Mueller-Thurgau and Chardonnay). Vines of these varieties were killed by the phytoplasma within 2 years. Experiments were carried out in the years 2005-2007 to analyze the nature of this severe outbreak and to find possible treatments against the disease.

Material and methods

Leaf veins from vines with typical disease symptoms (2003: 1, 2004: 7, 2005: 157, 2006: 126) were freeze-dried and analyzed by Maixner (BBA Bernkastel-Kues) using PCR-techniques for detection of subtypes of stolbur-phytoplasma (Maixner et al. 1995). Adult Hyalesthes obsoletus on stinging nettles were captured by yellow sticky traps and net-catching (2005:

121 122

164, 2006: 594). Furthermore, samples of stinging nettles from vineyards with disease symptoms were also analyzed (2005: 39, 2006: 34). In 2005 and 2006, symptomatic Lemberger vines of several vineyards and of different age were cut at two different times (summer - immediately after recognition of symptoms and winter). Furthermore, three different cutting intensities were employed: 10 cm above ground, cutting off all one-year-old wood and normal pruning. In the following season grapevines were checked for survival. In field experiments, herbicides were sprayed in November on spots of stinging nettles (Roundup (Glyphosate) 2,5 %, Garlon 4 (Triclopyr) 0,6 % and MCPA 1,0 %) either by use of a knapsack-sprayer and or by spot-application of highly-concentrated herbicides using a weed control stick. Survival of stinging nettles after 4 weeks and in April of the next year was evaluated. Scoring 0 -4; 0 = no effect; 1 = yellow discoloration; 2 = visible symptoms but not necrotic; 3 = partially necrotic; 4 = completely killed.

Results and discussion

In 2005, 67 % of the tested vines were Stolbur-positive. In contrast to other regions of Germany, most of the samples were tested as stolbur type A, only 14 % were tested as type B. In 2006, 96 % of the stolbur-positive tested vines were infected by Stolbur type A (Tab.1). Samples of Hyalesthes obsoletus and Urtica dioica were also tested positive for type A in several cases. In contrast to former findings in Germany the main type of phytoplasma here is stolbur type A hosting on Urtica dioica. This type seems to be much more aggressive than the type B originating from Convolvulus arvensis.

Table 1. Results of PCR analyses of grapevine-leaves (Mainer BBA Bernkastel-Kues).

Year Number of % stolbur-positive % Type A* % Type B* tested samples samples 2005 157 66,9 85,7 14,3 2006 126 76,3 96,1 3,9 *Percentage of stolbur-positive samples.

The flight of Hyalesthes obsoletus was about 16 days delayed in the years 2005, 2006 and 2007 (Fig. 1) compared to the model of Weber et al. (1996) and to the observed flight data of Maixner and Langer (2006). Most of the infected vines were observed in steep vineyards with walls or cross terracing or which were located near roads with asphalt surface. These observations and the delayed flight could be the consequence of a higher demand on temperature. Most of the catches of Hyalesthes obsoletus were made on stinging nettles (Figure 2). Larvae were found on this plant species, only. In contrast to our findings Hyalesthes obsoletus in Germany is described as polyphagous (Nickel and Remane 2002) feeding on different herbs. The delay of flight, the demand for obviously higher temperatures and the specialization on Urtica dioica argue for the spreading of a new subtype of Hyalesthes obsoletus. This subtype of Hyalesthes obsoletus is extremely effective in transmitting stolbur-disease, because of it’s specialization on one single host plant. Spreading of disease is supported whenever the host plant is not available (error-sucking).

123

160 140 120 100 Weinsberg 2005 80 Weinsberg 2006

Hyalesthes obsoletus 60 Weinsberg 2007 40 20 0 Number of 20 21 22 23 24 25 26 27 28 29 30 31 Week of the year

Fig. 1. Flight of Hyalesthes obsoletus found on stinging nettles in 2005, 2006 and 2007. Cursors refer to postulated start of flight of Hyalesthes obsoletus on Convolvulus arvensis in 2005, 2006 and 2007 according to Maixner and Langer (2006).

Table 2. Number of Hyalesthes obsoletus caught by yellow sticky traps on stinging nettles before and after herbicide-treatment of stinging nettles.

Number of Hyalesthes obsoletus

Site Before eradication of After eradication of Comparable location with stinging nettles stinging nettles preserved stinging nettles Brackenheim 125 4 Neipperg 209 0 218 Gundelsheim 60 10 Weinsberg I 363 0 216 Weinsberg II 121 0 Jungerberg 217 15 121 Uhlbach-Wiese 564 3 97

The herbicides based on Glyphosate and Triclopyr used in November are able to completely kill Urtica dioica (Tab.3). MCPA kills this weed only partially and could therefore increase the danger of error sucking on vines. Eradication of Urtica dioica by herbicides in November reduced the number of Hyalesthes obsoletus caught in the following year and reduced the number of infected vines two years after treatment (vegetation period one year!). Severest damage was observed when nettles were partly damaged by herbicide treatments or when nettles were being mowed during the flight period of Hyalesthes obsoletus (June and July). Larvae of Hyalesthes obsoletus are not able to get other host plants because of their specialization and because they are living in the soil. They do not survive, when their host plant stinging nettle is killed during 124

the first larval instars. At several sites, the number of Hyalesthes obsoletus caught by yellow sticky traps decreased after eradication of stinging nettles in contrast to the trend at other sites (Table 2).

Table 3. Effect of herbicides used in November 2005 on Urtica dioica

Mean effect of herbicides sprayed by use of knapsack sprayer Commercial product Genoxone ZX Garlon4 Durano U46M-Fluid Active substance Triclopyr + 2,4-D Triclopyr Glyphosate MCPA Concentration 1,25% 0,6% 1,5% 1,0% Effect after 3 weeks 3,0 3,8 3,5 1,7 Effect in April 4,0 4,0 4,0 2,0

Mean effect of herbicides, spot-application using a weed control stick Commercial product Genoxone ZX Garlon4 Durano U46M-Fluid Active substance Triclopyr + 2,4-D Triclopyr Glyphosate MCPA Product/water-relation 1:2 1:2 1:1 1:2 Effect after 3 weeks 4,0 4,0 2,4 0,5 Effect in April 4,0 4,0 4,0 2,0 0 = no visible symptoms 1 = yellow discoloration Scaling 2 = visible symptoms but not necrotic 3 = partially necrotic 4 = completely killed

Infected vines of susceptible varieties (Lemberger, Rhine-Riesling) only regain health if they are cut 10 cm above ground. Cutting immediately after identification of typical symptoms in summer had better results than cutting in winter or than regular pruning. We hope to get control of the disease by combining eradication of the host plant of phytoplasma and transmitting cicadidae with extreme cutting down of infected vines. Stinging nettles should be eradicated in November. Still existing host plants should not be damaged in Spring time in order to avoid infection of grapevines by adult Hyalesthes obsoletus searching new stinging nettles.

Acknowledgements

We thank M. Maixner, Bernkastel-Kues and his staff for analyzing a huge number of samples and the “Ministerium für Ernährung und ländlichen Raum, Stuttgart” for financial support

References

Kast, W.K. & Stark-Urnau, M. 2006: Schwarzholzkrankheit am Beispiel von Württemberg: Gefahrenherd Brennnessel. – Das Deutsche Weinmagazin 2006 (7): 12-14. 125

Maixner, M. 2007. Die Schwarzholzkrankheit der Rebe. – Südtiroler Zeitschrift für Wein- und Obstbau 3/2007: 100-103. Maixner, M. & Langer, M. 2006: Prediction of the flight of Hyalesthes obsoletus, vector of stolbur phytoplasma, using temperature sums. – IOBC/wprs Bulletin 29 (11): 161-166. Maixner, M.; Ruedel, M.; Daire, X. & Boudon-Padieu, E. 1995: Diversity of grapevine yellows in Germany. – Vitis 34 (4): 235-236. Nickel, H. & Remane, R. 2002: Artenliste der Zikaden Deutschlands, mit Angabe von Nähr- pflanzen, Nahrungsbreite, Lebenszyklus, Areal und Gefährdung (Hemiptera, Fulgoro- morpha et Cicadomorpha). – Beitr. Zikadenkde. 5: 27-64. Weber, A.; Maixner, M. & Seitz, A. 1996: Zur Biologie von Hyalesthes obsoletus Sign. (Auchenorrhyncha: Cixiidae) als Vektor der Vergilbungskrankheit der Rebe. – Mitteilun- gen aus der Biologischen Bundesanstalt für Land- und Forstwirtschaft, Berlin-Dahlem (321): 105. 126 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 127-136

Bois noir disease of the grapevine in Alsace: field transmission, observations made on symptomatology and reduction of transmission risk by the vector Hyalesthes obsoletus Signoret

Philippe Kuntzmann1, Emmanuel Bogen2, Claudia Renel1 1 Institut Français de la Vigne et du Vin, Colmar, Biopole, 28 rue de Herrlisheim, 68000 Colmar 2 probationer, student BTS viticulture-œnologie, Lycée Viticole, 16 avenue Charles Jaffelin, 21200 Beaune

Abstract: We present some results and discuss the way to reduce the transmission risk of stolbur isolate type I as defined by Maixner, by the vector Hyalesthes obsoletus in order to reduce frequency and/or severity of bois noir disease in Alsace vineyards.

Key words: Alsace, grapevine, bois noir, stolbur, Urtica dioïca, Hyalesthes obsoletus, disease incidence.

Introduction

Grapevine yellow indicated in the first half of last century in France and in Germany, the bois noir has been described on the french side of rhine by Caudwell (1961) and the Vergilbungs- krankheit on the other side by Gärtel (1965). It was only in the early nineties, that a phytoplasma, belonging to the group of the stolbur of solanaceous, has been identified in France (Daire et al., 1993) then in Germany (Maixner et al., 1995) as causal agent of both diseases. The phytoplasma incubation time in the vine is not well known, from thought to be higher as in the case of Flavescence dorée for producing vines (Walter et al., 2000), to 5 months for young grapevines (Osler et al., 1997). The role of the cixid Hyalesthes obsoletus in the transmission of the phytoplasma to the vine, was pointed out first in Germany (Maixner, 1994), and in the late nineties in France (Sforza et al., 1998). Possibility of other vectors can not be excluded, but their role has not been experimentally ascertained yet (Boudon-Padieu, 2005; Sabaté et al., 1997). Transmission can occur through infected plant material too, but this way of dissemination, although not neglectable, notably for long distance propagation, seems however to be of low incidence (Osler et al., 1997). Hyalesthes obsoletus presents five larval stages occuring on the roots of its host plants from september, to june of the following year. Second and third larval stages were described to be the overwintering stages (Sforza et al., 1999), but recently third and even fourth larval stages were found to be overwintering stages (Kuntzmann et al., 2007). As adults can be seen on various herbaceous or woody plants like Chenopodium album or Robinia pseudacacia (Kuntzmann, 2006; Kuntzmann et al., 2007) or Holcus lanatus (personal observation made 2007, not published), so it is very important to determine the host status of a plant by observing the larval stages on his roots. Stinging nettle, field bindweed, hoary cress and lavender are well known host plants on our climates. On these plants, if they are infected, the vector can acquire the phytoplasma at the time of his larval stages or as an adult, in order to transmit it then to other host plants or to

127 128

sensitive crops like vine by casual feeding trials. Recently some stolbur moleculary characterization works showed the existence of three isolates depending on the host plants on which Hyalesthes obsoletus had been catched, (Langer and Maixner, 2004). Isolate type I was found in the vine, in the stinging nettle and in Hyalesthes obsoletus captured on stinging nettle and hedge bindweed. Isolate type II was found in the field bindweed, in nightshade and blackthorn, and in Hyalesthes obsoletus captured on both bindweed species. A third isolate was found in a Mosel vineyard in vine, hedge bindweed and Hyalesthes obsoletus captured on the latter. These results seem to indicate a specializing of the vector to the host plant, although there is a lack of visible morphological differences between individuals. Since the early two- thousands, an outbreak of bois noir disease is related in different vineyards in France (Grosman et al., 2005; Grosman et al., 2006) as well as in other countries like Germany (Stark-Urnau et al., 2006; Maixner, 2006; Gilge et al., 2004 ), Austria (Riedle-Bauer et al., 2006) and Italy (Schweigkofler et al., 2006; Mazzoni et al., 2006; Riolo et al., 2006; Morandell et al., 2003; Alma et al., 2002). Captures of Hyalesthes obsoletus were sometimes low and in some cases other species of planthoppers were found positive to stolbur, but it is known not to be sufficient to incriminate a specie as a vector. In some cases Hyalesthes obsoletus had been found on the stinging nettle, in other cases on field bindweed, or on both at the same surveyed places. In 2006 we started an applied research programme for a better understanding of the recent outbreak of bois noir disease in Alsace, with the purpose to propose to vinegrowers some control measures. The first results acquired in 2006 showed the quite prevalence of type I in Hyalesthes obsoletus and the complete prevalence of it in the samples of vine tested (Kuntzmann, 2006; Kuntzmann et al., 2007), with an average value of positive insects about 12.5%. Moreover, in some plots with a disease incidence up 10% - about 30% - we showed that the diseased vines follow spatially a hearth distribution, in relation with the stinging nettle spots in or on the border of the plot. When the disease incidence is less than 10% , the distribution is hazardous, and there are in this case no hearthes of stinging nettle in the plot or in its close neighbouring. The disease severity increases with the viccinity of the stinging nettle spots, suggesting a possible relation between the risk of inoculation by the vector and the expression of symptoms. In addition, we checked for the presence of Hyalesthes obsoletus at his fifth larval stage and adult stage in the environment and found that cixid commonly in the surveyed viticultural area always on the stinging nettle, in some cases located on some hearthes on vine neighbouring embankments. Our observations suggest a spread all over the surrounding vineyards. Control measures for this disease, beside the sanitation of plant material, are constitued by the reduction of transmission risk by the vector Hyalesthes obsoletus, by acting against his host plants or against the overwintering stages in the soil, due to the non specific relation between this vector and the vine, on the contrary of the vector of Flavescence dorée. So it is advised to expose the larvae to the frost by ploughing the vines (Langer et al., 2003) or to saw some covering plants which will compete with the host plants. But in conventional viticulture, where some systemic herbicides are allowed, the better way to reduce the vector population is to suppress the host plants by this way, focusing on the spots of stinging nettle for example.

129

Material and method

Vine plots This research programme is carried out in 23 vineyards in Alsace, most of them located around Colmar in the towns of Turckheim and Herrlisheim. In Turckheim, we are monitoring 13 vine plots constituting an area from about 7 hectares, belonging to different producers and thus managed by different ways, situated on a hill side exposed from south east to south/south west. In Herrlisheim the plot is located on the west side of the Alsace production area, close to the plain devolved to cereals and corn. The plot is divided into two parts –north and south- planted at one year delay from each other. On one side of the plot, the vinerow is bordered by an embankment infested by the stinging nettle quite on its whole length. We started other trials in order to appreciate specifically the transmission risk by the plant material in the sites of Châtenois, Eichhoffen and Orschwihr, where we put into comparison hot water treated material vs. non treated material.

Table 1. Characteristics for some of further related vine plots

Planting Soil Location code variety Host plants year cultivation No stinging nettle Auxerrois and Chemical in the plot but on Turckheim RD 1990 Pinot blanc weed control embankments in the viccinity Stinging nettle, Riesling and 2001 and Natural green Turckheim KMRi suppressed Pinot gris 2003 cover autumn 2006 Sown green Turckheim D Auxerrois 1993 cover/chemical No stinging nettle weed control Sown green Stinging nettle, Turckheim BE Pinot gris 2003 cover/soil suppressed spring working 2007 Sown green Stinging nettle on Herrlisheim - Herrlisheim Chardonnay 2001 cover/chemical an embankment, nord weed control not suppressed

Checking for the presence of Hyalesthes obsoletus We have many observation points and for each one several probes are done, each one consisting by checking a nettle stem for the presence of the vector. The results were reported into three classes: no vector, from 1 to 10 individuals by probe, more than 10 individuals by probe. In 2006 we were looking for the vector on C. arvensis or L. draba at a very few places, in addition to U. dioica, but in 2007 it was only on U. dioica.

Monitoring of Hyalesthes obsoletus The monitoring of Hyalesthes obsoletus is done by yellow sticky traps positioned under the row just above the herbaceous cover so catching in a high of 18 to 33 cm in the vine plots, and in a high of 66 to 81 cm if in an embankment, due to the height reached by the cover. 130

These traps are rolled on a PVC tube in a vertical plan. The traps were placed on week 21 and checked weekly.

Disease incidence The disease incidence is assessed by notations as late as possible before the leaves fall down.

Results and discussion

Checking for the presence of Hyalesthes obsoletus Within the time of one year, the places colonised by the vector doubled, as it is possible to note it when considering the % of probes with larvae. This evolution would be still stronger if there would not have been taken out of account some places where stinging nettle has been suppressed by the grower (so we go from about 125 probes in 2006 to about 100 probes in 2007) and where the vector was present in 2006.

Table 2. Results of the investigations

% probes with … Larvae % probes with … Adults From 1 More From 1 More No one No one to 10 than 10 to 10 than 10 June 21,23 –2006 (129 probes) 82.2 10 7.8 July 17-2006 (124 probes) 38.7 48.4 12.9 June 6,8-2007 (101 probes) 67.3 21.8 10.9 June 29-2007 (100 probes) 50 39 11

That is one explanation for the reason that the % of probes with no larvae in 2007 is higher that the % of probes with no adults in 2006, when it should be quite the same. The other explanation is that all the places colonized by the adults in 2006 were not suitable for laying eggs or did not allow the development of the larvae. On one season scale, we can see the same, that the vector is colonizing the environment at his adult stage, so the % of probes with no indivuals are decreasing. But this effect is lesser in 2007 than in 2006, possibly due to the climatic conditions in this year much less favourable for this mediterranean insect, from the last decade of june to august, which were rainy and just in the average value or less, for temperature. In Alsace where the vector is out of his origine area, the ecological niches occupied by it, might change in relation with the climatic conditions of the year or of the long term period.

Monitoring of Hyalesthes obsoletus and evaluating efficiency of suppressing the stinging nettle In 2007 the flight began just as we had positioned the traps: we have no week with zero captures, but the observations of fresh insects made during the first reading let us ought the flight had just begun. The flight begins week 22 (last week of may) and lasts for 10 weeks, ending week 32 (first week of august) as we see in figure 1 showing the results in Turckheim for 82 traps. The flight peak occurs week 24, the descending part of the flight lasting for 8 weeks, with a second little peak around week 29, corresponding to a climatic very hot periode. 131

The general shape of flight curve is quite the same as 2006, except that the descending part of the flight lasted longer this year probably due to the formerly related climatic conditions. The flight began in 2007 quite at the same moment as in 2006, which is the end of flowering/begin of fruit set of the grape.

Turckheim : flight of the vector Hyalesthes obsoletus (2007 - yellow sticky traps in the vineyards N= 82)

600

500

400

300

capture (nb.) 200

100

0 week 22 week 23 week 24 week 25 week 26 week 27 week 28 week 29 week 30 week 31 week 32 BBCH BBCH BBCH 73 BBCH BBCH 79 BBCH 79 BBCH 79 BBCH BBCH 80 BBCH 69-70 71-72 77-79 79-80 81-90

Fig. 1. Flight of the vector in Turckheim

Characterization of the chromatic traping for the different plots - 2007 - (traps positioned in vineyards N=number of traps)

250 Trap which captured the less 226 Average 200 Trap which captured the more

150

99 100 74 72 capture (nb.) capture 59 59 50 47,00 19 18 23 19 21,09 19,85 4 20,27 5 22,73 10,33 15,44 11,67 0 5,63 5,58 2,50 3,00 6 13 6 7 1 5 0 3 00 11

FQ (N=8) KMRi DR (N=6) RD (N=6) RG (N=9) BE D (N=19) C (N=4) Châten. HS (N=3) KMAux Herrlis. (N=11) (N=13) (N=15) (N=3) (N=15) Plots

Fig. 2. Characterization of the chromatic catch in 2007 132

But what changes is the total amount of insects catched, which is less than a half in 2007 compared to 2006. As we had an upper average value of about 60 catches per trap in 2006, the higher value does not exceed 50 individuals per trap in 2007, as shown in figure 2. In a few plots the average value increased more or less, for example RD, Châtenois (Châten.) and Herrlisheim (Herrlis.), but in the most cases it did decrease. So for FQ, KMRi, DR and in a lesser extent RG and BE, we can observe a strong decrease, the strongest by 66% for KMRi. The eight first plots and the tenth and eleventh ones in the figure are all located in the same place, on the hill side surveyed in Turckheim. Except for Herrlisheim, the contrast between the more and the less captured is reduced compared to 2006. The influence of host plants in the vicinity of the trap (around 2 meters) on the capture level is given in figure 3. As in 2006, the presence of the stinging nettle favorizes the catch level of the vector, but this effect is about more than 3 times more captures in 2007 as it was only around 2 times in 2006.

Influence of host plants in the viccinity of the trap on capture level (traps positioned in vineyards and embankments - all sites - 2007 -N = number of traps) 47 50 45 40 35 35 26 30 25 20 12 9 15 8 10

average capture (nb.) 4 5 0 without with host with U. with C. with C. with U. with C. host plants plants dioica arvensis sepium dioica + C.sepium + (N=75) (N=59) (N=24) (N=23) (N=5) sepium C. arvensis and/or (N=2) C.arvensis (N=5) Traps

Fig. 3. Influence of the host plants on the catch level in 2007

Let us see if the suppression of the stinging nettle brought an improvement in the amount of insects catched. The datas are reported in figure 4, in which we compare the traps where no suppression did occur to those where the suppression was achieved by the grower. Rapidly overseen, we could imagine that the suppression of the stinging nettle brought roughly an efficacy of 40% on the catch level of the vector, because the reduction for untreated traps is about 17% and about 50% for treated ones. But this has to be corrected for the untreated ones: at a matter of facts, the datas in this case are falsed for one trap due to an increased capture caused by the casual mowing of the nettles (see figure 5) and so the average value for untreated traps in 2007 should rather be about 35 instead of 43. So the decrease for the control (stinging nettle kept) by 32% and not only 17%, it puts the efficiency of stinging nettle suppressing only to 26%, what is a pretty disappointing! One explanation of this poor consequence of suppressing the stinging nettle, on the catch level of the vector, could be that, especially in the area of Turckheim where the main suppression occurred, the infested plots were source of infestation for neighbour plots in the previous year. 133

Another reason not to be such disappointed by these results, is to consider in its globality the plot KMRi which showed last year the highest captures and where the suppression was quite effective, and for consequence a drastic decrease in the catch level for 2007. A last comment for the figure 4: the reason for what the average values for 2006 are higher in the case of control than treatment (stinging nettle suppressed), is because in a few cases the suppression was not complete when the nettle spot covered a great area, and a larger nettle spot in 2006 meant a higher amount of vector in 2006.

60 52 43 50 40 26 30 13 20

nb. captured 10 0 year 2006 year 2007 year 2006 year 2007 stinging nettle suppressed 2007 stinging nettle kept 2007 (N=29) (N= 26)

Fig. 4. Influence of suppressing stinging nettle on the catch level. N = number of traps.

Plot Herrlisheim : influence of mowing nettles on the flight (2007)

160

140 Embankment (N=2) Vine plot (N=15) 120

100

80 mowing of nettles beneath one trap

captured nb. 60

0 week 22 week 23 week 24 week 25 week 26 week 27 week 28 week 29 week 30 week 31 week 32 BBCH BBCH BBCH BBCH BBCH BBCH BBCH BBCH BBCH BBCH 69-70 71-72 73 77-79 79 79 79 79-80 80 81-90

Fig. 5. Influence of cultural practices on the flight 134

Disease incidence The disease incidence is shown in figure 6, where we compare 2007 with the previous year. In all surveyed plots the disease incidence did reduce but not for the same level. So the reduction rate ranges from up to twenty five (KMRi : 26) to less than two (Herrlisheim –nord with 1.7). This reduction in disease incidence is independant from any growers action, for exemple suppressing of diseased stocks. It is remarkable that in 2007 the symptoms were not only less frequent, but less severe too, than in 2006. But in 2007, on the contrary of 2006, we did not have the opportunity to note the severity of the disease which needs the presence of the grapes which were already picked up at the time of the notations. The datas shown in table 3 put into comparison both disease incidence and capture level for 2006 and 2007: RD and Herrlisheim which are the only plots which showed not a decrease in the capture level between 2006 and 2007, have the poorest reduction of disease incidence between 2006 and 2007. On the contrary the plots in which capture level decreased, show a decrease in disease incidence too. Nevertheless it is hardly possible to consider this as a consequence of the removing of the nettles, because of the quite not well known but probably longer than one year, incubation time.

Incidence of bois noir disease in a few vineyards - Comparison 2006/2007

30,0 28,8 26,2 25,0

20,0

15,5 15,0 % 10,0 6,4 7,7 6,7 5,3 5,0 2,1 1,9 0,8 0,9 1,7 1,1 1,6 0,0 20062007200620072006200720062007200620072006200720062007

Auxerrois Pinot blanc Pinot gris Auxerrois Riesling Pinot gris Chard.

RD BE D KMRi Herrlis.-nord

Fig. 6. Incidence of bois noir disease

Conclusions

In 2007 we observed a general decrease in flight activity of Hyalesthes obsoletus compared to 2006, and thus no great additional efficacy on the captured insects by suppressing the stinging nettles, according to the capture level of the traps. We could observe some differences in the disease incidence in 2007 compared to 2006, but hardly linkable to the reduction of flight activity of Hyalesthes obsoletus.

135

Table 3. Synthetic array comparing trapping vector and symptoms

Disease incidence 2007 Capture level 2007 Plot Variety versus 2006 versus 2006 Auxerrois :3 RD *1.5 Pinot blanc :2.4 Pinot gris :4.2 KMRi :3 Riesling :26 D Auxerrois :4.5 :3 BE Pinot gris :5.9 :2 *3 or *1 (without Herrlisheim –nord (trapping Chardonnay :1.7 casually influenced concerns both nord and sud) trap)

Acknowledgements

This project is supported by grants of VINIFLHOR and the Conseil Interprofessionnel des Vins d’Alsace.

References

Alma A., Soldi G., Tedeschi R., Marzachi C., 2002. Ruolo di Hyalesthes obsoletus Signoret (Homoptera Cixiidae) nella trasmissione del Legno nero della vite in Italia. – Atti II Incontro Nazionale sulle malattie da Fitoplasmi, 3-4 ottobre 2002, Roma, Italia: 57-58. Boudon-Padieu E., 2005. Phytoplasmes associés aux Jaunisses de la vigne et vecteurs potentiels. – Bulletin de l’O.I.V. 891-892: 299-320. Caudwell A., 1961. Etude sur la maladie du bois noir de la vigne: ses rapports avec la flavescence dorée. – Ann. Epiphyt 12: 241-262. Daire X., Clair D., Larrue J., Boudon-Padieu E., Alma A., Arzone A., Carraro L., Osler R., Refatti E., Granata G., Credi R., Tanne E., Pearson R., Caudwell A., 1993. Occurrence of diverse MLOs in tissues of grapevine affected by grapevine yellows in different countries. – Vitis. 32: 247-248. Gärtel W., 1965. Untersuchungen über das Auftreten und das Verhalten der flavescence dorée in den Weinbaugebieten an Mosel und Rhein. – Weinberg und Keller. 12: 347-376. Gilge U., Schwappach P., Herrmann J.V., Maixner M., 2004. Feldstudie zum Vorkommen der Schwarzholzkrankheit in Franken und Methoden zu ihrer Bestimmung. – Schweizerische Zeitschrift für Obst- und Weinbau Wädenswill. 19/2004: 10-13. Grosman J. et al., 2005. Bilan Phytosanitaire de la vigne en 2005. – Phytoma La Défense des Végétaux 587:18-23. Grosman J., Magnien C., Renaudin I., Retaud P., Trespaille-Barrau J.-M., 2006. Bilan phytosanitaire de la vigne en 2006. – Phytoma La Défense des Végétaux 598: 19-24. Kuntzmann P., 2006. Bois noir: progression inquiétante de la maladie dans certains vignobles français – l’exemple de l’Alsace: des relations très étroites entre les populations du vecteur Hyalesthes obsoletus et ses plantes hôtes permettent un début d’explication. – Compte rendu technique, Mondiaviti, 29 et 30 novembre 2006: 19-25. Kuntzmann P., Thill E., Marmonier A., Villaume S., Renel C., 2007. Recrudescence du bois noir de la vigne en Alsace – Les experts à Colmar mettent l’ortie au banc des accusés. – Phytoma La Défense des Végétaux. 607: 37-41. 136

Langer M., Maixner M., 2004. Molecular characterisation of grapevine yellows associated phytoplasmas of the stolbur-group based on RFLP-analysis of non-ribosomal DNA. – Vitis. 43: 191-199. Langer M., Darimont H., Maixner M., 2003. Control of phytoplasma vectors in organic viticulture. – IOBC/wprs Bulletin 26 (8): 197-202. Maixner M., 2006. Schwarzholzkrankheit: ein neues Problem? – Der Deutsche Weinbau. 16- 17: 44-49. Maixner M., 1994. Research note: transmission of German grapevine yellows (Vergilbungs- krankheit) by the planthopper Hyalesthes obsoletus (Auchenorrhyncha: Cixiidae). – Vitis. 33: 103-104. Maixner M., Ahrens U., Seemüller E., 1995. Detection of the German grapevine yellows (Vergilbungskrankheit) MLO in grapevine, alternative hosts and a vector by a specific PCR procedure. – European J. Plant Pathol. 101: 241-250. Mazzoni E., Aldini R. N., Pavesi F., Cravedi P., 2006. Surveys of the presence of Hyalesthes obsoletus Signoret (Rhynchota: Cixiidae) and other hoppers in Lombardy (Northern Italy). – IOBC/wprs Bulletin 29 (11): 183-186. Morandell A., 2003. Vergilbungskrankheiten im Südtiroler Weinbau. – Obstbau Weinbau Fachmagazin des Beratungsringes 11/2003: 320-325. Osler R., Vindimian M.E., Filippi M., Carraro L., Refatti E., 1997. Possibilità di propagazione del giallume della vite (legno nero) a mezzo del materiale vivaistico. – Informatore fitopatologico 11: 61-63. Riedle-Bauer M.,Tiefenbrunner W., Otreba J., Hanak K., Schildberger B., Regner F., 2006. Epidemiological observations on Bois Noir in Austrian vineyards. – Mitteilungen Klosterneuburg. 56: 166-170. Riolo P., Isidoro N., Nicoletti L., Riga F., Nardi S., Marozzi F.A., 2006. Potential leafhopper and planthopper vectors of phytoplasmas in wine vineyards of the Marche region (Central Italy). – IOBC/wprs Bulletin 29 (11): 193-198. Sabaté J, Lavina A., Batlle A., 2003. Potential vectors of grapevine bois noir phytoplasma in spain and evaluation of their transmission capacity. – Proceedings 14th ICVG Conference, Locorotondo, 12-17th september, 2003: 113. Schweigkofler W., Roschatt C., Baric S., 2006. Der Uberträger der Schwarzholzkrankheit der Rebe. – Obstbau Weinbau Fachmagazin des Beratungsringes. 05/2006: 141-143. Sforza R., Bourgoin T., Wilson W.S., Boudon-Padieu E., 1999. Field observations, laboratory rearing and description of immatures of the planthopper Hyalesthes obsoletus (Hemi- ptera: Cixiidae). – European Journal of Entomology. 96: 409-418. Sforza R., Clair D., Daire X., Larrue J., Boudon-Padieu E., 1998. The role of Hyalesthes obsoletus (Hemiptera: Cixiidae) in the occurrence of Bois noir of grapevines in France. – J. Phytopathology. 146: 549-556. Stark-Urnau M., Bleyer K., Kast W.K., 2006. Der Schwarzholzkrankheit vorbeugen. – Der Badische Winzer. Oktober 2006: 26-27. Walter B., Boudon-Padieu E., Ridé M., 2000. Maladies à virus, bactéries et phytoplasmes de la vigne. – Editions Féret. 191 pages: 151.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 p. 137-143

Time of treatment and selection of fungicides importance in control of Phomopsis cane and leaf spot disease of grapevine

Nedeljko Latinovic1, Petar Vuksa2, Zora Vucinic1, Jelena Latinovic1 1Biotechnical Institute, University of Montenegro, Kralja Nikole bb, 81 000 Podgorica, Montenegro, [email protected] 2Faculty of Agriculture, University of Belgrade, Nemanjina 6, 11 080 Beograd, Serbia

Abstract: Control of Phomopsis cane and leaf spot disease of grapevine (Phomopsis viticola Sacc.) in Montenegro represents a great problem, especially after withdrawing DNOC products from the market. Because of that there was a need to find out other adequate solutions in control of the disease. At the same time, it was of great importance to establish time of treatment correctly. During 2002 and 2003 the trial was set up in order to examine biological efficacy of fungicides that were applied in dormant period and at the begging of the vegetation. In winter dormant period DNOC and copper oxychloride were applied, while at the begging of vegetation fungicides based on azoxystrobin, benomyl, dithianon, fenarimol, folpet, mancozeb, prochloraz and triadimefon were used. Treatments at the beginning of vegetation were performed in two variants with two treatments each. In the first variant treatment was done in time of buds opening (BBCH 03/05) and when the shoots were around 10 cm long (BBCH 13). In the second variant first treatment was conducted when the shoots were around 10 cm long (BBCH 13), and second treatment when the shoots were around 50 cm long (BBCH 53). Results of the study showed that there are differences between applied fungicides, but time of treatment have a crucial role in attaining high efficacy. All applied fungicides have had better efficacy when they are applied in winter period together with the applications at the begging of vegetation and in the growth stage of shoots length of around 10 cm. It is also noted that the time of treatment was more important in part of the trial where copper oxychloride was applied for winter treatment, while the differences in efficacy were smaller if DNOC was applied. Folpet, mancozeb and dithianon showed the highest efficacy, followed by azoxystrobin and benomyl. Efficacy of triadimefon was weaker than the previous mentioned while prochloraz and fenarimol expressed the lowest efficacy of all fungicides applied.

Key words: Phomopsis viticola, fungicides, efficacy, time of treatment.

Introduction

Phomopsis cane and leaf spot, caused by pathogenic fungus Phomopsis viticola Sacc., is a widespread disease in almost all grapevine areas around the world: Africa, Asia, Australia and Oceania, Europe, North America (Hewitt & Pearson, 1990) and South America (CAB International, 2005). In Montenegro the disease was noted in 1955 but mass appearance of infected shoots and intensive spreading happened in the seventieths of the last century (Mijuskovic, 1975). Today it is one of the most important diseases in vineyards in Montenegro since it occurs every year and it causes more or less damages depending on attack intensity. The most represented domestic varieties Vranac and Kratosija belong to the group of the most susceptible varieties toward the disease (Latinovic et al., 2003). Until recently, control of the disease was quite successful if, together with sanitary measures, treatments were performed with products based on DNOC in dormant period followed by two treatments with some of the preventive fungicides at the beggining of

137 138

vegetation. Since the use of DNOC products banned in the meantime, there was a need to find out new adequate solutions of this problem. Therefore the trial with some new groups of fungicides was conducted with the accent on the research on time of treatment importance in the disease control.

Material and methods

The trial was performed in 2002 and 2003 on the field of Biotechnical Institute in Podgorica in vineyard with vertical cordon growth form, variety Kratosija, age 12 -13 years.Trial was carried out according to EPPO standards (1997) and with 15 repetitions in an accidental block system scheme (one repetition is one cane with ten shoots). In order to establish time of treatment importance in both years trial was done with two blocks or two variants of plant treatments: I – two treatments: BBCH 03/05 and BBCH 13 II – two treatments: BBCH 13and BBCH 53 The trial was carried out on the field after winter treatment performed with copper oxychloride or DNOC. Examined fungicides are shown in table 1. Control variant is also included in the trial, where there was no fungicides applied. Evaluation of the disease intensity was done in two times in both years: in June (BBCH 75) and in time of winter dormant period in January (BBCH 00). Ten shoots were checked on each cane and on each shoot four basal internodes were checked. In first check (June) degree of streaked depressions and bark cracking appearance on shoots was noted, while in the second check (January) pycnidia presence was recorded as well. Degree of the disease is expressed by scale in interval of 0-100 % (EPPO, 1997), and the efficacy is calculated according to Abbott formula. Data about fungicides efficacy are processed by standard statistical methods (analysis of varianse) and estimation of significance between group (F-test) and individual (lsd-test) differences was made.

Table 1. Review of examined fungicides

Applied Active ingredient Fungicide concentration (%) azoxysistrobin (250 g/kg) Quadris (Syngenta) 0,075 benomyl (500 g/kg) Benomil WP-50 (Zorka-Šabac) 0,06 dithianon (750 g/kg) Delan 750 SC (BASF) 0,035 fenarimol (120 g/l) Rubigan (Dow AgroSciences) 0,02 folpet (800 g/kg) Folpan 80 WDG (Makhtestim) 0,20 mancozeb (800 g/kg) Dithane M-45 (Župa-Kruševac) 0,25 prochloraz (450 g/l) Sportac 45-EC (Bayer) 0,04 triadimefon (250 g/kg) Bayleton 25-WP (Bayer) 0,02 copper oxychloride (500 g/kg) Bakarni kreč-50 (Zorka-Šabac) 0,75 DNOC (500 g/kg) Selinon (Bayer) 1,00

Results and discussion

Data about disease intensity on shoots and fungicide efficacy in 2002 and 2003 are presented in tables 2 and 3. Based on these data it could be stated that all variants with fungicides 139

applied are statistically much more significant then the control. In control variant disease intensity was 36,4 % in first reading and 69,7 % in second reading in 2002, while in 2003 it was 26,3 % and 65,2 % respectively. In 2002 in variant with DNOC application during winter dormant period, according to disease intensity, in first reading folpet application is better then the other preventive fungicides in phenophases BBCH 03/05 + BBCH 13 (1,7 %), as well as in BBCH 13 + BBCH 53 (2,3 %). Low level of the disease was also recorded in variants with applications of mancozeb (3,8 %, and 4,5 %), azoxystrobin (3,6 % and 3,8 %) and dithianon (3,8 % and 4,3 %). Weaker effects were achieved in applications of prochloraz (6,5 % and 6,7 %), fenarimol (7,7 % and 7,9 %) and tiadimefon (6,9 % and 8,3 %). In 2002 in variant with copper oxychloride application during winter dormant period, according to disease intensity, in first reading in phenophases BBCH 03/05 + BBCH 13 variants were more effective in applications of mancozeb (1,5 %), folpet (2,7 %), azoxystrobin (3,6 %) and dithianon (3,8 %) then with other fungicides applied in vegetation. The disease intensity is stronger when these fungicides were applied in phenophases BBCH 13 + BBCH 53 (7,1 %, 8,7 %, 7,7 %, and 6,5 % respectively). Weaker effects in reducing the disease intensity were noted in applications of triadimefon (7,1 % and 11,5 %), prochloraz (9,1 % and 12,5 %) and fenarimol (9,7 % and 12,3 %) in both phenophases. In second reading in variant with DNOC application, the disease intensity is greater in average for all fungicide application and in both moments of treatment. Similar like in first reading, again folpet (6,2 % and 7,0 %), mancozeb (5,8 % and 9,7 %), azoxystrobin (7,1 % and 8,0 %) and dithianon (8,2 % and 9,0 %) were better then prochloraz (9,6 % and 12,9 %), fenarimol (11,3 % and 14,7 %) and triadimefon (11,1 % and 15,5 %). According to disease intensity realized in variant with DNOC application in winter period, benomyl was similar to first above mentioned group of fungicides in first (4,3 % and 9,1 %) as well as in second time of treatment (6,2 % and 9,3 %) in spring applications in both readings. It is also similar to these fungicides in variant with copper oxychloride application in first time of treatment in spring (5,1 % and 9,1 %). In second time of treatment it is similar to «weaker» fungicides group (11,1 % and 14,1 %). Efficacy of preventive fungicides in 2002 in variants with DNOC application during winter period is also better in first reading in both times of spring treatments for folpet (95,3 and 93,7 %), mancozeb (89,6 % and 87,6 %), azoxystrobin (90,1 % and 89,6 %) and dithianon (89,6 % and 88,2 %). In second reading the situation is similar: for folpet (91,1 % and 90,0 %), mancozeb (91,7 % and 8,1 %), azoxystrobin (89,8 % and 88,5 %) and dithianon (88,2 % and 85,2 %). Lower efficacy is established in both time of treatments in first reading for fenarimol (78,8 % and 78,3 %), triadimefon (81,0 % and 77,2 %) and prochloraz (82,1 % and 81,6 %), and it is similar also in second reading for fenarimol (83,8 % and 78,9 %), triadimefon (84,1 % and 77,8 %) and prochloraz (86,2 % and 81,5 %). Taking in account efficacy, benomyl was also in between of previously mentioned groups of fungicides in both times of treatments in first (88,2 % and 83,0 %) and in second reading (86,9 % and 86,7 %). In 2003 in variant with DNOC treatment during dormant period, based on the disease intensity, there were differences between preventive fungicides applied in vegetation period. In first reading variants with folpet and mancozeb applied (0,0 % and 0,0%), then dithianon (0,1 %), and azoxystrobin as well (2,3 %) where better then variants with other applied fungicides in phenophases BBCH 03/05 + BBCH 13. However, they showed weaker effects if they were applied in phenophases BBCH 13 + BBCH 53 (4,8 %, 3,2 %, 4,8 % and 3,5 %, respectively). Higher degree of the disease in both times of application was recorded for benomyl (5,3 % and 6,8 %), prochloraz (6,9 % and 7,2 %), fenarimol (6,5 % and 8,1 %) and triadimefon (7,9 % and 8,1 %). Also in second reading lower disease intensity was noted for 140

folpet (5,0 % and 11,0 %), mancozeb (5,0, % and 7,9 %), azoxystrobin (6,1 % and 9,2 %) and dithianon (5,0 % and 12,1 %) in comparison to benomyl (9,0 % and 12,3 %), prochloraz (10,3 % and 13,2 %), fenarimol (10,4 % and 17,7 %) and triadimefon (12,7 % and 15,5 %).

Table 2. Disease intensity and fungicide efficacy in 2002

Treatment 1st reading 2nd reading Dormant Spring – T1; T2; T3 I E % I E % folpet + + - 1,7 95,3 6,2 91,1 mancozeb + + - 3,8 89,6 5,8 91,7 dithianon + + - 3,8 89,6 8,2 88,2 azoxystrobin + + - 3,6 90,1 7,1 89,8 benomyl + + - 4,3 88,2 9,1 86,9 prochloraz + + - 6,5 82,1 9,6 86,2 fenarimol + + - 7,7 78,8 11,3 83,8 triadimefon + + - 6,9 81,0 11,1 84,1 folpet - + + 2,3 93,7 7,0 90,0 DNOC mancozeb - + + 4,5 87,6 9,7 86,1 dithianon - + + 4,3 88,2 9,6 86,2 azoxystrobin - + + 3,8 89,6 8,0 88,5 benomyl - + + 6,2 83,0 9,3 86,7 prochloraz - + + 6,7 81,6 12,9 81,5 fenarimol - + + 7,9 78,3 14,7 78,9 triadimefon - + + 8,3 77,2 15,5 77,8 folpet + + - 2,7 92,6 6,5 90,7 mancozeb + + - 1,5 95,9 6,3 91,0 dithianon + + - 3,8 89,6 8,3 88,1 azoxystrobin + + - 3,6 90,1 8,2 88,2 benomyl + + - 5,1 86,0 9,1 86,9 prochloraz + + - 9,1 75,0 16,3 76,6 fenarimol + + - 9,7 73,4 17,1 75,5 triadimefon + + - 7,1 80,5 13,8 80,2 folpet - + + 8,7 76,1 11,9 82,9 mancozeb - + + 7,1 80,5 11,9 82,9 dithianon - + + 6,5 82,1 10,7 84,6 copper oxychloride azoxystrobin - + + 7,7 78,8 11,8 83,1 benomyl - + + 11,1 69,5 14,1 79,8 prochloraz - + + 12,5 65,7 17,3 75,2 fenarimol - + + 12,3 66,2 9,0 87,1 triadimefon - + + 11,5 68,4 17,8 74.5 Control 36,4 69,7 LSD0,05 2,49 2,93 LSD0,01 3,27 3,84

Note: T1– treatment performed in BBCH 03/05; T2– treatment performed in BBCH 13; T3– treatment performed in BBCH 53; I - disease intensity; E- efficacy 141

In 2003 in variant when copper oxychloride was applied during dormant period, based on the disease intensity observed in phenophases BBCH 03/05 + BBCH 13, variants with applications of folpet (0,0 %), mancozeb (0,1 %), dithianon (0,5 %), azoxystrobin (3,1 %) and benomyl (3,2 %) showed better results in first reading then other applied fungicides. Intensity of the disease was higher in phenophases BBCH 13 + BBCH 53 in applications of folpet (8,9 %), mancozeb (7,7 %), benomyl (9,3 %), azoxystrobin (9,5 %) and dithianon (11,6 %), and even more in applications of prochloraz (12,9 %), fenarimol (13,8 %) and triadimefon (13,3 %). In second reading in variant with applied copper oxychloride, the disease intensity was on the average higher in application of all examined fungicides in both times of treatments. Again folpet (5,3 % and 15,5 %), mancozeb (5,1 % and 14,7 %), azoxystrobin (8,9 % and 16,0 %), dithianon (5,0 % and 16,1 %) and benomyl (9,4 % and 20,9 %) expressed better results then prochloraz (16,9 % and 18,8 %), fenarimol (17,3 % and 20,5 %) and triadimefon (14,7 % and 21,9 %). Efficacy of fungicides in 2003 (table 3) in variant with DNOC applied in both readings was better in applications of folpet (100 % and 92,3 %), mancozeb (100 % and 92,3 %), dithianon (99,6 % and 92,3 %) and azoxystrobin (91,3 % and 86,2 %) then efficacy of benomyl (79,8 % and 86,2 %), prochloraz (73,8 % and 83,3 %), fenarimol (75,3 % and 84,0 %) and triadimefon (70,0 % and 80, %). In second term of spring-time applications of fungicides efficacy of folpet (81,7 % and 83,1 %), mancozeb (87,8 % and 87,9 %), azoxystrobin (86,7 % and 85,9 %) and dithianon (81,7 % and 81,4 %) was lower, but still better then with benomyl (74,1 % and 81,1 %), prochloraz (72,6 % and 79,8 %), fenarimol (69,2 % and 72,9 %) and triadimefon (69,2 % and 76,2 %). In variant with applied copper oxychloride and in first term of spring-time applications of fungicides, high efficacy was achieved in folpet (100 % and 91,9 %), mancozeb (99,6 % and 92,2 %), dithianon (98,1 % and 92,3 %), azoxystrobin (88,2 % and 86,3 %), and even so in benomyl (87,8 and 85,6 %) applications. These fungicides were more effective then prochloraz (67,7 % and 74,1 %), fenarimol (57,0 % and 73,5 %) and triadimefon (67,3 % and 77,5 %). In second term of spring-time application folpet (66,2 % and 76,2 %) and mancozeb (70,7% and 77,5 %) were significanly less effective then in first term, azoxystrobin (63,9 % and 67,9 %) and dithianon (55,9 % and 75,3 %) were similar with benomyl (64,6 % and 67,9 %), while prohloraz (51,0 % and 71,2 %), fenarimol (47,5 % and 68,6 %) and triadimefon (49,4 % and 66,4 %) were quite non effective. Study on time of treatment importance, as it could be seen from the above obtained results, revealed that applied protective fungicides are more effective if they are applied at the very beggining of vegetation rather then in later phases of grapevine growth. If the first treatment have been done in phenophase of 10 cm shoot lenght, efficacy is significantly lower then in case when the fungicides have been used at the very beggining of vegetation. Therefore fungicide treatments in period when vegetation starts is necessary in order to achieve high efficacy. This is confirmed in many reasearh work worldwide performed by numerous authors (Cucuzza and Sall, 1982; Pscheidt and Pearson, 1989; Smart, 1996; Emmett and Wiks, 1994; Nair et al., 2000; Wilcox et al., 1998). Correct time of fungicide applications is especially important if copper oxychloride is used for winter treatment of grapevine, which is the only fungicide allowed for application in this purpose in Montenegro. Concerning the adequate choice of fungicides, in the trial folpet, macozeb and dithianon showed the highest efficacy, followed by azoxystrobin and benomyl. Efficacy of triadimefon was weaker than the previous mentioned while prochloraz and fenarimol expressed the lowest efficacy of all fungicides applied. 142

Based on achieved results, it could be concluded that in control of Phomopsis cane and leaf spot disease of grapevine it is important to pay attention not only on the choice of the fungicides, but also on the correct time of treatment, since it is the only way to prevent realisation of infections on young shoots and disable appearance and development of the disease rightly, without application of DNOC fungicides.

Table 3. Disease intensity and fungicide efficacy in 2003

Treatment 1st reading 2nd reading Dormant Spring – T1; T2; T3 I E % I E % folpet + + - 0,0 100 5,0 92,3 mancozeb + + - 0,0 100 5,0 92,3 dithianon + + - 0,1 99,6 5,0 92,3 azoxystrobin + + - 2,3 91,3 6,1 90,6 benomyl + + - 5,3 79,8 9,0 86,2 prochloraz + + - 6,9 73,8 10,9 83,3 fenarimol + + - 6,5 75,3 10,4 84,0 triadimefon + + - 7,9 70,0 12,8 80,4 folpet - + + 4,8 81,7 11,0 83,1 DNOC mancozeb - + + 3,2 87,8 7,9 87,9 dithianon - + + 4,8 81,7 12,1 81,4 azoxystrobin - + + 3,5 86,7 9,2 85,9 benomyl - + + 6,8 74,1 12,3 81,1 prochloraz - + + 7,2 72,6 13,2 79,8 fenarimol - + + 8,1 69,2 17,7 72,9 triadimefon - + + 8,1 69,2 15,5 76,2 folpet + + - 0,0 100 5,3 91,9 mancozeb + + - 0,1 99,6 5,1 92,2 dithianon + + - 0,5 98,1 5,0 92,3 azoxystrobin + + - 3,1 88,2 8,9 86,3 benomyl + + - 3,2 87,8 9,4 85,6 prochloraz + + - 8,5 67,7 16,9 74,1 fenarimol + + - 11,3 57,0 17,3 73,5 triadimefon + + - 8,6 67,3 14,7 77,5 folpet - + + 8,9 66,2 15,5 76,2 mancozeb - + + 7,7 70,7 14,7 77,5 dithianon - + + 11,6 55,9 16,1 75,3 copper oxychloride azoxystrobin - + + 9,5 63,9 16,0 75,5 benomyl - + + 9,3 64,6 20,9 67,9 prochloraz - + + 12,9 51,0 18,8 71,2 fenarimol - + + 13,8 47,5 20,5 68,6 triadimefon - + + 13,3 49,4 21,9 66,4 Control 26,3 65.2 LSD0,05 2.52 2.90 LSD0,01 3.31 3.81 Note: T1– treatment performed in BBCH 03/05; T2– treatment performed in BBCH 13; T3– treatment performed in BBCH 53; I - disease intensity; E- efficacy 143

References

CAB International (2005): Distribution Maps of Plant Diseases. – October (Edition 1), Map 969. Cucuzza, J.D., Sall, M.A. (1982): Phomopsis cane and Leaf Spot Disease of Grapevine: effects of chemical treatments on inoculum level, disease severity, and yield. – Plant Disease 66 (9): 794-797. Emmett, R.W. and Wicks, T.J. (1994). Phomopsis. – The Australian and New Zealand Wine Industry Journal. 9: 197-225. EPPO (1997): Guidelines for the efficacy evaluation of plant protection products: Phomopsis viticola PP 1/55(2). – In: OEPP/EPPO Standards: Guidelines for the efficacy evaluation of plant protection products. Vol. 2, Fungicides and Bactericides, EPPO, Paris: 75-77. Hewitt, W.B. and Pearson, R.C. (1990): Phomopsis Cane and Leaf Spot. – In: Compendium of Grape Diseases. Aps Press – The American Phytophatological Society: 17-18. Latinović, N., Latinović, J., Zindović, J. (2003): Osjetljivost nekih sorti vinove loze prema prouzrokovaču crne pjegavosti vinove loze (Phomopsis viticola Sacc.). Poljoprivreda i šumarstvo 49 (1-2): 71-77. Mijušković, M. (1975): Crna pjegavost (ekskorioza) vinove loze. – Poljoprivreda i šumarstvo, 21 (2): 21-31, Titograd. Nair, N.G., Emmett, R.W., Wicks, T.J., Clarke, K., Sergeeva, V., Castillo-Pando, M. (2000): Dormant spray application facilitates management of Phomopsis on grapevines. – The Australian Grapegrower & Winemaker. (438a): 71-73. Pscheidt, J.W., Pearson, R.C. (1989): Time of infection and control of Phomopsis Fruit Rot of grape. – Plant Disease 73 (10): 829-833. Smart, R. (1996): Dealing with Phomopsis. – The Australian Grapegrower & Winemaker. (393): 82-83. Wilcox, W.F., Riegel, D.G., Burr, J.A. (1998): Evaluation of fungicides for control of Phomopsis cane and leaf spot of grapes. – Fungicide & Nematocide Tests 54: 109. 144

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 145-153

An expert-based crop protection decision strategy against grapevine’s powdery and downy mildews epidemics: Part 1) formalization

Leger B.1,2, Cartolaro P.2, Delière L.2, Delbac L.2, Clerjeau M.2, Naud O.1 1 UMR Cemagref - ENSAM – CIRAD - Information et Technologie pour les Agro-Procédés, 361, rue J.F. Breton - BP 5095 - 34196 Montpellier Cedex 5; 2INRA, UMR1065 Santé Végétale, 33183, Villenave d’Ornon, France. – [email protected]

Abstract: A formal model was built in the discrete event paradigm for dynamic systems 1) to test the feasibility of the spraying decision strategy “GrapeMilDeWS”, 2) to facilitate its transfer and usability, 3) to build a simulator that would allow testing climatic scenarios and scalability of the solution at the estate level. The knowledge elicitation process was made incrementally with the 4 designers of GrapeMilDeWS. The elicitation method uses the graphical language of Statecharts as a mediation tool between the experts and the knowledge engineer. Besides graphics, Statecharts are also a mathematically formal language. We present here the design principles that guided the experts in their design, how these principles have been formalized, the knowledge elicitation methodology that we set up and the questions that arouse from questioning formally the spraying strategy.

Key words: Decision, Statecharts, Elicitation, IntegratedIPM, viticulture

Introduction

With the growing consciousness of environmental and health issues among the public, sustainable agriculture is a major research topic. The present case study is about French vineyard. Vine growers consume about 20% of the pesticides used in France while vineyards’ area is 3% of the farmland (Aubertot et al., 2005). The development of fungal pathogens is climate dependant, and outbreaks are difficult to handle. Thus, growers have developed intensive and mostly preventive crop protection techniques. Integrated Pest Management (IPM) would then be desirable for sustainable viticulture. IPM aims at reducing the amount of inputs while keeping the revenue of farms, through the use of biological control as well as pesticides when necessary (Kogan, 1998). Unfortunately, the complexity of the vine- pathogens system and current lack of detailed epidemiological knowledge does not allow to calculate optimal solutions within a limited set of options, as it can be the case for cereal crops. So pest management procedures for viticulture need to be based on expertise. We consider here IPM as a process included in an overall agricultural production process. Design of production processes in agriculture has benefited from simulation for a long time now. Simulation models use a number of paradigms (Ascough II et al., 1997; Attonaty et al., 1994; Cros et al., 2003), including discrete event systems (Cournut and Dedieu, 2004). Rule based decision support systems are common in agriculture (Girard and Hubert, 1999; Shaffer and Brodahl, 1998). Our approach of decision making is process based. We present here a contribution to the design of decision procedures, with a representation that belongs to the family of Discrete Event Systems (DES) formalisms. An expert IPM solution was designed by our phyto-pathologists fellows (Clerjeau, 2004). We named it “GrapeMilDeWS” (Grape Mildews Decision Workflow System). GrapeMilDeWS aims at controlling two of the prevailing vineyard diseases: Powdery Mildew

145 146

(Erysiphe necator) and Downy Mildew (Plasmopara viticola). It is based on the following hypothesis: Expert knowledge, expertiseinformation and field observations can substitute numerous and systematic treatments. GrapeMilDeWS was experimented during two years on four plots, and it demonstrated its efficiency at this scale, with satisfactory harvest and greatly reduced number of crop protection operations(Cartolaro et al., 2007). Yet the process description was originally very informal, and its implementation still relied on the knowledge of the researchers. Formalising GrapeMilDeWS allowed all necessary hidden knowledge to be elicited. Modelling has indeed two objectives. First, the formal model would guaranty a better transfer of technology disambiguating the GrapeMilDeWS features. Second, a formal model can be included in a simulation model of a virtual farm, to check for usage of resources, induced costs, and to test the system against scenarios that were not experimented in the field. In the following sections we present an IPM solution for controlling the vine patho- system, designed using expert knowledge; statecharts which were identified as a pertinent formalism to model the system; and knowledge elicitation method which was conducted to build the formal model with the experts. Finally the teachings of this formalisation are discussed.

Design principles

The goal of GrapeMilDeWS is to avoid yield losses: disease symptoms are tolerated and monitored. This is achieved by the following strategy. Low epidemics can be controlled with a reduced number of systematic treatments applied at certain phenological stages (two against downy mildew and two against powdery mildew). Careful monitoring early in the season allows to identify potentially severe epidemics, in order to apply additional treatments (five optional sprayings are available against downy mildew and against powdery mildew, three extra treatments may be done). Another principle of GrapeMilDeWS is to make decisions plot by plot, according to their specific epidemics conditions. The main indicators are collected, at the plot scale, through observations of symptoms on the leaves as well as on the bunches. These observations give an estimate of the level of the epidemics. The observation results are then, translated into three discrete variable: M, O, Og. Variable M stands for downy mildew, O and Og stand for powdery mildew respectively on leaves and bunches. The number of symbolic values for each variable varies from two to three depending on the disease, and the observation date. These symbolic values encode the qualitative expert notion in the following manner: (‘0’) for absence of epidemics or low epidemics; (‘+’) for moderate to high epidemics; and eventually (‘++’) for very high epidemics. The values of the thresholds between these different modalities evolve with the phenological development of the vines. It allows adjusting the consequences of an epidemic to the evolution of the plant susceptibility during its development. Field observations are the only indications used as far a powdery mildew is concerned. Two extra indicators are used to assess downy mildew epidemics. • The local area risk level (ILM) gives information at a larger geographical scale than the plot, of the risk of development of the disease. It is based on a large disease monitoring network and on a climatic risk model. This information is provided through the official warning service bulletins (SRPV-Aquitaine, 2007). ILM is also a discrete variable, with two values: (0) low risk and (‘+’) medium to high risk. • The forecasted rain events from the weather forecast of MeteoFrance. 147

Thus ILM and rain forecasts are updated by exogenous sources of information at any time while O, Og and M are controlled by GrapeMildDeWS which triggers their update via requesting observations. When a spraying is required for one disease, decision about adding the treatment against the other will be taken in the same decision stage. Joining a much as possible the treatment against powdery and downy mildews, when both are needed, is consistent with the operational constraints that growers have to deal with. Still to alleviate the work load, GrapeMilDeWS is constrained on the number of observations. W.r.t the pathosystems at hand: grapevine powdery mildew and grapevine downy mildew , the majority of the field observations are done before flowering. Their objective is to detect the severe epidemics by quantifying the early symptoms of the diseases on the foliage, before the period of high susceptibility of the bunches. This allows, when required, to position treatments limiting the proliferation of inoculums on the foliage. During the period for which the bunches are most susceptible, protection is provided using systemic product against both target diseases. The application is done at the flowering stage and can be supplemented by an optional treatment depending on the early observation data as well as weather forecasts. The number of treatments past that period of susceptibility of the bunches is limited. A third observation is done in the field. It provides first an overview of the sanitary status of the plot and assesses the opportunity of more treatments.

GrapeMilDeWS

[BBCH>10] Stage_0 Season_start Evaluation_1 Stage_1 eNPSU[BBCH>=15 eNED AND No REI] [ exists(T1) eNPSU[BBCH>+60 AND dateIs(E1+15) AND ILM='+' AND No REI] AND NOT exists(T2)] Stage_4 Evaluation_2 [activePeriodEnded()] Stage_3 Stage_2 eNED eNPSU[BBCH>=65]

[dateIs(T3+28) AND No REI)] eRF[activePeriodEnded() AND BBCH>=81] Evaluation_3 Stage_5 Stage_6 eNED PreHarvest_Evaluation

eNPSU[BBCH>=85] eNPSU[grapeIsRipe()] eNE

eNPSU<=>event Phenological Stage Update Evaluations in the plot are ordered eNED<=>event Notify Evaluation Done Mandatory treatment decision stages against Powdery (1), Downy(6) or both(3) mildews eRF<=>event Rain Forecasted Optional treatment decision stages against Powdery, Downy or Decision reasoning Variable update Both mildews

Fig. 1. Synoptic view of the GrapeMilDeWS crop protection decision process with the interleaving of observations and decision stages. 148

A systematic treatment against downy mildew is ordered at the beginning of ripening to limit the development of foliar diseases at the end of the season, and thus ensure enough foliage for the maturation of the grapes. The process is represented with ist temporal constraints in Fig.1.

Formalisation methodology

In GrapeMilDeWS, the continuous phenomena that constrain the crop protection process (diseases incidence, local area downy mildew risks, rain forecasts and phenology) are discretized in an abstract time scale. Therefore the decision system can be assimilated with a DES. Beyond simulation, DES formalisms have been shown to be suitable for qualitative analysis and control of various systems (Cohen, 2007; Lunze, 2000). GrapeMilDeWS can be viewed as a “control system with humans in the loop”. The information from the vine plot is aggregated in synthetic discrete values. These are attached to a finite number of states, which can be reached during the season. Reacting to external events, transitions are fired, which generate internal and output events. The output events are decisions for actual sprayings on the plot executed by human operators. We chose the graphical language of Statecharts, introduced in (Harel, 1987), for our formal model.

The Statechart formalism The modeling work, intrisically, by the elicitation and formalization of the knowledge it implies, should help the designers of a decision process to specify it precisely. Statecharts which are now standard in Unified Modeling Language (UML 2.0) (OMG, 2007), combine finite state automata following two principles: "parallelism" (concurrent automata) and hierarchical “nesting”. “Nesting” means that each state can be broken down into a sub-automaton that describes the behavior of the system with a finer granularity. Transitions are labeled with triggering events, actions (which are generated events) and boolean conditions. The conditions are being tested on the variables defined in the Statechart. Finally, decision nodes allow to represent transitions with several options from the same state and within the same event (Harel and Kugler, 2004). Such expressiveness allows for complex synchronization between automata. Transitions are deemed to be instantaneous, allowing statecharts to be seen as reactive systems .After implementation into executable code, the behavior of these visual specifications can be tested with different initial conditions and external stimuli. See an illustration of the syntax in Fig.2. From these Statecharts, we can simulate the result of the decision-making process. These results are characterized by the number of treatment and the distribution of these treatments over time. Within the group project "Vin et Environnement” (Soler, 2004), we have chosen the number of treatment as a simple and relevant indicator of environmental performance. Our aim is to substitute reasoned risk management to conventional approaches which are mostly preventive and systematic.

Eliciting the knowledge from the experts Our elicitation method uses “intermediate” knowledge models, able to ensure the mediation between the phyto-pathologists, designers of the GrapeMilDeWS, and the knowledge engineer (KE) in an iterative elicitation process. From these “intermediate models of knowledge”, correctly formed automata and calculable models are then elaborated. 149

Knowledge acquisition process. The main elicitation method consists in iterative individual interviews of about an hour each. The KE prepares the subject of the interview and the documents needed. Each expert is interviewed over statechart diagrams and finally a synthesis of all editions is done to close the round when each expert has been interviewed. Group sessions may also be used at the end of the round to clarify eventual divergences between the experts. Rounds are repeated until more interviews does not improve the knowledge acquired and consensus is high enough. We will call hereunder diagrams the informal statecharts used during experts’ interview and models the formal statecharts which we simulate with Rhapsody from I-Logix. Diagrams are informal statecharts oriented towards communication. They have to be understood by both the KE who designs them and the experts who are interviewed. On the other hand, the models are the formal synthesis of the diagrams: an executable software. The informal and formal Statecharts have the same structure of states and events. The differences lie mostly in the labelling. Syntax of function calls and manipulation of variables of the formal Statecharts is better replaced by more natural language in the informal diagrams. Interviews are divided in two parts. The first half hour is dedicated to validating the diagram from the previous synthesis, checking that the experts and the KE have a common understanding of the GrapeMilDeWS, but also that expert agree on the process. Contradictory positions would be quickly observed and addressed via group sessions. The changes, additions and modifications from each expert are recorded on the diagram. The goal of the second part of the interview is to foray deeper into the details of the process. It allows the expert to build up on the latest progress of the other experts. Group sessions. There is no specific procedure defined for group sessions. All the experts are invited. The goal of such session is to quickly settle points of discrepancies which could not be synthesized using the interview information.

Short description of elicited model for GrapeMilDeWS As shown in Fig.1, GrapeMilDeWS’s process breaks down in seven Stages which alternate with three evaluations. The statechart has been simplified in this figure and only the up most level is shown. The sequence of decision and temporal constraints are visible, where as the decision logic is hidden in the sub-statechart. An evaluation is an action state during which GrapeMildDeWS waits for data to be refreshed. A Stage is a global decision state. The sub-states of a stage display the decision logic of the GrapeMilDeWS. Stage_4 is detailed as an exemple in Fig.2. A stage is also a temporal period defined by the phenology (we use the BBCH scale (Lorenz et al., 1995)) and temporal conditions relative to the previous sequence of actions ordered by GrapeMildDeWS (i.e. end of active period , legal restricted entry interval after the spraying of chemical products on the plot or fixed delay after an observation or an application). For example, the activation of Stage_5 is triggered by the execution of the third evaluation , which is itself positionned about one month after the flowering treatment during Stage_3. The exit of Stage_5 is controlled by one of the following: (i) the phenology reaching mid rippening (BBCH>=85) or (ii) after the beginning of rippening, if the crop is not protected and if rain is forecasted then in that later condition the last treatment at Stage_6 is ordered earlier. This event between Stage_5 and Stage_6 is thus dependent of the fourth treatment and the length of active period of the products used in Stage_5. Product choices have been modeled but again are hidden here for clarity. The treatment decisions in each stages are taken according to the estimators O, Og, M and ILM. A Stage is entered and the current values of the four variables are used to “route“ 150

through the decision nodes and select the correct decision state. Activation of the decision state generated a treatment order if necessary. On Fig.2 : variables [O==’0’] and [M==’++’] activate state High_dmildew. Entry into that state generates the doDMildewTreatment order. GrapeMildDeWS does not capture the operational resources and response of the vineyard. It throws requests, which the grower may fulfil at its will and receives notifications about execution of decisions.

Stage_4

[O<>'0']

[O=='0'] [M>'0' OR ILM=='+'] [M=='0' AND ILM=='0'] Watch_ILM High_pmildew_AND_ANY_dmildew [M=='++'] doMixedTreatment. [M>'0'OR| ILM=='+'] evILMChange [M=='0' AND ILM=='0']

High_dmildew PMildew doDMildewTre... doPMildewTreat... Dmildew_risk_OR_dmildew evNotifyTreatment evNotifyTreatment Wait_for_next_rain watchWeatherForecasts(); evNotifyTreatment evNotifyTreatment evRainForecasted Rain_forecasted

[date()>=treatmentDate("T3")+28]

Fig. 2. Decision logic detail in Stage_4’s sub-statechart – berries starting to touch

Discussion

We have presented the aim of GrapeMilDeWS, the formalism chosen for the model, and the way we elicited the formal model. We shall now discuss some of the insight it gave us over GrapeMildeWS. Elicitation itself, was very teaching. Working with a software engineering approach for instance allowed questioning about what would happen in case a treatment order could not be carried out before a forecasted rain. New contaminations are then most likely for downy mildew during the rain. The expert then provided a corrective solution for that eventuality: use of a curative product in the first 24 to 48 hours following the rain. If the application cannot be done during the period, then curative treatment is considered useless useless and a preventive product would be again preferable so as to avoid further contaminations. Therefore, it provided new ideas to reduce even more the number of application with more intelligent use of curative and protectant products. Thus formalisation permitted to gain a more detailled view of the process and to create new work hypothesis that then should be experimented in the field. On a more critical note, The formalism has the advantage of being quite intuitive (Cruz- Lemus et al., 2005)but it is best at representing Reactive Systems. Typically, crop protection is a lot about forecasting and adjusting to uncertain future events. Indeed, during the 151

experimental crop protection period, the experts were adjusting to the weather forecast and anticipating the future. This is quite difficult to represent with statechart. Had we tried to formalise this type, of knowledge, we would have lost the benefit of a humanly accessible statechart program. We have found comparing the experimental run of GrapeMilDeWS and simulated runs, that there is a cascade of automatic consequences from the second application to the third observation, which are taken into account by the expert while preparing the second spraying they already had in mind the third and fourth treatment in mind. If anticipation is hard to represent using statechart, such formalism is great to control the temporal dependancies if variables in connection to the decision taken. Namely, our primary analysis showed there is a link between (i) the timing of the first evaluation which update powdery and downy mildews in field variables, (ii) the moment the first symptoms are found in the area (i.e. ILM changes to ’+’), (iii) the timing of the first application and (iv) the decision concerning the powdery mildew application during stage_2. Observed using simulated phenological data, we cannot yet conclude if this particular scenario is possible in real life, but pointing it out is a relevant information that may have been hard to identify in a language did not enhance time computing so well. Instead of using systematic simulation to analyse the sensitivity of the model to its variable variations, we will prefer formal model checking methods (Hélias, 2003; Largouët, 2000) to control the quality of this design. These methods permit to guarantee mathematically, sequential or temporal properties of a process. Making sure the desired situation allways happen or that accident can never exist. Further expert elicitation will thus be necessary to identify the nature of these desired properties for crop protection decision systems. Along with mathematical work to assess the scalability of GrapeMilDeWS at the whole vineyard scale, these works are to lead to transfert. In the mean time, large scale experimentations with a prototype decision support system (DSS) will be starting next year to evaluate the process independantly from its expert designers, with volontary vine growers. In this contest, the rightness of the expert human decision doesn’t hold anymore. The GrapeMilDeWS will be tested in a context where it will hold more knowledge on epidemics and crop protection than most vine growers do. The result of this interactions will be interresting us.

Conclusion

In this article, we have presented the novel approach to modelling an IPM decision system for tactical decisions. We represented our Grapevine powdery and downy Mildews Decision Workflows system, both in its principle and through a short presentation of the modeled result. The elicitation of the knowledge from the IPM expert was an important aspect of the work. We modelled GrapeMilDeWS as a discrete event system, using Statechart, an automaton based graphical programming language. The resulting model has both the advantage of being an accesssible transfert document, as well as a working simulator. Our future work will be focusing on ameliorating the model in a conception loop with the experts; proving the scalability of the solution at the estate scale; and experimenting a DSS prototype software to assess acceptance within the vine growers population of such drastic change in practices.

References

Ascough II, J.C., J.D. Hanson, M.J. Shaffer, G.S. McMaster, and L.A. Deer-Ascough. 1997. Great Plains Framework for Agricultural Resource Management (GPFARM): A 152

decision support system for whole farm/ranch strategic planning Paper. – American Society of Agricultural Engineers, Vol. 2. Attonaty, J.M., M.H. Chatelin, J.C. Poussin, and L.-G. Soler. 1994. OTELO, un simulateur de connaissance pour raisonner équipement et organisation du travail. – Cahier des chambres d'agriculture 66: 37-48. Aubertot, J.N., J.M. Barbier, A. Carpentier, J.J. Gril, L. Guichard, P. Lucas, S. Savary, I. Savini, and M.E. Voltz. 2005. Pesticides, agriculture et environnement: réduire l'utilisation des pesticides et en limiter les impacts environnementaux. – INRA et Cemagref (France). Cartolaro, P., B. Léger, L. Delière, L. Delbac, M. Clerjeau, and O. Naud. 2007. An expert based crop protection decision strategy against grapevine's powdery and downy mildews epidemics: Part 2) Experimental design and results. – IOBC/WPRS Working Group on “Integrated Control in Viticulture” (pers. communication). Clerjeau, M. 2004. Le problème de la décision des interventions phytosanitaires en protection intégrée de la vigne. – Innovigne et Vin, Gruissan (France). Cohen, I.R. 2007. Real and artificial immune systems: computing the state of the body. – Nature Reviews Immunology 7: 569-574. Cournut, S., and B. Dedieu. 2004. A discrete events simulation of flock dynamics: a management application to three lambings in two years. – Animal Research 53: 383- 403. Cros, M.J., M. Duru, F. Garcia, and R. Martin-Clouaire. 2003. A biophysical dairy farm model to evaluate rotational grazing management strategies. – Agronomie 23: 105-122. Cruz-Lemus, J., M. Genero, M. Manso, and M. Piattini. 2005. Evaluating the Effect of Composite States on the Understandability of UML Statechart Diagrams. – Proc. of ACM/IEEE 8th International Conference on Model Driven Engineering Languages and Systems (MODELS/UML 2005). Montego Bay, Jamaica. Girard, N., and B. Hubert. 1999. Modelling expert knowledge with knowledge-based systems to design decision aids: the example of a knowledge-based model on grazing management. – Agricultural Systems 59. Harel, D. 1987. Statecharts: A Visual Formulation for Complex Systems. – Science of Computer Programming 8:231--274. Harel, D., and H. Kugler. 2004. The RHAPSODY semantics of statecharts (or, on the executable core of the UML) - (Preliminary version). –Integration of Software Specification Techniques for Applications in Engineering 3147: 325-354. Hélias, A. 2003. Agrégation/abstraction de modèles pour l'analyse et l'organisation de réseaux de flux : application à la gestion des effluents d'élevage à la Réunion. – Thèse de doctorat, ENSA-M, Montpellier. Kogan, M. 1998. Integrated pest management: Historical perspectives and contemporary developments. – Annual Review of Entomology 43: 243-270. Largouët, C. 2000. Aide à l'interprètation d'une séquence d'images par la modèlisation du système observé. Application à la reconnaissance de l'occupation du sol. – Thèse de doctorat, Université de Rennes I. Lorenz, D.H., K.W. Eichhorn, H. Bleiholder, R. Klose, U. Meier, and E. Weber. 1995. Phenological growth stages of the grapevine (Vitis vinifera L. spp. vinifera) - codes and descriptions according to the extended BBCH scale. – Australian Journal of Grape and Wine Research 1. Lunze, J. 2000. Process supervision by means of qualitative models. – Annual Reviews in Control 24: 41-54. OMG. 2007. Unified Modelling Languages 2.0 [Online]. Available by OMG www.uml.org. 153

Shaffer, M.J., and M.K. Brodahl. 1998. Rule-based management for simulation in agricultural decision support systems. – Computers and Electronics in Agriculture 21: 135-152. Soler, L.-G. 2004. Quelles interventions publiques et privées pour réduire l'utilisation de traitements phytosanitaires dans le secteur du vin? – [Online] http://www.inra.fr/internet/Projets/add-vin/index.htm. SRPV-Aquitaine. 2007. Avertissements Agricoles. – [Online] http://www.srpv-aquitaine.com/_publique/default_publique.asp (posted July; verified oct).

154 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 155-159

Detection of endophytic bacteria in leaves of Vitis vinifera by using Fluorescence in Situ Hybridization (FISH)

Sandra Lo Piccolo1, Gaetano Conigliaro1, Danilo Ercolini2, Livio Torta1, Santella Burruano1 and Giancarlo Moschetti1 1 Dipartimento S.En.Fi.Mi.Zo., Sezione di Patologia vegetale e Microbiologia agraria, Università degli Studi di Palermo, Viale delle Scienze 2, 90128 Palermo 2 Dipartimento di Scienza degli Alimenti, Università degli Studi di Napoli “Federico II”, Via Università 100, 80055 Portici (NA)

Abstract: Previous investigation on five cultivars of healthy Sicilian grapevine allowed the isolation of endophytic bacteria belonging to Bacillus genus from different organs (bud, leaf, stalk and shoot). The aim of this work was to use fluorescence in situ hybridization (FISH) experiments in healthy and damaged leaf tissues of Vitis vinifera to visualize and localize bacteria associated with plant materials. The leaves were cleared to minimize the autofluorescence of the plant fragments. The use of fluorescently labelled bacterial probe Eub338 in FISH experiments on discoloured grapevine leaf disks allowed the estimation of the spatial distribution of different bacterial colonies. At the same time, one cleared disk each foliar sample was minced on Plate Count Agar to detect cultivable endophytic bacteria. The combined microscopic approach showed a differential location of microbial colonies within the leaf tissues examined. Particularly, the bacterial colonies were found in the veins, cells, hairs, intercellular spaces and more in the cut edges of leaf disk. Several bacterial species were isolated from leaf tissue. The high presence of microbial colonies in leaf tissue suggests a potential use of these endophytic bacteria as plant growth promoting and sources of resistance against pathogenic agents, such as fungi. Further studies are required to understand the role of bacterial endophytes in plant-microbe interactions. Besides, the DGGE (Denaturing Gradient Gel Electrophoresis) analysis from foliar disks will be carried out to detect uncultivable endophytic bacterial populations.

Key words: fluorescence in situ hybridization, endophytic bacteria, Vitis vinifera.

Introduction

Endophytic bacteria are ubiquitous in most plant species, residing latently or actively colonizing plant tissues locally as well as systematically. Recently, several endophytic bacteria have been discovered to have beneficial effects on their host plants or to have no observable effects (Hallmann et al, 1997). In fact, bacterial endophytes may act as plant growth promoters (Thomas, 2004); they have also been shown to exhibit strong anti-fungal activity (Hinton et al., 1995), antagonism towards bacterial pathogens (Van Buren et al., 1993) and control plant parasitic nematodes (Hallmann et al., 1995). Moreover, these microorganisms are potential sources of novel natural products for exploitation in medicine, agriculture and industry (Strobel et al., 2004). These endophytes are known to reside in the roots, stems and leaves of plants and bacteria from 82 genera have been found in a broad range of plants (Lodewyckx et al., 2002). The occurrence of endophytic bacteria belonging to Pseudomonas, Enterobacter and Bacillus genera in grapevine has been reported (Bell et al., 1995; Thomas, 2004).

155 156

Previous investigation on five cultivars of healthy Sicilian grapevine allowed the isolation of endophytic bacteria belonging to Bacillus genus from different organs: bud, leaf, stalk and shoot (unpublished data). The aim of this work was to visualize and localize bacteria in healthy and damaged leaf tissues of Vitis vinifera by fluorescence in situ hybridization (FISH) experiments.

Materials and methods

Foliar sampling Four different typologies of grapevine leaves were collected in Sicilian vineyards located in Palermo and Trapani: asymptomatic and Acremonium byssoides infected leaves of five cultivars (Merlot, Nero d’Avola, Insolia, Tempranillo and Cabernet); Plasmopara viticola and esca fungi infected leaves (cv. Catarratto). Twenty-five leaves of each above mentioned typology were harvested and five foliar disks (0,7 cm diam) of each leaf were analyzed.

Clearing and isolation The decolourization of leaf tissue was carried out to minimize the autofluorescence of the plant fragments by using a Phillips and Haymann’s (1970) modified clearing procedure. Leaf disks were soaked for two weeks in 80% ethanol in water, boiled in a KOH aqueous solution (10%) for 20 min, transferred to a 30% H2O2 solution at room temperature until complete clarification (20-40 min), acidified with rapid immersion in 10% HCl in water and rinsed in distilled sterile water. At the same time, one cleared disk of each foliar sample was minced on Plate Count Agar (Oxoid) to detect cultivable endophytic bacteria.

Oligonucleotide probe and FISH The oligonucleotide universal probe EUB338, which hybridizes with a conserved region of all eubacteria, was used. The probe was synthesized by Invitrogen and was labelled with Fluorescein at the 5' end (Table 1). The fluorescence in situ hybridization protocol used was as described by Amann et al. (1990). Leaf disks were placed onto poly- L- lysine-coated slide and fixed by using 4% paraformaldehyde. Fixed foliar disks in PBS (Oxoid) were dried in an oven at 46°C for 10 min and dehydrated successively in 50, 80 and 100% ethanol solutions for 3 min each. Once the ethanol had evaporated, the specimens were used for the hybridization. Pretreatment of all specimens with proteinase K (10mg ml-1; Sigma) was performed at 37°C for 5 min, followed by washing step with ice-cold 1X PBS. The hybridization buffer (25% formamide, 0.9 M NaCl, 0.01% SDS, 20 mM tris-HCl pH 7.2) (30 µl) containing 10 ng of 16S rRNA probe was spotted onto the dry specimens, and the slides were incubated in a dark humid chamber at 46°C overnight. After hybridization, a washing step in a prewarmed washing buffer (20 mM Tris-HCl pH 7.2, 0.01% SDS, 40 mM NaCl, 5 mM EDTA) was performed at 46°C for 30 min. The slides were then rinsed with distilled water and stored in the dark.

Microscopy Microscopic observations were carried out by an epifluorescence microscope (Axiophot, Carl Zeiss, Oberkochen, Germany) coupled to a AxioCam MRc5 (Zeiss) camera. The images were captured using the software AxioVs40 V 4.6.1.0.

157

Results and discussion

The suitability of the clearing of foliar disks was checked by FISH experiments performed in leaf tissues of grapevine (Fig. 1).

bacterial cells hair a b c

Fig. 1. Fluorescence in situ hybridization (FISH) on cleared foliar disk: comparison between the microscopic observations in transmitted light (a), in fluorescent light (b), in both transmitted and fluorescent light (c).

The conditions described in Materials and Methods were found to be the optimum. Leaf tissue disks from healthy and damaged grapevines were examined. The universal bacterial probe Eub338 was used for the detection of the microbial colonies. On asymptomatic grapevine leaves numerous microcolonies (50%) were found. Particularly, many colonies were observed in the veins (Fig. 2a), cells, hairs (Fig. 2b), intercellular spaces (Fig.3a) and more along the cut edges (Fig. 3b) of leaf fragments. Most of the cells showed clear fluorescence signals. At the same way, on Acremonium infected leaves several bacterial colonies (50%) were detected (Fig. 4a). Instead, on P. Viticola (Fig. 4b) and esca fungi (Fig. 4c) infected leaves few epiphytic bacterial cells (3%) were observed.

bacterial cells

hair

vein bacterial colony a b

Fig. 2. Bacterial colonies from the vein detected by using FISH on healthy leaf (a); bacterial cells in the hair detected by using FISH on healthy leaf (b). 158

bacterial cells

bacterial colony vein edge a b

Fig. 3. Bacterial colonies in the intercellular spaces detected by using FISH on healthy leaf (a); bacterial cells from the cut edge detected by using FISH on healthy leaf (b).

mycelium of bacterial P. viticola cells

bacterial colony bacterial vein colony a b c

Fig. 4. Bacterial colony detected by using FISH on: (a) Acremonium infected leaf; (b) Plasmopara viticola infected leaf; (c) esca fungi infected leaf.

For each leaf disk that was positive at the FISH experiments different cultivable bacterial colonies were detected. The high presence of microbial colonies in leaf tissues of asymptomatic grapevine suggests a potential use of these endophytic bacteria as plant growth promoting. By contrast, the lower presence of microcolonies in P. viticola and esca fungi infected leaves suggests a possible competition between bacterial endophytes and pathogenic fungi, therefore a their use as sources of resistance against pathogenic agents. Further studies are required to understand the role of bacterial endophytes in plant- microbe interactions. Besides, the DGGE (Denaturing Gradient Gel Electrophoresis) analysis from foliar disks will be carried out to detect uncultivable endophytic bacterial populations.

Table 1. Oligonucleotide probe used in this study.

Probe Sequence Label Reference

EUB338 (universal for 5'-GCT GCC TCC CGT AGG AGT-3' Fluorescein Amann et eubacteria) al. (1990)

159

References

Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R. & Stahl, D.A. 1990: Combination of 16S rRNA-target oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. – Appl. Environ. Microbiol. 56: 1919-1925. Bell, C.R., Dickie, G.A., Harvey, W.L.G. & Chan, J.W.Y.F. 1995: Endophytic bacteria in grapevine. – Can. J. Microbiol. 41 (1): 46-53. Hallmann, J., Kloepper, J.W., Rodriguez-Kabana, R. & Sikora, R.A. 1995: Endophytic rhizobacteria as antagonist of Meloidogyne incognita on cucumber. – Phytopathology 85: 1136. Hallmann, J., Quadt-Hallmann, A., Mahaffee, W.F. & Kloepper, J.W. 1997: Bacterial endophytes in agricultural crops. – Can. J. Microbiol. 43 (10): 895-914. Hinton, D.M. & Bacon, C.W. 1995: Enterobacter cloacae is an endophytic symbiont of corn. Mycopathologia 129: 117-125. Lodewyckx, C., Vangronsveld, J., Porteous Moore, F., Moore, E.R.B., Taghavi, S. & van der Lelie, D. 2002: Endophytic bacteria and their potential applications. – Crit. Rev. Plant Sci. 21 (6): 583-606. Phillips, J. M. & Haymann, D. S. 1970: Improved procedures for clearing roots and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection. – Trans. Br. Mycol. Soc. 55: 158-161. Strobel, G., Daisy, B., Castillo, U. & Harper, J. 2004: Natural products from endophytic microrganisms. – J. Nat. Prod. 67: 257-268. Thomas, P. 2004: Isolation of Bacillus pumilus from in vitro grapes as a log-term alcohol- surviving and rhizogenesis inducing covert endophyte. – J. Appl. Microbiol. 97: 114-123. Van Buren, A., Andre, C. & Ishmaru, C.A. 1993: Biological control of the bacterial ring rot pathogen by endophytic bacteria isolated from potato. – Phytopathology 83: 1406. 160

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 161-165

Biodiversity in the grapevine: rhizosphere arthropods and mycorrhizal fungi

Alessandra Martorana1, Livio Torta 1, Gabriella Lo Verde 2, Ernesto Ragusa 2, Santella Burruano 1 and Salvatore Ragusa 2 Dipartimento S.En.Fi.Mi.Zo., 1Sezione di Patologia vegetale e Microbiologia agraria, 2Sezione di Entomologia, Acarologia e Zoologia Università di Palermo, Viale delle Scienze 2, 90128 Palermo. – [email protected]

Abstract: A study was carried out on rhizosphere of grapevine (Vitis vinifera L.), in order to know the AM fungi populations and arthropods associated with grapevine roots. One vineyard in Palermo in state of neglect and two vineyards in Alcamo (Trapani), one organically managed and one traditionally managed, were investigated. Three samples of soil and three samples of roots were collected from each locality in spring, summer and autumn 2007. The index of root mycorrhization (IM = percentage of fungal colonization/cm of root), the whole population of both AM fungi (nr of spores/g of soil) and arthropods (nr arthropods/kg of soil), were evaluated using Phillips and Haymann’s technique, Jenkins’ methodology, both opportunely modified, and by the Berlese-Tullgren’s extractor, respectively. Our preliminary results seem to confirm the variability of both AM fungi population and arthropods in the investigated grapevine rhizosphere. Further studies will be carried out to evaluate the possible interaction between AM fungi and arthropodofauna in rhizosphere of grapevine.

Key words: grapevine, rhizosphere, arthropods, AM fungi, biodiversity.

Introduction

Grapevine (Vitis vinifera L.) establishes a mutualistic relationship, defined arbuscular mycorrhiza (AM), between roots and fungi belonging to Glomeromycota (ex Endogonaceae, genera Glomus, Acaulospora, Gigaspora, Scutellospora, ecc.; Stahl, 1900; Petri, 1907; Nappi et al., 1980-81). This symbiosis improves several physiological processes, helping the host growth (Possingham and Groot Obbink, 1971). The establishment and the biological cycle of AM fungi in the soil are directly or indirectly influenced by ecological features. In particular, recent studies have shown that the fungal and arthropods populations in mycorrhizosphere can interact in different ways; mites, collembola, ants, earthworms, ecc. can allow or limit the spreading of both spores and mycelium in the soil as well as infection of the roots (McIlveen and Cole, 1976; Moore et al. 1985; Rabatin and Stinner, 1988). Our research aims to detect the variability of both AM fungi population and arthropods in the investigated grapevine rhizosphere.

Materials and methods

A vineyard in Palermo in state of neglect and two vineyards in Alcamo (Trapani), one organically managed and the other traditionally managed, were investigated (Fig. 1). In spring, summer and autumn 2007 three samples of soil and three samples of roots were collected from each locality.

161 162

Palermo Trapani Alcamo

Fig. 1. Different Sicilian stands sampled

Detection of endophytic AM structures and evaluation of mycorrhization index (IM) The presence of the AM fungi in the roots was verified using the Phillips and Haymann’s technique (1970), opportunely modified by Torta et al. (2003): the root samples were cleared by a treatment in a 10% KOH aqueous solution (for an overnight at 90 °C), treated with H2O2 12 vol. (at room temperature), rinsed in a 10% HCl aqueous solution, simmered for 5’ by immersion in 0,1% acid fuchsine lactophenol. After the treatment, the stained roots were rinsed in clear lactophenol and observed at a stereomicroscope: the index of root mycorrhization (IM = percentage of stained tissue on length of root) was evaluated by a scale sight. The typical mycorrhizal structures were observed dissecting longitudinally and transversally root fragments and mounting slides to observe them with an optical microscope.

Spore extraction and evaluation of AM fungi population The whole population of AM fungi was evaluated using Jenkins’ methodology (1964), opportunely modified by Walker et al. (1982): the soil samples were soaked in water, shaken, poured through a 250 and 50 µm sieves, centrifuged, re-suspended in 40% sucrose solution and filtrated by a polycarbonate filtration system, provided with glass microfibre filter. The extracted spores were counted using a stereomicroscope, mounted in slides and observed by optical microscope for their identification.

Arthropod extraction from soil and evaluation of the populations In order to extract arthropods from soil, Berlese-Tullgren’s extractor was used and the material fallen down was collected and preserved in a jar containing ethanol 70%. Afterwards, the various specimens were separated according to the different order they belonged to.

Results and discussion

The observations on the mycorrhizal status showed that in all vineyards and in all seasons, grapevine roots were infected by AM fungi (Figg. 2a,b), producing clear coils (Fig. 2c), vesicles (Fig. 2d) and arbuscules (Fig. 2e); the evaluation of IM showed similar values in vineyards organically managed and in state of neglect, while lower values were observed in the traditionally managed one (93-38 % and 80-25 %, respectively; Tab. 1). No substantial differences were observed on both composition and density of Glomeromycota populations, even if the number of spores/g of soil increased from spring to 163

autumn in all vineyards (Tab. 1). A high biodiversity of AM fungi (belonging to the genera Glomus, Figg. 3b,c,d,e; Gigaspora, Figg. 3a,f; Acaulospora; Scutellospora, Fig. 3g) was detected. The total number of arthropods, detected during the research, was 2802; the majority of them was recorded in the organically managed vineyard, the lowest in the traditionally managed one (Tab. 2). On the whole, the most represented taxa were mites (78%), while all other taxa showed values below 10%.

b d

Fig. 2. Endomycorrhizal structures in roots of V. vinifera: mycelial development (a,b) and coils (c), vesicle (d) and arbuscule (e), observed at optical microscope.

a b c

def g

Fig. 3. Spores of AM fungi (Glomus b,c,d,e; Gigaspora a,f; Scutellospora g) extracted from V. vinifera rhizosphere, observed at optical microscope.

164

Table 1. Variation of mycorrhizal status (Spores = n° spores/g of soil; IM = percentage of root infection) and of arthropods population (Tot = n° arthropods/kg of soil; N° Taxa = n° arthropods taxa/kg of soil).

Vineyard in state of neglect Vineyard organically Vineyard traditionally Seasons managed managed AM fungi Arthropods AM fungi Arthropods AM fungi Arthropods Spores IM Tot N° Spores IM Tot N° Spores IM Tot N° Taxa Taxa Taxa Spring 4 57 % 145 10 3 90 % 335 11 2 80 % 157 15 Summer 4 93 % 240 13 5 38 % 392 12 5 25 % 269 15 Autumn 8 47 % 573 11 6 80 % 423 11 4 58 % 268 12

Table 2. Total number of arthropods isolated from soil of each investigated vineyard.

Vineyard N° Arthropods In state of neglect 958 Organically managed 1150

Traditionally managed 694

Total 2802

Remarks

The differences found in composition and in population density of both Glomeromycota and arthropods are probably correlated to tillage system and seasonal trend, confirming the previous observations. Further studies will be carried out to evaluate the possible interaction between AM fungi and arthropodofauna in rhizosphere of grapevine.

References

Jenkins, W.R. 1964: A rapid centrifugal-flotation technique for separating nematodes from soil. – Pl. Dis. Reptr. 48: 692. McIlveen, W.D. & Cole, H. Jr. 1976: Spore dispersal of Endogonaceae by worms, ants, wasps and birds. – Can. J. Bot. 54: 1486-1489. Moore, J.C., St. John, T.V. & Coleman, D.C. 1985: Ingestion of vesicular-arbuscular mycorrhizal hyphae and spores by soil microarthropods. – Ecology 66 (6): 1979-1981. Nappi, P., Jodice, R. & Kofler, A. 1980-81: Micorrize vescicolo-arbuscolari in vigneti dell’Alto Adige sottoposti a differenti tecniche di lavorazione del suolo. – Allionia 24: 27-42. Petri, L. 1907: Su le micorrize endotrofiche della vite. – Atti Accad. naz. Lincei Rc. 16: 789. Phillips, J.M. & Haymann D.S. 1970: Improved procedures for clearing roots and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection. – Trans. Br. Mycol. Soc. 55: 158-161. Possingham, J.V. & Groot Obbink, J. 1971: Endotrophic mycorrhiza and the nutrition of grape vines. – Vitis 10: 120-130. 165

Rabatin, S.C. & Stinner, B.R. 1988: Indirect effects of interactions between vam fungi and soil-inhabiting invertebrates on plant processes. – Agricult. Ecosys. & Environ. 24: 135- 146. Stahl, E. 1900: Der Sinn der Mycorrhizenbildung. – Jb. wiss. Bot. 34: 539. Torta, L., Mondello, V. & Burruano, S. 2003: Valutazione delle caratteristiche morfo- anatomiche di alcune simbiosi micorriziche mediante tecniche colorimetriche usuali e innovative. – Micol. It. 2: 53-59. Walker, C., Mize, C.W. & McNabb, H.S. 1982: Populations of endogonaceous fungi at two locations in central Iowa. – Can. J. Bot. 60: 2518-2529. 166 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 167-174

Early detection of selection for resistance in Plasmopara viticola populations treated with organically based fungicides

Caterina L. Matasci1, Davide Gobbin1, Hans-Jakob Schärer2, Christian Stutz1, Lucius Tamm2 and Cesare Gessler1 1 Plant Pathology, Institute of Integrative Biology (IBZ), ETH Zürich, CH-8092 Zürich, Switzerland; 2 Research Institute of Organic Agriculture FiBL, Ackerstrasse, CH-5070 Frick, Switzerland

Abstract: Plasmopara viticola is considered one of the most important grape pathogens worldwide and shows a high risk of resistance development to fungicides. In organic viticulture copper treatments represent nowadays the unique valid control measure against this pathogen. However, the permitted amounts of copper allowed in agriculture are reduced stepwise in Switzerland and Europe. The European project REPCO aims to contribute to the replacement of copper fungicides in organic agriculture by studying and developing new organically based fungicides. We applied a high throughput method based on neutral specific SSR markers for early detection of selection toward resistance in P. viticola natural populations treated with organically based fungicides. Treated and untreated populations shared a comparable number of genotypes, a high percentage of single genotypes, a low occurrence of clones derived from the most frequent genotype and high genetic diversity. We concluded that selection pressure was not exerted on downy mildew populations by Agat-25k, Chitoplant, Novosil, Sonata, Tri-40, Yukka Extrakt, Timorex, Sonata/Chitoplant/KBV 99- 01, Mycosin/Stulln-S/Kocide DF, and the control product Kocide DF; but could not be completely excluded for the organically based fungicide Tecnobiol and Aliette, the latter used in the experiment as control product.

Key words: downy mildew, SSR, copper replacement, monitoring

Introduction

Plasmopara viticola ((Berk. and Curt.) Berl. and de Toni), the causal agent of downy mildew, is one of the most important grape pathogens worldwide and is included in the list of plants pathogens showing a high risk of development of resistance to fungicides (EPPO 1999). Control of the pathogen in organic agriculture strongly depends on copper. Due to the deleterious effects on living organisms in soil, the permitted amounts of copper implemented in agriculture are reduced stepwise in Switzerland and Europe and alternatives are sought. The European project REPCO (Replacement of Copper Fungicides in Organic Production of Grapevine and Apple in Europe) aims to contribute to the replacement of copper fungicides in organic agriculture by studying and developing new organically based fungicides and potentiators of resistance, new biocontrol agents and new integrated management systems. New products or strategies have a chance only if their efficacy is durable in time. Monitoring methods for detecting resistance toward fungicides are traditionally based on bioassay procedures. For P. viticola, tests are performed on detached leaves or in microtiter plates, determining dose response curves for different fungicide classes. These methods allow investigating of possible development of resistance, determining if and when resistance develops, and consent to identify resistant isolates (FRAC 2005).

167 168

In a few cases where the mutation is known, molecular analysis may also reveal the presence of resistant individuals (Hamamoto et al. 2001, Proffer et al. 2006). This method is only applicable if a single, or a combination of few mutations, leads to resistance. In Matasci et al. (2008a) a high throughput method based on neutral specific SSR markers for detecting selection in P. viticola populations is presented. The method is based on the hypothesis formulated by Grünwald et al. (2006) stating that genotypic diversity would progressively decrease in a pathogen population exposed to a fungicide imposing selection pressure toward resistance. In the present work the method was implemented for investigating if selection pressure is exerted on P. viticola populations by a broad palette of fungicides with different and unknown modes of action, which are already in use or are proposed to be used in organic viticulture.

Material and methods

Vineyard and treatments The screening-vineyard was established at Fibl in Frick (Switzerland) in 1997 and consists of 576 vines of cultivars Müller-Thurgau and Chasselas planted alternately. The distance between rows was 2 meters; distance within the rows was 1.1 meters. Treatments were arranged in a „Randomized Block Design“, with four replicates, each consisting of six vines per variety. The plants were treated on schedule (Tab. 1) with an air assisted knapsack sprayer or a compressor assisted sprayer until near run-off. The experiments were carried out following the EPPO-guidelines (EPPO).

Disease severity Disease severity (proportion of diseased leaf area) per plant was calculated by counting the number of leaves with disease from a subset of 20 to 50 leaves per plant and estimating the mean disease severity on the leaves with disease.

Samplings Collected samples, consisting of half a sporulating lesion (about 1 cm2, including some healthy leaf tissue) were excised with a cutter. Coordinates were assigned to the samples to locate their exact position in the vineyard (Gobbin et al. 2003). In 2005 maximally 10 lesions per vine per treatment were collected from six vines of cultivar Müller-Thurgau and six vines of cultivar Chasselas after important infective events. In 2006 maximally 16 lesions per vine per treatment were collected from six vines of cultivar Müller-Thurgau at a single sampling date.

Genotyping Samples collected in 2005 were analyzed following the protocol of Gobbin et al. (2003). To improve amplification efficiency samples collected in 2006 were analyzed using newly designed primers targeting the four polymorphic P. viticola-specific SSR loci, BER, CES, GOB and ISA (Matasci et al. 2008b). Fragments were analyzed as described in Matasci et al (2008b).

169

Table 1. Product name, code, producer, active ingredient, and applied product concentration of the treatments performed in 2005 and 2006 in the experimental field at FiBL in Frick (Switzerland) (Schärer, personal communication).

Product Code Producer Active Applied product name ingredient concentration(%)a 2005 Aliette ALI Aventis Crop Science, 69263 Lyon, F Fosetyl-Al 0.25-0.50 Chitoplant CHI CHIPRO GmbH, Fahrenheitstr. 1, Chitin, Chitosan 0.10-1.00 28359 Bremen, D Sonata COMb AgraQuest, Inc., 1530 Drew Avenue, Bacillus 0.40 Davis, CA 95616, USA pumilus Strain QST 2808 Chitoplant CHIPRO GmbH, Fahrenheitstr. 1, 0.10-1.00 28359 Bremen, D Chitin, Chitosan

KBV 99-01 KBV: Koppert Biological Systems BV, 0-0.15 2650 AD Berkel en Rodenrijs, NL Lactoperoxidase Kocide DF COP Griffin Corp. Valdosta, Georgia, USA Cu-Hydroxide 0.10 Sonata SON AgraQuest, Inc., 1530 Drew Avenue, Bacillus 0.40 Davis, CA 95616, USA pumilus Strain QST 2808 Mycosin STRc Andermatt Biocontrol, Grossdietwil, CH Acidified clay 0-0.80

Stulln-S Gebr. Schaette KG, Sulfur 0-0.50 88332 Bad Waldsee, D Agrostulln, Werksweg 2, 92551 Stulln, D

Kocide DF Griffin Corp. Valdosta, Georgia, USA Cu-Hydroxide 0-0.10 Tri-40 TRI Trifolio-M GmbH, Sonnenstrasse 22, Citrus-extract 0.25 35633 Lahnau, D Yukka Extrakt YUK Deru Ned bv., Natuurlijke Saponin 1.00 Gewasbescherming, Bleiswijk, Bestebreurtje, NL

2006 Agat-25k AGA BIO BIZ Company, 107140 Moscow, RU Inactivated 1.00 Pseudomonas extracts Kocide DF COP Griffin Corp. Valdosta, Georgia, USA Cu-Hydroxide 0.10 Novosil NOV Vorozhtsov Novosibirsk Institute of Extract of 0.50 Organic Chemistry SB RAS, 630090 Siberian pine Novosibirsk, RU tree Tecnobiol TEC Tecnotrea srl, Potassium salt 1.00 Crema, I of linoleic and oleic acid Timorex TIM Biomor Israel Ltd, 16 Menachem Begin Tea Tree Oil 1.00 St. Gama House, Ramat Gan 52521, IL a Concentrations of the products were changed during the season. Values indicate minimal and maximal concentration applied b Treatments 1-6: Sonata+Chitoplant, treatments 7-17: Sonata+Chitoplant+KBV 99-01 c Treatments 1-5 and 11-15: Mycosin+Stulln-S, treatments 6-10 and 16-17: Kocide DF

170

Multilocus genotypic diversity and population structure Genotypic diversity indexes, including Shannon-Wiener’s index H, Shannon’s equitability index EH, evenness index E5 and the genotypic richness were calculated. (Shannon 1949, Pielou 1966, Ludwig et al 1988, Grünwald et al. 2001, 2003, Gotelli & Entsminger 2007, Gobbin et al. 2006, Matasci et al. 2008a). The genotype occurring at the highest frequency was defined dominant genotype (DGEN), while genotypes occurring only once were defined as single genotypes (SGEN) (Gobbin et al. 2005). The contributions of dominant and single genotypes to the total disease incidence (DI) were calculated by dividing the number of lesions generated by DGEN and SGEN by the sample size and were expressed as a percentage (DI%DGEN and DI%SGEN). Population genetic structure was examined by testing the null hypothesis that the allele distribution is not significantly different across populations collected on different host cultivars and treated with different fungicides. Pairwise tests for allelic differentiation were performed by using the overall loci G-statistic, and their significance was evaluated after applying the sequential Bonferroni correction for multiple tests (Gobbin et al. 2006). The degree of differentiation among the plots was quantified using Weir and Cockerham‘s estimator (theta) of Wright‘s FST, as calculated in FSTAT (Weir & Cockerham 1984, Goudet 1995).

Results

In the season 2005 the first P. viticola symptoms were observed on 30/05/2005. In the untreated plot (KON), median disease severity values by the last assessment date (23/08/2005) reached 90.0% and 98.0% on Müller-Thurgau and Chasselas, respectively. Values observed for the two control treatments COP and ALI and for the copper based strategy (STR), were 7.0% and 8.0%, 15.0% and 7.0%, and 10.0% and 7.5%, for the tree treatments and the two cultivars Müller-Thurgau and Chasselas, respectively. Median severity for CHI, COM, YUK SON and TRI were between 47.5% and 90.0% for Müller-Thurgau, and between 66.0% and 95.0% for Chasselas by the last assessment date. The first symptoms of downy mildew in 2006 were observed on 07/06. Median disease severity values in the untreated control plot (KON) reached 8.6% on 28/06. In the copper treated control plot (COP), the median disease severity values were 0.4% on the 28/06. The highest median disease severity values were observed for AGA (8.6%) and NOV (7.8%). Median disease severity values observed for TEC and TIM reached 3.1% and 2.9% on 28/06 (Schärer, personal communication). Data from the five samplings performed in 2005 were pooled. A total of 673 and 419 individuals were collected and genotyped in 2005 and 2006, respectively. A maximum of 115 (KON, in 2005) and 90 (TIM, in 2006), and a minimum of 37 (COM, in 2005) and 36 (TEC, in 2006) individuals were considered per population (Tab. 2). Non significant different FST values (P<0.05) were obtained for P. viticola populations collected on Müller-Thurgau and Chasselas, exposed to the same treatment (year 2005, data not shown). Therefore individuals collected on the two cultivars and exposed to the same treatment were pooled. The lowest diversity was observed in the ALI and in the TEC- populations in 2005 and 2006, respectively. The expected number of genotypes in a sample size of N=36 isolates (corresponding to the largest common population size) ranged from 30 to 35. The lowest g36 values were observed for the ALI (30) and the TEC (31) populations, the highest for COM (35), CHI, SON and the control copper treated COP-population (all three 34). 171

Table 2. Number of individuals completely genotyped in the seasons 2005 and 2006, samples collected at five (17/06, 28/06, 04/07, 20/07 and 24/08/05) and one (29/06/06) sampling dates at FiBL in Frick (Switzerland).

Code 17/06/05 28/06/05 04/07/05 20/07/05 24/08/05 Code 29/06/06 KON 2 31 5 49 28 KON 73 COP 5 32 52 COP 71 ALI 31 43 CHI 3 21 26 AGA 63 COM 11 26 NOV 86 SON 4 12 2 23 12 TEC 36 STR 2 1 17 62 TIM 90 TRI 1 28 5 10 22 YUK 11 51 4 17 24 Total 18 124 25 211 295 Total 419

Table 3. Number of individuals genotyped at all four SSR loci, number of different genotypes, expected number of genotypes calculated for the largest common sample size, dominant- and single genotype-derived disease incidence, Shannon-Wiener’s index, Shannon’s equitability index and evenness calculated for P. viticola populations collected in 2005 and 2006 at FiBL in Frick (Switzerland).

a b c d e f g h Code N g g36 DI%SGEN DI%DGEN H EH E5 2005 KON 115 91 32 66.96 6.96 4.37 0.92 0.77 COP 89 81 34 83.15 3.37 4.36ns 0.97 0.95 ALI 76 59 30 69.74 13.16 3.86*** 0.89 0.63 CHI 50 47 34 90.00 6.00 3.82 ns 0.98 0.95 COM 37 36 35 94.59 5.41 3.57 ns 0.99 0.98 SON 53 49 34 84.91 3.77i 3.87 ns 0.97 0.96 STR 80 68 33 76.25 5.00 4.14 ns 0.94 0.88 TRI 66 53 31 68.18 7.58 3.86 ns 0.92 0.84 YUK 107 81 32 61.68 4.67j 4.26 ns 0.91 0.83 2006 KON 73 61 32 72.60 6.85 4.02 0.94 0.86 COP 71 60 32 76.06 7.04 4.00 ns 0.94 0.85 AGA 63 53 32 77.78 11.11 3.83 ns 0.92 0.74 NOV 86 74 33 76.74 4.651 4.23 ns 0.95 0.88 TEC 36 31 31 83.33 16.67 3.28*** 0.92 0.72 TIM 90 78 33 80.00 5.56 4.27 ns 0.95 0.85 a Individuals genotyped at all four SSR loci b Number of genotypes observed c Expected number of genotypes calculated for the largest common sample size (N=36) estimated using the rarefaction method (Grünwald et al. 2003, Gotelli & Entsminger 2007) d Percentage refers to the SGEN-derived disease incidence [DI%SGEN (NSGEN/N)] e Percentage refers to the DGEN-derived disease incidence [DI%DGEN (NDGEN/N)] f Shannon-Wiener’s index (Shannon & Weaver 1949, Grünwald et al. 2003). The value followed by *** indicates that H in this population was significantly different (P<0.001) from the H of the KON-population of the same year; ns: not significant, negative variance component g Shannon’s equitability (Pielou 1966, Gobbin et al. 2006) h Evenness (Ludwig & Reynolds 1988, Grünwald et al. 2003) i Four different DGEN in the treated population j Two different DGEN in the treated population 172

The Shannon-Wieners’s index H was significantly different between the untreated KON- and the ALI- and TEC- treated populations (Tab. 3). The lowest Shannon’s equitability and evenness index values were observed in the ALI (EH = 0.89, E5 = 0.63) and in the TEC- treated populations (EH = 0.92, E5 = 0.72) (Tab. 3). The proportion of disease incidence by the dominant genotype (DGEN) was more than 10% in the TEC- (16.67%), ALI- (13.16%) and AGA- (11.11%) treated populations. The lowest values were observed in the COP- (3.37%, in 2005), SON- (3.77%), NOV- (4.65%) and YUK- (4.67%) treated populations. Disease incidence derived by single genotypes (SGEN) was equal or higher than 90% in the COM- (94.59%) and CHI- (90.00%) treated populations (Tab. 3).

Discussion

Both years were characterized by high disease pressure. The best protection from P. viticola was obtained with Kocide DF (COP), with the copper based strategy (STR) and with Aliette (ALI). The organically based fungicides Tecnobiol (TEC), Chitoplant (CHI), Timorex (TIM), Yukka extract (YUK) and the combination of Sonata, Chitoplant and KBV99-01 (COM) provided only limited protection against the disease. Agat-25k (AGA), Novosil (NOV), Tri- 40 (TRI) and Sonata alone (SON) totally failed to control P. viticola (Schärer, personal communication). The population diversity indexes H, HE and E5 calculated for untreated and CHI-, COM-, SON-, STR-, TRI-, YUK-, COP-, AGA-, NOV-, TIM treated populations of P. viticola were very similar. HE was close to 1 indicating a high proportion of genotypes occurring only once (SGEN). Treated and untreated populations presented a comparable number of genotypes (estimated by rarefaction), a high percentage of single genotypes and a low occurrence of clones derived from the MFG. This picture is not expected in case of a strong selection pressure toward resistance, where few selected highly frequent genotypes are supposed to colonize the vines. Tecnobiol (TEC) and Aliette (ALI) engendered a significant but very low reduction in diversity in the P. viticola populations. To our knowledge no resistance to products based on potassium salts of linoleic and oleic acids (active ingredients of Tecnobiol) are reported in literature. Instead, resistance to fosetyl-Al (active ingredient of Aliette) was reported for Pseudoperonospora cubensis in Israel and potential risk for resistance development were observed also in the Czech Republic (Cohen & Samoucha 1984, Urban & Lebeda 2007). Insensitivity to fosetyl-Al was reported for Bremia lactucae populations infecting Lactuca sativa in California and for Phytophthora cinnamomi populations infecting the ornamental Chamaecyparis lawsoniana in France (Brown et al. 2004, Vegh et al. 1985). Further tests are needed to determine if fosetyl-Al and Tecnobiol truly exerts selection pressure on P. viticola. We concluded that selection pressure was not exerted on downy mildew populations by the tested organically based fungicides in accord with the general opinion that fungicides with indirect mode of action and/or with multisite activity are unlikely to engender resistance in the target pathogen’s population (Gullino et al. 2000, Urban & Lebeda 2006).

Acknowledgements

The authors are grateful to Rocchina Abbas-Pennella for the labor work, Thomas Amsler for the maintaining work in the vineyard, Niklaus Grünwald for performing the analysis with the algorithm , Marcello Zala, Paolo Galli and Giovanni Broggini for helpful discussion regarding improvement of PCR. This work was funded by SBF 03.0485-1 (Project 501542 REPCO). 173

References

Brown, S., Koike, S.T., Ochoa, O.E., Laemmlen, F. & Michelmore, R.W. 2004: Insensitivity to the fungicide fosetyl-aluminum in California isolates of the lettuce downy mildew pathogen, Bremia lactucae. – Plant Dis. 88: 502-508. Cohen, Y, & Samoucha, Y. 1984: Cross-resistance to four systemic fungicides in metalaxyl- resistant strains of Phytophthora infestans and Pseudoperonospora cubensis. – Plant Dis. 68: 137-139. EPPO Guidelines PP 1/31 (3); PP 1/152 (2); PP 1/181 (2). EPPO/OEPP. 1999: EPPO Standard PP 1/213(1) Resistance Risk Analysis. – EPPO Bulletin 29: 325-347. FRAC. 2005: http://www.frac.info/frac/index.htm. Gobbin, D., Pertot, I. & Gessler, C. 2003: Identification of microsatellite markers for Plasmo- para viticola and establishment of high throughput method for SSR analysis. – Eur. J. Plant Pathol. 109: 153-164. Gobbin, D., Jermini, M., Loskill, B., Pertot, I., Raynal, M. & Gessler, C. 2005: Importance of secondary inoculum of Plasmopara viticola to epidemics of grapevine downy mildew. – Plant Pathol. 54: 522-534. Gobbin, D., Rumbou, A., Linde, C. & Gessler, C. 2006: Population genetic structure of Plasmopara viticola after 125 years of colonization in European vineyards. – Mol. Plant Pathol. 7: 519-531. Gotelli, N.J. & Entsminger, G.L. 2007: EcoSim: Null models software for ecology. Version 7. – Acquired Intelligence Inc. & Kesey-Bear. Jericho, VT 05465. http://www.garyentsminger.com. Goudet, J. 1995: FSTAT (Version 1.2) A computer program to calculate F-statistics. – J. Hered. 86: 485-486. Grünwald, N.J., Flier, W.G., Sturbaum, A.K., Garay-Serrano, E., van den Bosch, T.B.M., Smart, C.D., Matuszak, J.M., Lozoya-Saldaña, H., Turkensteen, L.J. & Fry, W.E. 2001: Population structure of Phytophthora infestans in the Toluca Valley region of Central Mexico. – Phytopathology 91: 882-890. Grünwald, N.J., Goodwin, S.B., Milgroom, M.G. & Fry, W.E. 2003: Analysis of genotypic diversity data for populations of microorganisms. – Phytopathology 93: 738-746 . Grünwald, N.J., Sturbaum, A.K., Romero Montes, G., Garay Serrano, E., Lozoya-Saldaña, H. & Fry, W.E. 2006: Selection for fungicide resistance within a growing season in field populations of Phytophthora infestans at the center of origin. – Phytopathology 96: 1397- 1403. Gullino, M.L., Leroux, P. & Smith, C.M. 2000: Uses and challenges of novel compounds for plant disease control. – Crop Prot. 19: 1-11. Hamamoto, H., Hasegawa, K., Nakaune, R., Lee, Y.J., Akutsu, K. & Hibi, T. 2001: PCR- based detection of sterol demethylation inhibitor-resistant strains of Penicillium digitatum. – Pest Manag. Sci. 57: 839-843. Ludwig, J.A & Reynolds, J.F. 1988: Statistical Ecology: A primer on methods and computing. – John Wiley and Sons, New York. Matasci, C.L., Gobbin, D., Jermini, M. & Gessler, C. 2008: Plasmopara viticola in a mixed cultivars trial. In prep. Matasci, C.L., Gobbin, D., Schärer, H.-J., Tamm, L. & Gessler, C. 2008: Selection for fungicide resistance throughout a growing season in populations of Plasmopara viticola. – Eur. J. Plant Pathol. 120: 79-83. 174

Pielou, E.C. 1966: The measurement of diversity in different types of biological collections. – J. Theor. Biol. 13: 131-144. Proffer, T.J., Berardi, R., Ma, Z., Nugent, J.E., Ehret, G.R., McManus, P.S., Jones A.L. & Sundin, G.W. 2006: Occurrence, distribution, and polymerase chain reaction-based detection of resistance to sterol demethylation inhibitor fungicides in populations of Blumeriella jaapii in Michigan. – Phytopathology 96: 709-717. Shannon, C.E. & Weaver, W. 1949: The mathematical theory of communication. – University of Illinois, Urbana. Urban, J. & Lebeda, A. 2006: Fungicide resistance in cucurbit downy mildew-methodological, biological and population aspects. – Ann. Appl. Biol. 149: 63–75. Urban, J. & Lebeda, A. 2007: Variation of fungicide resistance in Czech populations of Pseudoperonospora cubensis. – J. Phytopathol. 155: 143-151. Vegh, I., Leroux, P., Le Berre, A. & Lanen, C. 1985: Detection on Chamaecyparis lawsoniana ‘Ellwoodii’ of a strain of Phytophthora cinnamomi resistant to fosetyl-Al. – Rev. Hortic. 262: 19-21. Weir, B.S. & Cockerham, C.C. 1984: Estimating F-statistics for the analysis of population structure. – Evolution 38: 1358-1370.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 175-180

Incidence and development of Esca disease in Trentino Province (Northern Italy)

Lorenza Michelon, Chiara Pellegrini, Ilaria Pertot SafeCrop Centre, Istituto Agrario di S. Michele all’Adige, via Mach 1, S. Michele all’Adige TN, 38010, Italy, e-mail: [email protected]

Abstract: Since ancient times Esca represents a problematic disease, which affects grape almost worldwide. This disorder, caused by three main pathogens (Phaeomoniella chlamydospora, Phaeo- acremonium aleophilum and Fomitiporia mediterranea), can have a chronic or acute evolution. In the chronic phase we observe growth reduction of canes and shoots and typical yellowing and necrotic spotting of leaves; in the acute phase a fast dieback (apoplexy) happens. Causes and epidemiological aspects of Esca are not yet fully understood. This work, which is part of a national research project coordinated by University of Firenze (MESVIT), presents a three-year (2005-2007) monitoring in Trentino Province (Northern Italy). The aims of this study were to assess the incidence of esca in the Province; to study the development of symptoms in vineyards during the summer and to verify the presence of airborne spores of P. chlamydospora (Pch) and P. aleophilum (Pal) in infected vineyards and to determine the conditions of their release. In this Province during the monitored period the disease incidence (symptomatic plants in each year) was low and stable. The number of plants showing symptoms at least in one year of the survey was very slowly increasing. Differences in incidence were found between cultivars and age of the plants. The cultivar and weather conditions during summer can influence the development of disease symptoms in the same season. In three experimental vineyards the release of spores was seldom recorded; this fact, combined with the low incidence and slow increase of the disease, implies that the spores of three pathogens have probably a modest role in the spread of disease under Trentino’s conditions.

Key words: Esca, grapevine, monitoring, symptoms, airborne spores

Introduction

Esca is a complex disease, which is present in almost all grape-growing areas of the world, Italy included, and causes significant losses in both young and old vineyards. Three species of fungi have been identified as the main causal agents of Esca disease in Italy: Phaeomoniella chlamydospora (Pch), Phaeoacremonium aleophilum (Pal) and Fomiti- poria mediterranea (Fomed). Airborne inoculum seems to play an important role in the epidemiology of these pathogens in some countries (Eskalen & Gubler, 2001; Eskalen et al., 2004; Larignon & Dubos, 2000), but little is know on the environmental conditions, which promote the formation of fruiting-bodies and influence the patterns of spore dispersal. Studies carried out on spore diffusion of these fungi and the influence of climatic factors on this process show different and, sometimes, contradictory results. In France spores of Pch were trapped all over the year, while the Pal spores were found only during the vegetative period (Larignon & Dubos, 2000). In California only Pch spores were captured, especially from October-November to April (Eskalen & Gubler, 2001). On the contrary, in South Africa spores of these pathogens were not trapped (van Niekerk et al., 2006) as well as in Australia, where sporulation was rarely observed on transversely-cut surface of naturally-infected rootstocks and cordons (Edwards et al., 2001).

175 176

Some doubts on symptoms expression still exist. In fact, the infected vines may not show visible symptoms for several years after their first appearance and the cause of this fluctuation is still unknown. It seems that a cool and rainy summer induces the expression of chronic symptoms, while a warm and dry summer promotes the acute syndrome (Surico et al., 2000). For these reason, in order to obtain a precise evaluation of Esca incidence and evolution in the vineyard, it is necessary to carry out the surveys over a long period (several years). This work presents the results of a three-year monitoring of Esca disease in Trentino Province (Northern Italy).

Material and methods

Incidence in Trentino Province The evaluation of Esca incidence in Trentino Province was carried out in the main wine- producing areas. A representative sample of 100 vineyards with different cultivars, ages and trellis systems was chosen. In each vineyard an experimental plot (200 plants) was observed from the end of August to the mid of September: the vines with chronic or acute symptoms on leaves were recorded, marked and for each plot the incidence of manifest Esca (percentages of plants showing symptoms of Esca in the year of survey) and cumulated Esca (plants that showed Esca’s symptoms at least once during the period 2005-2007) were calculated.

Expression of symptoms From June to September (2005, 2006 and 2007), two experimental vineyards (1 and 2) located in S. Michele all’Adige (TN) and Faedo (TN) were weekly monitored to verify the date of first symptom appearance and to study the evolution of Esca’s symptoms during the season. The aim was to possibly correlate disease expression with weather conditions. In Vineyard 1, two cultivars were monitored, Traminer (25-year-old) and Incrocio Manzoni (8-year-old); in Vineyard 2, 22-year-old Chardonnay vines were examined. In each vineyard symptom assessments were carried out in an experimental plot of about 1000 plants.

Monitoring of spores From March 2005 to September 2007, the presence of airborne spores of Pch and Pal was weekly verified in the two selected vineyards. For this study two different sampling methods were used. In Vineyard 1 a continuous volumetric spore trap (Burkard Spore Trap) was placed; the tape, coated with silicon solution, was weekly replaced. In Vineyard 2 the spores were collected on microscope slides with sticky surface (vaseline) placed at 1-2 cm from the surface of pruning wounds and changed every two weeks. From October 2006 to September 2007 we checked a third additional vineyard (vineyard 3); in this field, the incidence of disease observed in the past years was high and, in October 2006, fruiting bodies of F. mediterranea were found on the trunk of two vines. In this case the spores were trapped on microscope slides with sticky surface as in Vineyard 2. The spores collected on tape and slides were rinsed with 10 ml of sterile water, this water was double filtered (5 µm and 0.45 µm, cellulose acetate syringe filters; ALBET-JACS, Spain); the second filter was then washed with 1 ml of sterile distilled water, which was put (200 µl) in Petri dishes containing potato dextrose agar (PDA; Oxoid) amended with 0.1 g/l of chloramphenicol (Sigma) and incubated at 25°C (Eskalen, personal communication; Eskalen & Gubler, 2001; Larignon & Dubos, 2000). The plates were observed every 5 days to verify the presence of colonies of the pathogens.

177

Results and discussion

Incidence in Trentino Province The results of the three-year monitoring in Trentino Province (Figure 1) show that the incidence of Esca disease in this area can be considered low (1.1% in 2005, 1.4% in 2006 and 1.6% in 2007) and stable in the time. The cumulated Esca shows that new symptomatic plants can appear each year; this fact remarks the importance of evaluating the disease for several years in order to estimate the real incidence of disease in the vineyard.

Manifest Esca Cumulated Esca 5

3 Incidence (%) 2

0 2005 2006 2007

Fig. 1. Average incidence of Esca (manifest and cumulated) in Trentino Province in the three-year monitoring (2005-07). Bars represent standard error.

5 R2 = 0.9963 4

2 Cumulated Esca (%) 1

0 <10 years old 10-20 years old >20 years old

Fig. 2. Cumulated incidence in relation with age: a linear trend can be observed with the increase of vines age. Bars represent standard error.

178

The oldest vineyards are the most affected while in the youngest vineyards the percentage of Esca is much lower (Figure 2). The 13 surveyed cultivars show similar incidence of manifested Esca among the three years of monitoring, but differences were found between cultivars (Table 1). The results confirm the high susceptibility of Nosiola, Sauvignon Blanc, Műller Thurgau and Traminer for the white varieties, of Cabernet Sauvignon and Marzemino for the red varieties. On Teroldego only one symptomatic plant was found during surveyed period, even if the average age of the observed six vineyards is high. Therefore we can consider this cultivar to be not very susceptible to the disease in Trentino Province.

Table 1. Manifest and cumulated (2005-07) Esca incidence in the different cultivars monitored in Trentino Province. Plant’s age represent the average age of plants in the surveyed vineyards of each variety in 2007.

Cultivars Incidence (%) Plants’age 2005 2006 2007 Cumulated (average years) Pinot grìs 0.1 0.3 0.2 0.6 15 Traminer 2.3 2.0 3.8 6.6 17 Sauvignon Blanc 3.1 1.4 5.3 9.8 18 Cabernet Sauvignon 2.1 2.1 4.1 7.0 18 Pinot noir 1.5 1.8 2.0 1* 19 Chardonnay 0.9 1.3 1.2 3.3 21 Merlòt 0.8 0.9 0.8 1.8 21 Marzemino 1.8 2.1 1.7 4.9 21 Műller Thurgau 1.2 3.3 3.2 6.7 21 Lagrein 0.0 0.5 0.0 0.5 22 Nosiola 4.3 8.0 4.0 13.0 24 Teroldego 0.0 0.1 0.0 0.1 25 Schiava 0.8 0.8 0.8 2.2 28 *Cumulated incidence of Pinot noir is lower than manifest incidence because, in 2006, one vineyard was renewed.

Expression of symptoms The three-year summer monitoring in Vineyard 1 and 2, indicates different times of appearance of symptomatic plants in relation to weather conditions and cultivars. In 2007, probably because of the high mean temperature in spring, the first symptomatic vines appeared at the beginning of June; in 2005 and 2006 the symptoms appeared only from the end of June (Vineyard 1, cultivar Traminer) to the middle of July (Vineyard 2, cultivar Chardonnay) (Figures 3-4). During the three years in both vineyards the incidence quickly increased during the first part of the season and then remained steady from the end of August to the end of September. In 2006 and only in Vineyard 2, new additional symptomatic plants appeared in the second week of September, probably because of a rainy August (Marchi et al., 2006; Surico et al., 2000). In Vineyard 1, only chronic symptoms with typical tiger-stripes on leaves were observed; instead, in vineyard 2, vines with acute symptoms (apoplexy) were also recorded, especially in 2005. In 2007, the incidence of disease was higher than in 2005 and 2006 in the two vineyards; in the same year we can also observe the highest values of cumulated rain. 179

Cumulated rain 2005 (mm) Cumulated rain 2006 (mm) Cumulated rain 2007 (mm)

Incidence 2005 (%) Incidence 2006 (%) Incidence 2007 (%)

450 6 Vineyard 1 400 5 350 300 4 250 Rain (mm) 3

200 Incidence (%) 150 2 100 1 50 0 0 Jun July Aug Sep

Fig. 3. Evolution of Esca incidence in relation with cumulated rain in the three years of survey in Vineyard 1 (cultivar Traminer).

Cumulated rain 2005 (mm) Cumulated rain 2006 (2006) Cumulated rain 2007 (mm)

Incidence 2005 (%) Incidence 2006 (%) Incidence 2007 (%)

450 Vineyard 2 6 400

5 350 300 4 250 Rain (mm) 3

200 Incidence (%) 150 2 100 1 50 0 0 Jun July Aug Sep

Fig. 4. Evolution of Esca incidence in relation with cumulated rain in the three years of survey in Vineyard 2 (cultivar Chardonnay).

180

Monitoring of spores From April 2005 to January 2007 only a very low number of spores of the two main fungi involved in the disease (Pch and Pal) were found. In Vineyard 1 (spore trap) six spores of Pch were captured (four spores on the 8th of March 2006 and 2 spores on the 22nd of June 2006); in Vineyard 2 (microscope slides) two spores of Pal were trapped (one spore during the last week of July 2006 and one spore during the second week of April 2007). In Vineyard 3 (microscope slides) two spores of Pal were collected only during the third week of March 2007. The most plausible hypotheses to explain these results are: i) in Vineyards 1 and 2 the production of spores is too low to be detectable; the low Esca incidence (see previous paragraph) and the absence of fruiting bodies of the two fungi support this hypothesis; ii) in Vineyard 3 the incidence of disease is higher than in the other two (19.5% of Esca cumulated from 2005 to 2007), but only a little number of spores was collected probably because of the limited sporulation activity. In both cases we can consider that the level of spores of Esca causal agents is very small in the air; this implies that the role of spores of three pathogens has probably little significance in the spread of disease under Trentino’s conditions.

Acknowledgements

We thank the Meteorological Unit of Istituto Agrario di S.Michele a/A, CAT and “Cantina Endrizzi” for the collaboration in the experimental trials. The research was supported by the project MESVIT funded by ARSIA and SafeCrop Centre, a project funded by Fondo per la Ricerca, Autonomous Province of Trento.

References

Edwards, J., Laukart, N. & Pascoe, I.G. 2001: In situ sporulation of Phaeomoniella chlamydospora in the vineyard. – Phytopathol. Mediterr. 40: 61-66. Eskalen, A. & Gubler, W.D. 2001: Association of spores of Phaeomoniella chlamydospora, Phaeoacremonium inflatipes, and Pm. aleophilum with grapevine cordons in California. – Phytopathol. Mediterr. 40 (Supplement): 429-432. Eskalen, A., Latham, S.R. & Gubler, W.D. 2004: Spore release of Phaeomoniella chlamydospora associated with grapevine cordons in California. (Abstr.) – Phytopathology 94: S28. Larignon, P. & Dubos, B. 2000: Preliminary studies on the biology of Phaeoacremonium. – Phytopathol. Mediterr. 39: 184-189. Marchi, G., Peduto, F., Mugnai, L., Di Marco, S., Calzarano, F. & Surico, G. 2006: Some observations on the relationship of manifest and hidden esca to rainfall. – Phytopathol. Mediterr. 45: 117-126. Surico, G., Marchi, G., Braccini, P. & Mugnai, L. 2000: Epidemiology of esca in some vineyards in Tuscany (Italy). – Phytopathol. Mediterr. 39: 190-205. van Niekerk, J.M., Halleen, F. & Fourie, P.H. 2007: Spore dispersal patterns of grapevine trunk pathogens in South Africa. – Phytopathol. Mediterr. 46: 115-116.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 181-187

Preliminary studies on “esca” disease in Sicilian vineyards

V. Mondello, G. Conigliaro, A. Alfonzo, V. Ferraro, L. Torta , S. Burruano Dipartimento S.En.Fi.Mi.Zo., Sezione di Patologia vegetale e Microbiologia agraria, Università di Palermo, Viale delle Scienze 2, 90128 Palermo;

Abstract: Observations on the epidemiology of the “esca” disease and on fungi associated with xylematic symptoms were carried out for two years in two Sicilian vineyards with different cultivars (Insolia, Catarratto, Italia). The incidence of plant showing “esca” symptoms, similar in both investigated localities, increased during summer; at the same time sporadic grapevines with apoplexy were noted in July. Moreover a bit difference seems emerge regarding severity of disease among the cultivars. P. clamydospora (Pch) was associated with all internal symptoms; Phaeoacremonium aleophylum (Pal) was only isolated from pink wood, while, Fomitiporia punctata (Fomed) was especially isolated from white rot. These results, while confirming the etiological plurality of “esca” in Sicily, show also two vascular fungi (Pch and Pal) and one wood decaying (Fomed) as responsible of the disease.

Keywords: wood, disease, esca, aetiology, epidemiology

Introduction

“Esca” is a grapevine disease, spread in many viticultural areas of the world (Europe, America, South Africa, Australia), causing both the death of plants and loss of production. It is usually present in old vineyards. In the last 10-15 years, an evident recrudescence of “esca” occurred in different areas, due to several factors: changes in vineyard management and in cultural practices; reduced sanitary care of propagation materials; banning of sodium arsenite for the control of the disease and its substitution with less efficient fungicides. The “esca” symptoms can be internal or external: the external ones appear on leaves in summer and consist in chlorotic areas between veins and/or leaf margins, then turning in necrotic spots (“chronic esca”, fig 1a); sometimes, they may consist in a rapid necrosis of any or all the shoots (“acute esca” or “apoplexy”, fig.1b, 1c). The internal symptoms, visible on wood’s trunk, are several: brown streaks, pink-brown areas and white rot (fig. 1d, 1e). Recent studies (Graniti, 1999, Mugnai et al, 1996, 1999) have identified Phaeoacremo- nium aleophilum (Pal), Phaeomoniella chlamydospora (Pch) and Fomitiporia mediterranea (Fomed) as fungi involved, in combination or in succession, in the disease development. In reason of the different interactions occurring among fungi, hosts and environment, the expression of esca shows different syndromes and discontinuity in time and space. In 1998 a decline of young grapevines of cv. Victoria caused by Phaemoniella clamydospora has been observed in a vineyard in south-eastern Sicily; this was the first report for Italy (Sidoti et all., 2000). For this reason it was considered opportune to investigate on the presence of esca in some Sicilian vineyards.

181 182

Materials and methods

Observations on the epidemiology of the disease and on fungi associated with xylematic symptoms were carried out for two years (2005, 2006). In particular, the study was conducted in two vineyards, located in Alcamo (TP) and in Campobello di Licata (AG), respectively, and on three cultivars: Insolia, Catarratto (Alcamo) espalier trained and not irrigated, and Uva Italia (Campobello di Licata,) grown (managed) as “pergola” and irrigated. At the beginning of June, for the two years, a block of 1000 vines for each cultivar was monthly surveyed for a total of 5 surveys/year for Insolia and Catarratto and 3 surveys/year for Uva Italia. At each survey, all the plants were observed to individuate the external acute or chronic esca symptoms. Affected plants were visually rated on a scale based on the percentage of symptomatic leaves or shoots (fig. 2). For each survey, a two-dimensional map was generated showing the health status of the block at that moment. Temperature and rainfall data, collected by the Servizio Informativo Agrometeorologico Siciliano (SIAS) in weather stations near the observed vineyards, were utilized to correlate weather to disease incidence and intensity. At the same time, for isolation of the fungi associated with xylematic symptoms, for each survey and near the vines blocks, one affected plant for cultivar was uprooted and examined in laboratory. In detail, each trunk was first sectioned in three parts (upper, middle and lower), then stripped the bark off and surface sterilized by flame. Each section was further sectioned in lengthwise in aseptic conditions and observed to recognize internal wood symptoms: for each internal symptom (brown streaks, pink-brown areas, white rot), small wood pieces were placed on Petri dishes with Malt extract Agar (MEA). The inoculated Petri dishes were kept at 25° C. The developed colonies were singly transferred in pure culture for identification.

White rot

Brown streaks

a b d

pink-brown wood c e Fig. 1. “Esca” external and internal symptoms: a) “chronic esca” on leaves; b) necrosys of some shoots (“acute esca”); c) necrosys of all shoots (apoplexy); d,e) brown streacks, pink-brown wood and white rot. 183

Class of description severity 1 acute or chronic symptoms on 1-10% of leaves or shoots 2 acute or chronic symptoms on 1-10% of leaves or shoots 3 acute or chronic symptoms on 11-25% of leaves or shoots 4 acute or chronic symptoms on 26-50% of leaves or shoots 5 acute or chronic symptoms on 51-75% of leaves or shoots

Fig. 2. Scale to evaluate the esca severity in field

Results and discussion

Epidemiology The observation showed the esca presence and development in both vineyards. In Alcamo, either in 2005 and 2006, the disease incidence increased gradually up to percentages of 9,6%, showing a similar course in the two cultivars (fig. 3d, 3f). Differences between Insolia and Catarratto were detected in disease intensity of “acute” and “chronic” esca: even if both cultivars showed the prevalence of chronic symptoms, Insolia seemed to be more susceptible to “acute” esca than Catarratto (fig. 3e, 3g ). In Campobello di Licata the disease incidence in 2005 and 2006 was lower than in Alcamo, 7,3% and 5,8% respectively; differently from Alcamo, the values of “acute” and “chronic” esca intensity were similar in both years, with no prevalence of a particular syndrome (fig. 3).

cv. Catarratto - Alcamo cv. Insolia- Alcamo cv. Italia – C. di Licata

Fig. 3. Development of esca incidence (upper graphics) and severity (lower graphics) in the three cultivars during the two years of observation.

The comparison of the last 2-D survey maps of the two years put in evidence the discontinuity of esca expression in field: the percentage of surveyed vines that showed esca 184

symptoms in both the years was only the 2,3% in Insolia, 1,4% in Catarratto and 1,3% in Uva Italia (figg. 4a, 4b and 4c); Correlation of disease incidence and intensity with climatic data (temperature, R.U, rainfall) evidenced no apparent relation between disease expression and weather.

Isolation assays Observations on wood internal symptoms revealed how the most common alterations in xyleme of affected vines were: brown wood-streaking and vascular gummosis, pink-brown wood and white rot of soft consistency, often bordered by a black line. The identification of associated fungi with each xylematic symptom and the relative I.F. (reported in tab.1) shows that P. clamydospora (Pch, fig, 5a) was associated with all internal symptoms; Phaeoacremonium aleophylum (Pal, fig.5b) was isolated only from pink wood, while Fomitiporia mediterranea (Fomed, fig. 5c) was especially isolated from white rot. These results, while confirming the etiological plurality of “esca” in Sicily and discontinuity of symptomatic expressions in field, show also two vascular fungi (Pch and Pal) and one wood decaying (Fomed) as constantly associated with the disease.

Acknowledgements

Research study commissioned from ARSIA-Toscana (Regional Agency for Development and Innovation in Agriculture and Forest) by fourteen administrative Regions and one autonomous province, and financed with funds provided by the Ministero per le Politiche Agricole e Forestali (Ministry for Agriculture and Forestry Policy) to implement the inter-Regional Project “Grapevine esca: research and experiment in the nursery and in the field for prevention and cure.”

References

Crous P.W., Gams W., 2000: Phaeomoniella chlamydospora gen et comb nov., a causal agent organism of Petri grapevine decline and esca. – Phytopat. Medit. 3: 112 -118. Graniti G., Surico G., Mugnai L., 1999: Considerazioni sul mal dell’esca e sulle venature brune del legno della vite. – Informatore Fitopatologico 49(5): 6-12. Mugnai l., Imbriani R., Surico G., 1996: Indagine sulla diffusione e gravità del “mal dell’esca” in alcuni vigneti della Toscana. – Informatore Fitopatologico 46(6): 50-56. Pollastro S., Dongiovanni C., Abbatecola A., Faretra F., 2000: Observation on the fungi associated with esca and on spatial distribution of esca-symptomatic plants in Apulian (Italy) vineyards. – Phytopat. Medit. 39: 206-210. Sidoti A., Buonocore E., Serges T., Mugnai L., 2000: Decline of young grapevines associated with Phaeoacremonium chlamidosporum in Sicily (Italy). – Phytopat. Medit. 39: 87-91. Sparapano L., Bruno G., Ciccarone C., Graniti G., 2000: Infection in grapevines by some fungi associated whit esca. II interaction among Phaeoacremonium chlamidosporum, P. aleophilum and Fomitiporia punctata. – Phytopat. Medit. 39: 53-58 Surico G., Marchi G., Braccini P., Mugnai L., 2000: Epidemiology of esca in some vineyards in Tuscany (Italy). – Phytopathologia Mediterranea 39: 190-205

185

third survey fourth survey fifth survey

A= symptomatic plant in 2005; B=symptomatic plant in 2006; C= apopleptic plant in 2005 ; D= apopleptic plant in 2006; AB= symptomatic plant both 2005 and 2006 = dead plant before surveys. Fig. 4. Discontinuity of esca expression in field: two-dimensional maps of the last three surveys of cv. Catarratto.

186

third survey fourth survey fifth survey

A= symptomatic plant in 2005; B=symptomatic plant in 2006; C= apopleptic plant in 2005 ; D= apopleptic plant in 2006 AB= symptomatic plant both 2005 and 2006; = dead plant before surveys.

Fig. 4b. Discontinuity of esca expression in field: results of the comparison of the two years 2D-maps of the last three surveys on cv. Insolia.

187

third survey A= symptomatic plant in 2005; B=symptomatic plant in 2006; C= apopleptic plant in 2005 ; D= apopleptic plant in 2006 AB= symptomatic plant both 2005 and 2006; = dead plant before surveys.

Fig. 4c. Discontinuity of esca expression in field: results of the comparison of the two years 2D-maps of the last survey on cv. Italia.

a b c

Fig. 5. Pure culture colony and microphotography of : a) P. chlamydospora; b) P. aleophilum and c) F. mediterranea. 188

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 189-192

“Coptimizer”: a decision support system to reduce copper in organic viticulture

Daniele Prodorutti1, Alberto Pellegrini2, Shoham Simon3, Yochai Gafni3, Tsvi Kuflik3, Andrea Frizzi2, Ilaria Pertot2 1 Plant Protection Department, 2 Safecrop Centre, Istituto Agrario di San Michele all’Adige, via E. Mach 1, 38010 S. Michele all’Adige (TN), Italy, email: [email protected]; 3 MIS Department, University of Haifa, Mount Carmel, Haifa, 31905, Israel

Abstract: “Coptimizer” is a web-based decision support system (DSS), which can help growers and to optimize copper treatments against downy mildew on grapevine. The system assists the user in optimizing dosage and timing of the treatments and calculates the amount of copper applied in each treatment in a specific vineyard. It monitors the accumulated quantity of copper used during the season, giving a warning when the grower is close to a pre-defined threshold. In 2007 the DSS prototype was tested in two locations in northern Italy (Trentino Province where two vineyards were treated with copper according to the system’s recommendations. During the growing season the incidence of the disease was monitored and compared to untreated and weekly treated plots (standard practice). Coptimizer’s recommendations allowed a good control of downy mildew while reducing significantly the amount of copper applied per hectare compared to the standard practice.

Key words: software, downy mildew, copper, grapevine

Introduction

“Coptimizer” is a decision support system (DSS) developed to help growers and advisors in optimizing copper treatments against downy mildew on grapevine, especially when there is a limitation on the maximal amount of copper allowed to be used. This is the case of copper in organic agriculture, where a fixed amount per year per hectare (or an average in a number of years per hectare) is set in Europe (Commission Regulation EC 473/2002, http://eur- lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32002R0473:EN:HTML). Once applied on leaves, copper is washed away by rain. Copper is active only at application site, new plant growth results in unprotected leaves. Moreover downy mildew damages vary according to phenological stage of grapevines. Conditions for the infection are simplified in a model, which assumes, with concomitant suitable temperatures for the infection, the risk of rain in the following day equal to the risk of disease infection. The decision “to treat or not to treat” is taken based on a combination of two indices: “plant still protected by the previous treatment” and “risk of infection in the following day”. Coptimizer recommends copper applications according to: growing stage, plant growth (shoot length, new leaves since the last treatment), weather data (rain, average temperature), copper residues on leaves (as a function of copper concentration of the previous treatment and amonut of rain since the last treatment), weather forecast (risk of rain in the following days). Therefore the treatment is applied only when it is needed and copper dosages are adapted according to residues assumed to be present on leaves, forecasted amount of rain and plant phenological stage. The system has two complimentary functional components: a web based decision support system that allows growers to plan and record treatments during the growing season and a

189 190

web based system that allows extension services to provide growers necessary weather information (needed for the decision model). In addition the system assists the grower in calculating copper concentrations for the treatments and the amount of copper applied in each treatment. It monitors the copper quantity already used, giving a warning when the grower is close to the fixed threshold (Figure 1). For a more detailed system description, the reader is referred to Kuflik et al. (2007).

Fig. 1. Coptimizer web page for vineyard management. For each vineyard it is possible to insert or to see automatically several parameters and the system provides recommendations on the treatments. Coptimizer calculates the total amount of copper applied per hectare during the growing season giving a warning when this value is close to the fixed threshold.

Material and methods

In 2007 the DSS prototype was validated in two vineyards of Trentino (northern Italy): S. Michele all’Adige (cv. Schiava) and Navicello (cv. Cabernet Sauvignon). The selected vineyards, having respectively a low and high downy mildew infection risk, were treated with copper hydroxide (Kocide® 2000, DuPont) according to Coptimizer’s recommendations (86, 145, 200 g/hl corresponding to 30, 50 and 70 g Cu2+/hl respectively). During the growing season the incidence of downy mildew was assessed on leaves and bunches and compared to the untreated control and weekly treated plots (standard practice). Four replicates (plots) were used for each treatment. The amount of metallic copper sprayed per hectare was calculated. In the standard practice from 145 to 200 g/hl of Kocide® were used, depending on the growing stage. In each treatment the quantity of solution applied was 12 hl/ha. Sulphur applications (Thiovit®, Syngenta, 300 g/hl) were used to control powdery mildew infections. Ripe grapes were collected from the two vineyards to measure the sugar content (Brix degrees), the total acidity (g/l tartaric acid) and the copper residue (mg/kg). 191

Results and discussion

In 2007 downy mildew symptoms appeared especially on developed clusters as brown rot. In the two vineyards disease incidence was very similar between Coptimizer and weekly treated plots (P>0.05, Kruskal-Wallis test) and approximately 10 % on leaves and 20 % on bunches (Fig. 2).

S. Michele a/A (cv. Schiava) B 100 On leaves 90 On 80 bunches 70 b 60 50 Incidence 40(%) 30 A A 20 a a 10 0 Cu standard practice Coptimizer Untreated

Navicello (cv. Cabernet sauvignon) 100 On leaves b B 90 On 80 bunches 70 60

50 Incidence (%) 40 30 A A 20 a a 10 0 Cu standard practice Coptimizer Untreated

Fig. 2. Downy mildew incidence in the last assessment before harvest on leaves and bunches of the two vineyards (S. Michele and Navicello). Different small and capital letters indicate significant differences of downy mildew incidence respectively on leaves and bunches at P≤0.05 (Kruskal-Wallis test).

192

Untreated plots showed, instead, a significantly higher incidence of downy mildew (P≤0.05, Kruskal-Wallis test), reaching 90 % on bunches at harvest (Figure 2). Following the system recommendations it was possible to reduce the number of treatments and the amount of copper applied per hectare in both vineyards (Table 1). At the end of the growing season the total amount of metallic copper applied following Coptimizer DSS was lower than the fixed threshold of 6 kg/ha/year and almost half quantity if compared to the standard practice (Table 1, Figure 1).

Table 1. Number of copper applications and total amount of copper sprayed per hectare at the end of the growing season in the standard practice and according Coptimizer DSS.

Copper applications Copper amount Vineyard Copper treatment (No.) (Kg/ha) Standard practice 13 9.96 S. Michele Coptimizer 10 4.56

Standard practice 14 9.87 Navicello Coptimizer 9 4.92

Regarding the effects on quality, grapes from Coptimizer and standard reference treatments had the same sugar content (20-21° Brix) and acidity (7-8 g/l tartaric acid) while lower sugars and a higher acidity (17-18° Brix and 10 g/l tartaric acid) were observed on grapes of untreated plots. The copper residues on bunches of Coptimizer plots were definitely lower than the maximum residue level (20 mg/kg for wine grapes in Italy, Ministero della Salute, D.M. 27 August 2004, http://www.ministerosalute.it/alimenti/sicurezza/sicApprofondimento.jsp?lang= italiano&label=pro&id=400). Residues were 1.8 and 1.4 mg/kg for the samples collected in S. Michele and Navicello respectively. In conclusion, Coptimizer recommendations allowed a good control of downy mildew, significantly reducing the amount of copper applied per hectare and staying under the threshold of 6 kg/ha/year. On the basis of these results, in 2008 Coptimizer will be available on-line and accessible to growers and advisors to validate the system in different growing areas of Trentino Province. A large scale use of the system is expected in the future years.

Acknowledgements

The research was supported by SafeCrop Centre, funded by Fondo Unico per la Ricerca, Autonomous Province of Trento (trial in S. Michele) and by the project “Studi finalizzati ad ottemperare alle limitazioni dei quantitativi di rame o mediante l’impiego di formulazioni a basso dosaggio o con l’adozione di mezzi alternativi” funded by MiPAF, Italy (trial in Navicello). The authors thank Mario Ramponi (IASMA Research Centre) for copper residue analysis.

References

Kuflik, T., Gessler, C., Pertot, I., Simon, S., Gafny, Y., Soffer, E. & Hoch, D. 2007: Towards Agriculture: Supporting Growers Online – the Case Study of SafeCrop. – In: Proceedings of MCIS 2007, October 4-8, 2007, Venezia, Italy.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 193-201

Impact de l'estimation des pluies par radarsur la representation spatiale de la modelisation du risque d'epidemie de mildiou sur le vignoble de Bordeaux

Marc Raynal 1, Christian Debord 1, Katell Griaud 2, Serge Strizyk 3, Denis Boisgontier 4, Jean Congnard 5, Didier Grimal 5 1 Institut Français de la Vigne et du vin (IFV) 39, rue Michel Montaigne, 33290 Blanquefort; 2 Faculté œnologie, Master viticulture œnologie environnement, université Bordeaux II, 351 cours de la libération, 33405 Talence; 3 SESMA : 40 rue des frères Flavien, 75020 Paris; 4 CAP2020 / NOVIMET : 2 allée du chemin neuf, 91720 Gironville sur Essone; 5 Météo France, 7 avenue Roland Garros, 33700 Mérignac

Abstract: Le vignoble bordelais, particulièrement soumis aux conditions océaniques et aux risques cryptogamiques qui en découlent, contribue pour une part significative à la consommation nationale d’intrants phytosanitaires. Ces pratiques sont aussi bien néfastes pour l’environnement que pour l’image du produit et les finances du viticulteur ! Dans l’objectif de maîtriser ces intrants, l’IFV s’est très tôt intéressé aux outils de modélisation tels que les modèles Potentiels Systèmes de S. Strizyk. L’amélioration de ces outils se heurte aujourd’hui à l’imprécision des variables climatiques et notamment à l’approximation de la pluviométrie. Afin de remédier à cet état de fait, l’IFV a testé, au cours de la campagne 2007, le produit Antilope de Météo France qui s’appuie sur la technologie RADAR et qui fournit une donnée pluviométrique tous les kilomètres. L’évaluation strictement météorologique du produit Antilope montre que ces données ont tendance à sur-estimer les faibles précipitations et, à l’inverse, à sous-estimer les fortes précipitations. Néanmoins, Antilope met en évidence certaines cellules pluvieuses, non perçues par les stations météorologiques au sol, et pouvant être à l’origine de certaines contaminations. L’utilisation d’Antilope dans le modèle Potentiel Système mildiou version 2007 ne donne pas de meilleurs résultats que la méthode de référence actuelle. Néanmoins, Antilope paraît apporter une précision supérieure dans la représentativité spatiale des épidémies et reste, de ce fait, un facteur de progrès, structurant pour la démarche de modélisation.

Key words : modélisation, agro-météorologie, pluie, radar, maladies de la vigne, mildiou

Introduction

Les vignes couvrent seulement 3% de la SAU (Surface Agricole Utile), mais l’activité viticole utilise pourtant à elle seule près d’un tiers des pesticides en France ! En plus du risque de pollution, ces apports coûtent très cher aux viticulteurs et ne sont pas forcément utiles : en 1995 par exemple, certains viticulteurs de la région bordelaise ont effectué de 7 à 8 traitements contre le mildiou. Or, dans les mêmes zones, les témoins non traités n’avaient pas été attaqués (Raynal 1995). Dans le contexte de viticulture durable actuel, il apparaît de plus en plus important pour le viticulteur de savoir mieux maîtriser ses intrants phytosanitaires. L’objectif ainsi affiché par les pouvoirs publics est d’une réduction de 50% de cette consommation !

193 194

La solution OGM n’étant pas pour l’instant d’actualité en Europe, la meilleure maîtrise des interventions phytosanitaires reste une voie dans laquelle des progrès considérables doivent être accomplis. L’absence de mesure climatique à l’échelle de l’exploitation, la dynamique des variations climatiques, l’interaction de ces itinéraires climatiques variés sur le développement des parasites, constituent à n’en pas douter une des causes de ces méconnaissances et limitent la portée du travail de modélisation. L’IFV participe à la validation des modèles Potentiels Systèmes depuis leurs premières versions en 1992. La finalisation de ces outils se focalise actuellement sur l’utilisation de ces modèles à l’échelle de la petite région viticole, soit à une maille se rapprochant de la taille de l’exploitation viticole. L’ambition affichée est ainsi de faire des modèles un véritable outil d’aide au pilotage de la stratégie de traitement intégrant notamment les particularismes de la météorologie locale. L’exploitation des modèles se heurte depuis maintenant quelques années aux problèmes de la représentativité des sites météorologiques et de la fiabilité des mesures. L’amélioration des modèles est ainsi tributaire d’une amélioration des données pluviométriques. C’est pourquoi, l’IFV s’est s’intéressé dès les années 2000 aux possibilités offertes par la mesure RADAR et par ses applications en météorologie. En janvier 2007, Météo France nous a proposé de tester le produit ANTILOPE provenant de la fusion des données issues du réseau RADAR ARAMIS et de celles des stations météorologiques au sol. Antilope fournit ainsi une donnée pluviométrique à la maille de 1km. Le travail exposé porte sur l’évaluation des données Antilope et leur utilisation dans le modèle Potentiel Système mildiou version 2007. La première partie est consacrée à l’évaluation météorologique du produit Antilope tandis que la deuxième partie s’intéresse à l’impact des données Antilope dans l'utilisation du modèle Potentiel Système Mildiou 2007.

Matériels et méthodes

Nous avons comparé les données Antilope avec les données de 40 stations météorologiques. La période d’étude est de 4 mois, du 1er avril au 31 juillet 2007. Les données Stations sont issues du réseau DEMETER aquitain. Ce réseau comprend actuellement plus de 80 postes en Gironde. Pour la campagne 2007, l’IFV dispose des données provenant d’un réseau comportant 40 stations météorologiques réparties comme illustré dans la Figure 1. Le réseau DEMETER utilise des stations de type CIMEL à auget basculeur 0,5mm et à 4 capteurs (température, précipitations, humidité et humectation). DEMETER vérifie leur conformité puis somme ces précipitations horaires pour obtenir un cumul journalier de 6h jour j à 6h jour j+1. L’IFV dispose donc d’une donnée journalière de pluie pour chacune des 40 stations météorologiques. La donnée Antilope est donnée pluviométrique issue de la combinaison entre des données RADAR et des données des pluviomètres au sol. Antilope est calculée au niveau français grâce aux différents RADAR et pluviomètres répartis sur l’ensemble du territoire. En Gironde, Antilope provient principalement de la fusion des données d’une vingtaine de pluviographes pour les précipitations de grande échelle et des données du RADAR de Bordeaux-Mérignac pour les précipitations de petite échelle. Les stations utilisées par Météo France pendant la campagne 2007 sont différentes de celles utilisées par le réseau DEMETER et sont réparties en Gironde comme illustré dans la Figure 1. Ces appareils sont principalement du type CIMEL à auget basculeur 0,2mm ou 0,5mm.

195

Fig. 1 : Localisation des stations météorologiques DEMETER (gauche) et météo France (centre), et des parcelles témoins non traités (droite) exploitées en 2007

Le RADAR de Bordeaux-Mérignac est un RADAR Mélodi en bande S, mis en service en 1975. Ce RADAR explore le ciel selon 3 sites différents selon des angles de 0,8 - 1,2 et 1,8 degrés. Il fonctionne tous les jours 24h/24h et 7j/7j et accomplit un tour complet en 1 min environ. On considère que ce radar donne des informations utiles dans un rayon de 80 à 100km. Météo France étudie la concordance des données Antilope et transforme ces données horaires en données journalières de 6h jour j à 6h jour j+1. L’IFV dispose d’une donnée journalière tous les kilomètres sur une zone de 6000km2. Contrairement aux données Stations qui représentent la pluviométrie en un point précis, les données Antilope représentent la pluviométrie moyenne sur un pixel d’environ 1km2. Les données Antilope issues de pixel à proximité des stations météo sont interpolées aux coordonnées précises du poste météorologique : La donnée Antilope utilisée est donc interpolée par la méthode de l’inverse distance à partir de 4 données Antilope les plus proches du poste météorologique étudié. Un exemple de sélection est représenté à la Figure 3. Un cercle virtuel, d’environ 1km de diamètre, autour des postes météorologiques permet de sélectionner les 4 centres des pixels Antilope les plus proches du poste météo étudié. La donnée Antilope interpolée représente la hauteur de pluie moyenne sur une surface de 1km2 tout autour de la station.

Fig. 3. Exemple de sélection des 4 données Antilope à proximité du poste météorologique étudié

Méthodes statistiques : Les populations de données sont comparées et caractérisées par les critères statistiques suivants : 196

Le biais : moyenne des différences entre la donnée ANTILOPE et la donnée STATION L’écart moyen arithmétique (EMA) : moyenne des différences, en valeur absolue, entre la donnée ANTILOPE et la donnée STATION. L’écart-type : estime la précision de ei observée entre les données ANTILOPE et STATION corrigée du biais. La racine carrée de l’erreur moyenne quadratique (Root Mean Squared Error RMSE) : cet indicateur estime la précision du système sans correction du biais. Le RMSE est le plus souvent utilisé pour estimer et qualifier la justesse du produit Antilope.

Les données épidémiologiques Le modèle Le modèle utilisé dans le cadre de cette étude est le modèle Potentiel Système Mildiou version 2007, élaboré par S. STRIZYK (Société SESMA). Le modèle fonctionne à partir des données de températures horaires et de pluviométries journalières, issues du réseau de stations météo DEMETER ou des données Antilope. Une des hypothèses de base de ce modèle repose sur l'échange permanent entre le parasite et son environnement : la maturation des oospores est ainsi variable suivant les conditions climatiques hivernales. De plus, une grande importance est accordée aux contaminations primaires tout au long de la période végétative de la vigne; elles jouent un rôle prépondérant dans la gravité des épidémies et dans le modèle Potentiel Système mildiou. Une autre des hypothèses fondatrices est que le parasite est adapté aux conditions climatiques locales : La Gironde est ainsi partagée en 12 petites régions climatiques qui possèdent chacune son référentiel historique établi sur une dizaine d'années. Chaque simulation établie à partir d'une station météo ou d'un pixel Antilope est associée à l’un de ces 12 référentiels en fonction de sa distance au référentiel.

Le réseau des parcelles témoins Ce réseau de parcelles mises à disposition par des viticulteurs volontaires, permet d’effectuer un suivi épidémiologique de l’évolution naturelle des maladies cryptogamiques sur des témoins non traités. En 2007, Le réseau est constitué d’une quarantaine de sites suivis par l’IFV et ses partenaires; sur chacun de ces sites, 6 à 8 rangs ne reçoivent aucune protection fongicide le plus longtemps possible au cours de la campagne. Un monitoring hebdomadaire permet de relever les dégâts occasionnés par les maladies cryptogamiques. Les fréquences et intensités d'attaque relevées sur feuilles et grappes servent à apprécier en temps réel la fiabilité du modèle et des informations qu’il délivre. La période retenue dans notre étude débute au 9 mai, date des premiers dégâts conséquents et prend fin le 4 juillet, époque des premiers traitements réalisés sur les témoins. Les variables de sorties du modèle sont de différentes sortes : maturation des œufs en début de campagne, contaminations de début de campagne, et contamination épidémique classique : la fréquence théorique d'attaque (FTA) simulée par le modèle représente la quantité totale de contaminations primaires et secondaires exprimées après période d’incubation. Cette variable indique le pourcentage moyen d’organes, feuilles ou grappes, atteints par la maladie. Elle est comparée à la fréquence d'attaque observée (FAO) relevée lors du monitoring.

197

Résultats et discussion

Les données pluviométriques Le printemps et l’été 2007 ont été très pluvieux : sur les 4 mois de l’étude, du 1er avril au 31 juillet 2007, il a plu en moyenne 250 mm sur le vignoble bordelais. Le nombre d’observations étudiées est égal à 40 stations sur 122 jours soit 4880 couples de données Station/Antilope. Dans la plupart des résultats présentés, nous avons supprimé l’ensemble des couples (0/0) correspondant à l'absence de pluie détectée à la fois par la station et par Antilope. Sur la période du 1er avril au 31 juillet 2007, 56% des couples sont différents de (0/0).

Corrélation entre les données Antilope et les données Stations La Figure 4 présente le nuage de points des données Antilope en fonction des données Stations. Chaque point représente le couple d’un jour j. Elle montre l'existence d'une bonne corrélation entre les modalités : (y =0,8x ; corrélation r (Pearson) = 0,9). Ces données sont réparties selon 5 classes de pluies [0-1], ]1-5], ]5-10], ]10-20] et ]20-∞[ mm en fonction de la valeur de la donnée Station. La Figure 4 présente les différents nuages de points de ces données.

Fig. 4. Distribution globale et par classe de pluie des données Antilope en fonction des données Stations.

Les équations des courbes de tendance sont assez proches du cas parfait y = x sauf pour les deux classes extrêmes : les données Antilope sur-estiment globalement les données Stations faibles et sous-estiment globalement les données Stations fortes. Les coefficients de corrélation r sont médiocres pour les classes de pluie [0-1] et ]5-10] mm. Les données Antilope et Stations correspondantes sont peu corrélées lorsqu'elles sont comparées par classe de pluie.

Distribution des différences Globalement on observe que près de 90% des différences ei se situent entre -2,6mm et 2,8mm. La moyenne et la médiane sont proches de zéro, indiquant une distribution quasiment symétrique des différences. L’écart moyen arithmétique est de 1,6mm. 198

Près de 10% des couples étudiés donnent des différences ei extrêmes. Les différences minimale et maximale atteintes sont respectivement -28 et 26mm, soit près de 30mm d’écart entre les valeurs délivrées par ces deux méthodes de mesure ! Le RMSE, de 3mm permet de considérer qu’une donnée Antilope est égale à la donnée Station à plus ou moins 3mm près. Cet écart, faible en valeur absolue est en fait relativement important pour de faibles pluies. Cette observation conduit à regarder plus attentivement ces différences au niveau des cinq classes de pluies. La Figure 5 présente les box plots des différences ei entre données Antilope et Stations ainsi que le tableau des caractéristiques statistiques pour chacune des 5 classes de pluie.

Classe de 0 ˆ 1 mm Classe de 1 ˆ 5 mm Classe de 5 ˆ 10 mm Classe de 10 ˆ 20 mm Classe sup. ˆ 20 mm Nombre et % d'observations 1336 49% 820 30% 299 11% 197 7% 83 3% Nombre et % de valeurs atypiques 131 10% 34 4% 22 7% 7 4% 6 7% Maximum 18 22 26 9 18 Limite sup rieure ou Lsup. 1,9 3,5 3,6 5,9 11,2 3 me Quartile ou Q3 0,8 0,9 0,2 -0,2 -0,2 Moyenne ou Biais 0,6 0,3 -0,7 -2,3 -5,6

M diane ou Q2 0,3 0,0 -1,2 -2,6 -6,6 1er Quartile ou Q1 0,0 -0,8 -2,5 -4,3 -9,3 Limite inf rieure ou Linf. -1,0 -3,4 -6,2 -10,3 -20,7 Minimum -1 -5 -8 -13 -28 Ecart absolu moyen ou EMA 0,8 1,3 2,2 2,7 6,3 Ecart-type (n) 1,5 2,2 3,6 3,5 8,5

RMSE 1,6 2,2 3,7 4,2 10,1 Fig. 5. Box plots et caractéristiques statistiques des 5 classes de pluie.

Les parties centrales des box plots, proportionnelles à l’augmentation des précipitations, traduisent la diminution des observations et l'augmentation en valeur absolue des différences. Les distributions sont moins symétriques et non centrées en zéro. Les différences moyennes diminuent progressivement passant de 0,6mm pour la classe [0-1] mm à près de -6mm pour la classe ]20-∞[ mm. Les classes ]5-10], ]10-20] et ]20-∞[ mm, montrent des différences moyennes négatives, confirmant ainsi la sous évaluation des données Antilope par rapport aux données Station. Les RMSE augmentent avec les hauteurs de pluie. Le RMSE de la classe [0-1] mm est de 1,6mm, soit 160% de la limite supérieure de cette classe. De même, le RMSE de la classe ]10- 20] mm est de 4,2mm, soit plus de 20% de la limite supérieure de cette classe. Le pourcentage de valeurs atypiques est de 10% pour la classe [0-1] mm et de 4 à 7% pour les autres classes. 199

L'approche par classe de pluie révèle les données Antilope ont tendance à sur-estimer les données de précipitations de faible importance et à les sous-estimer sur les fortes précipitations.

Les données simulées par le modèle

La Figure 6 présente le nuage de points des données modélisées FTA Antilope (triangle rouge) et FTA Station (rectangle bleu) en fonction des données observées.

100 FTA (%) y = 0.4x FTA 'STATIONS' r = 0.6 80 FTA 'ANTILOPE' y = 0.2x r = 0.5

40 eistat eirad Biais -7 -10

20 RMSE 22 25 Mini -83 -91 Maxi 33 53 0 0 20406080100 FAO (%) Fig. 6. Corrélation entre les données modélisées par Antilope (triangle) ou Démeter (rectangle) et les données observées sur le terrain

Impact du référentiel météo Foyer mildiou orage du 24/5

FTA stations Démeter FAO témoins non traités FTA Antilope Référentiels météo

Fig. 7. FTA mildiou simulée le 4 juillet à partir des référentiels météo sur les données Démeter et Antilope, comparées à la FAO sur le réseau de témoins non traités au stade fermeture de la grappe.

Les données modélisées FTA Stations et FTA Antilope donnent des différences ei globales à peu près équivalentes. Les écarts ei négatifs montrent la sous-estimation générale des données terrains, tant par les données issues des stations météo que celles Antilope du Radar pluviométrique. Les données mildiou modélisées à partir des données Antilope apparaissent au contraire encore plus éloignées de la réalité du terrain que celles obtenues à partir des stations. Nous devons donc constater que le produit Antilope n’apporte pas de gain 200

de précision au modèle en comparaison de la méthode utilisée actuellement. Quoi qu’il en soit, la pluviométrie n’est pas la seule variable qu’il faut savoir bien détecter et mesurer. La diversité des sols, de la topographie, du milieu, de la réserve utile en eau, etc... sont autant de critères que les modèles Potentiels Systèmes ne prennent pas en compte, et qui peuvent en partie expliquer les différences observées et la sous évaluation globale des attaques cette année. Les observations suivantes nous conduisent en effet à nuancer ce constat critique. La figure 7 compare la spatialisation des dégâts simulés et observés sur le terrain début juillet au stade de fermeture de la grappe. La figure 7 confirme la sous évaluation globale du modèle Potentiel Système par rapport aux dégâts observés sur les témoins non traités, que les données pluviométriques proviennent de postes météo ou du radar. La carte de droite, obtenue à partir des données Antilope montre clairement : - Le motif qui apparaît sur le Médoc au nord du département est à rapprocher de la délimitation du zonage des référentiels météorologiques (carte de droite) et montre ainsi l'impact très marqué du référentiel météorologique utilisé sur le Médoc. Ce motif apparaît bien plus nettement que sur la carte établie par krigeage; même si ce secteur du vignoble s'avère être en définitive le plus proche de la réalité de l'épidémie de mildiou en 2007, cette carte manifeste un constat établi depuis plusieurs campagne de l'impact trop marqué, voire maintenant trop délimité de ces référentiels. - La donnée Antilope permet de visualiser à l'Est du vignoble bordelais, en Dordogne, un secteur très délimité sur lequel le niveau d'attaque simulé est nettement supérieur aux dégâts avoisinants. Le modèle traduit bien une attaque, réelle semble-t-il en dépit de l'absence de témoin non traité sur ce secteur du vignoble, qui n'apparaît pas sur le zonage établi par Démeter en raison de l'absence de pluviomètre à cet endroit. La couverture Radar permet ainsi d'obtenir une maille de calcul beaucoup plus fine, sur laquelle le modèle paraît apte à évaluer de fortes variations d'attaques sur de faibles distances. Cette observation confirme ainsi les résultats obtenus lors d'une étude préalable réalisée sous forme d'une maquette établie en 2006 avec des données fictives issues du Radar Hydrix de la société Novimet. - Enfin, cette maille plus fine des calculs montre, à l'intérieur du cercle de la figure 7, sur le secteur de l'Entre Deux Mers, une zone de moindres dégâts qui partiellement tout au moins a pu peut être matérialisée par l'absence de mildiou constatée sur l'un des témoins non traités du réseau. Ces observations révèlent ainsi le caractère structurant de ce type de données issues d'un Radar météorologique et matérialisent tout l'intérêt de leur apport dans notre démarche de prévision des risques épidémiques.

Conclusion et perspectives

La première étude, essentiellement consacrée à l’évaluation strictement météorologique des données Antilope en comparaison des 40 données Stations journalières jusque-là utilisées par l’IFV, révèle des différences significatives entre ces deux types de données. Antilope a ainsi tendance à sur-estimer les faibles précipitations et, à l’inverse, à sous-estimer les fortes précipitations ce qui conduit à une uniformisation globale de la pluviométrie alors que nous souhaitons au contraire affiner et préciser ces données. Il est, pour autant, certain que cette comparaison est discutable quand on sait que la donnée Antilope représente la hauteur moyenne des précipitations sur une surface de 1 km2 alors que la donnée Station est une mesure ponctuelle de la pluviométrie. Néanmoins, le produit Antilope pourrait très 201

certainement être encore amélioré par l’emploi conjoint d’un réseau pluviométrique au sol plus important et d’un RADAR plus précis notamment dans sa capacité à atténuer les échos fixes. La deuxième étude, consacrée à l’utilisation du produit Antilope dans le modèle Potentiel Système mildiou 2007 a profité d’une année fortement marquée par le mildiou. L’utilisation des données Antilope n’a guère suffi à mieux évaluer les fréquences d’attaque du champignon par le modèle. Néanmoins, la représentativité spatiale des épidémies modélisées à partir des données Antilope paraît intéressante car sans doute plus précise que celle fournie par interpolation des données Stations. Il sera souhaitable, à l'avenir de disposer d’un plus grand jeu de parcelles témoins pour vérifier cette interprétation des cartes. Ce travail nous permet d'évaluer la pertinence du produit Antilope proposé par Météo France et le gain de précision possible apporté par ce type de données dans la modélisation du mildiou. L’IFV envisage de poursuivre cette démarche de viticulture de précision afin de répondre à la demande croissante des viticulteurs qui souhaitent disposer de données précises adaptées et adaptables à leur exploitation.

Remerciements

Nous tenons à remercier les équipes IFV ainsi que toutes les structures et partenaires associés, viticulteurs ou techniciens qui participent à ce projet et rendent cette étude possible.

Bibliographie

Bois B. 2006. Spatialisation des précipitations, Investigations sur l’utilisation des lames d’eau HYDRAM issues de l’imagerie radar comme outil de spatialisation des champs de pluie en Gironde. Etude préliminaire. Météo France, ENITA Bordeaux, INRA. 30p. Cigogna A., Dietrich S., Gani M., Giovanardi R. et Sandra M. 2005. Use of meteorological radar to estimate leaf wetness as data imput for application of territorial epidemiological model (downy mildew – Plasmopora viticola). – Physics and Chemistry of the Earth 30: 201-207. Collectif Météo France 2005. La mesure RADAR : principe et description. Documentation technique. – Météo France. 10p. Strizyk S. 2007. Potentiel Système généralisé chez la vigne, Décider dans l’incertain. – SESMA. 44p. 202

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 203-206

Powdery mildew of grape on leaves – is it a problem?

Zahavi Tirtza1 and Reuveni Moshe2 1 Shaham, Ministry of Agriculture, Israel, 2Golan research institute, Univ. of Haifa, P.O. Box 97 Katzrin, Israel

Abstract: Powdery mildew of grapes, caused by Uncinula necator, is widespread in all grape growing areas of the world and attacks all green parts of the vine. The greatest damage is caused when the fungus develops early in the growing season and covers the inflorescence or small berries just after fruit set. Because of the possible high lose, growers focus on taking good control measures against the fungus starting soon after bud burst. New research on the etiology of the disease and its interaction with the vine at different phonological stages showed that the berries are very susceptible until they reach about pea size; at that point they develop ontogenic resistance. This finding, combined with the goal of minimizing pesticide usage, led to a reduction in the number of sprays, concentrating on the early vine phonological stages – from 5-20 cm. shoot length until the berries reach pea size. In the last few years, in some of the vineyards in Israel high levels of powdery mildew developed on grape leaves. Most of these vineyards were harvested late in the season, in October-November; this means that the leaves were covered with mycelium long before the harvest. The objectives of the present work are to find out whether fruit quality is reduced by the disease that develop on the leaves, whether leaf PM affects the organoleptic properties of the juice and to test whether high disease level on the leaves at the end of the season affects the disease incidence in the following year. The present report will focus on the effect of different spraying tactics on disease severity and the effect of the diseased leaves on juice quality and harvest parameters. Two types of experiments were set up, each in two vineyards; in the "early" experiment we tested the effect of the mode of spraying (covering the whole canopy Vs concentrating in the cluster zone) and that of an additional spray after "pea size stage" on disease development on the leaves. In the "late" experiment vines were sprayed with a commercial sprayer till the beginning of verizon when the experiment was set up, and were then sprayed with either Sulphur or Timorex – a tea tree oil based product – to suppress disease development on the leaves. Disease assessment on the leaves close to harvest time showed the importance of spraying the canopy at the early stages of the season. 2006 results showed that although the last spray was applied in mid June, three months before the disease assessment in mid September, 71.3% of the leaves in the "cluster zone treatment" had mycelium of U. necator (average severity – 12.6% of the leaf area) as compared to six percent of the leaves (0.23% severity) in vines that were completely covered with spray. In the "late experiment" – both Sulphur and Timorex gold were effective in suppressing the disease (seven and six percent of the leaf area covered with PM as compared to 30% in the untreated control). No effect was detected on sugar accumulation, pH or TA in either experiment. We got the same results in 2007 and they are presented in this article.

Key words: Erysiphe necator, wine aroma, flavor, sugar accumulation

Introduction

Powdery mildew of grapes, caused by Erysiphe (Uncinula) necator, is widespread in all grape growing areas of the world and attacks all green parts of the vine. The greatest damage is caused when the fungus develops early in the growing season and covers the inflorescence or small berries just after fruit set. This can cause a complete yield lose (2). Later fungal development can cause scaring of the berry peel and then cracking of the berry at verizon. Another problem caused by powdery mildew is an undesirable taste that it gives to wines (3).

203 204

Experienced winemakers were able to detect organolepticly wines with as little as four percent of infected berries. Because of the possible high lose, growers focus on taking good control measures against the fungus starting soon after bud burst. New research on the etiology of the disease and its interaction with the vine at different phonological stages showed that the berries are very susceptible until they reach about pea size; at that point they develop ontogenic resistance (1). This finding, combined with the goal of minimizing pesticide usage, led to a reduction in the number of sprays, concentrating on the early vine phonological stages – from 5-20 cm. shoot length until the berries reach pea size. When sprays are applied properly, using good chemicals and a well adjusted sprayer, effective disease control is accomplished and no symptoms appear on the grapes. In the last years severe epidemics of PM develop late in the season in many vineyards in Israel. This phenomenon is very common in Merlot and Chardonnay vineyards and less in Cabernet sauvignon or Carignan (personal impression). The primary concern with mildewed leaves is that this will lead to reduced sugar, flavour and aroma development in grapes. The second concern with "late" PM is that high disease levels at the end of the season can lead to higher disease levels in the following year.

Material and methods

Field experiments Field experiments were carried out in 2006 on Merlot and in 2007 both on Merlot and Cabernet sauvignon vines. All experiments were set in randomized block design with five replicates of six vines per treatment. Vines, or vine parts were sprayed to run-off with gun sprayers. Two types of experiments were set up: 1. "Early" experiment. Vines were sprayed every 2 weeks from shoot length of 20 cm (totaling 5-6 sprays). Sprays were applied to cover either the whole canopy or only the cluster zone. Unsprayed vines served as control. 2. "Late" experiment. Vines were sprayed uniformily with a commercial sprayer until "veraizon" when we started the differential spraying; vines were sprayed either with Helisulfur (wetable sulfur, Makteshim Ltd.), Timorex gold (an essential Tea tree oil, Biomor Ltd., Israel) or Dorado (Fenbuconazol, Syngenta) each at the recommended rate, unsprayed vines served as control. Sprayes were applied three times, every 2 weeks. Disease incidence and severity were assesed on clusters just before veraizon and on leaves from veraison to harvest. For disease severity the percentage of leaf area coverd by PM mycilium was evaluated. 30 full grown leaves at eye level, from both sides of the vine row were scored per plot. Statistical analysis was done using SAS software, GLM procedure with SNK test after Arcsin transformation of the percent of infected leaves and percent coverage.

Fruit analysis Must parameters were analysed for the "early" experiments on healthy berries from the two differentially sprayed treatments. Butches of 100 berries per plot were used to measure TSS and pH level of the fruit. Berries were picked from 25 random clusters and hand pressed.

Results and discussion

Effect of spraying on disease development 1. "Early" experiments. Results in both years and both varieties demonstrated the importance of good spray coverage from the beginning of the season (Fig.1). In 2007, disease level on the 205

clusters, that were sprayed to run off, was low (2-4 and 0 percent as compared to 11 and 35 percent in control plots in Merlot and Cabernet sauvignon, respectively). PM colonies were visible on leaves soon after "veraison" in the untreated control and in the plots where only the cluster zone was sprayed in the early stages but not in vines where the whole canopy was covered with spray. Those results suggest that although PM colonies became visible and well established only after veraison, the infection takes place at early growth stages. Thus in order to keep the canopy clean of PM growers must ensure a good spray coverage from the beginning of the season.

20% 60% Disease severity severity Disease Merlot 50%

15% (leaves) 40% 10% 30%

(clusters) 20% 5%

Disease severity 10% 0% 0% 2/7/07 1/8/07 31/8/07 30/9/07 Date

40% 75% Disease severity severity Disease Cabernet 60% 30% (leaves) 45% 20% 30% (clusters) 10% 15% Disease severity 0% 0% 14/7/07 4/8/07 25/8/07 15/9/07 6/10/07 Date

Whole canopy Control Cluster zone

Fig. 1. Development of powdery mildew on clusters and leaves of cv. Merlot (top) and Cabernet sauvignon (bottom) treated plants from the beginning of the season ("early" experiment). Solid lines represent disease on leaves while empty shapes represent disease severity on the clusters before veraizon. Bars represent STD.

2. "Late" experiments. Disease level on leaves in control plots was very different for the two varieties. PM was neglible in Cabernet sauvignon (data not shown) while in the Merlot plot 50 percent of the leaf area was covered with PM mycilium (Fig. 2). Timorex Gold and sulfur effectively inhibited disease development on leaves, compared to Dorado and control untreated vines. The fungicides tested in this experiment are very different in their mode of action and their effect on different developmental stages of PM. In earlier studies we have shown that sulfur is much better then DMI or Strobilorin fungicides in inhibbiting the development of 206

existing PM colonies on grape berries. Our present results suggest that the situation on leaves is the same. Spraying with Fenboconazole after "veraison" had no effect on the development of the disease that probably infected the leaves at earlier stages. Sulfur and the vegetal oil extract (Timorex gold) on the other hand, which act on multiple stages of the fungus significantly supressed fungal development.

80% 70% 60% 50% 40% 30% 20% 10% Disease severity (leaves) Disease 0% 18/08/2007 08/09/2007 29/09/2007 Date

Timorex gold Control Heliosulfur Fenboconazole

Fig. 2. The effect of protective treatment of Heliosulfur and Timorex Gold applied at the "late" stage (after vereizon) on powdery mildew development on leaves of cv. Merlot. Bars represent STD.

Fruit analysis Similar levels of sugar accumulation (Brix) and pH were observed on vines of both treatments when spray was on cluster zone or on whole plants. These results were found for both Merlot and Cabernet sauvignon (data no shown). It is concluded that powdery mildew on leaves did not affect sugar accumulation in berries.

Acknowledgements

We thank Y. Kotzer for assistance in field trials

References

Ficke, A., Gadoury D.M., Seem, R.C. and Dry I.B. (2003). Effects of ontogenic resistance upon establishment and growth of Uncinula necator on grape berries. – Phytopath. 93: 556-563. Pearson R.C., and Gadoury D.M. (1992). Grape powdery mildew.– In: Plant disease of international importance. Vol. 3, Diseases of fruit crop: 129-146. Stummer B.E., Francis I.l., Zanker T., Lattey K.A. and Scott E.S. (2003). The effect of powdery mildew on the sensory properties and composition of Chardonnay juice and wine when grape sugar ripeness is standardized. – Aust. J. of grape and wine research: 9: 28-39. Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 207-214

Competitive colonisation of Penicillium expansum and Botrytis cinerea on grapes

Ruth Walter, Marco Harms, Heinrich Buchenauer

Dienstleistungszentrum Ländlicher Raum – Rheinpfalz, Department of Phytomedicine, 67435 Neustadt an der Weinstraße, Germany; University of Hohenheim, Institute of Phytomedicine, 70593 Stuttgart, Germany

Abstract: Laboratory investigations on the colonisation of wounded berries (var. Riesling) demonstrated, that Penicillium expansum and Botrytis cinerea reached strongest disease severities and similar colonisation rates between 20°C and 25°C. Inoculation of a conidial mixture of B. cinerea and P. expansum on wounded berry skin first led to a colonisation of different parts of the berry until its surface was completely covered. With longer incubation time P. expansum was able to overgrow B. cinerea at each temperature tested (15°C, 20°C, 25°C, 30°C). P. expansum was not able to infect berries with intact berry skin whereas B. cinerea penetrated the skin layers of Pinot blanc without visible wounds. Secondary infections caused by P. expansum could be observed after the infection of B. cinerea through the berry skin, that clarified the role of B. cinerea as a wound causing pathogen. The treatment of wounded berries with botryticides (Boscalid, Cyprodinil + Fludioxonil, Fenhexamid) in the laboratory in most cases led to a reduction of B. cinerea. P. expansum was hardly affected by the compounds tested. On berries inoculated with a conidial mixture of both fungi, P. expansum profited from the fungicide application and inhibition of B. cinerea. In the field the results of the laboratory studies could only partly be confirmed. In the vineyard the fungicide treatments reduced B. cinerea while P. expansum was not able to take benefit from the inhibition of B. cinerea. The analysis of the data from different field trials showed a positive correlation between the disease severities of blue and grey mold. The correlation implicated, that an inhibition of B. cinerea was not resulting in a higher infection rate of in general and that both pathogens prefer similar conditions for their development. In conclusion best reduction of both pathogens was detected with cultivation methods that led to a reduction of wounds, the most important infection pathway of both pathogens. For example horizontal bunch dividing decreased the disease severity of both pathogens significantly. The effect of this kind of yield regulation was associated by the loosening of the bunch structure and thereby a reduced number of wounds.

Introduction

Penicillium expansum and Botrytis cinerea are important bunch rot pathogens in German vine growing regions causing blue and grey mold on grapes (Mohr 2005). Since the end of the 90th increasing problems with secondary bunch rot invaders, especially Penicillium expansum, can be observed (Harms et al. 2004). Even low levels of Penicillium-infected bunches may lead to persistent off-flavours in the wine by volatile compounds (e.g Geosmin) (La Guerche 2004). Therefore infected bunches have to be discarded carefully before harvest. While B. cinerea colonizes grape berries by direct skin penetration and wounds (Coertze & Holz 1999, McClellan et al. 1973, Nelson 1956), P. expansum depends predominantly on wounds for the infection process (Latorre & Rioja 2002, Seyb 2004). The reasons that lead to an improved development of P. expansum during the competitive colonisation by B. cinerea on grape berries are only poorly understood. Because the first problems occurred in the end of the 90th, there are lot of presumptions on the impact

207 208

of higher temperatures in autumn and the intensive use of new specific fungicide compounds. But evidences are still lacking. There is also no knowledge about the relevance of wounds from Botrytis infections for the infection pathway of Penicillium. The aim of this study was to evaluate the factors that might have an important influence on the competitive colonisation of P. expansum and B. cinerea on grape berries. Therefore the impact of temperature, botryticides and the role of B. cinerea as a possible wound source for Pencillium infections was investigated. As most problems with Penicillium infections occur in grape varieties having compact bunch structure the impact of different yield regulation methods was also tested (Harms et al. 2004).

Material and methods

Fungal strains For the different experiments six strains of P. expansum and three strains of B. cinerea were used. Three reference strains of P. expansum (DSM 62841, DSM 1994, DSM 1282) were obtained from DSMZ (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH). Further three strains of P. expansum and three strains of B. cinerea have been isolated from grapes. The inoculum was adjusted to a spore concentration of 200.000 conidia/ml. The conidial suspensions consisted of a mixture of the six Penicillium strains respectively three Botrytis strains and contained same portions of the different isolates. All fungal strains were cultivated on malt extract agar (2 %) at 21°C.

Impact of temperature (laboratory) Detached berries of the variety Riesling (19° Brix; n=20) were washed, surface sterilised according to Coertze & Holz (1999) (30 sec. 70 % Ethanol, 2 min. 0,35 % Na-hypochlorit, 30 sec. 70 % Ethanol) and wounded with a fine needle. Wounds were inoculated with 10 µl of conidial suspensions of P. expansum, B. cinerea and a mixture of both pathogens containing same portions of both fungi (200.000 conidia/ml each). The berries were incubated in humid boxes at 15°C, 20°C, 25°C and 30°C for 20 days. Each three to four days the disease development of both fungi on the berries was estimated.

Time of inoculation and effect of wounds (laboratory) To evaluate the role of wounds for the fungal development and whether it was possible for one fungus to colonize a wound that was already infected by the other fungus, further experiments on detached berries with different inoculation orders on wounded and unwounded berries were conducted (table 1). Different inoculation orders ensured a developmental advance to one of the two pathogens for three days. Also a simultaneous inoculation with a conidial mixture of both pathogens on wounded and intact berry skin was assessed. The experiments were conducted with berries of the variety Pinot blanc (16 Brix). Surface sterilization, inoculum preparation and inoculation was done as described. Inoculated berries (n=20) were incubated in humid boxes at 21°C for 20 days. Each three to four days the disease development of both fungi on the berries was estimated.

Impact of botryticides (laboratory) The impact of different botryticides on the development of P. expansum and B. cinerea on detached berries (Riesling, 21° Brix, n = 20) separately or in mixture was investigated in the laboratory. Washed and surface sterilised berries were wounded with a fine needle. Except for the water treated check the berries were dipped in suspensions of the commercial fungicides Switch, Teldor and Cantus (table 2). The fungicides were used with the recommended field concentrations. After the fungicides had dried on the berry skin, wounds were inoculated with 209

10 µl of conidial suspensions of P. expansum, B. cinerea or a mixture of both pathogens. The berries were incubated in humid boxes at 21°C for 29 days.

Impact of botryticides (field experiments) To confirm the results obtained in the laboratory, field trials were set up on the the impact of botryticides on the competitive colonisation of the two pathogens in the year 2006. 50 bunches (12-13 bunches per vine, four vines per variant) of the variety Riesling were wounded with a needle (three berries per bunch) and treated with the botryticides (table 2) at veraison (BBCH-GS 81). Inoculation followed with mixed conidial suspension (200.000 conidia/ml) of P. expansum and B. cinerea. The disease development of Penicillium and Botrytis was evaluated in regular intervalls.

Table 1: Inoculation order on wounded or Table 2: Tested fungicides in laboratory and intact berry skin (Pinot blanc) field trials

Var. Berries Inoculation order Recommended Product Active ingredient wounded Start + 3 days field conc. [%] 1 Yes B. cinerea P. expansum Cyprodinil (375 2 No B. cinerea P. expansum Switch g/kg) + Fludioxonil 0,06 3 Yes P. expansum B. cinerea (250 g/kg) Fenhexamid (510 Teldor 0,1 4 No P. expansum B. cinerea g/kg) 5 Yes Mixture Cantus Boscalid (500 g/kg) 0,075 6 No Mixture

Impact of yield regulation on disease development (field experiments) The impact of different kinds of yield regulation (bunch thinning and bunch dividing) on the disease development of Botrytis and Penicillium was investigated in six field trials in the years 2004 to 2006. Conventional bunch thinning to one or two bunches per shoot and horizontal bunch dividing (Hafner 2001) were conducted in vineyards with the variety Riesling at near bunch closing (BBCH-GS 77-79). Trials were set up in randomised block design with four replications per variant. Each plot had a size of about 30 m² (12 to 15 grape plants). Disease development of Botrytis and Penicillium was evaluated on 4 x 100 bunches per variant.

Results and discussion

Impact of temperature (laboratory) P. expansum and B. cinerea obtained the strongest disease severities and similar colonisation rates between 20°C and 25°C (table 3). Both fungi were also able to colonise the berries at 15°C and 30°C but the colonisation rates were slower. With the mixed inoculum especially the development of Botrytis was interfered in contrast to the individual inoculation. For example individual inoculations resulted in disease severities of 81,8 % (Penicillium) and 100 % (Botrytis) six days after inoculation at 20°C. With mixed inoculum Penicillium only reached a disease severity of 56 % and Botrytis 32 %. In the variants with mixed inoculum the two pathogens first colonised separate parts of the berry until the surface of the berry was completely covered. With longer incubation time P. expansum was able to overgrow B. cinerea at any temperature tested (table 3). From the 210

present data it can not be concluded that higher temperatures in autumn are the only responsible factor for the improved development of Penicillium in the last years.

Table 3: Impact of temperature on the disease severity (%) of P. expansum (Pen.) and B. cinerea (Bot.) using separate and mixed inoculum (Riesling berries, n = 20)

Separate inoculum P. expansum B. cinerea Temperature 15°C 20°C 25°C 30°C 15°C 20°C 25°C 30°C Days after inoculation 3 10,8 a 14 a 27,5 b 28,6 b 20,3 a 39 b 63,5 c 24,0 ab 6 53,6 a 81,8 b 94,7 b 88,3 b 99,5 a 100 a 100 a 99 a 9 100 a 100 a 100 a 94,4 a 100 a 100 a 100 a 100 a 20 100 a 100 a 100 a 100 a 100 a 100 a 100 a 100 a Mixed inoculum Temperature 15°C 20°C 25°C 30°C Days after Pen. Bot. Pen. Bot. Pen. Bot. Pen. Bot. inoculation 3 1,4 a 1,7 a 7,9 b 3,5 a 16,8 c 10,6 b 9,6 bc 11,2b 6 49,8 a 15,3 a 56 a 32 b 65 a 33 b 52 a 24,0ab 9 81,5 a 18,5 a 67,5 a 32,5 a 67,5 a 32,5 a 75,5 a 16,5a 20 85 a 15 a 71 a 29 a 69,5 a 30,5 a 77,5 a 16,5a Variants with same letter do not differ significantly within one pathogen and one date of evaluation (Dunns-Test, α= 0,05).

Time of inoculation and effect of wounds (laboratory) In the experiments with different inoculation orders both pathogens showed stronger disease severities and higher colonisation rates on wounded berries, compared to unwounded berries (table 4). P. expansum was not able to infect berries with intact berry skin, whereas B. cinerea penetrated the skin layers of Pinot blanc berries in many cases without visible wounds (variant 4, 6; fig. 3). Penicillium only established infections in variants with wounded berries or in variants where Botrytis penetrated the intact berry skin previously (table 2). P. expansum was able to colonize berries that had already infected by B. cinerea as a secondary invader (fig. 1,3). Even if P. expansum inoculation was delayed, the pathogen was able to overgrow B. cinerea afterwards. Botrytis was not able to develop disease severities as strong as Penicillium, acting as the the second pathogen.

211

Table 4: Disease severities (%) of P. expansum (Pen.) and B. cinerea (Bot.) on berries of Pinot blanc with different inoculation conditions (n= 20).

Variant 1 2 3 4 5 6 wounded yes no yes no yes no inoculum single fungus single fungus single fungus single fungus mixture mixture days after Bot. Pen.* Bot. Pen.* Pen. Bot.* Pen. Bot.* Pen. Bot. Pen. Bot. inoculation 32,0 0,0 0,0 0,0 1,0 0,0 0,0 0,0 6,0 15,0 0,0 0,1 5 bc a a a b a a a c b a a 93,0 0,0 0,0 0,0 1,0 2,0 0,0 0,0 28,0 65,0 0,0 9,0 8 d a a a b b a a c c a ab 98,0 3,0 0,0 0,0 7,0 0,2 0,0 0,2 39,0 57,0 0,0 24,0 13 c ab a a b a a a c b a a 80,0 20,0 0,0 0,0 14,0 3,0 0,0 0,2 41,0 55,0 0,0 26,0 16 d bc a a b ab a a cd cd a b 49,0 51,0 0,0 0,0 22,0 0,1 0,2 0,2 89,0 11,0 11,0 48,0 20 d c a a bc a a a d c ab bc * inoculation 3 days later Variants with same letter do not differ significantly regarding one pathogen and one date of evaluation (Dunns- Test, α= 0,05)

140 50 120 PenicilliumP. expansum BBotrytis. cinerea P. expansum B. cinerea 40 Penicillium Botrytis 100 80 30

60 B P 20 P B 40 10 20 disease severity [%]disease severity

[%] severity disease 0 0

0 2 4 6 8 10 12 14 16 18 20 22 0246810121416182022 days after inoculation days after inoculation

Fig. 1: Disease development on berries of Pinot Fig. 2: Disease development on berries of blanc in variant 1: wound inoculation of B. Pinot blanc in variant 3: wound inoculation of cinerea, P. expansum followed three days later P. expansum, B. cinerea followed three days on the same site. Arrows indicate the time of later on the same site. Arrows indicate the inoculation of Penicillium (P) and Botrytis (B). time of inoculation of Penicillium (P) und Botrytis (B).

The Botrytis development following Penicillium was interfered. As a reason for this P. expansum might produce metabolites during the early developmental period, which may have an inhibitory effect on the development of B. cinerea (Samson & Frisvad 2004).The trials showed, that P. expansum benefited from the direct skin penetration by B. cinerea. Penicillium also showed a significantly stronger development on wounded berries that had been affected by Botrytis compared to berries where P. expansum has to colonise the wounds on its own.

212

70 From the present data B. cinerea 60 PenicilliumP. expansum BotrytisB. cinerea has to be regarded as a source of 50 wounds, that allows P. expansum to 40 colonise the berry. From this point of 30 P+B view Botrytis should be excluded. 20 The experiments lead to the 10 conclusion that an effective disease seveity [%] seveity disease 0 protection strategy against B. cinerea 0 2 4 6 8 10 12 14 16 18 20 22 should also result in low levels of P. days after inoculation expansum. It was interesting to see that Fig. 3: Disease development on berries of Pinot berries of the variety Riesling exhibi- blanc in variant 6: inoculation of a conidial ted a higher sensitivity to Penicillium mixture of P. expansum and B. cinerea on the colonisation than berries of the intact berry skin. Arrow indicates the time of variety Pinot blanc. inoculation of Penicillium (P) und Botrytis (B).

Impact of botryticides (laboratory investigations) The treatment of wounded berries with botryticides (Switch, Teldor, Cantus) in the laboratory led to a reduction of B. cinerea in most cases, whereas P. expansum was hardly affected by the compounds (data not shown). On berries inoculated with a conidial mixture of both pathogens P. expansum profited from inhibition of B. cinerea by the fungicide application (table 5). As both fungi had to share the space on the berries in the untreated check, on treated berries the disease severity of B. cinerea was reduced while at the same time the severity of P. expansum increased.

Table 5: Impact of different botryticides on the disease severity (%) of P. expansum (Pen.) and B. cinerea (Bot.) with mixed inoculum (Riesling berries, n= 20)

Variant Check Switch Teldor Cantus Pen. Bot. Pen. Bot. Pen. Bot. Pen. Bot. Days after inoculation 4 18,0 2,84,5 * 0,0 * 16,80,0 * 7,9 0,0 * 7 63,0 30,854,65,0 * 94,5 * 1,0 * 80,1 0,0 * 13 69,6 30,584,6 * 6,0 * 99,2 * 0,8 * 95,5 * 0,3 * 20 65,3 30,383,42,0 * 98,5 * 1,5 * 99,0 * 1,0 * 29 80,0 18,387,1 *2,8 * 100 * 0,8 * 100 * 1,8 * * Variants differ significantly from the untreated check regarding one pathogen and one date of evaluation (Dunnett-Test, α= 0,05)

Impact of botryticides (field experiments) In the field the fungicides reduced the disease severity of B. cinerea between 30 % and 50 % (fig. 4). The disease severity of P. expansum was not reduced by the fungicide treatments. The results obtained in the laboratory studies could only partly be confirmed. P. expansum was not able to take benefit from the inhibition of B. cinerea in the field. From 213

60 the available data it can be concluded Penicillium Botrytis 50 b that a rise in Penicillium infections ab following botryticide treatments 40 a should be, if any, only a rare a 30 development in the field. These conclusions are supported by the 20 analysis of the data from different ab bc c 10 a field trials that showed a positive correlation (r= 0,67) between the diesease severity [%] severity diesease 0 disease severities of blue and grey check Switch Teldor Cantus mold (figure 6). The correlation indicated, that in most cases a Fig. 4: Impact of different botryticides on the disease reduction of B. cinerea should result severity of Penicillium and Botrytis on grapes in the in lower infection levels of P. field (29.9.06). Variants with same letters do not expansum and that both pathogens differ significantly regarding to one pathogen prefer comparable conditions for (Dunns-Test, α= 0,05). Error bars show standard their development. deviation

Impact of yield regulation on the disease development (fie ld experiments) The field trials showed, that the disease development of Botrytis and Penicillium was affected by the different methods of yield reduction. Especially bunch dividing reduced the disease severities of both pathogens on grapes significantly (Figure 5). In 2005 and 2006 the degrees of efficiency using horizontal bunch dividing varied between 67 % and 87 % regarding Penicillium and between 25 % and 80 % regarding Botrytis. The disease reduction by this kind of yield regulation was clearly associated with the loosening of the bunch structure and thereby a reduced number of wounds. From the present state a successful control of P. expansum with botryticides seems not to be possible. An improved development of blue mold following the use of botryticides should also be a rare event. For a successful protection strategy main attention should be placed on the prevention of wounds as the main source for Penicillium infections. Especially the impact of different cultivation techniques on the compactness of the bunches, on the emergence of wounds and on the Penicillium inoculum in the vineyards should be carefully regarded.

25 12 a Botrytis Penicillium r= 0,67 (Pearson) 20 10 a 8 15 6

10 a 4 b b 2 disease severity 5 c Penicillium [%] 0 disease severity [%] 0 check bunch bunch 0 10203040506070 thinning dividing disease severity Botrytis [%] Fig. 5: Impact of bunch thinning and bunch Fig. 6: Correlation between Botrytis and dividing on the development of Botrytis and Penicillium disease severities on bunches. Penicillium on grapes. (14.10.05). Variants (Data from 18 trial sites (436 values) from with same letters do not differ significantly 2004 to 2006). r = Correlation coefficient regarding to one pathogen (Dunns-Test, α= referring to Pearson. 0,05). 214

Acknowledgements

The authors would like to thank their colleagues from the Department of Phytomedicine for technical assistance. The project was funded by the “Forschungsring des Deutschen Weinbaus” (FDW).

References

Hafner, P. (2001): Cluster splitting a possibility to reduce acetic acid rot. – Obstbau-Weinbau, Fachblatt des Südtiroler Beratungsringes, 38 (6): 190-191. Harms, M., M. Rothmeier, M. Mayer (2004): Untersuchungen zum Auftreten der Grünfäule (Penicillium spec.) in pfälzischen Rebflächen. – Mitteilungen aus der Biologischen Bundesanstalt für Land- und Forstwirtschaft 396: 114-115. La Guerche, S. (2004): Recherches sur les déviations organoleptiques des moûts et des vins associées au développement de pourritures sur les raisins. Etude particulière de la géosmine. – Thèse pour le doctorat de l`Université Bordeaux 2. Latorre, B.A and M.E. Rioja (2002): The effect of temperature and relative humidity on conidial germination of Botrytis cinerea. – Ciencia e Investigacion Agraria 29: 67–71. McClellan, W.D., W.B. Hewitt, P. La Vine, J. Kissler (1973): Early Botrytis rot of grapes and its control. – American Journal of Enology and Viticulture 24 (1): 27-30. Mohr, H.D. (2005): Farbatlas: Krankheiten, Schädlinge und Nützlinge an der Weinrebe.– Ulmer. Nelson, K.E. (1956): The effect of Botrytis infection on the tissue of Tokay grapes. – Phytopathology 46: 223-229. Samson, R.A. and J.C. Frisvad (2004): Penicillium subgenus Penicillium: new taxonomic schemes, mycotoxins and other extrolites. – Centraalbureau voor Schimmelcultures, Utrecht, Netherlands. Seyb, A.M. (2004): Botrytis cinerea inoculum sources in the vineyard system. – Lincoln, Lincoln University, PhD Thesis.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 215-219

Characteristics of wine and table grapevine hybrids tested for cultivation in Trentino (northern Italy)

Luca Zulini, Antonella Vecchione, Luigi Antonelli, Marco Stefanini IASMA Research Center - Agricultural Resources Department. Via E. Mach 1, I-38010 San Michele all’Adige (TN), Italy. e-mail: [email protected]

Abstract: All traditional European grapevine varieties (Vitis vinifera L.) are susceptible to the most destructive fungal diseases (downy mildew, powdery mildew and grey mould), while most interspecific varieties are more or less resistant toward fungal diseases, particularly towards mildews. Resistant hybrid vines have the potential to strongly reduce the application of plant protection compounds, leading a substantial and effective contribution for the protection of the environment. Despite of these evidences, in many viticultural areas the cultivation of hybrid varieties is still prohibited for the production of high quality wines. Thus, most of the countries involved in the breeding of new interspecific varieties in the last 20-30 years are located in northern parts of the grapevine production area, causing a lack of data about the use of new hybrids for warmer climate. In this work a field trial was carried out beginning from 2004, in Rovereto (province of Trento), with the aim to enlarge the knowledge of the agronomical characteristics of twenty-five hybrid vines bred in different countries. We evaluated fungus-resistance, yield quantity and quality (soluble solids, titratable acidity and pH) of the varieties. After four years of observation, 50% of the varieties showed good resistance against downy mildew (<5% of both leaf and bunch infected area). High levels of resistance were showed also against powdery mildew and grey rot. Most varieties revealed satisfactory adaptability characteristics regarding yield and fruit quality. We conclude that varieties Regent, Seyval blanc, Bianca, Nero and Palatina could be suitable for cultivation in our areas with good quality of grapes both for table and for wine production.

Key words: viticulture, grapevine interspecific hybrids, Trentino region, mildews, table grape, wine grape, disease resistance.

Introduction

In the EU, a major share of plant protection compounds are used to control fungal diseases in viticulture, mainly versus downy mildew, powdery mildew and grey mould. All traditional European grapevine varieties (Vitis vinifera L.) are susceptible to these fungal diseases, while most interspecific varieties are more or less resistant toward fungal diseases, particularly towards mildews. Resistant hybrid vines have the potential to strongly reduce the application of plant protection compounds, leading a substantial and effective contribution for the protection of the environment. These cultivars produce wine with some characteristics which are not appreciate by European consumers (Guedes de Pinho and Bertrand, 1995) but, especially in the last years, based on blind wine tastings of such wines we see that it is possible to make wine of equal quality with these cultivars as with traditional ones (Basler and Pfenninger, 2003). Despite these evidences, in many viticultural areas the cultivation of hybrid varieties is still prohibited for the production of high quality wines. Conversely, for the cultivation of hybrid varieties for table grapes there are no limitations. Thus, most of the countries involved in breeding and studies (Kozma, 1998; Eibach and Töpfer, 2003; Bényei et al., 2003) of new interspecific varieties in the last 20-30 years are located in northern parts of the grapevine production area (Germany, Hungary), causing a lack of data about the use of

215 216

new hybrids for warmer climate. The aim of this work is to enlarge the knowledge of the agronomical characteristics of twenty-five hybrids vines bred in different countries.

Material and methods

Examination were done between 2004 and 2007. The vineyard collection of hybrids is located in “Navicello” (Rovereto, 171 m a.s.l.), Experimental Farm of the “Istituto Agrario di San Michele all’Adige”. Vines are trained on “Pergola trentina semplice” and plant distance is 3 m between rows x 1 m between plants. All varieties are grafted onto SO4 rootstock. The vineyard is organically managed and no fungicide treatments were applied for the first three years of observations, while the last year (2007) five treatments with copper and sulphur were applied. The main characteristics of the 25 interspecific hybrids used are listed in table 1.

Table 1. Parentage, berry color and use of the hybrid varieties. Berry Variety Parentage Use color § Eger 2 x Halili krasznij white table Aurore Seibel 788 x Seibel 29 white wine Bianca Bouvier x Seyve Villard 12375 white wine Birstaler Muskat Seyval Blanc x Bacchus white wine/table Buffalo Herbert x Watkins black table Campbell Moore Early x (Belvidere x M. Hamburg) black table Chambourcin Seyve Villard 12417 x Seibel 7053 black wine Fanny Eger 2 x (Téli muskotàly x Olimpia) white table Glenora Ontario x Russian seedless black table Himrod Sultanina x Ontario white table Kyoho Centennial 4N x Ishihara Wase black wine/table Leon Millot (V. riparia x V. rupestris) x Goldriesling black wine Lidi Seyve Villard 12375 E.2 x Magaracsi csemege rosè table Lilla Eger 2 x (Pannonia kincse x Mathias Jànos diadala) white table (Seyve Villard 18315 x Müller Thugau) x Seyve Villard Muskat Bleu black table 20347 (Medoc Noir x Perle von Csaba) x (Seyve Villard 12375 Nero black table x Gardonyi) Ontario Winchell x Diamond white wine/table Palatina Königin der Weingärten x Seyve Villard 12375 white table Perle von Zala Seyve Villard 12375 x Csaba Gyoengye white table Regent Diana x Chambourcin black wine Romulus Ontario x Thompson Seedless black table Seyval blanc Seibel 5656 x Seibel 4986 white wine Suzi Villard Blanc x Pannonia Kincse white wine Therese Eger 2 x Olimpia white table Verdelet Seibel 5455 x Seibel 4938 white wine/table

217

At harvest, grape yield per vine, bunch weight, and berry weight were measured, as well as soluble solids (Brix), titratable acidity (g L-1 tartaric acid) and pH on the must. Resistance to downy mildew and powdery mildew was evaluated during the season based on the percentage of affected area on leaves and on bunches. Resistance to grey rot was determined at harvest time based on the surface of infected bunches.

Results and discussion

Productive parameters and grape quality parameters (table 2) showed a great variability among the 25 hybrid varieties.

Table 2. Grape yield and quality of the hybrid varieties.

Yield Bunch Berry Soluble solids Titratable Variety pH (kg/vine) weight (g) weight (g) (Brix) acidity (g/L) Angela 3.6 248 4.3 16.3 5.74 3.12 Aurore 1.9 70 1.6 19.9 7.89 3.41 Bianca 4.1 128 2.4 20.4 7.66 3.27 Birstaler Muskat 2.9 142 2.1 16.9 3.11 3.58 Buffalo 3.9 132 3.9 20.9 3.76 3.44 Campbell 4.6 226 7.5 14.9 3.96 3.42 Chambourcin 1.7 103 2.0 19.7 8.97 2.99 Fanny 2.2 239 3.5 15.4 4.49 3.23 Glenora 3.8 179 2.8 17.8 7.23 3.13 Himrod 0 – – – – – Kyoho 1.8 180 6.6 15.3 5.82 3.28 Leon Millot 2.7 61 1.2 22.0 8.96 3.37 Lidi 2.3 162 3.5 19.7 6.92 3.28 Lilla 4.5 241 4.2 17.5 6.92 3.15 Muskat Bleu 1.9 93 3.6 18.4 5.20 3.28 Nero 1.5 69 3.1 18.7 8.70 3.12 Ontario 0.4 88 2.4 23.7 5.18 3.43 Palatina 3.2 152 3.2 19.0 7.64 3.14 Perle von Zala 5.3 227 3.2 19.3 7.20 3.11 Regent 3.1 109 2.3 18.9 5.83 3.24 Romulus 3.6 169 1.7 19.0 6.89 3.00 Seyval blanc 3.4 128 1.2 22.0 6.90 3.32 Suzi 3.9 257 4.7 13.9 5.54 3.28 Therese 5.8 431 5.2 13.4 5.86 3.15 Verdelet 1.7 141 2.6 22.8 4.93 3.38

Table varieties Therese, Perle von Zala, Campbell and Lilla resulted in the higher values of yield, bunch weight and berry weight, while among the wine varieties, the most productive were Bianca, Suzi, Seyval blanc and Regent. Most of varieties obtained very good values of sugars, titratable acidity an pH (table 2), while some of hybrids did not reach an optimal ripening with low values of sugars (Therese, Suzi, Campbell, Kyoho, Fanny). Resistance to Plasmopara viticola, Uncinula necator and Botrytis cinerea can be observed in table 3. Against downy mildew, 14 hybrids showed good resistance with less of 218

10% of infected area both on leaf and bunch. Among the examined hybrids, varieties Himrod, Lilla, Ontario and Perle von Zala had the lowest resistance.

Table 3. Percentage of infection due to fungal diseases of the hybrid varieties. Downy Powdery Powdery Grey rot on Downy mildew Variety mildew on mildew on mildew on bunches on bunches (%) leaves (%) leaves (%) bunches (%) (%) Angela 0 0 3 3 10 Aurore 2 0 0 3 5 Bianca 1 0 0 0 0 Birstaler Muskat 3 2 0 5 0 Buffalo 13 9 0 0 0 Campbell 0 0 0 0 0 Chambourcin 4 12 0 5 0 Fanny 10 6 0 3 20 Glenora 8 7 3 8 5 Himrod 34 80 0 0 - Kyoho 18 4 3 0 0 Leon Millot 3 1 0 0 0 Lidi 10 3 10 15 35 Lilla 12 28 0 3 0 Muskat Bleu 3 0 0 13 0 Nero 1 0 0 0 0 Ontario 32 10 0 0 0 Palatina 3 0 3 3 0 Perle von Zala 21 2 3 15 10 Regent 1 0 0 0 0 Romulus 7 6 0 5 0 Seyval blanc 3 0 0 0 5 Suzi 10 3 5 5 0 Therese 3 3 0 0 10 Verdelet 5 10 0 0 0

Varieties Lidi, Muskat bleu and Perle von Zala showed the higher susceptibility to powdery mildew, while the remaining hybrids had good resistance. To grey rot infection, hybrids Fanny and Lidi were highly susceptible whereas the varieties Angela, Perle von Zala, Therese, Seyval blanc and Glenora were slightly sensitive to this disease. Remaining hybrids showed very high resistance. To the three fungal diseases only the variety Campbell showed a complete resistance, this hybrid is a table grape with a strong foxy flavour. Among the remaining hybrids examined in our trial, wine varieties Aurore, Bianca, Leon Millot, Regent, Seyval blanc and table varieties Nero and Palatina showed a high resistance to downy and powdery mildew as well as to grey rot. Summarizing resistance to diseases, grape yield and grape quality, the best performing hybrids in our environment seems to be Regent, Seyval blanc and Bianca among wine varieties and Nero and Palatina among table varieties. 219

Study on grape hybrids will continue in the future with a further evaluation of yield quality (in particular, flavour content and characteristics of wine). In conclusion interspecific varieties could allow to strongly reducing the application of plant protection compounds and the production costs, with important ecological benefit.

References

Basler, P. & Pfenninger, H. 2003: Disease-resistant cultivars as a solution for organic viticulture. – Acta Hort. (ISHS) 603: 681-685. Bényei, F., Lőrincz, A., Gácsi, T., Lukácsy, G. 2003: The results of the crossbreeding of the Department of Viticulture. – Acta Hort. (ISHS) 603: 697-699. Eibach, R. & Töpfer, R. 2003: Success in resistance breeding: “Regent” and its steps into the market. – Acta Hort. (ISHS) 603: 687-691. Guedes de Pinho, P. & Bertrand, A. 1995: Analytical determination of furaneol (2,5-dimethyl1- 4-hydroxy-3(2H)-furanone). Application to differentiation of white wines from hybrid and various Vitis vinifera cultivars. – Am. J. Enol. Vitic. 46(2): 181-186. Kozma, P. 1998: Evaluation of fungus-resistant wine grape varieties. – Acta Hort. (ISHS) 473: 93-103.

220 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 221-231

The impact of spiders (Araneae) on Lobesia botrana (Denis & Schiffermüller) population density

R. Addante 1, S. Di Gioia 1, C. Calculli 2, A. Pollice 2 1 Dipartimento di Biologia e Chimica Agro-Forestale ed Ambientale, Università degli Studi di Bari, Via Amendola 165/A, 70126 Bari (BA), Italy, [email protected]; 2 Dipartimento di Scienze Statistiche, Università degli Studi di Bari, Via C. Rosalba 53, 70124 Bari, Italy

Abstract: The role of spiders in regulating phytophagous pest populations is very difficult to investigate in open fields. In 2001 we began to observe the activity of these predators upon grapevine moth, and we continued this study in 2006-2007 with research to quantify their impact on the key pest of vineyards in Southern Italy, i.e. Lobesia botrana (Denis & Schiffermüller) (Lepidoptera Tortricidae). The research was realized in an experimental screenhouse vineyard completely covered by one single piece of insect-proof net. The screenhouse was subdivided into two isolated equal parts by a vertical net. The experiment consisted in collecting the spiders from each vine of one half of the vineyard and transferring them into the second half, then evaluating the variation of population density both of L. botrana and spiders over the seasons. In order to evaluate the spider population density, we randomly selected three plots in each half vineyard; each plot consisting of 10 cv Italia vines. Spiders were collected weekly from each vine of the plot using an entomological umbrella. From bloom to harvest, grapevine moth infestation was assessed by collecting one inflorescence or one cluster per vine and counting the numbers of infested berries and live larvae. Grapevine moth adults were collected weekly using sex pheromone traps and then counted. The regular transfer of spiders from one half of the vineyard to the other caused a noticeable variation in their density. The total number of spiders in the enriched vineyard was about double that in the depleted vineyard in 2006, and more than four times the number in 2007. As a consequence of spider enrichment in the release plot, the grapevine moth adults diminished, reaching the lowest density in the second year of study, when the number of adults trapped was less than one half of the number trapped in the plot where spiders were removed. The usefulness of spider activity in limiting L. botrana adults was not followed by a similar effect on phytophagous larvae, mainly because predators have poor possibilities of reaching the endophytic larvae of L. botrana carpophagous generations.

Key words: spiders, grapevine moth, vineyard

Introduction

Spiders are among the most important invertebrate predators in agroecosystems, and they play a remarkable role in limiting insect population densities (Nyffeler & Benz, 1987; Wise, 1993). Although these arthropods are generic predators, they show an indirect selectivity as they usually hunt for preys associated to their own ecological niche (Brignoli, 1983). In the past, studies on spider morphology, taxonomy, and biology were preferred to studies on their ecology and role as biological control agents in agroecosystems (Agnew & Smith, 1989). The reason for this preference lies in the difficulties in carrying out field studies on spiders; difficulties due to the cryptic behaviour of some species, the nocturnal activity of others, the problematic identification of spiderlings, etc. (Costello & Daane, 1995). Furthermore, because the life cycle of spiders is normally longer than that of their prey, it is very difficult to demonstrate the predator response to changes in prey population density (Costello & Daane, 1995).

221 222

Nevertheless, in the last few years, interest in ecological studies on spiders has increased. A lot of works have investigated the role of spiders on natural pest control in different agroecosystems. Studying the effects of cover crop and spider density on Erythroneura variabilis Beamer population in Californian vineyards Hanna et al. (2003) demonstrated the negative impact of predators on the leafhopper and furthermore the positive effect of cover crops in increasing the spider density, though no reduction in E. variabilis density was found on vines in cover crop plot as compared with vines in bare ground plots. Costello & Daane (1995) studied the spider species composition and diversity indices in seven grape vineyards in San Joaquin Valley (California) also describing the seasonal abundance of the eight most commonly collected spider species. The influence of ground cover on the leafhoppers Erythroneura spp. and spider populations was investigated by Costello & Daane (1998, 2003). Nicholls et al. (2000) also included Frankliniella occidentalis (Pergande) in their research on spider predation. Addante et al. (2003) studied the relationships between spiders and grapevine moth in Apulian vineyards, also reporting the spider species composition and specifying the dominant species. Further investigations in the same vineyards conducted to the description of the seasonal fluctuations of the most representative spider species, and pointed out the negative effect of covering nets on the spider fauna, particularly on Araneidae, Theridiidae and Miturgidae (Addante et al., 2006, 2007). The aim of the present work was to evaluate the variations of Lobesia botrana (Denis & Schiffermüller) (Lepidoptera Tortricidae) populations in correlation to varying spider densities in an Apulian vineyard.

Materials and methods

The research was carried out in an experimental “tendone” vineyard, established in 1990 with a 2x2 m planting density. The area of about 750 m2 was completely and permanently covered by one single piece of insect-proof net installed in 2000. At the beginning of March 2006 the screenhouse vineyard was subdivided into two equal parts by a vertical net to prevent both insects and spiders passing from one unit to another. Spider transfer From mid-March 2006 until the end of August 2007, the spiders were removed weekly from each vine of one half of the vineyard and released into the second half in order to evaluate the impact of spiders on L. botrana population. We expected a reduction in spider density, and a corresponding increase in grapevine moth population in the half vineyard where spiders were removed. Spiders were collected by placing an entomological umbrella under the canopy of the vines, and beating canes three times to make the spiders fall. This procedure was applied to the whole canopy of the removal plot. The collected spiders were released uniformly over the vines in the second half vineyard. Evaluation of L. botrana and spider population densities In each unit many plots with different cultivars were present, but measurements were made only on cv Italia. We randomly selected three plots of 10 cv Italia vines in each half vineyard. In order to evaluate the spider population density, spiders were collected weekly from the beginning of March until the first week of September of each year from each vine of the plot using a 1 x 1 m entomological umbrella. The spiders were identified up to the family, genus or species level according to the developmental stage of each single specimen. From pre-blooming to harvest, grapevine moth larval infestation was assessed by collecting one inflorescence or one cluster per vine and counting the numbers of infested berries and live larvae. Grapevine moth adults were collected weekly by sex pheromone traps and then counted. 223

No insecticidal treatment was applied during the experimental period. Statistical analysis Generalized linear models (GLM) have been applied to the collected data. This model was chosen considering the nature of the collected data: since the response variable Yi expresses a count, its distribution can be reasonably assumed as the Poisson distribution. The systematic component of the model is a function of a set of explanatory variables x1 ,K, x p , either quantitative or qualitative (factors). The models applied to this study have the following form: p

log(E[Yi ]) = α + ∑ β j xij for i = 1,...,n j=1 where α represents the intercept. Generalized Estimating Equation models (GEE, Liang and Zeger, 1986) were developed to extend GLM’s when observations cannot be considered independent. As such the use of GEE’s implies the assumption of a specific correlation structure for the observed data. Separate analysis (available from the authors on request) lead to the assumption of a common order one autoregressive time correlation structure (AR(1)) for the counts of spiders and of the grapevine moths: each observation (count) is thus assumed to depend on the previous one and to determine the next. Notice that the documented seasonality of both population densities could not be sufficiently appreciated (i.e. quantified) due to the limited length of the study. Nevertheless the explicit consideration of the correlation among succeeding observations allows to pool the data for year without incurring in a pseudo-replication issue (Hurlbert, 1984). In other words, the absence of replications in the experimental design is accounted for in the data analysis by testing treatment differences considering subsequent observations as replications, making an explicit assumption for the lack of independence. GEE models and GLM’s were respectively used to analyze two different kinds of relations: 1. the numerical variations of L. botrana adults, total number of spiders, total number of Icius, and total number of Cheiracanthium in relation with variations in the plot and year; 2. the numerical variations of L. botrana adults in relation with variations in the total number of spiders, total number of Icius, total number of Cheiracanthium, plot, and year. The estimates of the model parameters were carried out using functions glm and geeglm, in the R statistical software, implementing the standard procedures respectively based on the maximum likelihood method and on generalized estimating equations (Liang & Zeger, 1986; McCulloch et al., 2001). In the following sections we show the results of modelling experimental data by GEE’s and GLM’s. Only the models with corresponding significant estimates are reported.

Results and discussion

Spider species We collected 786 spiders in 2006 and 1802 in 2007, making a total of 2588 individuals grouped in 21 families and 56 genera (Tab. 1). The largest number belonged to the Salticidae family, with 11 genera, more than 12 species, and 1247 individuals. The most abundant genus was Icius with 34% of the total number of spiders, followed by Cheiracanthium with 14% and Theridion with 9%. Since all the collected Icius adults belonged to the species I. hamatus (C.L. Koch), we assumed that the Icius spiderlings also belonged to the same species. Similarly, since all the Cheiracanthium adults were found to be C. mildei L. Koch, we 224

assumed that the whole population detected belonged to this species. So we can conclude that the most abundant species collected in the vineyard were the Salticidae I. hamatus and the Miturgidae C. mildei.

Table 1. Distribution in families and genera of the collected spiders.

Family No. of genera No. of spiders Agelenidae 1 2 Amaurobidae 1 1 Anyphaenidae 1 21 Araneidae 4 75 Clubionidae 1 78 Dictynidae 1 7 Gnaphosidae 5 157 Linyphiidae 6 94 Lycosidae 2 3 Mimetidae 1 1 Miturgidae 1 354 Oxyopidae 1 43 Philodromidae 2 119 Pisauridae 1 9 Salticidae 11 1247 Sparassidae 2 8 Theridiidae 8 247 Thomisidae 4 103 Uloboridae 1 1 Zodaridae 1 1 Zoridae 1 17 Total 56 2588

Effect of spider transfer on spider population densities Considering that the vineyard was covered by net in 2000 and that it was subdivided into two parts in 2006, we can assume that at the beginning of the experiment the spider community and L. botrana population densities were almost the same in the whole experimental unit. The assumption is supported by the very similar pest and predator densities observed in the two plots at the first samplings in 2006 (Figs. 1, 2, 3, 4, and 5). In 2006, due to spider transfer, the mean density of the whole spider community observed in the release plot was 25.14, almost double the mean density of the removal plot which was 12.29 (Tab. 2). In 2007, the density of spiders in the release plot was 52.64, more than four times the removal plot density of 11.71. Comparing the spider density between years, whereas in the release plot the spider number was more than double in 2007 as compared to 2006, in the removal plot only a slight reduction in spider density was observed in 2007 (Tab. 2).

225

Table. 2. Mean density of spiders and grapevine moth per field and per year. 2006 2007 Variables removal release removal release Spiders (all species) 12.29 25.14 11.71 52.64 Icius hamatus 1.86 10.19 4.29 18.79 Cheiracanthium mildei 2.81 3.24 1.79 6.32 L. botrana adults 27.33 20.33 14.42 6.62 L. botrana larvae 19.43 20.48 19.96 15.37

The application of the generalized estimating equations (GEE) to the total number of spiders led us to estimate the coefficients β0 , β1 , α reported in Table 3. The estimated coefficient for the removal plot is negative, thus proving an increase in the total number of spiders in the release plot.

Table 3. Results of generalized estimating equations applied to the total number of spiders as a function of the variable plot. Effects Estimated coefficients SE p-value

Intercept ( β0 ) 5.96 0.43 <0.00

Removal plot ( β1 ) -1.28 0.43 <0.00 Estimated Correlation Parameter (α ) 0.88 0.14 -

The estimated coefficients β0 , β1 , β2 ,α obtained by applying GEE to I. hamatus and to C. mildei counts are reported in Table 4 and 5 respectively.

Table 4. Results of generalized estimating equations applied to the total number of I. hamatus as a function of the variables plot and year. Effects Estimated coefficients SE p-value

Intercept ( β0 ) 4.55 0.01 <0.00

Removal plot ( β1 ) -1.54 0.04 <0.00

Year 2007 ( β2 ) 0.67 0.04 <0.00 Estimated Correlation Parameter (α ) 0.55 0.14 -

Table 5. Results of the generalized estimating equations applied to the total number of C. mildei as a function of the variable plot. Effects Estimated coefficients SE p-value

Intercept ( β0 ) 3.85 0.39 <0.00

Removal plot ( β1 ) -0.85 0.42 0.04 Estimated Correlation Parameter (α ) 0.62 0.16 -

226

As for the total number of spiders, estimated coefficients for the effects of the removal plot are negative for the I. hamatus. The expected number of I. hamatus and C. mildei counts is thus greater in the release plot than in the removal. Notice that the estimated effect of year 2007 is significant and positive for the I. hamatus. The model again confirmed that the highest density was predicted during the second year. The effectiveness of spider transfer was also confirmed in this case. Based on the previous considerations, we can say that spider transfer was effective in increasing the spider density in the release plot; this was verified for the whole spider community (Fig. 1) and for the most important spider species, I. hamatus (Fig. 2) and C. mildei (Fig. 3). The less effective transfer of C. mildei than for I. hamatus (Tab. 2), probably due to the difficulty of dislodging Cheiracanthium adults from foliage, was not confirmed by the adopted models.

vegetation 20

0 mean number of spiders per sq. m 9/3 6/4 4/5 1/6 7/9 23/3 20/4 18/5 15/6 29/6 13/7 27/7 10/8 24/8

date

2006 removal 2006 release 2007 removal 2007 release

Fig. 1. Cumulative number of the total number of spiders in the observed plots and years.

On the whole, spider density increased over the seasons, reaching the highest values in the second year of study, thus indicating that spider transfer needs to be repeated many times and over a long period in order to be effective. Although spiders are able to predate on other spiders up to cannibalism, their density was artificially increased in the release plot. This means that the environmental carrying capacity for these predators could be increased in the studied vineyard, thus improving their use as biological control agents.

Effect of spider transfer on grapevine moth population trend Statistical analysis applied to L. botrana adults showed a significant and positive estimated coefficient for the removal plot, implying smaller expected counts of L. botrana in the release plot, and a negative estimated effect of year 2007 (Tab. 6). We can conclude that the spider transfer to the release plot corresponds to a reduction of the counts of L. botrana and that this is far more evident in the second year, due to cumulative effects (Fig. 4).

227

20 18 per sq. per 16 14 12

I. hamatus 10 8

m vegetation 6 4 2 0 mean number of 9/3 6/4 4/5 1/6 7/9 23/3 20/4 18/5 15/6 29/6 13/7 27/7 10/8 24/8

dates

2006 removal 2006 release 2007 removal 2007 release

Fig. 2. Cumulative number of I. hamatus in the observed plots and years.

per sq. m per 4

3 vegetation

C. mildei C. 2

mean n. 0 9/3 6/4 4/5 1/6 7/9 23/3 20/4 18/5 15/6 29/6 13/7 27/7 10/8 24/8

dates

2006 removal 2006 release 2007 removal 2007 release

Fig. 3. Cumulative number of C. mildei in the observed plots and years.

Table 6. Results of the generalized estimating equations applied to the total number of L. botrana adults as a function of the variables plot and year. Effects Estimated coefficients ES p-value

Intercept ( β0 ) 2.86 0.10 <0.00

Removal plot ( β1 ) 0.46 0.12 <0.00

Year 2007 ( β2 ) -0.81 0.13 <0.00 Estimated Correlation Parameter (α ) 0.39 0.07 - 228

700

600

500

400 adults per trap 300

200

L. botrana 100 n. 0 9/3 6/4 4/5 1/6 7/9 23/3 20/4 18/5 15/6 29/6 13/7 27/7 10/8 24/8

dates

2006 removal 2006 release 2007 removal 2007 release

Fig. 4. Cumulative number of L. botrana adults in the observed plots and years.

Regarding the total number of grapevine moth larvae collected over the whole season, no difference was observed between the two plots in 2006, and only a slight decrease was seen in the release plot in 2007 (Tab. 2 and Fig. 5). The GLM model applied to the total number of larvae did not show any statistically significant difference either between plots or years.

20 18 larvae 16 14 12

L. botrana 10 8 per cluster 6 4 2

men number of 0 7/6 5/7 2/8 9/8 6/9 17/5 24/5 31/5 14/6 21/6 28/6 12/7 19/7 26/7 16/8 23/8 30/8

dates

2006 removal 2006 release 2007 removal 2007 release

Fig. 5. Cumulative number of L. botrana larvae in the observed plots and years.

Impact of spiders on L. botrana The impact of the spider populations on that of L. botrana is analyzed including spiders counts as explanatory variables within models for the grapevine moth counts. The AR(1) time 229

correlation structure of spiders counts (for the two species and for the total) is very similar to that of L. botrana. For this reason inclusion of both the spider effect and time correlation within GEE is redundant and leads to unsatisfactory results. The consideration of spiders counts as explanatory variable with no time correlation effect lead to the use of Poisson GLM’s. Concerning the relations between L. botrana and spiders accounting for plot and year effects, GLM gave highly significant estimated coefficients except for the whole community of spiders, which resulted less significant (P<0.01) (the effect of the interaction between the variables plot and year on the expected number of L. botrana was not included in the model as it was not found to be statistically significant). This last less significant coefficient was presumably due to the limited effect of many of the collected spider species on grapevine moth. The negative spider coefficient demonstrates the limiting effect of predators on grapevine moth adults (Tab. 7).

Table 7. Generalized linear model applied to the number of L. botrana adults as a function of the variables spiders, plot and year. Effects Estimated coefficientsp-value Intercept 3.103 <0.001 Spiders -0.006 <0.01 Removal plot 0.322 <0.001 Year 2007 -0.753 <0.001

The reduction in L. botrana adults observed in the release plot, and at the same time the noticeable increment detected in I. hamatus and C. mildei densities (Tab. 2), allow us to suppose an important role of both spider species in controlling the pest. This evidence was supported by the applied models, given that the estimated coefficients for the two species were both negative (Tabs. 8 and 9), indicating that an increase in spider density caused a reduction in grapevine moth adults (the effect of the variable year was not included in the two models as it was not found to be statistically significant).

Table 8. Generalized linear model applied to the number of L. botrana adults as a function of the variables Icius and plot. Effects Estimated coefficientsp-value Intercept 2.722 <0.001 Icius -0.012 <0.001 Removal plot 0.322 <0.001

Table 9. Generalized linear model applied to estimate the number of L. botrana adults as a function of the variables Cheiracanthium and plot. Effects Estimated coefficientsp-value Intercept 2.628 <0.001 Cheiracanthium -0.017 <0.05 Removal plot 0.415 <0.001

230

Since no shifting of grapevine moth was possible between the two plots, and considering that climatic and cultural conditions were the same on the whole vineyard, we can reasonably attribute the lower L. botrana population in the release plot to the effect of the spiders. The lower grapevine moth density in 2007 compared to 2006, particularly in the release plot, can be attributed mainly to the greater number of spiders observed in the second year of study. Since GLM applied to spiders and L. botrana larvae did not give any statistically significant estimated coefficient it seems that spiders were not very effective in controlling grapevine moth larvae, this was probably due to the difficulties spiders encountered in reaching the larvae of the carpophagous generations feeding and living in the berries.

Conclusions

The most abundant spider species collected in the studied vineyards was I. hamatus, followed by C. mildei. Weekly transfer of spiders from one half of the vineyard to the other caused a noticeable increase in their density in the release plot, and a less conspicuous impoverishment in the removal plot. Nevertheless, even though spiders were regularly collected and transferred, their community was never extinguished in the half vineyard from which they were removed. Effectiveness in spider transfer varied according to the species, causing a greater enrichment in I. hamatus population density than in C. mildei. The total number of spiders in the enriched plot was almost double that of the impoverished plot in 2006, and more than four times the number of the impoverished plot in 2007. As a consequence of spider enrichment in the release plot, the grapevine moth adults diminished, reaching the lowest density in the second year of study, when less than one half the number of adults were caught in comparison to the plot from which spiders were removed. The significant action of spiders in limiting L. botrana adults corresponded to only a slight effect on the phytophagous larvae, mainly due to the difficulties for spiders of reaching the endophytic larvae of L. botrana carpophagous generations. Spiders, and particularly I. hamatus and C. mildei, seem to be able to limit L. botrana populations on vineyards, but further long-term investigations will need to evaluate the effects of predators both on pest adults and larvae, and their interaction with the whole beneficial arthropod community.

Acknowledgements

The authors wish to express their gratitude to Dr Carlo Pesarini of the Museo Civico di Storia Naturale in Milan (Italy) for having identified some of the spiders collected.

References

Addante R., Moleas T. & Ranieri G. 2003: Preliminary investigations on the interaction between spiders (Araneae) and grapevine moth (Lobesia botrana (Denis et Schiffer- müller)) populations in Apulian vineyards. – IOBC/wprs Bull. 26(8): 111-115. Addante R., Pesarini C. & Ranieri G. 2006: Faunistic and ecological aspects of spiders (Araneae) on vineyards. – IOBC/wprs Bull. 29(11): 221-234. Addante R., Ranieri G. & Pesarini C., Moleas T. 2007: Indagine sulle relazioni tra ragni e Lobesia botrana (Denis & Schiffermüller) (Lepidoptera Tortricidae) in vigneti pugliesi. – Atti XXI Congr. naz. it. Entomol., Campobasso 11-16 giugno 2007: 214. 231

Agnew C.W. & Smith Jr. J.W. 1989: Ecology of spiders (Araneae) in a peanut agroeco- system. – Environ. Entomol. 18(1): 30-42. Brignoli P.M. 1983: I ragni quali predatori di insetti: il loro potenziale ruolo negli agroecosistemi (Araneae). – Atti XIII Congr. naz. it. Entomol., Sestriere–Torino: 591- 597. Costello M.J. & Daane K.M. 1995: Spider (Araneae) species composition and seasonal abundance in San Joaquin Valley grape vineyards. – Environ. Entomol. 24(4): 823-831. Costello M.J. & Daane K.M. 1998: Influence of ground cover on spider populations in a table grape vineyard. – Ecol. Entomol. 23: 33-40. Costello M.J. & Daane K.M. 2003: Spider and leafhopper (Erythroneura spp.) response to vineyard ground cover. – Environ. Entomol. 32(5): 1085-1098. Hanna R., Zalom F. & Roltsch W. 2003: Relative impact of spider predation and cover crop on population dynamics of Erythroneura variabilis in a raisin grape vineyard. – Entomol. Exp. Appl. 107: 177-191. Hurlbert S.H. 1984: Pseudoreplication and the design of ecological field experiments. – Ecol. Monogr. 54(2): 187-211. Liang, K.Y. & Zeger S.L. 1986: Longitudinal data analysis using generalized linear models. – Biometrika 73: 13-22. McCulloch C.E. & Searle S.R. 2001: Generalized, linear and mixed models. – John Wiley & Sons, Inc. New York. Nicholls C.I., Parrella M.P. & Altieri M.A. 2000: Reducing the abundance of leafhoppers and thrips in a Northern California organic vineyard through maintenance of full season floral diversity with summer cover crops. – Agric. for. entomol. 2(2): 107-113. Nyffeler M. & Benz G. 1987: Spiders in natural pest control: a review. – J. Appl. Ent., 103: 321-339. Wise D.H., 1993: Spiders in ecological webs. – Cambridge University Press, Cambridge, U.K. 232

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 233-236

Olfactory responses of Eupoecilia ambiguella (Hübner) (Lepidoptera Tortricidae) females to volatiles from grapevine*

Gianfranco Anfora1, Marco Tasin1, Anna-Carin Bäckmann1, Elisabetta Leonardelli1, Antonio De Cristofaro2, Andrea Lucchi3, Claudio Ioriatti1 1 SafeCrop Centre and Plant Protection Department, IASMA Research Centre, Via E. Mach 1, 38010 San Michele a/A (TN), Italy, [email protected]; 2 Department of Animal, Plant and Environmental Sciences, University of Molise, Via De Sanctis, 86100 Campobasso, Italy; 3Dipartimento Coltivazione e Difesa Specie Legnose “G. Scaramuzzi”, Sezione Entomologia agraria, Università di Pisa, Italia

Abstract: The activity of volatiles compounds emitted by a highly susceptible grape variety (Chardonnay) was assessed in a dual choice oviposition test on Eupoecilia ambiguella (Hübner) (Lepidoptera Tortricidae) behaviour, in which females could choose by using only olfactory cues. Volatiles released by the grape clusters were collected through headspace technique. Electrophysio- logical analyses of the volatiles released by the grapes were conducted for the identification of antennally active compounds, supposed to be involved in the host plant and oviposition site selection by females. Results revealed the impact of the grape odour on total oviposition highlighting a major role played by olfactory cues in driving females to the oviposition site. GC-EAD analyses of the headspace collections detected several active compounds on antennae of E. ambiguella mated females.

Key words: grapevine moth, Vitis vinifera, oviposition bioassay, GC-EAD, GC-MS

Introduction

The grape berry moth Eupoecilia ambiguella (Hübner) (Lepidoptera Tortricidae) is one of the key pests of vine in Europe. The larvae feed on the clusters and favour the spread of grey mould, Botrytis cinerea. Its control commonly relies on insecticide treatments, whilst mating disruption is the most widespread strategy based on semiochemicals. One of the main drawbacks of mating disruption is that it cannot affect the behaviour of ovipositing females. Hence, the effectiveness of pheromone-based methods could be significantly enhanced by identification of kairomonal compounds. Despite its relevant economic importance, the influence of olfactory cues in female host-finding and egg-laying selection is nowadays ignored. This study aimed at investigating the activity of volatiles compounds emitted by a highly susceptible grapevine variety (Chardonnay) on E. ambiguella behaviour. First, we studied the influence on the E. ambiguella egg-laying behaviour of bunches volatiles by means of a dual choice oviposition assay, borrowed from one previously set up for the potato tuber moth (De Cristofaro et al., 2005) and herein improved for the grapevine moths. In the assay the choice of females is restricted only to volatile cues. Moreover, volatiles released by the grape clusters were collected through headspace technique; such extracts were analysed by a gas chromatograph coupled to an electroantennodetector (GC-EAD). Electrophysiologically active compounds were identified by a gas chromatograph equipped with a mass spectrometer (GC-MS).

233 234

Material and methods

Insects Moths were selected from a colony of E. ambiguella reared on a semi-artificial diet at IASMA. Insects were reared under a L16:D8 photoperiod, 60±10 R.H. and 22°C. Newly formed pupae were sexed and emerging adults were kept in plastic Petri dishes provided with sucrose solution (10% w/v in H2O).

Oviposition bioassay Assays were conducted in ventilated cylindrical cages (25 cm ID x 45 cm; metallic net, mesh 2mm wide). One bunch (cv Chardonnay, phenological stage "green berries", in correspon- dence of the second moth flight) was confined into a plastic conical glass (bottom 61 mm, top 88 mm, height 130 mm). Thirty holes (1.5±0.2 mm ID) were pierced in the side wall of the glass to allow volatiles flowing out. A second glass was left hermetically sealed as control. Three mated females were released into the centre of each cage. After 72 hours at 25°C and L16:D8 photoperiod, the eggs laid on each glass were counted (n=20). Results were compared using Student’s t-test.

eggs/female 8

0 odour source control

Fig. 1. Number of eggs (mean ± SD) laid per female of E. ambiguella in the oviposition bioassay (n=20) using Chardonnay green berries as odour source. Vertical bars represent standard deviation. Student’s t-test: t = 12.9, P < 0.01, d.f. = 38.

Headspace collection Volatiles released by the grape clusters (phenological stages "green berries") belonging to the cv Chardonnay were collected through headspace technique (Tasin et al., 2005). Bunches were placed in a 25x38 cm polyacetate bag for volatiles collection. Air from the bag was drawn at 150 ml/min through an adsorbent cartridge (75 mg Super Q, Sigma-Aldrich, Milan, Italy) connected to a vacuum pump. Charcoal filtered air was pulled back simultaneously in the bag by the same pump to maintain a constant pressure. Volatiles collection lasted 6 hours. 235

Volatiles were eluted from the cartridge by solvent desorption at room temperature using 400 µl of hexane (>99% purity, Sigma-Aldrich, Milan, Italy). Extracts were stored in 2-ml vials at -18°C until use.

FID 1 2 3 4 5 6 7 EAD

Fig. 2. Electrophysiological (GC-EAD) responses of an E. ambiguella antenna to headspace collection from Chardonnay green berries. Compounds eliciting antennal responses: 1) 1-octen-3-ol, 2) limonene, 3) 4,8-dimethyl-1,(E)3,7-nonatriene, 4) linalool, 5) β-caryophyllene, 6) (E,E)-α-farnesene, 7) methyl salicylate

Electrophysiological and chemical analysis Plant extracts were analysed by a Hewlett Packard (Palo Alto, CA) 5890 gas chromatograph (GC) with a flame ionization detector (FID) and a Innowax column (30 m x 0.2 mm x 0.32 µm) programmed from 60°C (3 min hold) at 8°C/min to 220°C (10 min) coupled to a Syntech electroantennodetector (GC-EAD, Hilversum, The Netherlands) (n=3). An excised E. ambiguella antenna was suspended between two glass electrodes filled with Kaissling solution. The mounted antenna was placed into the air stream carrying the volatiles compounds eluted from the GC column. Compounds eliciting antennal responses in all the recordings were scored as active. Antennally active compounds were identified by coupled gas chromatography and mass spectrometry (GC–MS). Analyses were performed on a Hewlett-Packard 5890 GC, with a polar Innowax column (30 m x 0.32 mm; J & W Scientific, Folsom, CA) interfaced with a Hewlett-Packard 5970B MS with electron impact ionization (70 eV). Identity of compounds was verified by comparison with synthetic compounds purchased from Sigma-Aldrich and Fluka Chemie (Buchs, Switzerland). 236

Results and discussion

Females significantly preferred to lay their eggs on the glass releasing the volatiles bouquet from the berries rather than on the control (Figure 1). It shows that olfactory cues are a central factor in the host plant location by mated females and are sufficient to indicate a suitable oviposition site. GC-EAD analyses of the headspace collections detected several active compounds on antennae of mated females. On the basis of antennal responses and relative amounts in the blend, the most active compounds were 1-octen-3-ol, limonene, 4,8-dimethyl-1,(E)3,7-nonatriene, linalool, β-caryophyllene, (E,E)-α-farnesene and methyl salicylate (Figure 2). We did not find differences in EAD responses between males and females, both unmated and mated. Single compounds or blends, responsible of the plant attractiveness to the moth, could be used for enhancing the current pheromone-based strategies and for novel methods in female monitoring and controlling (Büsser & Guerin, 2007).

Acknowledgements

Many thanks are due to Gessica Tolotti for her technical assistance.

This study was partially supported by MIUR (PRIN 2006 funds).

References

Büsser, D. & Guerin, P.M. 2007: Attraction of the European grape berry moth to its pheromone is affected by plant volatiles. – 10th European Symposium for Insect Taste and Olfaction, Roscoff, France, 15-22 September 2007, http://www.esito-symp.org. De Cristofaro, A., Anfora, G., Germinara, G.S., Cristofaro, M. & Rotundo, G. 2003: Olfactory and behavioural responses of Phthorimaea operculella (Zeller) (Lepidoptera, Gelechiidae) adults to volatile compounds of Solanum tuberosum L. – Phytophaga XIII: 53-61. Tasin, M., Anfora, G., Ioriatti, C., Carlin, S., De Cristofaro, A., Schmidt, S., Bengtsson, M., Versini, G. & Witzgall, P. 2005: Antennal and behavioural responses of grapevine moth Lobesia botrana females to volatiles from grapevine. – J. Chem. Ecol. 31: 77-87.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 237-243

Comparison of two methods for the agrochemicals side effect evaluation on Phytoseiid mites in vineyards.

Mario Baldessari, Romano Maines, Gino Angeli IASMA Research Center, Plant Protection Department, Via E. Mach 1. I-38010 - San Michele all’Adige (TN) Italy ; e-mail: [email protected]

Abstract: Two assessment techniques, washing method and direct counting, were compared for estimating population size of the phytoseiids Amblyseius andersoni and Kampimodromus aberrans in vineyard. Applying the two techniques, trials on side effects of insecticides and estimation of phytoseiids on different grape variety were performed. Results show that washing method provide an efficient estimation of phytoseiid population and it was comparable with the direct counting, both with high and low population density. The mites were easily washed off and well preserved for classification. Moreover, washing method allow to reduce at least 70% the assessment time spent in laboratory and so could be considered the better way to handle an extensive evaluation programme on side effects of agrochemicals on Phytoseiids that inhabit the vineyards. Concerning the selectivity levels of the tested insecticides, Actara, Cascade, Confidor and Calypso resulted slightly harmful according the IOBC evaluation classes, compared to the higher toxicity of reference Trebon. High differences concerning Phytoseiid populations checked comparing three grape cultivars should be opportunely considered in planning side-effects field studies on grape varieties.

Key words: Amblyseius andersoni Chant, Kampimodromus aberrans Oudemans, washing method, direct counting, side effects of insecticides, vineyards.

Introduction

Predatory phytoseiid mites are indisputable effective management components of vineyards. Amblyseius andersoni (Chant) and Kampimodromus aberrans (Oudemans) are two of the major type III biocontrol agents of European red mite Panonychus ulmi Koch (Angeli & Ioriatti, 1994; Duso et al., 2006). Their effectiveness in an integrated pest management system of vineyard is assured, provided they are resistant to agrochemicals (Angeli et al., 1996; 1997; 2001; Candolfi et al., 2000). Thus the identification of selective pesticides in IPM programs is crucial factor. Side effect studies are traditionally evaluated according to the guidelines developed by EPPO and IOBC/WPRS WG “Side effect and beneficial organism” for laboratory, semi-field and field toxicity. The use of standard methods will allow to exchange the results from one country to another. Before the agrochemical application in field trials, phytoseiid mites density should reach at least a mean number of 30 mites per sample (~ 1 mobile form/leaf), in order to minimize the consequences of assessment and counting errors (Blümel et al., 2000). If the mean number of phytoseiid mites is less than one mobile form per leaf (e.g. due to ecological reasons) the sample size expressed as number of leaves per plot has to be increased, with the consequence of an increasing time need for the correct assessment of the population density. One of the main problems in every population ecology study of predacious and phytophagous mites in vineyards is the development of accurate and precise assessment techniques. Sampling and counting methods should be as accurate as possible, simple and rapid. Besides, it is useful if

237 238

the same techniques can be applied to survey predator and prey intensity. This is possible with phytoseiid mites and their tetranychid prey, since the main area of interaction are the leaf surface. Therefore, some techniques suitable for phytoseiid assessment are equally applicable for spider mite. Several approaches could be adopted to assess mite density as direct counting, washing method, brushing and heat extraction method (Boller, 1984; Sabelis, 1985; Nachman, 1985). The efficiency of the different methods when applied to assess mite density on grape leaves has been rarely studied. Generally the side-effect studies of agrochemicals include many treatments and as a consequence they are know to be time consuming, due to the high number of leaves that have to be checked. Two of these sampling techniques, washing method and direct counting were compared to evaluate the side effects of some new insecticides in vineyard and also in a dynamic population study of phytoseiids on three grape varieties characterized by different leaf pubescence. This paper discusses the obtained results.

Materials and methods

The two assessment approaches, direct-counting and washing method, were evaluated in the following trials. In 2006 one vineyard cultivated with the grape cultivars Merlot and inhabited by A. andersoni (>90%) was sprayed with three neonicotinoids, Actara 25 WG (a.i. thiamethoxam; 200 ml/ha), Confidor 200 SL (a.i. imidacloprid 17.8%; 1125 ml/ha) and Calypso (a.i. thiacloprid 40.4% SC; 375 ml/ha) and the chitin-inhibitors Cascade 50 DC (a.i. flufenoxuron 4.7%; 2250 ml/ha), in a randomized experimental design (4 replicates per treatment); the reference Trebon (a.i. etofenprox 15% ME; 1500 ml/ha) and a water control completed the tested treatments. In 2007 three nearby vineyards of the cultivars Chardonnay, Merlot and Pinot grigio, characterized by different leaf morphology (unpublished data) but inhabited by the same strain of K. aberrans (>93%) and exposed to same agrochemicals were chosen for the comparison of the two sampling techniques. In both trials Phytoseiid densities were estimated by randomly collecting 120-200 leaves per treatment and date. In laboratory each leaf sample was divided into two homogenous groups and undergo at two assessment approaches. Washing method Sampled leaves were dipped into a soap solution (Agral® 5%, 30 cc/hl) and kept overnight (12±2 hours) at room temperature. Then, leaves were washed by shaking them gently and were discarded later. The solution was then filtered through a 15 µm mesh cloth and the number of mobile mites (young and adults) was subsequently estimated under a stereobinocular microscope. The mite load was expressed as mite per leaf by dividing the total number of mites of filter by the number of sampled leaves. Direct counting The mobile mites (young stages and adults) confined both on the over and undersite leaves were numbered under a stereobinocular. The mite load was expressed as mite per leaf. Statistical analysis The number of phytoseiids were compared across the two monitoring techniques using Student’s t-test. The standard error (SE) was also calculated. The side-effects of the pesticides on phytoseiids were evaluated using the Henderson & Tilton formula and the data were correlated to the classification established by IOBC (Boller et al., 2005). The phytoseiids cheeked across the three grape variety were compared using one-way ANOVA followed by Tukey test (SPSS Programme).

239

Results

Figure 1 shows the estimated relative density of A. andersoni obtained using the two different monitoring techniques. Results show that washing method provide an efficient population estimation of mite population and it’s comparable with the direct counting, both with high and low population density. Overall, of 42 comparisons there were 9 cases of statistic differences between the two assessed techniques, five in favour of direct counting and four favourable at the washing method.

UNTREATED ACTARA

5 5

4,5 4,5 p ≤ 0,05 4 4 3,5 3,5 3 3 2,5 2,5 2 2 mobil stage/leaf mobil stage/leaf mobil 1,5 1,5

1 1 0,5 0,5

0 0 T - 1 T + 4 T + 7 T + 14 T + 21 T + 28 T + 42 T - 1 T + 4 T + 7 T + 14 T + 21 T + 28 T + 42 Direct counting method Washing methods Direct counting method Washing methods

CASCADE CONFIDOR

5 5

4,5 4,5 p ≤ 0,05 4 4 p ≤ 0,05 p ≤ 0,05 3,5 3,5 3 3 2,5 2,5

2 2 mobil stage/leaf mobil stage/leaf mobil 1,5 1,5 1 1 0,5 0,5 0 0 T - 1 T + 4 T + 7 T + 14 T + 21 T + 28 T + 42 T - 1 T + 4 T + 7 T + 14 T + 21 T + 28 T + 42 Direct counting method Washing methods Direct counting method Washing methods

CALYPSO TREBON

5 5 4,5 4,5 p ≤ 0,05 p ≤ 0,05 4 4 p ≤ 0,01 3,5 3,5

p ≤ 0,05 p ≤ 0,05 p ≤ 0,05 3 3

2,5 2,5

2 2 mobil stage/leaf mobil mobil stage/leaf mobil 1,5 1,5 1 1 0,5 0,5 0 0 T - 1 T + 4 T + 7 T + 14 T + 21 T + 28 T + 42 T - 1 T + 4 T + 7 T + 14 T + 21 T + 28 T + 42 Direct counting method Washing methods Direct counting method Washing methods Fig. 1. Mean density of A. andersoni mobile forms estimated with the two monitoring technique in the side-effects study.

240

Concerning the side effects of agrochemicals (Table 1), the estimated phytoseiid population density did not change whatever the two monitoring techniques used. Both the three tested neonicotinoids Actara (39.54-47.56%), Confidor (17.97-18.03%) and Calypso (26.72-25.88%) and the ICI Cascade (39.48-39.74%) resulted slightly harmful, while a negative effect was more pronounced for the reference Trebon (89.61-86.01%), that should be classified as harmful (T).

Table 1. Side-effects of the pesticides on A. andersoni expressed as % of reduction evaluated by Henderson & Tilton formula. The toxicity class referees to the classification established by IOBC (N=harmless/slightly harmful 0-50%; M= moderately harmful 51-75%; T= harmful >75%).

Toxicity effect Toxicity effect Toxicity Treatment “Direct counting” “Washing method” class Actara 39.54 % 47.56 % N Cascade 39.48 % 39.74 % N Confidor 17.97 % 18.03 % N Calypso 26.72 % 25.88 % N Trebon 89.61 % 86.01 % T

The Table 2 and figure 2 shows the estimated relative density of K. aberrans obtained applying the two sample techniques on leaves of the three varieties Chardonnay, Merlot and Pinot grigio. The estimated phytoseiid population density did not substantially change whatever the technique of monitoring used. Significant differences on the mite populations were observed between the three grape cultivars.

Table 2. Mean population of K. aberrans (mobile forms/leaf) on the cultivars Merlot, Chardonnay and Pinot grigio applying the two monitoring technique.

Varieties Washing method Direct counting

Merlot 2.16 a 2.35 a

Chardonnay 1.31 b 1.21 b

Pinot grigio 0.56 c 0.54 b

F. Tukey 25.206 19.682

Sign. Tukey 0.000 0.000

The figure 3 shows the relationship between the overall density of K. aberrans on the foliage as estimated by the sum of two sampling techniques and the number of phytoseiids sampled by washing method (Shrewsbury & Hardins 2004). The coefficient of correlation (R2) of 0.959 confirms a positive relationship. 241

4 p ≤ 0,05 3,5

2,5

mobil stage/leafmobil 1,5

0,5

0 T 1 T 2 T 3 T 1 T 2 T 3 T 1 T 2 T 3 Pinot grigio Chardonnay Merlot

Direct counting method Washing method

Fig. 2. Mean density of K. aberrans mobile forms on the cultivars Merlot, Chardonnay and Pinot grigio estimated with the two technique “direct counting” and “washing method”.

8 y = 2,108x - 0,13 7 R2 = 0,959

2 Direct counting washing + method

0 00,511,522,533,5 Washing method

Fig. 3. Relationship between the total number of K. aberrans on the foliage and the number of phytoseiids sampled by washing method.

Discussion

It’s undoubted the importance of an accurate assessment of phytoseiid density to achieve a correct side-effects evaluation studies. The results of the present study demonstrate that washing method could be used to estimate the phytoseiid densities on grape leaves. In fact, few cases of statistical difference between the results provided by the two methods whatever 242

the phytoseid species present (A. andersoni and K. aberrans) and grape leaf morphology (Merlot, Chardonnay and Pinot grigio). Using the washing method the phytoseiids were easily washed off, easily distinguish between young and adult stages and well preserved for further study (e.g. classification). Moreover using the direct counting method there is a risk of overestimating the number of phytoseiids, because handling leaves with live mites may increase the latter’s activity and some individuals could be counted more than once. This risk is higher with a high population density of mites and a low leaf surface. Similar hypothesis was supported by Hossain (1992) considering Panonychus ulmi. Furthermore the direct counting is a time-consuming procedure while the washing method allows to reduce at least 70% the assessment time spent in laboratory. In conclusion the washing technique could be considered the better way to handle an extensive evaluation programme on side effects of agrochemicals on Phytoseiids that inhabit the vineyards. Concerning the selectivity levels of the tested insecticides, Actara, Cascade, Confidor and Calypso resulted slightly harmful according the IOBC evaluation classes, compared to the higher toxicity of reference pyretroid Trebon. The Neoinicotinoids and the ICI tested are compatible in IPM viticulture programs, on the contrary Trebon is dangerous. Finally the high differences concerning the Phytoseiid populations checked between the three grape cultivars, due to different leaf hair density, should be opportunely considered in planning side-effects field studies on grape varieties.

References

Angeli G., Ioriatti C. 1994: Susceptibility of two strains of Amblyseius andersoni Chant (Acari: Phytoseiidae) to dithiocarbamate fungicides. – Exp. Appl. Acarol. 16: 669-679. Angeli G., Forti D., Maines R. 1996: Toxicity of a number of pesticides on mortality and reproduction of the predatory mite Amblyseius andersoni Chant (Acarina: Phytoseiidae). – In: Haskell P.T. & McEwen (eds) Proceeding of Welsh pest Management Forum conference: New Studies in Ecotoxicology: 1-4. Angeli G., Forti D., Maines R.1997: Effetti collaterali di fitofarmaci di interesse fruttiviticolo verso gli acari fitoseidi. – Informatore Agrario 14: 74-77. Angeli G., Forti D., Finato S. 2001: Extended laboratory methods to determine effects of plant protection products on two strains of Amblyseius andersoni Chant and their resistance level. – Bulletin OILB/SROP 24 (4): 53-60. Blümel S., et al. 2000: Guidance document to detect side effects of plant protection products on predatory mites (Acari: Phytoseiidae) under field conditions: vineyards and orchards.– In: M.P. Candolfi et al., 2000. Guidelines to evaluate side-effects of plant protection products to non-target arthropods, IOBC, BART and EPPO Joint Initiative: 145-153. Boller E. 1984: Eine einfache Ausschwemm-Methode zur schnellen Erfassung von Raub- milben, Thrips und anderen Kleinarthropoden im Weinbau. – Schweiz. Zeitschrift für Obst- und Weinbau 126: 16-17. Boller E., Vogt H., Ternes P., Malavolta C., 2005: Working document on selectivity of pesticides. Candolfi M., Bigler F., Campbell P., Heimbach U., Schmuck R., Angeli G., Bakker F., Brown K., Carli G., Dinter A., Forti D., Forster R., Gathmann A., Hassan S., Mead-Briggs M., Melandri M., Neumann P., Pasqualini E., Powell W., Reboulet J-N., Romijn K., Sechser B., Thieme Th., Ufer A., Vergnet Ch., Vogt H. 2000: Principles for regulatory testing 243

and interpretation of semi-field and field studies with non-target arthropods. – Anzeiger für Schädlingskunde / Journal of Pest Science 73(6): 141-147. Castagnoli M., Angeli G., Liguori M., Forti D., Simoni S. 2002: Side effects of Biopiren, Piresan, and Neemazal on predatory mite Amblyseius andersoni (Chant). – Anzeiger für Schädlingskunde / Journal of Pest Science 75 (5): 122-127. Duso C., Angeli G., Castagnoli M., Liguori M., Facchin P., Malagnini V., Pozzebon A. 2006: Pesticides and phytoseiid mites: a synopsis of research carried out on grapes, apples and vegetables in northern and central Italy. – Acarology XI, 113-126 [Proceedings of the 11th International Congress of Acarology, Merida, Mexico.]. Shrewsbury P., Hardins M.R. 2004: Beat sampling accuracy in estimating spruce spider mite (Acari: Tetranychidae) populations and injury on juniper. – J. Econ. Entomol. 97(4): 1444-1449. Hossain Sk. M. 1992: Comparison of sampling techniques for the eropean red mite, Panonychus ulmi (Koch) (Acari: Tetranychidae) and the apple rust mite, Aculus schlechtendali (Nalepa) (Acari: Eriophydae). – Acta Agric. Scand., Sect. B, Soil and Plant Sci. 42: 128-132. Zacharda M., Pultar O., Muska J. 1988: Whashing technique for monitoring mites in apple. – Exp. Appl. Acarol. 5: 181-183. 244

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 245-249

Olfactory cells responding to the main pheromone component and plant volatiles in Lobesia botrana (Den. & Schiff.): possible effects on monitoring systems*

De Cristofaro A.1, Vitagliano S.1, Anfora G.2,3, Germinara G.S.1, Tasin M.2,3, Lucchi A.4, Ioriatti C.3, Rotundo G.1 1 Dept. of Animal, Plant and Environmental Sciences, University of Molise, Via De Sanctis, 86100 Campobasso, Italy, [email protected]; 2 SafeCrop Centre, Via E. Mach 1, 38010 San Michele a/A, Italy; 3 IASMA Research Centre, Plant Protection Department, Via E. Mach 1, 38010 S. Michele a/A, Italy; 4 Dept. C.D.S.L., Section Entomologia agraria, University of Pisa, Via San Michele degli Scalzi 2, 56124 Pisa, Italy

Abstract: During the last years electrophysiological and behavioural studies have been carried out in order to establish the role of plant volatile compounds in the host-finding process and oviposition site selection by L. botrana. In previous papers, it has been shown that several compounds are able to stimulate the antennae of both sexes (virgin and mated) of the moth. Results of wind-tunnel studies showed that olfactory cues are greatly involved in the location of host plant and oviposition site. Since olfactory cells sensitive to pheromone components and plant volatiles have been recently found in various insect species, in the present study we applied a single cell recording (SCR) technique (surface contact) to find out olfactory neurons stimulated by the two categories of compounds. The differential saturation electroantennographic technique (DS-EAG) was employed to evaluate the possible decrease of the antennal sensitivity to a plant volatile when the antenna is continuously exposed to E7Z9-12:Ac and, alternatively, to E7Z9-12:Ac when a plant compound is supplied. A large variety of cellular types emerged, from the specific (relatively to the tested compounds) to the highly generalist ones. DS-EAG results showed that cells responding to E7Z9-12:Ac and at least to one plant volatile are widely represented in L. botrana and diverse substances induce a different reduction of the antennal sensitivity to E7Z9-12:Ac. The finding of these cells supports the observations reported by various authors about the ability of plant compounds to enhance or reduce the biological activity of a pheromone component. These “peripheral interferences” in odour perception need to be evaluated when setting up new blends for monitoring purposes.

Key words: kairomones, Vitis vinifera, olfactory cells, SCR, DS-EAG

Introduction

The main component of the sex pheromone of the European Grapevine Moth, Lobesia botrana (Den. & Schiff.) (Lepidoptera Tortricidae), (E,Z)-7,9-dodecadienyl acetate (E7Z9- 12:Ac), is widely used in monitoring systems and control strategies in the European vineyards. During the last years electrophysiological and behavioural studies have been carried out in order to establish the role of plant volatile compounds in the host-finding process and oviposition site selection (Tasin et al., 2005). In previous papers, using electrophysiological techniques (EAG, GC-EAD) it has been shown that several compounds are able to stimulate the antennae of both sexes (virgin and mated) of L. botrana. EAG dose- response curves were calculated for various plant compounds both in females (hexanal, (E,E)- 2,4-decadienal, hexanoic acid, heptanal, 1-octen-3-ol) and males (hexanal, (E,E)-2,4- decadienal, hexanoic acid, (E,E)-2,4-nonadienal, E2-nonenal) (Vitagliano et al., 2005).

245 246

Results of wind-tunnel studies demonstrated that olfactory cues are greatly involved in the location of host plant and oviposition site (Tasin et al., 2006). Since olfactory cells sensitive to pheromone components and plant volatiles have been found in various insect species, like Cydia pomonella (L.) (Lepidoptera Tortricidae) (De Cristofaro et al., 2004), in the present study we applied a single cell recording (SCR) technique (surface contact) to find out olfactory neurons stimulated by the two categories of compounds in L. botrana males and females. Using the most EAG-active doses previously found (Vitagliano et al., 2005), the differential saturation electroantennographic technique (DS-EAG) was employed to evaluate the possible decrease of the antennal sensitivity to plant volatiles when the antenna is exposed to E7Z9-12:Ac and, alternatively, to E7Z9-12:Ac when a plant compound is supplied.

Material and methods

Insects Moths were obtained from a colony of L. botrana reared on a semi-artificial diet at IASMA. Insects were reared under a L17:D7 photoperiod, 60±10 R.H. and 22°C. Pupae were separated by sex and isolated; the emerging adults were kept in plastic Petri dishes provided with a 10% sucrose solution. Electrophysiological analyses The SCR (surface contact) and the DS-EAG techniques have been already described (De Cristofaro et al., 2004). During SCR, spikes from individual cells were recorded by pressing the tip (i.d. < 2 µm) of the different glass electrode against the cuticle of the antenna; recordings were made from the rostral-ventral scale-less area of the antenna, at the distal end of a medial segment (28th-32nd), and the preparation was continuously flushed with charcoal- filtered, humidified air at room temperature (23°C).

Table 1. Action potential frequency (spikes/s ± SD) recorded by olfactory neurons (n=30) of L. botrana male antennae on stimulation with E7Z9-12:Ac and plant volatile compounds. Stimulus: 10 µl of a compound solution (0.1 µg/µl) in mineral oil. * Significant increase (Student t-test; P=0.01) of the spike frequency compared to the resting activity of the neuron.

spikes/s Compound ABCD E (E,Z)-7,9-12:Ac 26.0 ± 8.0 * 16.3 ± 4.7 24.3 ± 6.3 41.3 ± 10.3 * 41.8 ± 12.6 * hexanal 11.0 ± 3.0 39.3 ± 12.7 * 24.7 ± 8.7 43.7 ± 8.7 * 63.2 ± 14.0 * heptanal 10.5 ± 3.5 45.7 ± 14.7 * 24.7 ± 8.3 37.3 ± 6.3 * 52.4 ± 11.6 * E2,E4-nonadienal 11.0 ± 4.0 47.7 ± 13.3 * 23.3 ± 7.3 41.7 ± 7.3 * 46.6 ± 9.8 * E2,E4-decadienal 12.0 ± 3.0 46.3 ± 11.3 * 23.3 ± 6.7 36.3 ± 6.7 * 52.4 ± 11.8 * Z3-hexen-1-ol 13.5 ± 4.5 14.7 ± 3.7 51.3 ± 9.3 * 11.7 ± 2.7 46.0 ± 11.3 * 1-octen-3-ol 13.0 ± 2.0 15.3 ± 4.3 57.3 ± 9.7 * 15.3 ± 4.3 32.4 ± 6.6 * Mineral oil 11.5 ± 4.5 13.7 ± 3.3 25.3 ± 4.0 16.3 ± 4.3 16.6 ± 5.3 Resting activity 12.0 ± 3.0 14.3 ± 2.7 24.7 ± 6.3 16.7 ± 3.3 16.6 ± 6.6

Cell type (%) 7 33 30 10 20 Spike amplitude (mV) 1,3 0,7 1,4 1,1 0,6

247

During the DS-EAG measurements the continuous air stream originated from a bottle containing a mineral oil solution (1 µg/µl) of a chemical (E7Z9-12:Ac or hexanal).

Table 2. Action potential frequency (spikes/s ± SD) recorded by olfactory neurons (n=30) of L. botrana female antennae on stimulation with E7Z9-12:Ac and plant volatile compounds. Stimulus: 10 µl of a compound solution (0.1 µg/µl) in mineral oil. * Significant increase (Student t-test; P=0.01) of the spike frequency compared to the resting activity of the neuron. spikes/s Compound I II III IV V (E,Z)-7,9-12:Ac 42.8 ± 8.6 * 17.0 ± 4.0 29 41.3 ± 10.3 * 52.7 ± 9.7 * hexanal 15.8 ± 4.6 46.0 ± 11.0 * 25 51.3 ± 11.7 * 67.7 ± 12.7 * heptanal 16.6 ± 4.3 48.5 ± 12.5 * 24 48.7 ± 12.3 * 52.7 ± 6.3 * E2,E4-nonadienal 15.6 ± 3.8 52.0 ± 8.0 * 24 46.3 ± 11.7 * 48.3 ± 8.7 * E2,E4-decadienal 13.8 ± 5.3 44.5 ± 13.5 * 26 45.3 ± 10.3 * 44.3 ± 7.7 * Z3-hexen-1-ol 17.3 ± 4.3 18.0 ± 4.0 60 13.3 ± 4.7 51.3 ± 10.7 * 1-octen-3-ol 15.3 ± 3.8 18.0 ± 4.0 68 14.7 ± 4.9 42.3 ± 7.7 * Mineral oil 15.8 ± 4.6 16.0 ± 3.0 28 12.3 ± 4.7 18.3 ± 8.3 Resting activity 16.6 ± 4.3 18.5 ± 3.5 24 12.7 ± 2.3 18.6 ± 7.3 Cell type (%) 50 7 3 10 30 Spike amplitude (mV) 1,70 0,7 1,2 0,9 0,6

Results and discussion

Stimulation of single olfactory neurons with different compounds allowed to distinguish various cell types (Tables 1, 2) on the antennae of both sexes, with different percentages. As expected, in males (Table 1) specialist cells (I) sensitive only to the main component of the sex pheromone were observed in a high percentage (50%). Other olfactory neurons (IV, V; 40%) sensitive to plant volatile compounds were also stimulated by the pheromone component. A cell type (V; 30%) was highly generalist and stimulated by all tested compounds. Olfactory neurons (II, III) sensitive only to plant compounds were identified in a low percentage (10%). In females (Table 2), cells (A) sensitive only to the main component of the sex pheromone were also observed but in a very low percentage (7%). Olfactory neurons (B, C) stimulated only by plant volatile compounds are more present (63%) than in males. Cells (D, E; 30%) sensitive to the main pheromone component and plant compounds were found. Like in males, a cell type (E; 20%) stimulated by all tested compounds was detected. In this paper, information about the different neurons and their relative abundance are limited to: the more frequent cell types; the compounds tested; the recording sites of the action potentials, where trichoid and auricillic sensilla are present. A higher variety of cell types, in this and other antennal areas, is expected. A rough estimation of the relative weight of the olfactory neurons sensitive to the main component of the sex pheromone and to plant volatiles on the magnitude of the antennal response was given by DS-EAG tests (Figures 1, 2). A highly significant reduction of the EAG response to all tested compounds was recorded by the male antennae previously saturated with E7Z9-12:Ac or hexanal. The exposure to E7Z9-12:Ac induced a general decrease of the response to the other compounds 248

and the residual electrophysiological activity was quite similar (Figure 1). The exposure to hexanal drastically reduced the EAG response to the plant compounds; a relatively high residual EAG response was elicited on stimulation with E7Z9-12:Ac, probably due to the large number of specialist neurons sensitive to the main pheromone component.

without saturation exposed to E7,Z9-12:Ac -4,500 * * * * -4,000

-3,500 **

-3,000

-2,500 * * **

-2,000 * -1,500 EAG response (mV)

-1,000

-0,500

0,000 hexanal octanal E2-nonenal E2,E4-nonadienal E2,E4-decadienal hexanoic acid

Fig. 1. EAG response (mV±SD) of virgin male antennae (n=10) of L. botrana to plant volatile compounds (25 µl of a 100 µg/µl solution in mineral oil.) under saturation with E7Z9-12:Ac. Significant differences: * P=0.05; ** P=0.01 (Student t-test).

without saturation exposed to hexanal -5,000 * * -4,500 * * -4,000

-3,500 * *

-3,000

-2,500 * * * * -2,000 * * -1,500 EAG response (mV) -1,000

-0,500

0,000 E7,Z9-12:Ac octanal E2-nonenal E2,E4-nonadienal E2,E4-decadienal hexanoic acid

Fig. 2. EAG response (mV±SD) of virgin male antennae (n=10) of L. botrana to plant volatile compounds and E7Z9-12:Ac (25 µl of a 100 µg/µl solution in mineral oil.) under saturation with hexanal. Significant differences: * P=0.05; ** P=0.01 (Student t-test). 249

In addition to the well known integration of the different neural signals occurring at the central nervous system level, the finding of cells responding to plant volatiles and pheromone compounds supply a new basis for a better understanding of the reported ability of plant compounds to enhance or reduce the biological activity of a pheromone component. In the related moth, Cydia pomonella (L.) (Lepidoptera Tortricidae) (Light et al., 1993; Yang et al., 2004) and other insect species (Dickens et al., 1990) host-plant volatiles can synergize the attractant power of the synthetic pheromone. The interactions in odour perception at a peripheral level need to be carefully considered when setting up new blends for monitoring purposes; how and how much the behaviour of the moth and the capture efficiency of a trap baited with pheromone components and plant volatile compounds can be affected by the simultaneous exposure to substances of different origin interacting on the same olfactory neurons is still unpredictable.

References

De Cristofaro, A., Ioriatti, C., Pasqualini, E., Anfora, G., Germinara, G.S., Villa, M. & Rotundo, G., 2004: Electrophysiological responses of Cydia pomonella (L.) to codle- mone and pear ester ethyl (E,Z)-2,4-decadienoate: peripheral interactions in their perception and evidences for cells responding to both compounds. – Bulletin of Insectology 57 (2): 137-144. Dickens, J.C., Jang, E.B., Light, D.M. & Alford, A.R., 1990: Enhancement of insect pheromone response by green leaf volatiles. – Naturwisserschaften 77: 29-31. Light, D.M., Flath, R. A., Buttery, R.G., Zalom, F.G., Rice, R. E., Dickens, J.C. & Jang E.B., 1993: Host-plant green leaf volatiles synergize the synthetic sex pheromones of the corn earworm and codling moth (Lepidoptera). – Chemoecol. 4: 145-152. Tasin, M., Anfora, G., Ioriatti, C., Carlin, S., De Cristofaro, A., Schmidt, S., Bengtsson, M., Versini, G. & Witzgall, P., 2005: Antennal and behavioural responses of grapevine moth Lobesia botrana females to volatiles from grapevine. – J. Chem. Ecol. 31 (1): 77-87. Tasin, M., Bäckman, A-K., Bengtsson, M., Varela, N., Ioriatti, C. & Witzgall, P., 2006: Wind tunnel attraction of grapevine moth females, Lobesia botrana, to natural and artificial grape odour. – Chemoecol. 16: 87-92. Vitagliano, S., Anfora, G., Tasin, M., Germinara, G.S., Ioriatti, C., Rotundo, G. & De Cristofaro, A., 2005: Electrophysiological and olfactory responses of Lobesia botrana (Den. & Schiff.) (Lepidoptera Tortricidae) to odours of host plant. – IOBC wprs Bulletin 28 (7): 429-435. Yang, Z., Bengtsson, M. & Witzgall, P., 2004: Host plant volatiles synergize response to sex pheromone in codling moth, Cydia pomonella. – J. Chem. Ecol. 30 (3): 619-629.

* This study was supported by MIUR (PRIN 2006).

250 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 251-254

Notes on the distribution and the phenology of Erasmoneura vulnerata (Fitch) (Homoptera: Cicadellidae) in North-Eastern Italy

Carlo Duso, Renzo Moret, Gaia Marchegiani, Alberto Pozzebon Department of Environmental Agronomy and Crop Science - University of Padua - Viale dell’Università, 16 – 35020, Legnaro (PD), Italy

Abstract: The nearctic leafhopper Erasmoneura vulnerata (Fitch) was recorded for the first time on grapes in Europe (Italy, Veneto region) in 2004. Since then, it has been detected in four regions in North-eastern Italy. The phenology of this pest in North-eastern Italy was investigated in three subsequent years on unsprayed Vitis labrusca vines. Overwintered adults colonised vines in spring. Nymphs occurred from late May to late October. E. vulnerata densities increased progressively in mid and late summer. E. vulnerata was rare in commercial vineyards probably because of pesticide use.

Keywords: Leafhoppers, grapes, seasonal abundance, distribution

Introduction

The leafhopper Erasmoneura vulnerata (Fitch) has been reported as being associated with wild and cultivated grapes and other plant species in the USA, Canada and Mexico (McAtee, 1920; Beamer, 1932; Johnson, 1935; Metcalf, 1968; Dietrich and Dmitriev, 2006). In early literature, it was considered a serious grape pest (e.g. Robinson, 1926; Beamer, 1946) but recent studies suggest a low incidence of E. vulnerata in North-American vineyards (Martinson and Dennehy, 1995; Zimmerman et al., 1996; Paxton and Thorvilson, 1996). These discrepancies may be explained by controversial events related to the taxonomy of E. vulnerata.The life history of E. vulnerata in North America has been investigated in Western Colorado vineyards comprising European varieties and French hybrids, also colonised by Erythroneura ziczac (Walsh) (Zimmerman et al., 1996). Overwintered adults of E. vulnerata colonised vineyards in mid-May. Nymphs and adults were observed primarily on the upper leaf surfaces. Nymph densities peaked in June on most cultivars. An additional peak was detected in mid-summer on untreated cultivars, suggesting that E. vulnerata could complete two generations. In these vineyards E. ziczac was dominant over E. vulnerata. Zimmerman et al. (1996) mention three factors affecting the latter species: low adaptation to dry climate and high altitude, higher pressure from predators, low susceptibility to the pest by local cultivars. The ovipositional behaviour of E. vulnerata has been studied in Texas. Females oviposited in the vascular bundles on the leaf midrib in contrast with other leafhoppers occurring on grapes (Paxton and Thorvilson, 1996). This different ovipositional behaviour may be used as a criterion to identify leafhopper species occurring in vineyards. E. vulnerata nymphs and adults feed on leaf mesophyll. Feeding sites appear as pale speckled areas, which can involve the entire leaf. Active feeding is associated with black excrements. Severe symptoms are represented by curled and scorched leaves, and sometimes by premature leaf fall (Zimmerman et al., 1996). A different susceptibility among grape cultivars to E. vulnerata has been reported (Zimmerman et al., 1996; Martinson and Dennehy, 1995). The ovipositional preference of

251 252

leafhopper species may be affected by grape cultivars, in particular by phenotypic differences among them (e.g. leaf pubescence) (Martinson and Dennehy, 1995). In 2004, E. vulnerata was detected for the first time in Europe, in particular in North- eastern Italy (Duso et al., 2005). Its distribution appeared to be restricted to some areas of the Veneto (Treviso, Padova, Vicenza, Venezia) and the Friuli-Venezia Giulia (Pordenone) regions. The first record of E. vulnerata was related to isolated plants of Cabernet Sauvignon, but later, significant population densities were recorded on unsprayed plants of V. labrusca (cv. Isabella) and V. riparia x V. labrusca (cv. Noah). Since other grape pests, which are unimportant in North America (e.g. the planthopper Metcalfa pruinosa Say) have caused significant damage following their introduction to Europe (Duso, 1987), a study regarding the distribution and the phenology of E. vulnerata in Italy was planned. The first results are reported herein.

Materials and methods

The distribution of E. vulnerata was estimated by analysing leaf samples collected from sprayed and unsprayed vines in Northern Italy. Additional information by Italian and foreign entomologists has been included in this paper. The phenology of E. vulnerata was investigated considering a number of selected unsprayed V. labrusca vines growing in a garden close to a commercial vineyard. Observations were carried out on 40-50 leaves every 1-2 weeks (from May to October). In spring, the leaves were inspected by a magnifier to reduce the impact of leaf removing. Later, leaf samples were examined in the laboratory using a dissecting microscope. The number of E. vulnerata (and of co-occurring leafhoppers) active stages was recorded for each leaf. Yellow sticky traps were placed from May to October to assess the fluctuations of adult leafhoppers more easily. The traps were replaced every 1-2 weeks.

Results and discussion

Investigations on E. vulnerata distribution The distribution of E. vulnerata in Europe seems be limited to North-Italian regions. In 2006 and 2007 the pest was found throughout the Veneto and Friuli-Venezia Giulia regions, with the exception of Trieste. Rare specimens were found in an area situated 10 km from the border with Slovenia (F. Pavan, pers. comm.), but attempts to find E. vulnerata in Slovenia have been unsuccessful (G. Seljak, pers. comm.). In 2006, E. vulnerata was detected in the Trentino-Alto Adige region where its distribution was restricted to southern Trentino, the northernmost record being 170 km from the border with Austria. In late 2007, the pest was also recorded in Emilia-Romagna (P. Tirello, pers. comm.). The records were related mainly to unsprayed vines, in particular belonging to V. labrusca and French hybrids. A few individuals have been detected in commercial vineyards where other leafhoppers, e.g. Empoasca vitis Göthe and Zygina rhamni Ferrari, were dominant. Nymphs and adults of E. vulnerata were frequently detected on Parthenocissus spp. in accordance with data from North-America (Connecticut Experimental Station, 2004).

Notes on E. vulnerata phenology Observations on V. labrusca unsprayed vines revealed that E. vulnerata was frequently dominant over other leafhoppers, in particular E. vitis. Overwintered E. vulnerata adults were found on leaves from May onwards. The first nymphs were recorded from late May to early June depending on the season. In 2005, nymph numbers increased in late June. A peak of 253

adults was recorded on traps two weeks later. The dynamics of nymphs and adults in August and September suggest the development of two additional generations. One year later, nymph densities peaked in late June, early August and early September confirming the tendency seen in the previous season. During 2007, a few nymphs were detected in the second half of May but their numbers increased from late June to early July. Nymph peaks were noticed again in mid-August and mid-September.

Discussion

The distribution of E. vulnerata in North-eastern Italy has increased in the last two years probably due to the availability of unsprayed vines and ornamentals (e.g. Parthenocissus). The pest has been recorded in two new regions and several new areas with respect to 2005 data. In contrast, a few specimens of E. vulnerata were detected in commercial vineyards, even when fungicides were applied only. In North-eastern Italy most vineyards are treated with insecticides in order to control E. vitis and Scaphoideus titanus Ball., vector of the “Flavescence dorée” disease (Mori et al., 1999; Girolami et al., 2000). Therefore, pesticide use is an important factor affecting the spread of E. vulnerata in Italy. The data reported in the current study suggest that E. vulnerata can complete three generations per year in North-eastern Italy. Detailed studies are required to confirm this hypothesis. In Colorado, USA, two generations were suggested after observing E. vulnerata peaks on a number of treated and untreated varieties (Zimmerman et al., 1996). In our study, conducted on untreated V. labrusca vines, population densities reached significant values in mid and late summer. In this situation E. vulnerata appeared to compete successfully with E. vitis.

Acknowledgements

This work has been partially supported by a project of Veneto Region (Unità periferica per i servizi fitosanitari di Verona). We thank G. Seljak, F. Pavan and P. Tirello for information on pest distribution, and A. Di Muzio and D. Fornasiero for their support in field studies.

References

Beamer, R.H., 1932: Erythroneura collected on apple with description of a new species. – J. Kansas Entomol. Soc. 5: 62-64. Beamer, R.H., 1946: The Erythroneura of the vulnerata group (Homoptera - Cicadellidae). – J. Kansas Entomol. Soc. 19: 15-22. Connecticut Experimental Station, 2004: Virginia Creeper, Insect problems. – Plant Pest Handbook Homepage [www document]. http://www.caes.state.ct.us/PlantPestHandbookFiles/pphV/pphvirg.htm Dietrich, C.H. & Dmitriev, D.A., 2006: Review of the New World genera of the leafhopper tribe Erythroneurini (Hemiptera: Cicadellidae: Typhlocybinae). – Bull. Illinois Natur. Hist. Surv. 37: 119-190. Duso, C., 1987: A new pest of vine in Europe: Metcalfa pruinosa (Say) (Homoptera: Flatidae). – Proc. EC Experts' Meeting "Integrated pest control in viticulture", Porto- ferraio, 26-28 Sept. 1985, Ed. R. Cavalloro, Balkema/Rotterdam/Brookfield: 103-107. 254

Duso, C., Bressan, A., Mazzon, L., Girolami, V., 2005: First record of the grape leafhopper Erythroneura vulnerata Fitch (Hom., Cicadellidae) in Europe. – J. Appl. Entomol. 129: 170-172. Girolami V., Mori N., Pasini M., Tosi L., 2000: Probabile resistenza della cicalina verde della vite agli insetticidi fosforganici. – Informatore Agr. 15: 85-86. Johnson, D.M., 1935: Leafhoppers of Ohio. Subfamily Typhlocybinae (Homoptera: Cicadell- idae). – Bull. Ohio Biol. Surv. 6: 39-122. McAtee, W.L., 1920: Key to the nearctic species and varieties of Erythroneura (Homoptera; Eupterygidae). – Trans. American Entomol. Soc. 46: 267-321. Martinson, T.E. & Dennehy, T.J., 1995: Varietal preferences of Erythroneura leafhoppers (Homoptera: Cicadellidae) feeding on grapes in New York. – Environ. Entomol. 24: 550- 558. Metcalf, Z.P., 1968: General catalogue of the Homoptera. VI. Cicadelloidea. 17. Cicadellidae. – Washington: US Dep. Agr.: 1513 pp. Mori N., Posenato G., Sancassani G., Tosi L., Girolami V., 1999: Insetticidi per il controllo delle cicaline nei vigneti. – Informatore Agr. 15: 93-97. Paxton, DW & Thorvilson, HG, 1996: Oviposition of three Erythroneura species on grape leaves inWestern Texas. – Southwestern Entomol. 21: 141-144. Robinson, W., 1926: The genus Erythroneura north of Mexico (Homoptera, Cicadellidae). – Sci. Bull. Univ. Kansas. 16: 101-155. Zimmerman, R., Kondratieff, B., Nelson, E., Sclar, C., 1996: The life history of two species of grape leafhoppers on wine grapes in western Colorado. – J. Kansas Entomol. Soc. 69: 337-345. Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 255-258

Effects of irrigation on Empoasca vitis populations

Diego Fornasiero, Filippo Maria Buzzetti, Alberto Pozzebon, Carlo Duso Department of Environmental Agronomy and Crop Science, University of Padua, Agripolis – Viale dell’Università, 16 – 35020 Legnaro (PD) – Italy

Abstract: The effect of irrigation on the incidence of pests was investigated in two vineyards located in north-eastern Italy in 2006. Irrigation was performed 2-3 times during the experimental season by a micro-spray system. The leafhopper Empoasca vitis Göthe was the most frequent pest in the two vineyards. Its seasonal abundance was monitored by examining leaf samples. Empoasca vitis population densities were affected by irrigation, and higher levels of this pest were recorded on irrigated vines in particular.

Key words: Empoasca vitis, grape, irrigation

Introduction

Irrigation is a common practice in several Italian viticultural areas due to the hot and dry summer climatic conditions, in order to obtain adequate levels of production, as well as grape and wine quality (Cifre et al., 2005). Drought stress leads to low yields and shoot growth as well as poor fruit composition (low sugar, low or high pH, impeded nutrient uptake) with possible negative consequences for wine quality (low acid and atypical aging) (Lakso & Pool, 2001). Plant water-stress can affect herbivorous insects, including leafhoppers (Homoptera Cicadellidae). Leafhopper densities can be related to applied water amounts (Trichilo et al., 1990; Daane et al., 1995; Daane & Williams, 2003). Trichilo et al. (1990) reported that lower amounts of applied water and the associated reduction in vine vigor, resulted in decreasing leafhopper (i.e. Erythroneura elegantula Osborn) abundance. E. elegantula body size, nymph mortality and adult fecundity can be related to irrigation levels (Daane et al., 1995). Further studies conducted on Erythroneura variabilis (Beamer) showed that vine vigor can be lowered to reduce leafhopper densities, their fecundity and movements, without crop yield losses (Daane & Williams, 2003). The leafhopper Empoasca vitis Göthe (Homoptera, Cicadellidae) is an important grepevine pest in Europe (e.g. Vidano, 1958; Baggiolini et al., 1968; Moutos & Fos 1973; Cerutti et al., 1988; Baillod et al., 1990). Adults and nymphs feed on vascular tissues, mainly phloem, (Vidano, 1963; Carle & Moutous, 1965; Tavella & Arzone, 1992) causing a reduction in photosynthesis, mesophyll conductance and transpiration rate (Candolfi et al., 1993). Leaf margins become reddish or yellowish, depending on the cultivar, and symptoms progressively invade the leaf lamina. Xylem conduction can be damaged drying the leaf margins. The severity of symptoms is affected by the cultivar, dry conditions and plant stress factors (Vidano, 1963; Pavan et al., 1992). E. vitis females overwinter outside of the vineyards on coniferous and evergreen trees and shrubs (Vidano, 1958; Cerutti et al., 1989). Then, from late April–May, E. vitis colonizes grapevines. In northern Italy, the first generation develops in May-June, and an additional two generations are completed during the vegetative season. The highest economic importance is often related to the activity of the

255 256

second generation (Pavan et al., 1992; Pavan et al., 1988). Infestations can cause a decreased berry weight and sugar content (Pavan et al., 2000). Factors affecting E. vitis outbreaks were studied in Italy, Switzerland and France. The most important concern relationships between E. vitis and egg parasitoids, the management of wild vegetation, and the use of insecticides and fungicides (Vidano, 1963; Cerutti et al., 1989; Arnò et al., 1988; Pavan et al., 1992, 1997; Picotti & Pavan, 1993). The role of cultural practices on E. vitis populations has been not investigated in depth. In the present work, we have evaluated the effect of water management practices on arthropod pests in two commercial vineyards located in north-eastern Italy, with a particular focus on E. vitis, the most frequent pest in the experimental sites.

Material and methods

Experimental design This study was conducted in two vineyards located in the hilly area of the Vicenza province (Colli Berici) during 2006. The vineyards contained Chardonnay and Garganega cultivars and vines were grafted to rootstock SO4. The Chardonnay vineyard had a spacing of 4x1 m (2500 plants per hectare), whereas the Garganega vineyard had a spacing of 4x0,8 m (3100 plants per hectare). The “pergola” system was adopted in both. The control of Grape downy mildew and Grape powdery mildew required multiple fungicide applications from May to July. Fenitrothion was also sprayed (July 6) in both vineyards to control Lobesia botrana Den. & Schiff. and Scaphoideus titanus (Ball.). For each vineyard, two treatments where vines were irrigated or not irrigated were compared. Each treatment comprised two replications of 25 plants. The vines were watered with the micro-spray irrigation system. In the Chardonnay vineyard, irrigation was applied on July 1st and 15th and the water amount per season was 72.8 mm. In the Garganega vineyard, irrigation was applied three times (June 21st, 30th, and July 22nd) and the water amount per season was 94.2 mm.

Sampling methods and data analysis Grape pest (e.g. leafhoppers, spider mites and grape moths) abundance in vineyards was monitored by examining leaf and bunch samples. In particular, E. vitis populations were monitored from June to September by sampling 50 vines per treatment (one leaf per vine) every 15 days. In order to evaluate the effect of irrigation, the data were analyzed applying a repeated measures ANOVA with the Proc GLM of SAS (SAS Institute, 1999). Means were separated by the REGWQ test, and the significance level was set at a P = 0.05. Prior to the analysis data were log (y + 1) transformed in order to meet ANOVA assumption.

Results and discussion

E. vitis was the most important arthropod pest in both vineyards. However, its populations did not exceed the economic thresholds. In the Garganega vineyard, the first generation of E. vitis was negligible. As to the second generation, nymph populations peaked in the second half of July while adult numbers peaked in the first half of August. Low population densities were recorded in the third generation. Nymph densities were higher on irrigated than on non- irrigated vines (F1, 3= 11.87; p = 0.041). This tendency was confirmed for adults (F1, 3= 50.30; p < 0.01). The first generation of E. vitis was scarce; this was also found in the Chardonnay vineyard. In the second generation, nymph populations peaked in mid July while adults were relatively abundant in the first half of August. The third generation was clearly detected. Nymph numbers peaked in early September, adults increased in late September. Nymph 257

densities were higher on irrigated than on non-irrigated vines (F1, 3= 16.94; p = 0.026) but differences among adults were not significant (F1, 3= 8.34; p = 0.06). Therefore, it seems that water supplied by micro-spray systems positively affects E. vitis populations. Implications of this phenomenon for IPM require further investigations. Our results agree with those of previous studies showing that leafhoppers (E. elegantula and E. variabilis) prefer irrigated and thus vigorously growing vines (Trichilo et al., 1990; Daane et al., 1995; Daane & Williams, 2003). Restricted available water can affect the vine water status and alter the characteristics of the microenvironment (Trichilo et al., 1990). Daane et al. (1995) reported that well-irrigated grapevines had canopy temperatures up to 10°C cooler than deficit irrigated vines. E. elegantula nymph survival was markedly reduced in the latter. Adult leafhoppers were more attracted to and/or better arrested in the well-irrigated vines and also showed a higher reproductive potential. Similar results were obtained with E. variabilis (Daane & Williams, 2003). According to the latter authors, applied irrigation amounts to vines can be manipulated to suppress insect herbivore densities without negatively influencing crop yields.

Acknowledgements

This work was carried out as part of the Aquavitis project (Regione Veneto). This work was carried out during the Doctoral project in Viticulture, Oenology and Marketing supported by the Province of Treviso.

References

Arnò C., Alma A., Arzone A., 1988: Anagrus atomus as egg parasite of Typhlocybinae (Rhynchota Auchenorryncha). – Proc. 6th Auchenorrhyncha Meeting, Turin, Italy, 7-11 September 1987: 611-615. Baggiolini M., Canevascini V., Tencalla Y., Caccia R., Sobrio G., Cavalli S., 1968: La cicadelle verte, Empoasca flavescens F. (Homopt., Typhlocybidae), agent d’altérations foliaires sur vigne. – Rech. Agron. Suisse 7: 43-69. Baillod M., Jermini M., Schmid A., 1990: Essais de nuisibilité de la cicadelle verte, Empoasca vitis Goethe sur le cépage Merlot au Tessin et le cépage Pinot noir en Valais. – IOBC/WPRS Bull. 13 (7): 158-161. Candolfi M.P., Jermini M., Carrera E., Candolfi-Vasconcelos M.C., 1993: Grapevine leaf gas exchange, plant growth, yield, fruit qualità and carbohydrate reserves influenced by the grape leafhopper, Empoasca vitis. – Entomol. Exp. Appl. 69: 289-296. Carle P., Moutous G., 1965: Observations sur le mode de nutrition sur vigne de quatre espèces de cicadelles. – Ann. Epiphyties 16: 333-354. Cerutti F., Baumgärtner J., Delucchi V., 1988: Ricerche sull’ecosistema ‘vigneto’ in Ticino: I. Campionamento delle popolazioni di Empoasca vitis Goethe (Hom., Cicadellidae, Typhlo- cybinae). – Mitt. Schweiz. Entomol. Ges. 61: 29-41. Cerutti F., Delucchi V., Baumgärtner J., Rubli D., 1989: Ricerche sull’ecosistema ‘vigneto’ in Ticino: II. La colonizzazione dei vigneti da parte della cicalina Empoasca vitis Goethe (Hom., Cicadellide, Typhlocybinae) e del suo parassitoide Anagrus atomus Haliday (Hym. Mymaridae), e importanza della flora circostante. – Mitt. Schweiz. Entomol. Ges. 62: 253- 267. 258

Cifre J., Bota J., Escalona J.M., Medrano H., Flexas J., 2005: Physiological tools for irrigation scheduling in grapevine (Vitis Vinifera L.). An open gate to improve water-use efficiency? – Agr. Ecosys. Environ. 106: 159-170. Daane K.M., Williams L.E., Yokota G.Y., Steffan S.A., 1995: Leafhopper prefers vines with greater amounts of irrigation. – Calif. Agric. 49: 28-32. Daane K.M., Williams L.E., 2003: Manipulating vineyard irrigation amounts to reduce insect pest damage. – Ecol. Appl. 13: 1650-1666. Lakso A.N., Pool R.M., 2001: The effects of water stress on vineyards and wine quality in Eastern vineyards. – Wine East 29: 12-20. Moutous G., Fos A., 1973: Influence des niveaux de populations de cicadelles de la vigne (Empoasca flavescens Fab.) sur le symptome de la grillure des feuilles. – Ann. Zool. Ecol. Anim. 5: 173-185. Pavan F., Pavanetto E., Duso C., Girolami V., 1988: Population dynamics of Empoasca vitis (Göethe) and Zygina rhamni (Ferr.) on vines in northern Italy. – In: C. Vidano and A. Arzone (eds.): Proc. 6th Auchenorrhyncha Meeting, Turin, Italy, 7-11 September 1987: 517-524. Pavan F., Picotti P., Girolami V., 1992: Strategie per il controllo di Empoasca vitis Göthe su vite. – Informatore agr. 48: 65-72. Pavan F., Picotti P., Gregoris A., 1997: Studi su Anagrus atomus (Linnaeus) (Hymenoptera, Mymaridae) parassitoide oofago di Empoasca vitis (Göthe) (Homoptera, Cicadellidae) su vite. 3. Influenza di trattamenti insetticidi sulla dinamica di popolazione. – Boll. Lab. Entomol. Agr. “Filippo Silvestri” 53: 103-115. Pavan F., Stefanelli G., Villani A., Gasparinetti P., Colussi G., Mucignat D., Del Cont Bernard D., Mutton P., 2000: Danni da Empoasca vitis (Göthe) (Homoptera: Cicadellidae) in vigneti dell’Italia nord-orientale e soglie d’intervento. – Frustula entomol. 21: 109-124. Picotti P., Pavan F., 1993: Studi su Anagrus atomus (Linnaeus) (Hymenoptera, Mymaridae) parassitoide oofago di Empoasca vitis (Göthe) (Homoptera, Cicadellidae) su vite. 1. Dinamica di popolazione in assenza di trattamenti insetticidi. – Boll. Lab. Entomol. Agr. “Filippo Silvestri” 48: 105-115. SAS Institute Inc., 1999: SAS/STAT User’s Guide ver. 8. – SAS Institute Inc., Cary, NC. Tavella L., Arzone A., 1992: Aspetti nutrizionali in Zyginidia pullula (Boheman), Empoasca vitis (Göethe) e Graphocephala fennahi Young (Homoptera Auchenorrhyncha). – Boll. Zoll. agr. Bachic. 24: 137-146. Trichilo P.J., Wilson L.T., Grimes D.W., 1990: Influence of irrigation management on the abundance of leafhoppers (Homoptera: Cicadellidae) on grapes. – Environ. Entomol. 19: 1803-1809. Vidano C., 1958: Le cicaline italiane della vite. Hemiptera Typhlocibinae. – Boll. Zool. agr. Bachic. 1: 61-115. Vidano C., 1963: Alterazioni provocate da insetti in Vitis osservate, sperimentate e comparate. – Ann. Fac. Sci. agr. Univ. Torino 1: 513-644. Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 259-265

Simulation of Lobesia-botrana-egg-laying for autecological and insecticide studies

Christoph Hoffmann JKI, Institute for Plant Protection in Fruit Crops and Viticulture, Brüningstr. 84, 54470 Bernkastel-Kues, Germany. e-mail: [email protected]

Abstract: The rearing facilities of Lobesia botrana were modified in a way that moths laid their eggs on polyethylene sticks which were used for repeated and continuous artificial egg infestations of grapes in vineyards. In the context of a randomized insecticide field trial in a Riesling vineyard in the German vinegrowing region Mosel, inoculations were carried out up to three times a week during the vegetation period. Thus a constant egg laying of the Grapevine Moth was simulated. The infestation level varied considerably within the season. During a period of dry and hot weather between June and July of 2006 we found the highest levels of successful inoculations. The development of Lobesia botrana was not effective during the cool-wet periods before and after June and July. In this way survival of Lobesia botrana in the field underlies a high coincidence factor. For one insecticide the effective period was surprisingly long. Three insecticides showed different efficacy between the period of the first and the second generation of the Lobesia botrana field population. A possible reason for this is the changing architecture of the clusters during the season, as the second insecticide treatment was conducted after grape closure.

Key words: Lobesia botrana, Grapevine moth, artificial inoculation, infestation level, efficacy evaluation, control, effect of weather conditions.

Introduction

Measuring field efficacy of insecticides against grape berry moths by randomized field plots requests a homogenous distribution of the insects within a vineyard to get reliable results. In central Europe this distributions apparently occurs only in gradation years of the grape berry moth Lobesia botrana and the grapevine moth Eupoecillia ambiguella. In most years these two species appear in damage clusters, which often are located in depressed areas within grape territories. Thus in those years it is difficult to carry out insecticide or autecological studies on grape damaging moths in any vineyard. Investigating the impact of natural enemies on grape berry moths by artificial inoculations with reared eggs on inoculation cards, Hoffmann & Michl (2003) stated that the damage produced was homogeneous within one vineyard on the one hand but different in time of the inoculation and location on the other hand. Breuer & Huber (2006) advanced and simplified the production of egg cards. Hoffmann (2007) built up a rearing facility which allows continuous production of inoculation cards with eggs of a specific age. This allowed the following investigations upon the durability of insecticide efficacy and on weather effects on the development of the grape berry moth.

259 260

Material and methods

Rearing of Lobesia botrana and the production of egg stripes The rearing of the moths was done on an artificial diet according to MANI et al. (1978). Pupae where put into a paperbag with coarse surface, which was pulled through the round opening of a bunsen burner tripod and folded downwards. Thus the diameter of paper bag opening was of a certain diameter. Plexiglas tubes which fitted exactly in this opening where lined with white felt to prevent the moths from laying their eggs on the plexiglass itself. The lined plexiglass tubes where charged with round magazines of corrugated paper of 2.5 cm hight. The magazines could be charged with up to 60 polyethylen stripes of 10 cm lenghth. The stripes called “twisties” (Firm Genpack, Værløse, Denmark) consisted of a central wire shielded by a flat polyethylene lamella. Arranged in the magazine the stripes formed a kind of polyethylene curtain which coated the felt within the plexiglass tube. The moths preferred to lay their eggs on this sticks than on the felt and on the coarse paper. The top cover of the tube was also made by plexiglass and furnished with a hole that allowed holding a plastic tube. For the water supply of the moths the tube was filled with water and closed with a cotton plug. Changing the plexiglass tuber allows the production of stripes with eggs of a certain age.

Lobesia rearing Egg-laying equipment Egg-laying container lined with felt, charged with egg stripes Cluster 1 month after inoculation Æ Lab Æ dissection and counting of nests

Stripe charged with Lobesia eggs

Fig. 1: Rearing of Lobesia botrana egg stripe production and inoculation

Design of the field trial and spraying equipment In a Riesling vineyard near Bernkastel-Kues (Moselle–valley) RAK 1+2 was used to suppress wild populations of Lobesia botrana and Eupoecilia ambiguella. Using a caterpilllar recycling spayer (Firm Schachtner) the rows of the vineyard where treated with different insecticides (Table 1). In each case four randomly arranged rows where treated with the same insecticide. In the year 2007 there where additional trials with a product not registered for Lobesia control: 261

Neem Azal TS with the active ingredients Azadirachtin A+B. The timing of the spraying was achieved by using a BASF delta trap with Lobesia-female-Pheromone in a vineyard 500 m away from the mating disruption plot. Treatments where carried out 7-10 days after flight peak.

Table 1: Insecticides used in the 2006 Lobesia botrana field trial Product Active ingredient Mode of action Runner ® Methoxyfenozide molting enhancer, acts through contact and ingestion by larvae, ovicide action, active ingredient transported translaminarly within the plant. Mimic® Tebufenozid molting enhancer, acts through ingestion by larvae Steward® Indoxacarb Action on nervous system of insects, acts through contact and ingestion by larvae. Dipel ES® Bacillus thuringiensis- Paralysis of the intestine, acts through ingestion by toxin larvae.

The inoculations with the egg stripes where carried out up to three times a week during the vegetation period using the direction of the columns in a right angle to the direction of the insecticide applications. In one column each grape plant was inoculated with three egg stripes by fixing the stripes in the inflorescences, flowers or clusters (12 clusters per date and variant). Inoculations started at one end of the vineyard and where continued date by date until the other side was reached (see Fig 2). After three to four weeks the inoculated inflorescences, flowers and clusters where brought to the lab and where dissected. The number of nests or feeding sites where counted.

Fig. 2: Insecticide treatment- and inoculation plan of a Riesling vineyard near Bernkastel-Kues 2006.

262

Correlation with weather data An automatic weather station was used to record weather data, which was located in a distance of 500 m from the trial plot. As the measurement frequency was high (each 5 min) all kind of possible correlations could be tested. In this study we only took into account dailymeans of temperature and relative humidity. For rain the daily sums were taken into account.

Results and discussion

Is there a correlation between wild population phenology and the infestation levels achieved? Male flight in wild population was used as a measure for Lobesia botrana’s phenology. The infestation levels which resulted from the artificial egg-inoculation might represent the survival conditions of wild Lobesia botrana populations. Thus, the correlation of both parameters was tested. According to Fig. 3 there is no correlation between the male flight and the infestation levels after artificial egg inoculation in 2006.

700 120 600 100 500 80 400 60 300 40 200

20 100 infestation level [%] 0 0 No. of malemoths in BASF trap 1.5 31.5 30.6 30.7 29.8 28.9 male flight Lobesia Infestation level after one month egg infestation evaluated egg infestation not evaluated

Fig. 3 Bernkastel-Kues 2006: Infestation level in the untreated control after artificial egg inoculations in a Riesling vineyard versus male flight measured in a nearby vineyard with a BASF pheromone trap

Effect of weather conditions on the survival of the eggs and instars The relationship between weather conditions (temperature, precipitation) and the infestation levels, that resulted from artificial inoculations in the untreated control, is shown in Fig. 4. During the cool and wet conditions of May and August there was almost no development of the eggs. In the months of June and July, when temperature was high and precipitation was low, infestation levels where generally high (up to 600 %). Evaluation of the efficacy of insecticides The infestation levels, evaluated three to four weeks after the artificial inoculation for the four different insecticide applications and the untreated control, are shown in Fig. 5. The inoculations around the first insecticide treatment (29.05.2006) were not effective at all. One and a half week later Dipel ES® performed like an untreated control, whereas on the grapes treated with Runner® the infestation levels stayed constantly low compared to the untreated 263

check. Mimic® and Steward® ranged in the middle. There was an enduring efficacy of Runner® until the next treatment in the mid of July. That means that the efficacy of the product during this dry and hot summer persisted about six weeks. The inoculations around the second treatment showed different results for Mimic, Steward and Dipel ES. Dipel ES showed a performance similar or even better than Mimic and Steward (see Fig. 6).

700 30

] 600 25 500 20 400 15 300 10 200 infestation level [% level infestation 100 5

0 0 precipitation [mm] Temp [°C]; 1.5 9.8 precipitation 21.5 10.6 30.6 20.7 infestation29.8 level18.9 egg infestation evaluated egg infestation not evaluated Temperature

Fig. 4: Bernkastel-Kues 2006: Infestation level in the untreated control after artificial egg inoculations in a Riesling vineyard versus temperature (as dayly means) and precipitation (dayly sums).

700

600

500

400

300

Befallsstärke [%] [%] Befallsstärke 200 Infestation level [%] 100

6 6 6 6 6 6 6 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2 .07. .07. .08. .08. 4 9 3 8 15.05. 30.05. 14.06. 29.06. 1 2 1 2

ControlKontrolle Steward Mimic Runner Biologische Bundesanstalt für Land- und Forstwirtschaft Dipel ES Institut für PflanzenschutzInfestationAnsiedlung im Weinbau InsecticideBehandlung treatm. NotNicht to ausgewertetevaluate

Fig. 5: Bernkastel-Kues 2006: Infestation levels in four different insecticide variants three to four weeks after inoculation with Lobesia botrana eggs. The dates of the insecticide applications are indicated with arrows and diamonds. 264

Changing performance of insecticides during changing cluster architecture The observations led to the educated guess, that the changing performance of the insecticides during the vine growing season could be related to the changing cluster architecture of the grape. The absolute number of nests found in the different variants before and after grape closure (10.07.2006) is shown in Fig. 6. The Bacillus thuringiensis formulation Dipel ES performs like a control before and like an effective insecticide after grape closure. A shadowing-effect inside the closed clusters, is keeping away UV-radiation from the centre of the clusters. This could be a reason why in the second generation the UV-unstable Bt-toxin stayed effective for a longer period and why it didn’t do so in the early season. Therefore it would be interesting to compare the effect of Bt-formulations on eggs put on the outer surface and the centre of the clusters at the same time.

350 300 250 Before grape closure 200 150 Treatment 29.05.2006 100 50 0 Kontrolle Dipel ES Steward Mimic Runner

160 140 120 100 80 After grape closure 60 40 Treatment 17.07.2006 20 0 Kontrolle Dipel ES Steward Mimic Runner

No. of No. Lobesia nests in dissected grapes Fig. 6 Bernkastel-Kues 2006: Changing cluster architecture: Sum of Lobesia botrana nests counted before and after grape closure in the different insecticide variants and in the untreated control.

nontreated" ] Neem azal 1200 Runner 1000

800

600

400

200

infestation level [% infestation 7 7 7 07 0 00 007 00 2 20 .2007 2007 2007 2 20 5. 6 .05.2 .06. .0 07. .07. .07. 08. .08.2007 5.05.2007 5.06.2007 3. 7. 1 22 29.0 0 12 19 26.06.20070 10.07.200717 24 31.07.20070 14

265

Fig. 7 Bernkastel-Kues 2007: Infestation levels in two different insecticide variants three to four weeks after inoculation with Lobesia botrana eggs. The dates of the insecticide applications are indicated with arrows. Reproducibility of egg inoculation experiences The trial was repeated in 2007 with only one weekly inoculationand two insecticide variants. Runner®, which was known from 2006 as a product with a long term effect and Neem Azal TS®, a product derived from the Neem tree Azadirachta indica, were included. The possible effect of Neem-Azal TS® on Lobesia botrana was never tested reliably. Fig. 7 shows the infestation levels reached within the trial of 2007. Obviously Neem Azal TS®, applied at 5 L/ha had no effect.The fact that the line of the untreated control in the diagram and that of the Neem product are nearly the same indicates that reproducible results can be achieved with the method of artificial egg inoculations. Future prospects Efficacy evaluations become independent from mass outbreaks of Lobesia botrana using the method described in this paper. Different stages of the moth of defined age are treated at the same time. This facilitates the interpretation of field trial results and testing the durability of insecticides under differing field conditions is simplified. Effects of cultural practices on Lobesia survival might be tested too, e.g. defoliation of parts of the canopy. It is known that few individuals of Lobesia botrana can cause a great damage if weather conditions are favourable for this insect. The presence of a great number of Lobesia adults in a vineyard does not cause damage if weather conditions are unfavourable. As the conditions for Lobesia development change daily, long term predictions for future population density can’t be made. The described method may be a tool for a future forecast system as it allows the correlation of weather conditions with developmental rates of the moths. Collecting such kind of data over years may allow in the future a survival prediction based on field experience. Besides topological or micro climatical effects on the development of larvae could be tested as well as the suitability of presumptive alternative host plants for the development of the moths. Hoffmann et al. (2004) used the method for studies on the question how to enhance natural enemies of Lobesia botrana in vineyards by greencover management.

References

Breuer, M. & Huber, B. 2006: Künstlicher Traubenwicklerbefall – eine neue Möglichkeit zur Prüfung von Insektiziden im Weinbau. – Mitt. Dtsch. Ges. allg. angew. Ent. 15: 259- 262. Hoffmann, C. 2007: Simulation der Eiablage als Werkzeug zur Freilandprüfung von Insekti- ziden im Weinbau und zur Erforschung der Autökologie des Bekreuzten Trauben- wicklers Lobesia botrana Den. und Schiff. (Lepidoptera: Tortricidae). – Nachrichtenbl. Deut. Pflanzenschutzd. 59(6): 133-137. Hoffmann, C.; Doye, E.; Michl, G.; Wüstner, P. 2004: Innovationen des Ökologischen Weinbaus. – Der Badische Winzer 10/2004: 21-24. Hoffmann; C. & Michl, G. 2003: Parasitoide von Traubenwicklern – ein Werkzeug der natürlichen Schädlingsregulation?. – Deutsches Weinbaujahrbuch 2004; 55. Jhrg., Verlag Eugen Ulmer: 120-134. Mani, E.; Riggenbach, W., Mendik, M., 1978: Zucht des Apfelwicklers (Laspeyresia pomonella L.) auf künstlichem Nährboden 1976-1978. – Mitteilungen Schweizerische Entomologische Gesellschaft 51: 315-326.

266

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 267-272

Seasonal abundance and distribution of Planococcus ficus on grape vine in Sardinia

Andrea Lentini1, Giuseppe Serra2, Salvatore Ortu1, Gavino Delrio1 1 Dipartimento di Protezione delle Piante, Università di Sassari, Via E. De Nicola, 07100 Sassari, e-mail: [email protected]; 2 CNR, Istituto per lo Studio degli Ecosistemi, sede di Sassari, Via E. De Nicola, 07100 Sassari

Abstract: During 2006-2007, the seasonal spatial distribution of Planococcus ficus throughout the vine was studied in a “Tendone system” trained vineyard, situated in the North of Sardinia (Italy). Male flights were monitored by pheromone-baited traps, whereas eggs, crawlers, nymphs and females abundance was estimated by counting on trunk, canes, spurs, leaves and bunches. Vine mealybug developed throughout all the year in cryptic location under the bark of trunk and arms, overwintering principally as fecundated female, or eggs in ovisacs. In April, the majority of crawlers were found under the bark and, after bud break, also around the base of new shoots. Females of this generation were detected in June, and crawlers of the second generation moved toward basal leaves. Females of this generation peaked at the end of July, and the crawlers of the third generation colonised grape bunches in late July-August. The maximum female density was recorded in October and, after the leaf quality deterioration, the majority of mealybugs moved back under the bark. Males were captured by pheromone traps from May to December and their population density showed three peaks in June, late July and September-October, when the highest female density was recorded.

Key words: vine mealybug, spatial distribution, pheromone trap.

Introduction

The vine mealybug, Planococcus ficus (Signoret), is emerging as the major pest of Sardinian vineyards, as a consequence of cultural technique changes, such as the increasing use of nitrogen fertilizers and irrigation, which creates favourable conditions for the development of the mealybug (Dalla Montà et al., 2001). The control of this species normally relies on summer applications of organophosphate insecticides. However, treatment efficacy is not always fulfilling. In fact, vine mealybug develops on the trunk and arms in cryptic location under the bark and only progressively infests vegetation (Duso, 1989). To be effective, control applications need to be conducted when the majority of mealybug population reaches vegetation and is mostly represented by juvenile stages that are more sensitive to insecticides. Due to the importance of this pest and the lack of information on its biology in Sardinia, observations on spatial distribution and seasonal abundance of different P. ficus stages on the plant were carried out in order to establish also the optimal chemical treatment timing. This paper reports the results of two-year investigations in a vineyard in the North of Sardinia (Alghero (SS), Italy).

Material and methods

Investigations were conducted in a 0.5-ha vineyard of Vermentino cultivar trained as “Tendone system” with a 2-m spacing among plants. Annual soil fertilization was based on

267 268

600 kg complete fertilizer (15-10-15) per hectare, whereas no artificial irrigation was conducted because annual water requirement was already met by superficial groundwater availability. Grape harvest was complete at the end of August in 2006 and at the end of September in 2007. Observations on P. ficus population started at the beginning of May 2006 and continued uninterruptedly until the end of November 2007. For each sampling date, mealybug numbers and stages on trunk, arms, spurs, canes and shoots of one randomly chosen infested plant were recorded. Within each shoot a more detailed distinction included: the shoot base, main and lateral leaves in different nodes, and bunches. Bark was removed to permit the accurate inspection of trunk and arms for bug counting. Sampling, based on the inspection of a different plant for each date, was carried out weekly during the vegetative season and every 2- 3 weeks during winter. Data on population abundance and mealybug distribution on different parts of the plant are presented as monthly means, whereas mealybug age distribution is represented by a moving average based on three consecutive samplings. Male flights were monitored by three bottle traps baited with 0.1 mg sex pheromone of P. ficus (Ortu et al., 2006). Male counts were performed every week but pheromone dispensers were replaced monthly.

Results and discussion

Mealybug abundance in 2007 was lower than 2006, probably because of adverse weather conditions, such as drought and high summer temperature. In fact, although population density increased from May-June in both years, the 2006-peak was reached in August, whereas population growth apparently stopped in July 2007 (Fig. 1). Then, population density decreased considerably from November and during winter in both years.

4500 4000 3500 3000 2500 2000 1500 1000

Mealybugs/vine/month 500 0 MJ JASONDGFMAMJ JAS ON 2006 2007

Fig. 1. Mean number of mealybugs on vines.

As a result of male capture by pheromone traps, three peaks corresponding to the maximum density of pre-ovipositing females were recorded. This trend probably indicates the tendency of this insect species to involve three successive generations per year, under the climatic conditions in Northern Sardinia (Fig. 2). Vine mealybugs overwintered on the trunk mainly as fecundated females, or eggs in ovisacs. In March, the percentage of ovipositing females increased, and at the end of April, 269

egg was the most represented stage (Fig. 3). In this period, crawlers of the first generation were also detected, whereas females of this generation reached the peak in June. The second generation grew mostly on vegetation, and the development completed in about 1.5 months. Females of this generation peaked at the end of July, whereas the maximum female density of the third generation was reached in early October. Later, mealybugs moved back under the bark to overwinter.

900 300 800 250 700 600 200 500 150 400 Males/trap

Females/vine 300 100 200 50 100 0 0 MJ JASONDGFMAMJJAS ON 2006 2007

Fig. 2. Mean number of P. ficus males caught in pheromone traps and mean number of pre-ovipositing females recorded on vines.

Vine mealybugs developed throughout all the year in cryptic location under trunk and arms bark (Fig. 4). At the end of April, after bud break, some crawlers were seen around the base of new shoots even if the majority of mealybugs (74% in 2006 and 54% in 2007) were still found under the bark in May. In June, crawlers of the second generation spread on leaves, and in July, mealybug population was mainly detected on vegetation, although a significant percentage (about 30%) was still found under trunk and arms bark. In August, mealybug population reached the highest density on leaves and bunches (90% in 2006 and 80% in 2007). In autumn, after leaf quality deterioration, the majority of mealybugs moved back under trunk bark (Fig. 4). During the vegetative season, suckers (shoots with a direct insertion on the trunk) were the most infested sites, because more easily reached by crawlers of the first generation (Fig. 5). On canes, a population density gradient with a higher infestation on proximal shoots which tended to decrease toward distal ones, was recorded. Average mealybug density on shoots from spurs was similar to that recorded on the first shoot from canes. The mealybug distribution along shoots in different periods of the vegetative season is reported in figure 6. In June, most of the population was located at the base of shoots. By contrast, in July, only a small percentage of mealybugs (5%) was found at the base of shoots, whereas the first 4 main leaves were the most infested. The following month, mealybugs colonised consistently bunches and lateral leaves, although infestation was mainly concentrated on the first node bunch and on the beneath one. In autumn, mealybugs leaved the already too old basal leaves and stayed within the first 6 nodes of the shoot.

270

80 ovisacs 60 40 20 0 MJ JASONDGFMAMJ JAS ON 80 first instar 60 40 20 0 MJ JASONDGFMAMJ JAS ON 80 second instar 60 40 20 0 MJ JASONDGFMAMJ JAS ON Stage distribution (%) 80 third instar 60 40 20 0 MJ JASONDGFMAMJ JAS ON 80 pre-ovipositing females 60 40 20 0 MJ JASONDGFMAMJ JAS ON 2006 2007 Fig. 3. Seasonal distribution of P. ficus stages reported as a proportion of total mealybugs found.

100

80 Bunches

60 Leaves Base of shoots 40 Trunk and arms 20

Population distribution (%) 0 MJ J AS ONDGFMAMJ J AS ON 2006 2007

Fig. 4. Location of P. ficus on vines, represented as a percentage of total mealybugs found. 271

70 60 50 40 30 20

Mealybugs/shoot 10 0 1st 2nd 3rd 4th 5th 6th others

Shoots of canes Suckers trunk of Shoots spurs of

Fig. 5. Location of P. ficus on the different types of shoot, represented as average mealybug numbers during the period June – September 2007.

Bunch June 80 Lateral leaves Principal leaf 40 Base of shoot 0 July 40

0 August 40 Population distribution (%)

0 base12345678910others Node

Fig. 6. Location of P. ficus on shoot (June – August, 2006).

Vine mealybug development involves 3 generations per year in Sardinian vineyards, as reported for other Mediterranean areas (Franco et al., 2004) and Northern Italy (Duso, 1989), but it might include up to 5-6 generations per year in Sicily (Longo et al., 1991). Our observations on vine mealybug behaviour on grapevine give a contribution in relation to its infestation control. This insect colonises different part of the plant depending on the season, and it prefers sites more protected against sunbeams. The first generation of the 272

year is mainly protected under the bark, and consequently is hardly reached by insecticidal treatments. Only from June, crawlers of the second generation move to leaves and new shoots, so that they become an easier insecticide target. However, these upward movements of juveniles are gradual and part of the population is still under the bark in summer. P. ficus population monitoring methods might be simplified by using pheromone traps, which permits to focus observations on the period following male peaks, thus reducing the number of samplings needed for a prompt detection of crawler flows on leaves and bunches. Further advances in mealybug control approach might be achieved identifying an intervention threshold based on capture levels. In fact, observations conducted in California and South Africa demonstrated a positive correlation between the number of captured males and infestation level (Millar et al., 2002; Walton et al., 2004).

Acknowledgements

The authors are grateful to Dr L. Ruiu for his assistance in writing the English version of the manuscript.

References

Dalla Montà, L., Duso, C. & Malagnini, V. 2001: Current status of scale insects (Hemiptera: Coccoidea) in the Italian vineyards. – Boll. Zool. agr. Bachic. Ser. II, 33(3):343-350. Duso, C. 1989: Indagini bioetologiche su Planococcus ficus (Sign.) nel Veneto. – Boll. Lab. Ent. agr. Filippo Silvestri 46: 3-20. Franco, J.C., Russo, A., Suma, P., Neto, E., Zada, A. & Mendel, Z. 2004: Comparative biology of the citrus mealybug and the vineyard mealybug (Hemiptera: Pseudococcidae). – Proceeding of the X International Symposium on Scale Insect Studies. 19-23 April 2004, Adana, Turkey: 232-233. Longo, S., Mazzeo, G. & Russo, A. 1991: Note bioetologiche su Planococcus ficus (Homoptera: Coccoidea Pseudococcidae) in Sicilia. – Atti XVI Congr. Naz. It. Entomol., Bari-Martina Franca, 23-28 settembre 1991: 705-710. Millar, J.G., Daane, K.M., McElfresh, J.S., Moreira, J.A., Malakar-Kuenen, R., Guillén, M. & Bentley, W.J. 2002: Development and optimization of methods for using sex pheromone for monitoring the mealybug Planococcus ficus (Homoptera: Pseudococcidae) in California vineyards. – J. Econ. Entomol. 95(4): 706-714. Ortu, S., Cocco, A. & Lentini A. 2006: Utilisation of the sexual pheromones of Planococcus ficus and Planococcus citri in vineyards. – IOBC wprs Bulletin 29(11): 207-209. Walton, V.M., Daane, K.M. & Pringle, K.L. 2004: Monitoring Planococcus ficus in South African vineyards with sex pheromone-baited traps. – Crop Prot. 23: 1089-1096.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 273-277

Impact of the erineum mite Colomerus vitis on Muscat

Christian Linder1, Mauro Jermini2, Vivian Zufferey3 1 Station de recherche Agroscope Changins-Wädenswil ACW, CP 1012, 1260 Nyon 1, Switzerland, e-mail: [email protected]; 2 Station de recherche ACW Cadenazzo, 6594 Contone, Suisse; 3Station de recherche ACW Pully, 1009 Pully, Switzerland

Abstract: The erineum mite Colomerus vitis (Pagenstecher) is widespread in Swiss vineyards. Because damages of C. vitis are rarely serious, its pest status is still unclear. Even if acaricides are only occasionally applied these products are nonetheless moderately toxic to predatory mites. For a better understanding of the impact of erineum mites on vine, harmfulness of C. vitis has been studied from 2005 to 2007. We compared the effect of mites on plant damage, gas exchange, photosynthesis and plant growth on Muscat. In May 2006, the month of the most severe mite infestation, about 3% of leaves showed more than 60% damaged leaf area. Nevertheless, mites had almost no effect on transpiration rates measured. Photosynthesis and stomatal conductance rates slightly decreased on heavily infested leaves but the chlorophyll index was unaffected. The presence of mites had also no effect on shoot growth and foliation. Overall, plant damage was neither correlated with overwintering mite population nor plant damage in the previous year. In conclusion, the impact of C. vitis on photosynthesis is negligible and acaricide treatment can be abandoned at moderate pest infestation levels as observed in this study.

Key words: Colomerus vitis, plant growth, damages, gas exchanges, photosynthesis

Introduction

The erineum strain of Colomerus vitis (Pagenstecher) is a common pest species in Swiss vineyards. Its biology is well documented (Mathez 1965; Baggiolini et al., 1969; Baur, 2000) but its real impact on the plant has not been clearly determined (Baggiolini et al., 1969; Barnes, 1992; Hluchy & Pospisil, 1992). In the absence of any established thresholds, growers treat on the basis of damage in the previous season (Linder & Höhn, 2007). In an IPM point of view, this strategy is unsatisfactory. First, the necessity of spring treatments is unproven. Second, pesticides are applied on unknown mite densities. And third, pesticide use can be moderately toxic for predatory mites (Linder et al., 2005). For a better understanding of C. vitis' impact on vine, its harmfulness has been studied from 2005 to 2007. Interactions between pest and crop were investigated according to the study scheme proposed by Jermini et al. (2006).

Material and methods

The study was conducted in a vineyard at Mont-sur-Rolle (Vaud, Switzerland). A single line of 250 Muscat vines was studied. Vines were planted in 1987, grafted onto 3309, Guyot- trained and planted at a row distance of 0.8 m. In 2005 at the phenological BBCH stage 13-15, 20 mite-infested plants were randomly chosen. Each of these plants was regulated to 6 shoots per stock to provide greatest, possible homogeneity. Three representative shoots of the rootstock were then selected from each plant.

273 274

Growth of these shoots was repeatedly measured and damage was regularly assessed according to Horsfall & Cowling (1978). During winter 2006, overwintering mite densities were assessed by controlling 4 buds per selected shoot (level 2-3 and 7-8). Buds were dissected and then agitated in a solution of water + 0.1% wetting agent (Teepol). After 20 minutes the dissected buds were removed and the remaining solution was vacuum-filtered on 90 mm ∅ black paper filters (Schleicher & Schuell). Mites were counted under a dissecting microscope. In Spring 2006, half of the vines were treated with a mixture of rapeseed oil and diazinon to provide maximum protection against the development of C. vitis. The other half remained untreated. This design should allow us to follow a range of different mite infestation levels in a single vineyard. A total of 30 plants were randomly chosen. Vines were monitored and measured as in the year before. In addition, leaf gas exchange was assessed at the end of flowering (BBCH 69-71). It was measured with a portable infrared gas analyser LC4 on leaves of similar physiological age under standardised conditions, e.g. at a temperature of 24- 25°C between 9 and 11am under fully light-saturated conditions. Measurements were made on healthy leaves parts as well as on galls of C. vitis. Net photosynthesis, stomatal conductance and transpiration were calculated. Additionally, the chlorophyll index of 180 leaves was measured with a N-tester.

Results and discussion

Dynamic and intensity of mite damage In spring, symptoms were concentrated at the base of the shoots (Figure 1). Over the following weeks, C. vitis migrated towards young leaves in the top. Foliation and topping of vines decreased the level of damage per plant from 9.9% in May to 4.9% at the end of July.

24.05 28.06 27.07 22 22 22 19 19 19 16 16 16 13 13 13 10 10 10 7

level of main leaf 7 7 level of main leaf main of level leaf main of level 4 4 4 1 1 1 02040 0204002040 % of damages % of damages % of damages Fig. 1. Damage of C. vitis along the shoot on infested plants in 2006. Error bars = 1 SD.

At the end of the season, lateral leaves were more infested than the main leaves. However, there was no correlation between damage on main and lateral leaves. Erineum mite infestation was very variable between the years. In 2005, 35% of untreated main leaves were attacked, in the following year about 50% and in 2007 only 6.5%. This massive decline is probably explained by the hot and dry weather conditions in April 2007, which favoured shoot growth and the development of predatory Phytoseiid mites. However it should be noted 275

that even in 2006, only 3% of main leaves had more than 60% of the leaf surface damaged. Mite-attacked grapes were not observed over the whole study period.

Overwintering mites and damage Baur (2000) found a good correlation between the number of overwintering erineum mites and the percentage of leaves damaged the following spring. However, in this study we were unable to find a similar correlation between overwintering mites and severity of damages (Figure 2). This may be a result of C. vitis' mobility and of the large impact of biotic and abiotic factors on natural population dynamics. Overall, we are not in a position to make any recommendations for defining a valuable threshold for acaricide applications in spring.

12 25 y = 0.0025x + 2.7344 a b 10 r2 = 0.0096 20 y = -0.015x + 11.088 8 r2 = 0.0705 15 6 10 4

5 2 % of main leaves damaged % of main leaves damaged

0 0 0 100 200 300 400 500 0 100 200 300 400 500 Overwintering C. vitis per bud Overwintering C. vitis per bud

Fig. 2. Relation between the number of overwintering C. vitis per bud and the percentage of main leaves damaged in a) the previous summer (September 2005) and b) the following spring (May 2006).

Plant growth There was no difference between infested and uninfested vines regarding the number of main leaves, the length of main shoots, the number of lateral shoots per main shoots and the number of lateral leaves per main shoots. Thus, the impact of C. vitis on plant growth was probably negligible at the infestation level of this study. High damage levels on lateral leaves still have to be evaluated. They may affect the content of plant reserves as it is the case for downy mildew (Jermini et al., 2001).

Gas exchange The relation between the percentage of damaged leaf area and net photosynthesis, respectively stomatal conductance was analysed by multilinear regressions (Fig. 3). The moderate correlation (= r2) and the missing of measures at high damage (only 0.3% of leaves had galls on more than 80% of leaf surface) ask for a careful interpretation of the results. Nevertheless, there was a slight decline in net photosynthesis and in stomatal conductance when damages exceeded 70 to 80%. Thus, the impact of C. vitis on the overall gas exchange can be neglected at moderate mite infestation. Healthy leaves and healthy parts of damaged leaves did not differ in their chlorophyll index, in their net photosnythesis, in their transpiration and in their stomatal conductance (Figure 4). It can therefore be concluded that healthy parts of attacked leaves do not increase their photosynthetic activity to compensate for damage.

276

net photosynthesis stomatal conductance 21.2 0.54

2 y = -0.0015x + 0.0879x + 10.823 y = -5E-05x2 + 0.003x + 0.2825 2 r = 0.4183 r 2 = 0.3023 -1 s -1 -2 s -2 m 2

10.6 Om 0.27 2 mol H mol µmol CO

0 0 080080 % of damages % of damages

Fig. 3. Relation between damaged leaf surface in the assimilation chamber of the gas analyser and a) net photosynthesis and b) stomatal conductance. Straight lines correspond to the mean value of healthy leaves.

Photosynthesis, transpiration and stomatal conductance were slightly increased at 35% of leaf surface damaged (Figure 4). This might be linked to the fact that small C. vitis galls on Muscat leaves often appeared to be greener than healthy leaf surface. This may suggest a higher photosynthetic activity on galls. Photosynthesis and stomatal conductance was significantly reduced at higher damage levels. A reduction of nearly 30% was recorded at 85% of leaf surface damaged. However, transpiration was not affected. Our results suggest that very high damage levels have only a minor impact on total photosynthesis.

net photosynthesis transpiration stomatal conductance 14 6 0.5

0.45 12 5 0.4

10 0.35 4 -1 -1 s -1 s s

-2 0.3 -2

8 -2 m 2 O m 2 3 0 m 0.25 2 6 0.2 mol H mol µmol CO µmol mmol H mmol 2 4 0.15

0.1 1 2 0.05

0 0 0 HL H-D 35 85 HL H-D 35 85 HL H-D 35 85

Fig. 4. Impact of C. vitis on the net photosynthesis, transpiration and stomatal conductance of main leaves at the end of flowering. HL = healthy treated leaves, H-D = healthy parts of damaged leaves, 35 = 35% damaged leaf surface and 85 = 85% damaged leaf surface. Error bars = 1 SE.

Conclusions

Moderate C. vitis infestation during the phase of « bud break – end of flowering » does not affect plant growth and has only a limited effect on photosynthesis, transpiration and stomatal conductance. Damage is restricted to basal main leaves and there seems to be no or only a weak relationship between the number of overwintering mites and the severity of damage in the following spring. Biotic and abiotic factors may therefore play an important role in the 277

natural control of the pest populations. Furthermore, topping of lateral shoots in summer and the green harvest before vintage are probably both reducing the impact of erineum mites on vines. We therefore conclude that bud break treatments should be abandoned at moderate pest infestation levels as observed in this study. Further studies have to investigate the impact of summer mite attacks on lateral leaves, on grapes, on carbohydrate reserves and harvest quality. Moreover, it would be interesting to see if our result of C. vitis negligible impact on wine production is confirmed by studies on other varieties such as Pinot noir.

References

Baggiolini M., Guignard E., Hugi H., Epard S., 1969: Contribution à la connaissance de la biologie de l’érinose de la vigne et nouvelles possibilités de lutte. – Revue suisse de Vitic. Arboric. Hortic. 1 (3): 50-52. Barnes M.M., 1992: Grape Erineum mite. – In: Grape pest management. Eds. Flaherty D.L., Christensen W.T., Lanini W.T., Harois J.J., Philips P.A., Wilson L.T.. Univ. Calif. Publ. 3342 Oakland USA: 262-264. Baur R., 2000: Verteilung von überwinternden Pockenmilben auf Rebholz. – Schweiz. Z. Obst-Weinbau 5: 84-86. Hluchy M., Pospisil Z., 1992: Damage and economic injury levels of eriophyid and tetranychid mites on grapes in Czechoslovakia. – Exp. Appl. Acarol. 14: 95-106. Horsfall, J.G., Cowling, E.B., 1978: Pathometry: measurement of plant disease. – Academic Press, New-York: 120-134. Jermini M., Blaise Ph., Gessler C., 2001: Quantification of the influence of Plasmopara viticola on Vitis vinifera as a basis for the optimisation of the control. – Bulletin IOBC/ WPRS 24 (7): 37-44. Jermini M., Gessler C., Linder Ch., 2006: The use of know-how on the interaction between grapevine and pests or diseases to improve integrated protection strategies. – Bulletin IOBC/WPRS 29 (11): 95-102. Linder Ch., Bouillant S., Höhn H., 2005: Evaluation de l’impact de produits à base d’huiles et de diazinon sur les populations de Phytoseiidae en viticulture. – Revue suisse de Vitic. Arboric. Hortic. 37 (2): 113-117. Linder Ch., Höhn H., 2007: Le guide viti ACW: Principaux ravageurs: Acariens: – Revue suisse de Vitic. Arboric. Hortic. 39 (1): 48-49. Mathez F., 1965: Contribution à l’étude morphologique et biologique d’Eriophyes vitis Pgst., agent de l’Erinose de la vigne. – Bull. Soc. Entom. Suisse 37: 233-283. 278 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 279-281

Ecological infrastructures in a vineyard of Western Sicily

Lo Genco A., Fucarino A. and Lo Pinto M. Dipartimento Scienze Entomologiche, Fitopatologiche, Microbiologiche e Zootecniche (SEnFiMiZo), Università di Palermo, Viale delle Scienze 13, 90128 Palermo, Italy e-mail: [email protected] , [email protected]

Abstract. Cultural methods such as excessive use of fertilizers, pesticides and monocultures have decreased agrobiodiversity and created more favourable conditions to increase insect pests and diseases levels in agroecosystems. Monoculture is a concentrated resource for specialized pests, which increases the attraction and accumulation of these species, the time they spend in the system and their reproductive success. Furthermore, expansion of monocultures has decreased abundance and activity of natural enemies due to removal of food sources (i.e. nectar, pollen and often honeydew), refuges and hibernation sites. Some researchers have proposed to increase agro-biodiversity by creating an appropriate ecological infrastructure within and around the agro-ecosystem to favour the abundance and effectiveness of natural enemies. The maintenance and management of ecological infrastructures, or ecological compensation areas (ECAs), on rural farms is considered crucial in enhancing functional biodiversity for pest suppression (Boller et al., 2004). In autumn 2006, we start cooperating with a young winery (www.funaro.it) located in Marsala area. In order to enhance biodiversity we have introduced ecological infrastructures in 2 hectares vineyard of Merlot variety. We have planted several plants of bramble (Rubus spp.), oregano (Origanum vulgare L.), rosemary (Rosmarinus officinalis L.), wild rose (Rosa canina L.), oak (Quercus pubescens Willdenow) all around Merlot vineyard. These plants should provide refuges, hibernation sites, alternate host animals, prey for the juvenile stage of predators, resources for a successful immigration and reproduction of natural enemies of key pest in viticulture. The role of ecological infrastructures in vineyard is still not well understood and their use to manage vineyard pests is a controversial practice. This work aims to study the possible effects of ecological infrastructure on arthropod associated to a vineyard.

Key words: Hedges, vineyard, biological control

Introduction

Conservation biological control is a pest control strategy based on the manipulation of wild populations of natural enemies to enhance their impact on pest by habitat management and by maintaining ecological infrastructures (Boller et al., 2004).To improve conservation biological control requires a detailed basic knowledge of the trophic relationships among ecological non-crop plants, phytophages and beneficials (Burgio et al.,2006). Adequate ecological infrastructures provide refuges, hibernation sites, alternate host animals, prey for the juvenile stages of predators, and with flowering plants the essential food sources for maturation and reproduction of all adult parasitoids, many predators and pollinators. Thus, each plant species have an important function in natural pests regulation because act as important sources of natural enemies of key pest. In the agro-ecosystem vineyard, the influence of such hedgerows inside and outside production sites on grape pest management has been investigated by several authors (Nicholls et al., 2001; Boller et al., 2004; Ponti et al., 2005). For example, the egg parasitoid Anagrus spp.(Mymaridae), the most important natural enemy of the grape leafhoppers, spends the winter as egg in leafhopper eggs deposited in

279 280

bramble and wild roses (Boller et al., 2004). Research by Ponti et al. (1984) established that maintenance of the bramble hedge surrounding vineyard is therefore advisable to promote early Anagrus colonisation leading to an effective control of grape leafhoppers. The aim of the present research is to study the influences of ecological infrastructures on biodiversity and their role in grape pest management.

Materials and methods

Study region The cultivation of vine and the production of wine are spread in all the Sicilian territory and the ampelographic wealth of the island is pretty interesting. Sicily is Italy’s biggest wine producer and it has also the biggest area cultivated in vineyard.Efforts made in the last twenty years have allowed the revaluation of many autochthonous grapes of the island and now – after having concretely faced extinction – are considered among the most important grapes of Italy. In the region are also cultivated many international grapes, mainly used for blends with local grapes (Cabernet sauvignon, Merlot, Syrah, Chardonnay, etc.). Funaro Property has more than 50 ha of vineyards, between native and international varieties.The vineyards are located between Salemi and Marsala, right at the heart of this area with a marked vine- growing vocation. In this area an excessively aggressive viticulture has no permitted the conservation of some main natural area where the natural flora and fauna can grow up. Going across this area is easy to understand how the agro-ecosistem is altered and the necessity of a rapid restoring is vital for achieve a correct integrated or organic viticulture. The commitment and openness of Funaro family has permitted to start an interesting research on ecological infrastructures, planted them around and inside their property, and also using specific cover crops during the raining moths.Funaro family with our experience and help is converting all the vineyards from a conventional viticulture to a modern and organic viticulture.

Study site This experiment is conducted in a 2 ha Merlot organic vineyard, located in Salemi (situated 400 m above sea level), 10 Km from Marsala (Sicily), a typical wine growing area. The vineyard is affected yearly by various pathogens, insects, and mites. Downy mildew (Plasmopara viticola B. et C.), powdery mildew (Uncinula necator Schaw.) and bunchrot (Botrytis cinerea Pers.) are the most prevalent fungal pathogens of grapes in the area. Among insect pests, leafhopper (Jacobiasca lybica Bergevin & Zanon and Zygina rhamni Ferrari) is the most persistent one. Others local key pests are, grape moths, spider mites and thrips. Vineyard is protected yearly with preventive applications of two fungicides, Bordeaux mixture and sulphur. No insecticide is applied. In the autumn 2006, in order to enhance biodiversity, several young plants of bramble (Rubus spp.), oregano (Origanum vulgare L.), rosemary (Rosmarinus officinalis L.), wild rose (Rosa canina L.), oak (Quercus pubescens Willdenow) were planted all around the perimeter of the Merlot vineyard. Furthermore, winter cover crops (a mixture of barley, Hordeum vulgare L. and vetch, Vicia narbonensis L. cv velo) are planted yearly in an alternating row pattern. Brambles and wild roses are an important source for predatory mites and Mymarid parasitoids such as Anagrus sp. (Boller et al., 2004); Rosemary and oak are important plants for development of lepidopteran parasitoids and predatory mites (Ragusa & Tsolakis, 2005); oregano is a good nectar source for parasitoids and predators.

Sampling procedure In order to determine the species diversity and densities of prevalent herbivores and predators/ parasitoids associated with the ecological infrastructures, insects and mites on each hedge are 281

sampled by mechanical knock-down (MKD) every 15 days, by visual inspections (VIS) every 7 days(Burgio et al.,2004). Also, the entomofauna are sampled by randomly placing yellow and blue sticky traps in grape and in the hedges. Lobesia botrana Denis et Schiffermüller are monitored by direct count on inflorescences or bunches and by pheromone traps placed in the vineyard. All traps are returned to the laboratory and examined under a stereomicroscope to identify and count the number of phytophagous insects and natural enemies.

Results and discussion

Preliminary observation (periodical sampling of 100 leaves/hectare) conducted in this vineyard from April-September 2006, before planting ecological infrastructures, showed among the prevalent species of pests, species such as Z. rhamni and J. lybica. The leafhoppers population increased from May to the beginning of September reached the highest peak at the end of September with 4.1 leafhoppers/leaf. During the visual examination of the leaves in the field, some species of dragonflies were seen to prey leafhoppers. The tortricid moth L. botrana was also present with a low density, total yearly captures ranged from 0 to 3 males/ pheromone trap. Moreover, our preliminary data suggest that vineyard exhibit higher level of predator families such as Coccinellidae, Chrysopidae, Syrphidae and Araneae. No data are available about the presence of parasitoids species. At the present time, ecological infrastructures are one year old and, until now, only spider species were found in their young leaves. Our challenge is to identify the hedges that are desirable to maintain in the vineyards in order to carry out specific ecological services such as pest regulation. This study would investigate the presence of some parasitoid species and their alternative hosts in relation to seasons of the year, relationships between these species and entomofauna living on the vineyard, and comparisons of the population levels of phytophagous species detected in preliminary observations with data collected in presence of ecological infrastructures. Thus, it will be possible to make an evaluation of the role of insect biodiversity on the natural control of main pests of organic vineyard provided by each hedges.

References

Altieri M.A., 1993. Ethnoscience and biodiversity: key elements in the design of sustainable pest management systems for small farmers in developing countries. – Agriculture, Ecosystems and Environment 46: 257-272. Boller E.F., Hani F., Poehling H.-M., 2004. Ecological infrastructures. Ideabook on functional biodiversity at the farm level. – IOBC/wprs, Lindau, Switzerland. Burgio G., Ferrari R., Boriani L., Pozzati M., van Lenteren J., 2006. The role of ecological infrastructures on Coccinellidae (Coleoptera) and other predators in weedy field margins within northern Italy agroecosystems. – Bulletin of Insectology 59 (1): 59-67. Nicholls C.I., Parrella M. & Altieri M.A., 2001. The effects of a vegetational corridor on the abundance and dispersal of insect biodiversity within a northern California organic vineyard. – Landscape Ecology 16: 133-146. Ragusa S. & Tsolakis H., 2006. La difesa della vite dagli artropodi dannosi. – Università degli studi di Palermo. 282

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 283-288

Arthropods as biological soil quality indicators in a vineyard under different soil management

Gabriella Lo Verde1, Vanessa Palermo2, Antonino Santoro2, Luciano Gristina2 1 Dipartimetno SENFIMIZO, Sezione Entomologia, Acarologia e Zoologia 2 Dipartimento di Agronomia Ambientale e Territoriale - Università degli Studi di Palermo (Italy), viale delle Scienze, 90128 Palermo. [email protected]

Abstract: The impact from anthropogenic sources on the soil environment is almost exclusively assessed for chemicals, although other factors like covers crops and tillage practices have an important impact as well. Thus, the farming system as a whole should be evaluated according to its environmental benefits and impacts. A research was carried out in a vineyard located in south-west of Sicily in which five different soil managements were applied: conventional tillage, durum wheat cover crop, faba bean cover crop, permanent meadow (lolium 60% and berseem 40%) and permanent meadow (lolium 40 % and berseem 60%). Faba bean and durum wheat cover crops were managed using two different incorporation techniques. Three soil samples were collected for each treatment, every two months. Arthropods were extracted with a Berlese- Tullgren funnel and identified at Order level. QBS index (Soil Biological Quality index), arthropods abundance and Shannon diversity index were considered to analyze the differences among the studied plots, UPGMA distances among the cover crop systems were clustered. In plots with different cover crop incorporation technique arthropods abundance showed an inverse trend in comparison with the diversity index, lower in plots in which plowing was applied. The similarity cluster analysis emphasized the incidence of the cover crop species, both in purity and as prevalent species in mixed treatments, whereas QBS and the nitrates availability seem to be influenced by the management system more than by the cover crops.

Key words: vineyard, soil management, arthropods, soil quality.

Introduction

In agricultural soil, a suite of anthropogenic events shape the ecosystem processes and populations. However, the impact of anthropogenic sources on the soil environment is almost exclusively assessed for chemical factors only, although in agriculture mechanical factors like tillage and biological factors such as cover crops have a large impact as well (Steen, 1983). Based on data about the agricultural events arthropods communities, as biological indicators, should be included in soil quality classification to identify the best choice among the several techniques available to assess soil quality and to convert the information obtained from the biological soil indicator into a form relevant for policy makers. Soil microarthropods demonstrated to respond sensitively to land management practices and to be correlated with beneficial soil functions. Recently, a new approach (called QBS), based on the biological forms of edaphic microarthropods has been proposed to assess soil biological quality (Parisi, 2001; Parisi et al., 2005). Presently, several initiatives are directed toward elucidation of the relationship between land management and the soil quality (Angelini et al., 2002; Lo Verde et al., 2007; Martorana et al., in press).

283 284

In many Sicilian areas visible erosion problems induced farmers to introduce alternative soil management techniques. Till now the efficiency in soil erosion control was demonstrated by using legumes cover crop and graminaceous one as well; the efficacy of these management techniques on soil quality is still unknown. The aim of this study is to evaluate the influence of the different cover crops management on the biological soil quality, through the arthropods communities analysis.

Materials and methods

To fit the aim of this work a vineyard area in south-west of Sicily (37° 39’N-13°00’E) has been chosen. The vineyard was managed to vertical trellis system with cane pruning (cultivar "Insolia" grafted on 140 Ru), spaced 2.80 m x 1.00 m apart, planted on an approximately 20% hillslope. Five different soil managements were applied, as shown in tab.1. Two different cover crops incorporation techniques were used for faba bean and durum wheat cover crops biomass manure in April (Tab.1).

Tab. 1: Scheme of the treatments

Cover crop species incorporation techniques

permanent meadow 1 pm1 ---- (lolium 60 % and berseem 40%)

permanent meadow 2 pm2 ---- (lolium 40 % and berseem 60%)

plowing (P) fb faba bean rotary howing (H)

plowing (P) dw durum wheat rotary howing (H)

conventional tillage ct ---- (several tillage during the year)

Three soil samples were collected for each treatment, every two months, in the top, medium and bottom part of the slope, from March 2006 to January 2007, for a total of 84 samples. A simple and cheap Berlese-Tullgren funnel has been used for separating arthropods from soil. The metal funnel size used is 15 cm diameter, to a depth of 10 cm. Because of the size and water content of the samples a 20 Watt light bulb was used to create microarthropods reaction to the heat and desiccation. Specimens were collected and preserved in alcohol (70°) and observed under a stereomicroscope at low magnification (range 6–50X). As above-mentioned QBS index is based on microarthropod groups present in a soil sample. Arthropods present in each sample, were classified at Order level, evaluating the different adaptation levels to soil environment in some Orders. Each biological form found in the sample receives a score from 1 to 20 (Eco-Morphological Index, EMI), according to its 285

adaptation to soil environment. The QBS index sums up these scores, thereby characterising the microarthropod community of the sample being studied (Parisi, 2005). Arthropods abundance was analyzed using ANOVA followed by Tukey post-hoc test (p<0.05). For each treatment was calculated the Shannon diversity index (H’) with regard to the whole research period. Moreover, a cluster analysis based on the similarity index of Bray- Curtis was applied (UPGMA, software PRIMER 5.0).

Results and discussion

On the whole, 2186 Arthropods have been examined. Eighteen groups of mesofauna were found, wherein Acari and Collembola were dominant and Coleoptera, Psocoptera, Diptera and Hymenoptera were frequently recorded (Tab. 2).

Tab. 2: Arthropod taxa collected in the different plots Taxa Pm1 Pm2 Pfb Hfb Pdw Hdw ct Tot. Arachnida Acari 122 59 190 65 269 135 141 981 Insecta Collembola 63 25 19 24 38 79 26 274 Insecta Coleoptera 13 15 53 73 10 7 5 176 Insecta Psocoptera 45 8 8 37 18 23 23 162 Insecta Diptera (larvae) 8 10 4 4 5 16 21 68 Insecta Hymenoptera 4 8 21 8 4 5 6 56 Insecta Rhynchota 14 1 2 4 5 8 1 35 Insecta Protura 8 0 1 0 1 0 0 10 Symphyla 1 1 2 1 2 1 0 8 Chilopoda 0 2 0 1 1 0 3 7 Insecta Thysanoptera 0 2 0 0 2 1 0 5 Insecta Diplura 0 0 1 1 0 0 2 4 Arachnida Araneae 0 1 1 0 1 0 0 3 Insecta Dermaptera 0 0 0 0 0 0 1 1 Insecta Embioptera 0 0 0 0 1 0 0 1 Crustacea Isopoda 0 0 0 1 0 0 0 1 Insecta Holometabola 33 65 78 44 70 69 35 394 (larvae + adults) Total 311 197 380 263 427 344 264 2186

Arthropods abundance in each plot ranges from a minimum of 197 in pm2 to 380 in Pfb (Tab. 2). The diversity index ranges from a minimum of 1.34 in Pdw to 1.92 in pm2 (Tab.3). In the plots having the same cover crop species, the plowing seems to determine higher numbers of Arthropods in comparison with the howing technique; nevertheless, diversity index in these plots shows an inverse trend, being higher in the plots in which howing was applied. Samples collected in the different parts of the slope did not show significant differences (ANOVA, followed by Tukey test, p<0.05), thus data from each plot were cumulated before analyzing them. No statistical differences have been found among the different plots, probably due to the high variability of data (ANOVA, followed by Tukey test, p<0.05). 286

Tab. 3: Abundance and diversity of Arthropods in the different plots.

Soil management N Arthropods N taxa H' pm1 311 13 1.81 pm2 197 13 1.92 Pfb 380 11 1.59 Hfb 263 13 1.89 Pdw 427 15 1.34 Hdw 344 11 1.72 ct 264 12 1.62

As regards the cluster analysis, it is not really surprising to find a clear separation between the ct treatment and the other management techniques (Fig.1); among the latter ones, two groups were separated, the first containing both Pdw and Hdw plots, while the second containing both Pfb and Hfb plots, to remark the similarity of the cover crops species. Moreover, pm1 (containing only 40% of the legume) was in the same group of the durum wheat plots, while pm2 (containing 60% of the legume ) resulted more similar to the two faba bean plots. totali 75 80 85 90 Similarity 95 100 ct Pdwpm1Hdw pm2 Pfb Hfb T7 tot T3 tot 5F tot T4 tot 1F tot 5A tot 1A tot

Fig. 1: Similarity tree obtained from the cluster analysis based on the total number of arthropods collected in the different plots during the research.

As regards QBS values, the different cover crop species do not show heavy differences (Fig. 2); Pfb and Hfb show the highest QBS values (154 and 133, respectively), while ct has a 110 score, almost the same of pm2 (109) and slightly higher than Hdw (94). The low QBS value recorded in ct is probably due to the high number of tillage, that causes disturbance to Arthropods. Comparing Pfb and Hfb with Pdw and Hdw respectively, the legume faba bean seems to guarantee in both cased a slightly higher value of QBS. This trend is not confirmed by the index values of pm1, in which legume were only 40%, that resulted higher than pm2, in which they represented 60% of the cover crop. In relation to the two different cover crops 287

incorporation techniques used for faba bean and durum weat, plowed plots show higher QBS values, probably due to the fact that cover crop incorporation by plow occurs at higher depth than rotary howing one, leading to a longer availability of organic matter. As regards nitrate availability (Fig. 2) its value seems to be more correlated to the number of tillage than to cover crop species. Actuallyt, the highest values are recorded in ct (88.8 mg/l), while in all the other treatments it ranged between 19.4 (Hdw) and 41.1 (Hfb).

200 100

180 90

160 80

140 70

120 60 L- e

100 50

BSQ indBSQ 80 40 nitrates (mg 60 30

40 20

20 10

0 0 pm1 pm2 Pfb Hfb Pdw Hdw ct

BSQ index nitrates

Fig. 2 - QBS index and nitrates levels in the different soil management plots.

Remarks

Arthropods abundance in plots with different cover crop incorporation technique showed an inverse trend in comparison with the diversity index, showing that plowing leads to higher number of Arthropods collected but to a correspondent lower diversity value. According to the cluster analysis, the plots characterized by the same cover crop species, both in purity and as prevalent species in mixed treatments, showed higher similarity, independently from the different cover crop incorporation technique. On the other hand, both biological soil quality index (BQS) and the nitrates availability seem to be influenced by the management system more than the cover crops. Arthropod abundance and the QBS values were higher in Pfb and Pdw in comparison with Hfb and Hdw respectively. To confirm the relationships between QBS index with agronomic parameters (soil fertility and management), further applications are needed.

References

Angelini P., Fenoglio S., Isaia M., Jacomini, C., Migliorini M., Morisi A., 2002: Tecniche di biomonitoraggio della qualità del suolo. – Pubblicazioni ARPA Piemonte 2002: pp 106. Lo Verde G., Palermo V., Santoro A., Gristina L., 2007: Biological soil quality in a sustainable vineyard management in South-West of Sicily (Italy). – Proceedings of 5th Inter- 288

national Congress of the ESSC “Changing Soils in a Changing World: The Soils of Tomorrow”. 2007, Palermo (Italy). (Poster). Martorana A., Torta L., Lo Verde G., Ragusa E., Burruano S., Ragusa S., 2008: Biodiversity in the grapevine: rhizosphere arthropods and mycorrhizal fungi. IOBC/WPRS Bull. 36: 161-165. Parisi V., 2001: La qualità biologica del suolo. Un metodo basato sui microartropodi. – Acta Naturalia de “L’Ateneo Parmense” 37 (3/4): 97-106. Parisi V., Menta C., Gardi C., Jacomini C, Mozzanica E., 2005: Microarthropod communities as a tool to assess soil quality and biodiversity: a new approach in Italy. – Agriculture Ecosystem & Environment 105: 323–333. Steen, E., 1983: Soil animals in relation to agricultural practices and soil productivity. – Swedish J. Agric. Res. 13: 157–165. Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 289-293

Organizational traits for the grapevine moth control in Tuscany (Italy)

Andrea Lucchi1,, Elena Pozzolini1, Luciano Santini1 & Bruno Bagnoli2 1 Dipartimento Coltivazione e Difesa Specie Legnose “G. Scaramuzzi”, Sezione Entomologia agraria, Università di Pisa, Italia ([email protected]). 2 CRA, Istituto sperimentale per la Zoologia Agraria, Firenze. ([email protected])

Abstract: In Tuscany, in the last few years, there has been a growing interest for the adoption of pheromone mating disruption (MD) in the control of the grapevine moth Lobesia botrana (Den. & Schiff.). In 2007 the method has been applied on about 500 hectares with Isonet L (CBC/Shin-Etsu) dispensers in 27 estates of the provinces of Pisa, Livorno, Grosseto, Siena and Firenze. Most of the estates were applying MD for the first time. To help farmers in assessing the method efficacy a small technical team was formed, involving in the project some students of the Course “Viticulture and Enology” of the Agriculture Faculty (Pisa University). The team has been deeply involved in managing the method application and directly assessing MD efficacy in most of the farms along the three generations of the grapevine moth. In this paper are shortly highlighted the key points that enabled the method to attain good results in most vineyards and the loop holes that could compromise the MD success if not taken into account.

Key words: vine moths, Tortricidae, pheromones.

Introduction

In Tuscany, the European grapevine moth Lobesia botrana (Den. & Schiff.) is still considered as the main pest, in particular for the harmfulness of its carpophagous generations on vine varieties characterized by compact and late ripening bunches. The other tortricid Eupoecilia ambiguella (Hb.) is less frequent and less noxious with the exception of particular environments as, for example, those of the coastal strip of the province of Massa Carrara (Northern Tuscany), where the species can often reach high population levels (Bagnoli, 1990; Bagnoli & Lucchi, 2003). The control of L. botrana, conducted in the past with traditional neurotoxic insecticides, in recent years has been carried out in many cases with more selective compounds such as Bacillus thuringiensis ssp. kurstaki and aizawai, some insect growth regulators (CSIs and MACs) and new neurotoxic molecules (indoxacarb and spinosad) (Boselli et al., 2000; Pasquier & Charmillot, 2000; Charmillot et al., 2004). Also, in the last ten years even the pheromone mating disruption technique (MD) has become important in certain territories of Italy and outside Italy, proposing itself not only for the undoubted efficacy demonstrated in vine moth control but also because of its innovative characteristic in relation to the control programs of the “area wide” type (Ioriatti et al., 2005 and 2008). MD was tested in Tuscany from 1989 to 1994 and from 1999 to 2000 using materials of the Basf company, respectively in an area of the Chianti from the province of Siena (Bagnoli et al., 1993; Bagnoli & Goggioli, 1996) and in the province of Pisa (Bagnoli et al., 2001). In the first case the tests were carried out at the “Agricola S. Felice” estate of Castelnuovo Berardenga (Siena), in vineyards with Sangiovese grape variety, in areas of the estate which increased year by year (respectively 2, 2.5, 4, 8, 10 and 12 hectares) with Basf Rak2 dispensers. Following Basf protocols, 500 dispensers per hectare were distributed having as a

289 290

target the second flight adults in 1989 and the second and third flight in the next five years. In addition, in 1991 and 1992 an insecticide treatment was carried out with chlorpyriphos- methyl against the first generation. In Pisa the method involved the use of a new type of Basf dispenser, the Rak2 R (with repellent), applied experimentally on 10 hectares of the “Celaia” estate in Crespina in 1999 and on 5 hectares of the “Villa Vestri” estate in Cevoli in 2000. In these two tests the normal 500 dispenser per hectare were distributed just before the beginning of the first flight (first week of April) taking into consideration their presumed longer efficiency (Lucchi and Bagnoli, 2007). Even if in the districts of Pisa and Siena the method has been applied in different ways as regards treated surface, type of dispenser, installation period and target flights, the obtained results were very similar, showing a substantial method efficacy in relation to untreated vineyards but, at the same time, a phytosanitary validity not entirely acceptable. Aware that in Trentino South Tyrol (Northern Italy) and in other important vine growing regions of France, Switzerland and Germany this method has been applied on an ever increasing scale (Stockel et al., 1994; Charmillot & Pasquier, 2000; Charmillot et al., 2000; Varner et al., 2001), together with the availability of Shin-Etsu Isonet dispensers, which were considered able to guarantee a more regular and longer pheromone release, in 2001 we decided to start a new experimental cycle in the estate of Castello di Ama (Gaiole in Chianti, Siena), on a surface of about 23 hectares, that increased year by year progressively including other estates. In 2007 the method has been applied on about 500 hectares using Isonet L (CBC/Shin-Etsu) dispensers in 27 estates of the provinces of Pisa (3), Livorno (8), Grosseto (3), Siena (3) and Firenze (10), most of which (17) were applying MD for the first time. To help farmers in assessing the method efficacy and to promote the development of an area wide MD application in Tuscany, a small technical team was formed with students and ex- students of the “Viticulture and Enology” degree Course of the University of Pisa. Cost of the project was completely in charge of the dispensers’ provider (CBC-Europe, Milan Branch). In this survey we give a short draft of the work carried out by the team, highlighting the key points that enabled the method to attain good results in most vineyards and the loop holes that could compromise the project success if not taken into account.

Materials and methods

Materials and protocols used from 2001 to 2007 have always been of the company CBC/ShinEtsu. For each district it was decided to compare pheromone treated, untreated and chemical treated plots with appropriate spatial separation to limit plot interference. Unfortunately, it wasn’t always possible to compare plots of similar age, structure and comparable population density. The variables considered for the assessment of MD effectiveness were: males captured per trap, rate of infested clusters, nests per inflorescence, larvae per cluster and damaged berries per cluster. Plots were of the same variety (in our case Sangiovese for Siena, Florence and Pisa districts and Cabernet for Livorno district); both are late varieties, thus suitable in demonstrating effectiveness of MD across all generations. In each plot, at least 500-1000 clusters in 1st generation and 50-100 clusters in 2nd and 3rd generation have been sampled (1 bunch every 10 vines, choosing the bigger and more compact bunches). For an easier interpretation of results, in each district we used site maps (Google earth) and “climatic data collecting system” to have information on T°, UR and average wind speed. 291

For the vineyards and varieties considered more important from an economic and phytosanitary point of view, specific comparisons were carried out in order to evaluate the percentage efficacy of the treatment compared to a control (Abbott formula).

Results and discussion

The team involved in the project consisted of one graduate student (the project leader), four undergraduates majoring in entomology and two senior entomologists (as supervisors). As above mentioned, main purposes of the team were the farm assistance and the knowledge transfer to operators. Keeping in close contact with the farmers, from the very beginning the team has been deeply involved in managing the method application, taking care of dispensers positioning, placement of the monitoring traps and choice of the plots for the scouting activity. Moreover, the team carried out the task of directly assessing MD efficacy in most of the mentioned farms along the three generations of the grapevine moth and evaluated the dispensers' performance in the different environments. The schedule of the activities carried out by the team during the year embraced: 9 Meeting with the farmers involved in the project for general information and entrustment of the vine-growing areas to the team members (month of February). 9 Team members in the farms for dispensers positioning, choice of the plots for the scouting activity and placement of the monitoring traps in MD and control vineyards (end of March). 9 Team meeting at the beginning and at the end of each generation to discuss results and plan in detail what to do for the following generation (end of May, June and July). 9 Collection of dispensers and their conservation in freezer until the GC analysis (end of each flight). The number of bunches examined by the team in the first, second and third generations was respectively 66,333, 16,958 and 11,400 (94,691 in total). Accurate estimation of 2nd and 3rd generation larval population required destructive sampling and time consuming work, in particular for the main Tuscan grape variety (Sangiovese) characterized by large and compact bunches. The method allowed obtaining everywhere infestation levels decidedly lower than in the untreated vineyards and comparable to insecticide treated vineyards. Only in few cases the final infestation was high because of the favourable development reached by the species in some areas during the 3rd generation. In all the considered districts, at least one estate showed great interest for the method and became the reference leader in helping the others in the MD application. The experience gained by the students of Viticulture and Enology degree course has been very helpful for the farmers that, in most of the cases, did not have the basic knowledge necessary to recognize L. botrana and the other non target moths (not necessarily harmful) hosted in the bunches during the growing season. Among the critical points has to be firstly considered the little attitude of the farmers to cooperate: as a matter of fact, out of the 23 supported farms, only 13 were cooperative, 7 showed little interest, 3 were not cooperative at all. As a consequence, samplings of the first two generations have been conducted in all the farms whereas, for the 3rd generation 5 farms harvested grapes in the control plots without informing the team, very probably afraid of the bad phitosanitary state of their grapes. As for the knowledge transfer, the team aim was reached (at different level of depth) in 10 cases. Usually farmers prefer spending most of the time in the winery, delegating to others 292

the choices for the vineyard protection. Farm operators are, very often, not much motivated in learning and – unfortunately or luckily – “only those who want to learn, learn!” Factors that could slow down or hamper the future adoption of MD in Tuscany are mainly connected with socio-cultural conditions existing in the region: unlikely from the Trentino South Tyrol, in our region an individualistic approach is the rule; consortiums and producer associations are well organized for many aspects (marketing, brand promotion etc.) but cannot – or do not want – to have the control of the farms, at least for what concerns the vine growing and protection choices. Probably Tuscan growers are more strongly impacted by the perceived higher costs of MD, they are not prone to assume the risks initially associated with this technology and, last but not least, they are generally lacking in a structured and tight organization such as the cooperatives that have been the winning factor for the implementation of MD in the province of Trento, strongly pushing the members to the method adoption. In 2007 our team tried to be for Tuscan farms what the advisory service of the Istituto Agrario of San Michele all’Adige (IASMA) is for Trento farmers; IASMA consultants have provided and still provide a wise and free technical assistance to MD users. What is bizarre in our case is that the first Tuscan experience of “extension service” would not be possible without the funds of the dispensers Company CBC-Europe Ltd.

References

Bagnoli B., 1990: Incidenza delle infestazioni da artropodi e difesa dei vigneti in Toscana. – La difesa delle piante 13 (3-4): 89-112. Bagnoli B., Goggioli D., 1996: Application of mating disruption technique to control the grape moth Lobesia botrana (Den. and Schiff.) in Tuscany. – Proceedings of the XX International Congress of Entomology, Firenze, 25-31/08/1996, abstract (poster) 15- 194, p. 497. Bagnoli B., Lucchi A., 2003: European grapevine moth control in a Chianti vineyard by mating disruption technique. – IOBC/wprs Bulletin 26(8): 121-125. Bagnoli B., Cosci F., Santini L., Lucchi A., 2001: Mating disruption against Lobesia botrana in Tuscany: do local factors affect method efficacy? – IOBC/wprs Bulletin 24(7): 85-86. Bagnoli B., Goggioli D., Righini M., 1993: Prove di lotta con il metodo della confusione sessuale contro Lobesia botrana (Den. e Schiff.) nella zona del Chianti. – Redia 76(2): 375-390. Boselli M., Scannavini M., Melandri M., 2000: Confronto fra strategie di difesa contro la tignoletta della vite. – L'Informatore Agrario 56(19): 61-65. Charmillot P.-J., Pasquier D., 2000: Lutte par confusion contre les vers de la grappe: succès et problèmes rencontres. – IOBC/wprs Bulletin 23(4):145-147. Charmillot P.-J., Pasquier D., Bolay J.M., Jeanrenaud M., Zingg D., Zufferey E., 2000: Lutte par confusion et lutte classique contre les vers de la grappe dans les vignobles vaudois en 1999. – Revue suisse Vitic. Arboric. Hortic. 32(2): 83-88. Charmillot P.-J., Pasquier D., Verneau S., 2004: Efficacité larvicide de différents insecticides incorporés au milieu artificiel d’élevage sur les vers de la grappe. – Revue suisse Vitic. Arboric. Hortic. 36(3): 141-145. Ioriatti C., Lucchi A., Bagnoli B., 2008: Grape Areawide Pest Management in Italy. – In: Koul et al., Areawide Pest Management: Theory and Implementation, CAB Inter- national: 208-225. Ioriatti C., Bagnoli B., Lucchi A., Veronelli V., 2005: Vine moths control by mating disruption in Italy: results and future prospects. – Redia 87 (2004): 117-128. 293

Lucchi A., Bagnoli B., 2007: Seis años de interrupción del acoplamiento (confusión sexual) para el control de la polilla europea de la vid, en Toscana. – Ias Jornadas Internacionales sobre feromonas y su uso en agricultura, Murcia, 21 y 22 de noviembre de 2006, Consejeria de Agricultura y Agua, Comunidad Autónoma de la Region de Murcia (Eds.): 53-59. Pasquier D., Charmillot P.-J., 2000: Potentialité du métoxyfénozide, un agoniste de l'hormone de mue, pour la lutte contre l'eudémis Lobesia botrana. – IOBC/wprs Bulletin 23(4): 161-162. Stockel J.P., Schmitz V., Lecharpentier P., Roehrich R., Neumann U., Torres-Vila M., 1994: La confusion sexuelle chez l’Eudemis Lobesia botrana (Lepidoptera Tortricidae). Bilan de 5 années d’experimentation dans un vignoble bordelais. – Agronomie 2: 71-82. Varner M., Lucin R., Mattedi L., Forno F., 2001: Experience with mating disruption technique to control grape berry moth, Lobesia botrana, in Trentino. – IOBC/wprs Bulletin 24(2): 81-88. 294

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 295-300

Mating disruption technique vs Lobesia botrana (Denis & Schiffermüller) (Tortricidae): 3 years of experience in vineyards of Abruzzo (Italy)

Angelo Mazzocchetti, Antonio Zinni Regione Abruzzo - A.R.S.S.A. (Agenzia Regionale per i Servizi di Sviluppo Agricolo) - Via Nazionale, 38 - 65010 Villanova (Pe), Italy

Abstract: Since 2004 ARSSA is conducting a large scale trial (about 330 ha) with the application of mating disruption technique for the control of L. botrana. First three years results were positive on the whole in spite of the agronomical (cultivar, soil, etc.) and environmental (microclimate, wind, slope, etc.) variability typical of this grape growing area. Best results were achieved in large fields (>10 ha) with espalier training system or, anyhow, on training system lower than overhead. A correct dispensers application (number/ha and timing of application) was an important factor for the success of this method. Organic grape growing showed a favourable condition for the mating disruption technique thanks to a good agro-ecosystem equilibrium (greater abundance of predators and parasitoids). Some critical situations emerged in sloping vineyards and/or in presence of high pest populations but, at the end, mating disruption gave positive results with a generalized reduction of insect pest infestations in comparison with conventional treated ones. In view of a rational approach of plant protection in vineyards, that seems to be nowadays more and more complex, mating disruption could represent, also in Abruzzo, an interesting alternative to traditional techniques for an adequate protection from infestation of L. botrana, offering, at the same time, the respect of the environment, the protection of the consumer health without penalizing farmer income.

Key words: IOBC, mating disruption, viticulture.

Introduction

In this study we show the results obtained at the end of first three years (2004-’06) of the application of mating disruption (MD) method for the control of grapevine moth (Lobesia botrana) in Abruzzo (middle Italy). This region offers favourable conditions to apply this technique thanks to a wide viticultural territory. Nevertheless the variability of agronomical (cultivar, soil, training systems, etc.), environmental (microclimate, wind, slope, etc.) and structural (fragmentation of land holding) factors affect the results of MD application. For this reason we decided to evaluate the results obtained after a three years of employing of this techinque.

Material and methods

The total grape growing surface extension under mating disruption, mainly located in hilly areas with an altitude of about 300 meters above sea level, increased from 170 ha in 2004 to 330 ha in 2006 (figure 1). The maximum size of vineyards involved in the project was more than 30 ha. The Isonet L twist-tie dispensers manufactured by Shin-Etsu Chemical Co. (Tokyo, Japan) and imported by CBC (Europe) Ltd. were used for this experience (figure 2). Each dispenser contains 172 mg (guaranteed) of (E,Z)-7,9-dodecadienyl acetate.

295 296

Fig. 1. MD surface extension in Abruzzo (period Fig. 2. Isonet L dispenser used in the trials. 2004-2006).

The application rate was of 500 dispensers per hectare on espalier training system and 750 dispensers per hectare on overhead training systems with an increase of 8-10% of dispensers at edges on both training systems. Dispensers were applied in March, before the flight of over wintering generation of the target pest L. botrana. For each sampling unit (2-3 ha), assessments were done in 5 steps, through the acquisition of different parameters listed in figure 3. At least 100 bunches of grapes, randomly selected on 50 plants in the centre and at the edges of the vineyard, were inspected in order to determine the percentage (%) of infested bunches. Additional sprays, focused on 2nd and 3rd generation of the insect, were decided by evaluation of specific necessity. In places where it was possible, conventional vineyards located near to the MD vineyards were included as a control in order to compare data collected during the summer.

Fig. 3. Field assessment methods.

Results and discussion

70.000 bunches were sampled and collected data related to the harvest assessment are reassumed below. Vineyards at the third year of MD application gave positive results with 297

a generalized reduction of insect infestations. Best results were obtained in large fields (figure 4). Relevant results were achieved in middle Abruzzo, where infestation decreased from 30% at first year of application to almost zero at the third year, even reducing additional sprays number (figure 5). A significant reduction of infestation was also observed in vineyards with an higher insect pressure (southern Abruzzo), even if, in this situation, MD needs integration with insecticides. Nevertheless the technique provided a remarkable reduction of insect damages (figure 6). Moreover, the infestation rate was different between coastal and hilly vineyards, as well as in different altitude of the same farm: the infestation level was always higher at the hilltop sites due to the fact that pheromone is heavier than air and for this reason tends to drain down slopes (figure 6).

Fig. 4. MD fields in southern Abruzzo.

Fig. 5. Percentage of damages at harvest and number of treatments in MD vineyards in comparison to control ones in middle Abruzzo. 298

Fig. 6. Percentage of damages at harvest and number of treatments in MD coastal and hilly vineyards in southern Abruzzo.

Conversely, the infestation level in the centre of MD fields was always lower than along the edges because this area is less penalized by pheromone drift and external interference (edge effect) (figure 7). In many cases, minimum and maximum damage differential within the same vineyards notably reduced into three years to almost overlap with the average damage level (figure 8). This confirms that the continuative use of MD is an important factor to stabilize the results establishing a “flywheel effect” that gradually reduces the infestation levels. Finally we summarized the percentage of damages at harvest assessed in the sampling units during the three years project (figure 9). Average data show the decrease of the insect infestation over time and the reduction of additional sprays in MD vineyards in comparison with the conventional ones.

Fig. 7. Percentage of damages at harvest in the Fig. 8. Minimum, maximum and average % of sampling units compared with control plots. damages at harvest in middle Abruzzo.

299

Fig. 9. Percentage of damages at harvest in experimental fields in Abruzzo in the period 2004-2006: comparison between MD and control vineyards. Average number of insecticides/year is shown.

Conclusions

As shown in figure 10, the most significant results were achieved in large fields and in vineyards grown with an espalier training system, while some critical situations emerged in sloping vineyards and/or in presence of high insect pest populations. Mating disrupted vineyards gave positive results on the whole with a generalized reduction of insect pest infestations compared to conventional ones contributing to restore, in some cases, a good agro-ecosystem equilibrium (figure 11). Thus, data collected and evaluated in first three years project permit to express a positive judgement of the experience.

Fig. 10. Mating disruption performances Fig. 11. An example of an agro-ecosystem equilibrium in Abruzzo vineyards in the period 2004- restored in an organic farm in middle Abruzzo. 2006.

300

Acknowledgements

Special thanks goes to all technicians whose support enabled the realization of the project: M. Angelucci, D. D’Agostino, A..De Laurentiis, G. De Laurentiis, G. De Luca, L. D’Ercole, V. Di Camillo, O. Di Candilo, S. Di Candilo, M. Di Ciero, M. Di Domenico, L. Di Lello, D. Di Loreto, P. Di Paolo, A. Di Virgilio, R. Fattore, P. Fedele, S. Firmani, D. Giuliani, M. Lovato, A. Marcucci, N. Pellegrini, L. Pisani, M. Piucci, A. Puccella, A. Russo, L. Santoferrara, F. Savino, S. Tontodonati. Our thanks go to all wineries and farms who made available vineyards for experimental trials as well as CBC (Europe) Ltd. for the particular helpfulness.

References

Bagnoli B., Goggioli D. 1994: Cinque anni di applicazione della confusione sessuale contro la tignoletta della vite in una zona del Chianti. – Meeting “Innovazioni e prospettive nella difesa fitosanitaria”, Ferrara, Ed. Ist. Sper. Pat. Veg. Roma: 83-87. Bagnoli B., Lucchi A., Loni A., Santini L. 2002: Confusione sessuale contro Lobesia botrana in un’area viticola del Chianti. – Proceedings of Giornate Fitopatologiche, Baselga di Piné (Tn), vol. I: 437-444. Lucchi A., Giotti D., Bagnoli B. 2007: Efficacia della confusione sessuale contro la tignoletta. – Informatore Agrario 17: 58-61. Mazzocchetti A., Angelucci S., Casolani A., Di Lena B., Di Paolo E., Odoardi M. 2004: Applicazione del metodo della confusione sessuale nella difesa da Lobesia botrana (Denis & Schiffermüller) (Tortricidae) su vigneti allevati a tendone in Abruzzo. – Proceedings of: Giornate Fitopatologiche, Montesilvano (Pe), vol. I: 77-82. Scannavini M., Melandri M., Boselli M., Marani G. 2005: La confusione sessuale per il controllo della tignoletta. – Informatore Agrario 15: 81-84. Varner M., Lucin R., Mattedi L., Forno F. 1999: Confusione sessuale per controllare la tignoletta dell’uva. – Informatore Agrario 20: 81-84. Varner M., Mattedi L., Rizzi C., Mescalchin E. 2001: I feromoni nella difesa della vite. Esperienze nella provincia di Trento. – Informatore fitopatologico 10: 23-29.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 301-303

Daily “chores” of the Nearctic leafhopper Scaphoideus titanus Ball (Hemiptera Cicadellidae)

Valerio Mazzoni 1,2, Meta Virant Doberlet 3,4 , Luciano Santini 1 and Andrea Lucchi 1 1 Dipartimento Coltivazione e Difesa Specie Legnose “G. Scaramuzzi”, Sezione Entomologia agraria, Università di Pisa, Italia ([email protected]). 2 Safecrop Centre, Via Mach 1-88010 San Michele all'Adige (TN), Italy ([email protected]). 3 Department of Entomology, National Institute of Biology, Večna pot 111, SI-1000 Ljubljana, Slovenia. ([email protected]). 4 current address: School of Biosciences, Cardiff University, Main Building, PO Box 915, Cardiff, CF10 3TL, UK.

Abstract: In this paper we describe several daily chores performed by males and females of S. titanus, recorded with a laser vibrometer and a video camera. Among the activities carried out by the leafhopper during the 24 hours, some are strictly associated with reproduction (calling signal, courtship phrase and disturbance noise of the male and pulses of the female) while others are not closely related to mating (grooming, brochosome production and anointing, male aggressiveness and jumping).

Key words: bioacoustics, vibrational signals, brochosome anointing, jumping, male aggressiveness.

Introduction

Scaphoideus titanus Ball is a Nearctic leafhopper firstly recorded from Europe about fifty years ago, in the Southwest of France (Bonfils & Schvester, 1960). To date the species is present also in Italy, Switzerland, Slovenia, Croatia, Spain, Portugal and Serbia (Mazzoni et al., 2005). In its native region S. titanus is found on herbaceous vegetation, on different shrubs and trees (Maixner et al., 1993; Hill & Sinclair, 2000) and on the wild vine Vitis riparia. In Europe it is considered monophagous being exclusively associated with cultivated V. vinifera (Bonfils & Schvester, 1960; Vidano, 1964). The species does not cause any major direct damage to the grapevine but rather it is indirectly harmful as vector of a phytoplasma that causes Flavescence dorée (FD), one of the most threatening grapevine diseases in Europe (Bressan et al., 2006). As is known, mate recognition and localization in “Auchenorrhyncha” (with the exception of most cicadas) are mediated via acoustic signals transmitted through the substrate (Virant-Doberlet et al., 2006), whereas up to now there is no evidence that chemical communication plays a role in reproductive behaviour of these insects. With the aim to develop possible environmentally friendly control practices such as, for example, a “vibrational mating disruption”, we focused both on the acoustic behaviour of S. titanus recording, by the way and on several daily activities performed by one or both sexes during the 24 hours. In this paper we shortly described the stereotyped activities strictly associated with reproduction such as the calling song, the courtship phrase and the interference signal of the male and the simple pulses of the female in addition to some behaviours not necessarily related to mating, such as grooming, brochosome production and anointing, jumping and male aggressiveness.

301 302

Materials and methods

Tests were carried out on adults in an anechoic and sound insulated chamber (Amplifon Fa., Amplaid, Italy) at temperatures of 22-25°C and relative humidity between 70-75%. Males and females were placed on a cut grapevine stem with a leaf, inside a plexiglas cylinder (height 50 cm; diameter 30 cm) with a small opening at the top to let the laser beam pass through. The bottom of the stem was put into a vial filled with water to prevent withering and placed upright in a jar filled with moist artificial substrate. The behaviour of S. titanus was recorded with a video camera Canon MV1 miniDV camcorder. Vibrational signals were detected on the leaf lamina by the use of a laser vibrometer (OFV 353 sensor head with OFV-2200 vibrometer controller or PDV 100, both Polytec GmbH, Waldbronn, Germany). Signals were digitized with 48 kHz sample rate and 16-bit depth and stored directly onto a hard drive of a PC computer using Sound Blaster Audigy 4 sound card (Creative Labs Inc.) and Cool Edit Pro 2.0 (Syntrillium Software). Video recordings were transferred into the computer with Adobe Premiere 5.1. Signal recordings were analyzed using the computer software program Raven 1.2.1 with a FFT window length of 1024 samples.

Results and discussion

The vibrational communication connected with the mating strategy of the leafhopper is rather complex. The male repertoire includes a calling signal, a courtship phrase and a disturbance noise. Females emit pulses only in response to male signals, performing a duet with him. No other vibrational signals connected with mating behaviour were ever recorded from females. The establishment and the maintenance of the mating duet is strongly dependent from the respect of precise temporal features of vibrational signals which when distorted, inevitably led to the interruption of the communication and failure of localization of the female by male. All recorded vibrational signals have a dominant frequency below 900 Hz (Mazzoni et al., 2008). Grooming is performed several times a day: both genders use the legs to clean their antennae, wings and abdomen in a sort of stereotyped fast movements. Brochosome production is carried out by males and females, mainly during the night, in a typical vertical posture, with stylets inserted in the leaf tissues and legs lifted along the sides of their bodies. When emitted from the anal opening, brochosomes are immediately collected with the hind legs and distributed all over the six legs and the body surface in form of transparent fluid droplets. Male aggressiveness, consisting in a quick and strong abdominal stroke of a male against another male, was observed every time the latter positioned himself just behind the former, in a sort of pre-copulatory behaviour. Jumping is always preceded by a short phase of preparation, characterized by few repeated lateral movements of the insect along the longitudinal axis of its body, with bending of the fore legs.

Acknowledgments

The work has been supported by research program P1-0255 (Slovenian National Research Agency), by Fondi di Ateneo of Pisa University and by a grant of SAFE CROP Centre (IASMA). 303

References

Bonfils, J. & Schvester D. 1960: Les Cicadelles (Homoptera, Auchenorrhyncha) dans leurs rapports avec la vigne dans le Sud-Ouest de la France. – Annales des Epiphyties 3: 325- 336. Bressan, A., Larrue, J. & Baudon Padieu, E. 2006: Patterns of phytoplasma-infected and infective Scaphoideus titanus leafhoppers in vineyards with high incidence of Flavescence dorée. – Entomologia Experimentalis et Applicata 119: 61-69. Hill, G.T. & Sinclair, W.A. 2000: Taxa of leafhoppers carrying phytoplasmas at sites of Ash Yellows occurrence in New York state. – Plant Disease 84: 134-138. Maixner, M., Pearson, R.C., Boudon-Padieu, E. & Caudwell, A. 1993: Scaphoideus titanus, a possible vector of grapevine yellows in New York. – Plant Disease 77: 408-413. Mazzoni, V., Alma, A. & Lucchi, A. 2005: Cicaline dell’agroecosistema vigneto e loro interazioni con la vite nella trasmissione di fitoplasmi.– In: A. Bertaccini e P. Braccini (Eds), Flavescenza dorata e altri giallumi della vite in Toscana e in Italia. Quaderno A.R.S.I.A. 3, LCD srl., Firenze: 55-74. Mazzoni V., Prešern J., Lucchi A. and Virant-Doberlet M. 2008: Reproductive strategy of the Nearctic leafhopper Scaphoideus titanus Ball (Hemiptera: Cicadellidae). – Bulletin of the Entomological research (in press). Vidano, C. 1964: Scoperta in Italia dello Scaphoideus littoralis Ball cicalina americana collegata alla flavescence dorée della vite. – L’Italia Agricola 101: 1031-1049. Virant-Doberlet, M., Čokl, A. & Zorović, M. 2006: Use of substrate vibrations for orientation: from behaviour to physiology.– In: Drosopoulos, S. & Claridge, M. F. (Eds), Insect Sounds and Communication: Physiology, Behaviour and Evolution. Taylor & Francis Group, Boca Raton: 81-97. 304

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 305-307

Olfactory stimuli involved in host plant detection of Scaphoideus titanus Ball nymphs

Valerio Mazzoni1,2, Anfora G.1, Trona F.1, Lucchi A.2, Ioriatti C.1 1SafeCrop Centre and Plant Protection Department, IASMA Research Center, Via E. Mach 1, I-38010 San Michele a/A (TN), Italy; 2Dipartimento Coltivazione e Difesa Specie Legnose “G. Scaramuzzi”, Sezione Entomologia agraria, Università di Pisa, Italia ([email protected]).

Abstract: The responses of S. titanus nymphs to volatiles emitted by grapevine tissues (leaves and apical shoots) were studied. In first instance, behavioural tests were conducted with a self-made vertical glass Y olfactometer. Most nymphs showed significant preference for plant tissues over the blank control. Then, electroantennography experiments showed a response of the insect to plant odours. Headspace collections were made from fresh apical shoots with leaves and concentrated extracts were analyzed by coupled gas chromatography and mass spectrometry (GC–MS). The volatile fraction collected allowed the identification of several substances characterising the emission pattern released by the feeding sites of S. titanus.

Keywords: leafhopper, Vitis, olfactometer, electroantennography, headspace.

Introduction

Scaphoideus titanus Ball (Hemiptera Cicadellidae) is a Nearctic leafhopper, accidentally introduced into Europe, where it feeds only on Vitis spp. This species can transmit the phytoplasma of Flavescence dorée, one of the most threatening grapevine yellow-diseases (Mazzoni et al., 2005). It is documented that host selection and orientation in several leafhoppers are mediated by olfactory cues (Ranger et al., 2005; Patt & Sétamou, 2007). This study aimed to investigate on the response of S. titanus to volatiles emitted by grapevine tissues by behavioural and electrophysiological tests, in order to check whether odours can effectively drive the species to its host.

Material and methods

Insect rearing Eggs of S. titanus were hatched in climatic chamber (24±1°C, L16:D8, RH:75±5%) from canes collected in organic farms. Nymphs were reared in cylindrical plastic boxes (8X5cm). The base of the box was covered with a layer of technical agar solution (1.5%) on which was laid a disk of grapevine leaf, replaced twice a week; the top was covered by net. All tests were done with nymphs of 3rd-5th instars.

Olfactory tests Tests were conducted with a self-made vertical glass Y olfactometer (stem and arms 40 cm, diameter 5 cm). Charcoal-filtered air (0.3 l/ min) was forced through the apparatus. Each arm was connected to a glass flask (diameter 15 cm). Specimens were tested in groups of 5 or 6

305 306

for a maximum time of 1 hour. Either leaves (test 1) or apical shoots (test 2) were compared to the control (empty arm). The positions of the volatiles source and the control were switched around after each group was tested. In test 1 and 2 were employed respectively 35 and 22 specimens. Data were analysed using G-test (after William's correction).

Odour collection and chemical analysis Headspace collections (n = 3) were made from fresh shoots and leaves of Vitis riparia x rupestris 101/14. Ten shoots (ca. 100 g) were placed into a 25x38 cm polyacetate bag. Charcoal-filtered air was pumped into the bag at 0.15 l/min and over a Porapak Q cartridge containing 50 mg of adsorbent (Sigma-Aldrich, Milan, Italy). Collections were made for 6 hours in a climatic chamber at 25±2 ºC, 60±10% RH, 1000 lux. Volatiles were desorbed by eluting the cartridge with 500 µl of redistilled hexane. Extracts were concentrated to 50 µl with a slow stream of nitrogen, sealed in glass vials and stored at -80 ºC. Five samples of extracts were analysed by coupled gas chromatography and mass spectrometry (GC–MS). Analyses were performed on a Hewlett-Packard 5890 GC, with a polar Innowax column (30 m x 0.32 mm; J & W Scientific, Folsom, CA) programmed from 60°C (hold 3 min) at 8°C min-1 to 220°C (hold 7 min), interfaced with a Hewlett-Packard 5970B MS with electron impact ionization (70 eV). The identity of most compounds in volatile collections was verified by comparison with synthetic compounds purchased from Sigma-Aldrich and Fluka Chemie (Buchs, Switzerland).

Electroantennography (EAG) Alive insects were placed in plastic pipette tips with head and antennae protruding from the cut-off tip. The ground electrode was an electrolytically sharpened (in 10% NaNO2 solution) tungsten wire inserted into the insect’s head. The different electrode was a glass capillary pipette, filled with Kaissling solution containing 5 g l-1 polyvinylpyrrolidone K90 (Fluka Chemie), brought into contact with the distal end of the antennae. Each stimulus was prepared by absorbing 20 µl of the concentrated plant extract on a piece of filter paper (1.5 cm2) inserted into a Pasteur pipette. Hexane (20 µl) was applied to filter paper to serve as a blank. The interval between two next stimulations was 2 minutes. The antennal signals were amplified and recorded by Syntech (Hilversum, The Netherlands) software. Mean (n=5) responses (mV) to the plant extract were compared to those of the blank by unpaired Student’s t-test.

Results and discussion

Results of olfactory tests are reported in table 1. In test 1, 60.0% of nymphs showed preference for leaves, 25.7% for the control and 14.2% did not choose. In test 2, 81.8% of nymphs chose the way to the shoots whereas only 9.1% chose the control, as much as the specimens that did not make a decision. In both cases the choice of vine tissues was significant (test 1: G = 4.86, P < 0.05, d.f. = 1; test 2: G = 14.36 P < 0.001, d.f. = 1). Antennae of S. titanus nymphs gave a significantly higher EAG response to the plant extract than the blank control (t = 4.9, P < 0.01, d.f. = 6). The volatile fraction collected allowed the identification of 25 substances from the headspace that characterise the emission pattern released by the feeding sites of S. titanus. The extract was composed of monoterpenes (8), sesquiterpenes (7), aromatics (3), esters (3), hydrocarbons (1), alcohols (1), ketones (1) and irregular terpenoids (1). In conclusion, S. titanus nymphs were shown to respond to compounds emitted by grapevine tissues, such as leaves and apical shoots. Such host plant volatile compounds are 307

able to elicit both antennal and behavioural responses. In particular, apical shoots, which in our tests were chosen by more than 80% of the specimens, can be considered as an important source of attractive volatiles for young stages of the species. It descends that odours can constitute important cues in host selection of S. titanus. This knowledge may point the way toward further research, such as, for instance, the identification of single or blended active compounds potentially employable in field in pest control strategies.

Table 1. Responses (%) of S. titanus nymphs in controlled trials in a Y tube olfactometer. Test 1: leaf versus air control comparison; test 2: apical shoot versus air control (Williams corrected G-test).

Volatile source Control (air) No choice G-test Test 1 60.0% 25.7% 14.3% G = 4.856 P < 0.05 Test 2 81.8% 9.1% 9.1% G = 14.363 P < 0.001

Acknowledgements

We are grateful to thank Umberto Malossini and Luca Zulini (IASMA) for providing the grapevines, Silvia Carlin (IASMA) for chemical analysis and Elisabetta Leonardelli (IASMA) for technical assistance. The research was funded by Autonomous Province of Trento (SafeCrop and HOST Projects).

References

Mazzoni, V., Alma, A. & Lucchi, A. 2005: Cicaline dell’agroecosistema vigneto e loro interazioni con la vite nella trasmissione di fitoplasmi. – In: Flavescenza dorata e altri giallumi della vite in Toscana e in Italia, Quaderno A.R.S.I.A. 3, LCD srl., Firenze eds. Bertaccini and Braccini: 55-74. Patt, J.M. & Sétamou, M. 2007: Olfactory and visual stimuli affecting host plant detection in Homalodisca coagulata (Hemiptera: Cicadellidae). – Environ. Entomol. 36: 142-150. Ranger, C.M., Winter, R.E.K., Backus, E.A., Rottinghaus, G.E., Ellersieck, M.R. & Johnson, D.W., 2005: Discrimination by the potato leafhopper (Hemiptera: Cicadellidae) of host volatiles from resistant and susceptible Alfalfa, Medicago sativa L. – Environ. Entomol. 34: 271-280. 308

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 309-319

Comparison between downy mildew fungicides in a vineyard of the Avellino province (Campania South Italy) and their influence on the population of Phytoseid mites (Parasitiformes, Phytoseiidae).

Mariano Nicòtina, Giulio Cesare Capone Department of Agricultural Science Entomology and Zoology, University of Naples “Federico 11”, Via University, 100 80055 Portici (NA) e-mail: [email protected]

Abstract: Numerous studies have shown that the treatment of anti downy mildew (control) with copper based products, widely used in viticulture, are one of the causes of high levels of contamination in the soil. However, in the ecosystem of the vineyard, copper based products have always proved (very) efficient against (anti) downy mildew and are, in fact, the only products authorized for biological production in viticulture. For three consecutive years (2005, 2006 and 2007) experiments have been carried out in a vineyard located in Montefredane at 350 meters above sea level on a soil of medium consistency where a 24 year cultivar Fiano of Avellino is grown. The experiment puts in evidence the differences of copper content in the cultivated soil of the vineyard and the soil in the immediate surroundings. The objective of the test was to sample the selectivity of some low dosage copper or copper based products on Phytoseiid mites. At the same time these results were compared with the effects of products organic of synthesis commonly used in viticulture. Furthermore, the effectiveness of the active substances in controlling the downy mildew was tested. The higher presence of Phytoseiids in the vineyard were Kampimodromus aberrans (Oudemans). None of the plots revealed total eradication of the Phytoseiids from the vineyard even if the products tested showed different degrees of selectivity as well as different levels of control of the downy mildew. As regards downy mildew control the most efficacious plots were those treated with Cu- oxychloride and Zoxamide + Cu-oxychloride.

Key words: powdery mildew, selectivity, phytoseiids, copper, biological production

Introduction

In recent years downy mildew (Plasmopara viticola (Berk, et Curt) Berlese et De Toni) of grapes has become a dreaded cryptogam also in the vineyards of Central and Southern Italy. This situation has been triggered off largely because of the changing rainfall conditions. In addition there is a greater than ever awareness in Southern Italy regarding the importance of the choice of the location of vineyards for the production of high quality wines. In fact there is a growing trend to plant vineyards in hilly areas where the rainfall in spring and summer is frequent and profuse. Pesticides for downy mildew control disturb and influence the population of useful arthropods (Nicòtina, 2003). In fact, phytoseiid mites, internationally recognized as the most efficient deterrent against various families of both phytophagous mites and some insects, are also valuable biological indicators (Athias-Henriot, 1975; Zacharda, 2001). Indeed, their existence in agro ecosystems guarantee the use of less toxic pesticides for downy mildew control. After only one century from their introduction in the viticulture field, copper based products have proved to be valid in the control of downy mildew. Furthermore, they have encouraging collateral effects against other diseases of the vine.

309 310

Besides, in biologically run vineyards copper based products are the only defence strategy against downy mildew. The use of copper in viticulture dates way back and its positive effects have been acknowledged; in fact it does not select strain of resistant fungicides, it favours the desiccation of rachilla, it has a secondary activity against powdery mildew and botrytis, it is non toxic to useful organisms that predate harmful mites (little red spiders, etc) and in dry weather conditions it is discreetly resistant on vegetation. The likely inconveniences caused by an excessive use of copper in the vineyard are connected to the fact that copper can be deposited on the actual grapes, on the neighbouring vegetation and in part falls directly onto the surrounding ground (Mescalchin e Pertot, 2003; Egger e D’Arcangelo, 2004). The objective of this research is to study the efficacy of the means used to control downy mildew fungicides and selectivity on Phytoseiids of some pesticides used in the vineyard for the control of downy mildew, among which adjuvant copper molecules to reduce quantities and create an equally efficient alternative using new products for pesticide control (Scannavini et al., 2003; Cravero et al., 2004).

Materials and methods

This research has been carried out in the province of Avellino in the Vadiaperti farm of the Montefredane vineyards. These vineyards are located in typical areas of high quality wines produced in traditional areas of the Irpinia region (VQPRD). The vineyard covers an area of 2 hectares on a medium clayey soil. It is situated in a hilly location at an average altitude of 350 metres above sea level with a south-eastern exposition. As mentioned before, pesticide treatment with cupric based products represents the major cause of copper contamination of the soil (Mazzini et al., 2003). When the soil is heavily impregnated with this metal there is a negative impact on the soil fauna causing a reduction of carabidae and earth worms (Paoletti and Bertocelli Brotto, 1985). Moreover, there is an alteration of the microbiological and enzymatic composition of the soil leading to acidification. Previous to research, tests were carried out to check the copper levels of the soil in the vineyard, the adjacent hazelnut grove and the nearby grazing field.

Table 1. Differences of copper content among cultivated soils. Clearly shows the relationship between the copper content and the different soils analysed.

Analysed Soil Copper content

Experimental vineyard (25 years) 112 mg Cu/Kg Neighbouring Hazelnut grove (20 years) 90 mg Cu/Kg

Grazing-field 40 mg Cu/Kg

The results are undoubtedly due to pesticide treatment and can in no way be underestimated. For three consecutive years (2005-2007) tests for some downy mildew control were carried out in the company vineyards with the aim of testing the selectivity of the population of useful mites of the Phytoseiidae Family and assess their ability in controlling certain cryptogams. 311

For the experiments, on the basis of the calendar treatment to be effected, a trial field of about 2000 sq m of the Vadiaperti vineyard was chosen. For this test each plot underwent a specific calendar treatment. Applications were effected with a motor pump sprayer manually. A mixture, previously calibrated, equal to about 1000 litres/ha was applied to the plots and after each treatment the quantity distributed was carefully checked (EPPO Bulletin, 1988). To avoid the drifting effect treatment was only effected on the central rows of the plots. Spraying was carried out by wetting both sides of the vine rows up to drip level. Each application was monitored by a technician who controlled the application methods with the objective of reducing the danger of the water drifting to the adjacent rows (Nicòtina et al., 2002). To control the Phytoseiids, leaf samples were taken immediately before each application with the exception of the last sampling when spraying was not effected. For each sample, 40 leaves were collected at random from the middle rows of the plot and the leaves were picked from the central part of the plants. This at 10 to 12 day intervals according to the plot. The leaves were placed in sealed plastic bags and immediately taken to the laboratory where the Phytoseiids were collected by shaking the leaves (frappage method, Burgio, 1999) on to a black tray. Samples were then clarified in Nesbitt liquid, mounted on slides with Hoyer liquid and successively put in an oven for 5-6 days at a temperature of 30°C. Ultimately, the samples were observed using a phase contrast microscope to classify and identify the species (Tsolakis & Ragusa, 1999). The experimental field consisted of 14 rows with an average of 20 vine plants, the rows were divided into two sampling stations (repetition) of 10 plants each, recognized by an Arabic number and by a letter of the alphabet, which indicated the plot and the repetition; therefore for each of the 7 plots there were 4 repetitions. In this test, new commercialized products were also used as well as products authorized in biological agriculture made up of a mixture of seaweed and copper and copper “gluconate” (Tab 2). To evaluate the data the untreated plot (7) was sprayed with water alone and represents negative control, whereas for positive control azinphos methyl was used in Plot 6 in order to have a plot wherein the products were absolutely non selective to Phytoseiids.

Table 2. Programme of application of downy mildew products from 2005-2007

Rate a.i. Dose Copper/Ha* Plots Active ingredients % (gr/ml/100lt) (gr) 1 Copper oxychloride 50 300 1500 2 Zoxamide + Copper oxychloride 4.3 + 28.6 350 1000 3 Copper Gluconate 8 300 240 4 Seaweed + Copper oxychloride 10 + 50 300 + 200 1000 5 Folpet 80 150 - Azinphos methyl + Copper 6 25 + 35 150 + 200 700 oxychloride 7 Untreated - - -

The quantity of copper per hectare was calculated considering the use of 10 hl of water per hectare

312

As revealed from the above table, copper based pesticides determine different ratios according to the percentage of the metal content and the doses that are applied. For example, Copper oxychloride contains a percentage of 50% copper and consequently an application effected in the vineyard distributing 10 hl/Ha determines a ratio of 1,5 kg of copper per hectare of the plot. The results of the various tests were analysed for variances with the Duncan test (P=0,05). All the procedures were effected with the Statistics programme version 6.

Results

During the three years of the experiment the species of Phytoseiids found in the vineyard of Montefredane (AV), in Campania, was Kampimodromus aberrans (Oudemans) (Ragusa di Chiara and Tsolakis, 1994). However, other species were also sporadically found.

2005 Test The graph 1 clearly shows the trend of the population of Phytoseiid mites. As can be seen the population was somewhat low at the preliminary control on 24th May. However, right from the first application on 31st May there is a reduction in the population of phytoseiids in all the plots but this is particularly noticeable in Plot 6 (Azinphos methyl + Copper oxychloride) in which the variation is more apparent. On the other hand, it must be considered that the decrease of Phytoseiids is also connected to environmental factors. During the experimental period, in fact, there were abundant rainfalls which presumably contributed to the reduction of the population. The statistical elaboration indicates significant differences among the plots sampled in terms of selectivity (F=6,45; p<0,0001).

Downy mildew 2005 18

6 average number of phytoseiids of number average

0 24- 2 28- 30-5-051-6 3- 5 7- 9-6-0511 1 15- 17-6-0519 21- 2 25- 27- 29 1- 3- 5-7-057- 9-7 1 13-7-0515- 17- 1 21- 23- 25- 27 29- 31-7-052-8-05 6-5 -6-05 3-6 3-6 1-7 9-7 6-05 6- -6 -6 -6 7-05 7- 7-05 -7 5 5 -05 0 6 6 6 6 0 -05 7 7 7 7 7 7 -05 -05 -05 5 -05 -05 -05 -05 -0 -05 -05 -05 -05 5 -05 -05 -05 -05 -05 -05 -05 -05 -0 5 5

Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Plot 7

Fig. 1. Trend of Phytoseiid Population during downy mildew control.

313

In particular, it has been noticed that plot 5 (Folpet), 2 (Zoxamide + Copper oxychloride), 1 (Copper oxychloride), 4 (Seaweeds + Copper oxychloride and 3(Copper Gluconate) are statistically all the same and generally show good selectivity. Furthermore, as expected, Plot 7 (untreated) was more selective whereas plot 6 (Toxic) was the worse in terms of selectivity (Graf. 2). As far as the efficacy of the treatment in all the various plots in the control of downy mildew, as shown in Table 3, the pesticide has not caused serious damage to the foliage; the lowest value was recorded in plot 5 (Folpet) and 2 (Zoxamide + Copper oxychloride); followed in decreasing order by plot 1 (Copper oxychloride), plot 6 (Azinophos methyl + Copper oxychloride, plot 4 (Seeweed + Copper oxychloride), plot 3 (Copper Gluconate) and plot 7 (untreated) (Table 3). Plot 2 and 5 also showed a greater efficacy regarding the protection of the bunches of grapes, with 0% hit and 0% of single grapes attacked, followed in decreasing order, by plot 1, plot 6, plot 4 and lastly plot 3 achieving better results than the untreated plot.

Selectivity 2005

average number of phytoseiids 2

1 a ab ab ab ab b c 0 Plot 7 Plot 5 Plot 2 Plot 4 Plot 1 Plot 3 Plot 6

Fig. 2. Selectivity on phytoseiids mite of the different pesticides (year 2005) (The average marked with the same letter do not differ amongst each other statistically according to the Duncan test (P=0,05))

2006 Tests The initial population of phytoseiids found in the vineyard was very low. In fact, on average not more than 6 phytoseiids per sampling were collected. However, towards the last decade of May a variation was recorded in the population of these mites. Only the untreated plot, which had initially presented a steady trend not having undergone any interferences, showed an increase in the populations from the second week (Graph 3).

314

Table 3. Efficacy of downy mildew control treatment 2005

% leaves % foliage % bunches %grapes Plots Active ingredients hit attacked hit attacked 1 Copper oxychloride 9 a 12 c 2 b 5 b Zoxamide + Copper 2 8 a 7,5 a 0 a 0 a oxychloride 3 Copper Gluconate 15 b 8 ab 5 c 7 c Seaweed + Copper 4 14 b 6 a 3 b 6 bc oxychloride 5 Folpet 7 a 6 a 0 a 0 a Azinfos methyl + Copper 6 10 a 10 bc 3 b 5 b oxychloride 7 Untreated 18 c 15 d 10 d 10 d

(The average marked with the same letter do not differ amongst each other statistically according to the Duncan test (P=0,05))

Downy mildew 2006 14

average number of phytoseiids of number average 4

0 16- 23-5- 30 6- 13-6-0 20-6-0 27-6- 4- 11 18-7-06 25-7-0 1- 8- 15- 22 29 5- 6 7 8 8 9 5 -5- -06 -0 -7- -06 -0 8 -8- -8- -06 -06 06 06 06 6 06 6 -06 06 06 6 6 6

Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Plot 7

Fig. 3. Trend of the Phytoseiid Population during the 2006 tests

The statistical elaboration showed significant differences between the various plots (F=15,48 P<0,0001). Considering this the experiment continued in order to determine the most selective plot amongst those tested. The results obtained at the Duncan test, as graph 4 shows, reveal that the plots all show discreet selectivity, plot 6 (Azinphos methyl + Copper oxychloride) confirmed the expected toxicity. 315

As far as the efficacy of products for downy mildew control is concerned, it was noticed that during the 2006 vegetative season downy mildew showed very low infection rates. The best results were obtained by plot 2 (Zoxamide + Copper oxychloride), followed in decreasing order by plot 5 (Folpet), 4 (Seaweeds + Copper oxychloride), 3 (Copper gluconate), 1 (Copper oxychloride) and plot 6 (Azinphos methyl + Copper oxychloride). As shown in Table 4, the worst damage due to downy mildew was verified in the plot treated by water alone (plot 7). Selectivity 2006

average number of phytoseiids 2

1 a b b b b b c 0 Plot 7 Plot 5 Plot 3 Plot 2 Plot 1 Plot 4 Plot 6 Fig. 4. Selectivity on phytoseiids mites of the different pesticides (year 2006) (The average marked with the same letter do not differ amongst each other statistically according to the Duncan test (P=0,05))

Table 4. Efficacy treatment downy mildew control 2006

% leaves % foliage % bunches % grapes Plots Active ingredients hit attacked hit attacked 1 Copper oxychloride 9 bc 25 c 6 c 15 c

2 Zoxamide + Copper oxychloride 5 a 6 a 2 a 4 a

3 Copper Gluconate 8 b 22 bc 4 ab 16 c

4 Seaweed + Copper oxychloride 7 b 20 b 5 b 12 b

5 Folpet 4 a 6 a 3 a 6 ab Azinphos methyl + Copper 6 10 c 18 b 5 b 14 c oxychloride 7 Untreated 18 d 38 d 8 d 25 d

The average marked with the same letter do not differ amongst each other statistically according to the Duncan test (P=0,05) 316

2007 Tests As Graph 5 shows, before the tests the population of phytoseiids in the vineyard was quite low and constant in all the plots; in fact the first sampling oscillates around 2-4 phytoseiids. With the applications the population did not submit negative influences and manifested a very similar trend in all the various plots, with the exception of plot 6 (Azinphos methyl + Copper oxychloride) which was toxic. In particular, during the period of applications the population of phytoseiids underwent two variations, the first was recorded at the beginning of June and the second after about a month around the 1st July. After having statistically analysed to verify the absence of significant differences among the repetitions of the same plot in terms of selectivity, the tests continued to determine whether there were significant difference among the various sampled tests. The results obtained show highly significant results among the plots tested (F = 16,48 p< 0,00001). Comparing the results at the Duncan test, plot 3 (Copper gluconate) recorded a very low selectivity to phytoseiids, in fact just a little less than plot 7 which had been treated with water (untreated) (Graph 6).

Downy mildew 2007 12

4 average number of phytoseiids of number average

0 18-5- 20- 2 2 2 28-5-0730-5-071- 3-6 5-6 7 9- 11-6-0713 15 17-6-0719- 21- 23-6- 25-6-027-6-029-6-01-7-073-7-075-7 7- 9- 11-7- 1 15-7-017-7-07 2- 4- 6- -6 3- 6 6 - - 7 7 5- 5- 5- 5- -07 -07 -07 -07 -07 6- 6- 6 6 -07 -0 -0 7- 07 07 07 07 07 07 07 -07 -07 07 7 7 07 07 7 7 7 7 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Plot 7 Fig. 5. Trend of the Phytoseiid Population during the 2007 tests

Plot 2 (Zoxamide + Copper oxychloride), 1 (Copper oxychloride) and 5 (Folpet) show adequate selectivity; less selectivity was recorded in Plot 4 (Seaweed + Copper oxychloride). As expected during the planning phase, the less selective plot was 6 (Azinphos methyl + Copper oxychloride). At the end of the season the efficacy of the defence strategies for the control of downy mildew was assessed. As far as the efficacy of the applications carried out in the various plots in the control of downy mildew as illustrated in table 5, the pesticide used did not determine any serious damage to the foliage. The best results were recorded in plot 2 (Zoxamide + Copper 317

oxychloride), where leaves attacked were nearly always less than 2 % and infection was low, also thanks to the curative action of the active ingredients used. Good result were also obtained in plot 5 (Folpet). In decreasing order follow, plot 1 (Copper oxychloride), plot 4 (Seaweed + Copper oxychloride), plot 6 (Azinphos methyl + Copper oxychloride) and plot 3 (Copper Gluconate). Plot 2 showed the greatest efficacy regarding the protection of bunches, with 0 % of bunches and 0 % of grapes hit, and plot 5, followed, always in increasing order by plot 4, plot 3, plot 6 and finally by plot 1, which recorded slightly better results than the untreated plot.

Selectivity 2007

2 average number of phytoseiids

1 a ab bc bc bc c d 0 Plot 7 Plot 3 Plot 2 Plot 1 Plot 5 Plot 4 Plot 6

Fig. 6. Selectivity on phytoseiid mites of the different pesticides (year 2007) (The average marked with the same letter do not differ amongst each other statistically according to the Duncan test (P=0,05))

Considerations and conclusions

The different strategies adopted have allowed an analysis of the impact of the fungicides used both in controlling the diffusion of downy mildew and the populations of species of predatory mites found in the experimental field, and in particular on K. aberrans (Ragusa and Ciulla, 1991; Nicòtina, 1998a, b; Nicòtina & Cioffi, 2000; Nicòtina &Cioffi, 2001). It is important to highlight that the species mostly present was found constantly, with different trends according to the calendar treatments used. In no case, however, was any species totally eradicated from the vineyard, confirming the resistance and the tolerance of the predators to the pesticides used (Nicòtina et al., 2002). The research results not only give indications connected to the selectivity of useful mites but also favourable indications on the efficacy of the active ingredients tested for downy mildew control. Copper based products are still today, after a century from their introduction as pesticides, valid tools in the fight for downy mildew control and have a secondary effect against some pathologies in viticulture. 318

Table 5. Efficacy of downy mildew control 2007

% leaves % foliage % bunches % grapes Plots Active ingredients hit attacked hit attacked 1 Copper oxychloride 6 b 22 c 5 b 11 c

2 Zoxamide + Copper oxychloride 2 a 4 a 0 a 0 a

3 Copper Gluconate 8 c 22 c 4,5 b 10 c

4 Seaweed + Copper oxychloride 7 bc 20 c 4 b 9,5 c

5 Folpet 3 a 5 a 1 a 2 b

Azinfos methyl + Copper 6 8 c 15 b 5 b 10 c oxychloride

7 Untreated 14 d 35 d 8 c 20 d

(The average marked with the same letter do not differ amongst each other statistically according to the Duncan test (P=0,05))

A fundamental argument in favour of limiting the use of cupric products is that this metal accumulates in the soil causing devastating effects. In the meantime, it is indispensable to alternate the use of copper based products with other non cupric pesticides; there are numerous synthetic molecules which can be used as an alternative to copper based products but there is very little experience relative to “biological” goods used as an alternative to copper based products. The need is felt to reduce the presence of copper and to have a mixture that is efficient against downy mildew but at the same time selective for useful arthropods. In conclusion research results show that the plots with the greatest efficacy in 2005 and 2007 are Zoxamide + Copper Oxychloride and Folpet. In 2006, when there were more frequent rainfalls, the best outcome was recorded for the mixture Zoxamide +Oxychloride.

References

Athias-Henriot, C. 1975: Nouvelles notes sur les Amblyseiini. II – Le relevé organotaxique de la face dorsale adulte (Gamasida protoadeniques, Phytoseiidae). – Acarologia 17 (1): 20-29. Burgio, G. 1999: Tecniche di campionamento e raccolta di insetti ed acari entomofagi. – Inf. Agr. 12: 3-18. Cravero, S.; Ferrari, D., Crovella, P., Bassignana, E. 2004: Confronto tra diversi fungicidi rameici impiegati a basso dosaggio contro Plasmopara viticola con lo scopo di ridurre l’apporto di rame in viticoltura biologica. – Atti Giornate Fitopatologiche 2: 171-176. Egger, E.; D’Arcangelo, M. 2004: Strategie di difesa antiperonosporica per una riduzione degli apporti di rame nel vigneto. – Atti Giornate Fitopatologiche 2: 177-184. EPPO 1988: Guideline for the biological evaluation of fungicides. – EPPO Bulletin 18: 605- 612. 319

Mazzini, F.; Rossi, R., Spada, G., Scannavini, M. 2003: Il rame nella difesa delle colture alla luce delle limitazioni di impiego. – L’Informatore Agrario 59 (14): 75-79. Mescalchin, E.; Pertot, I. 2003: La riduzione del rame in viticoltura biologica. – Bioagricoltura 81: 27-29. Nicòtina, M., Cioffi, E., Capone, G.C. 2002: Defence strategies with fungicides and impact on Typhlodromus exhilaratus Ragusa, Kampimodromus aberrans Oudemans and Phyto- seius finitimus Ribaga sensu Denmark (1966) (Parasitiformes, Phytoseiidae) populations in Italian vineyards. – XI International Congress of Acarology, Merida (Mexico): 8-13 September 2002: 241-242. Nicòtina, M. 2003: Side effects of some fungicides on the population of Typhlodromus exhilaratus Ragusa and Phytoseius finitimus Ribaga sensu Denmark (1966) (Parasiti- formes, Phytoseiidae) on grapevines in Tuscany. – Advanc. in Horticul. Science 17 (2): 72-76. Ragusa di Chiara, S.; Tsolakis, H. 1994: Revision of the genus Kampimodromus Nesbitt, 1951 (Parasitiformes, Phytoseiidae), with a description of a new species. – Acarologia 35: 305-322. Scannavini, M.; Spada, G., Mazzini, F., Bortolotti, P. 2003: Efficacia antiperonosporica di vari composti rameici a basse dosi e aspetti qualitativi delle uve. – Informatore Fito- patologico 59 (15): 69-72. Tsolakis, H.; Ragusa di Chiara, S. 1999: Overwintering of phytoseiid mites (Parasitiformes, Phytoseiidae) on hazelnut (Corylus avellana L.) in Sicily (Italy). – In: Bruin, Van Der Geest, Sabelis (eds.): Ecology and Evaluation of the Acari, Kluwer Academic Publiscers: 625-635. Zacharda, M. 2001: Predatory phytoseiid mites (Acari: Phytoseiidae) as bioindicators of stress impact on a farmland and butresses of the farmland revival. – Ekológia (Bratislava) 20 (1): 47-56. 320 Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 321-323

Population dynamics of grapevine moths reflect weather conditions over the past decade

Denis Pasquier, Pierre-Joseph Charmillot, Patrik Kehrli Station de recherche Agroscope Changins-Wädenswil ACW, CP 1012, CH-1260 Nyon, Switzerland

Abstract: The two grapevine moths Eupoecilia ambiguella and Lobesia botrana are the most important pest insects in Swiss vineyards. As a result of their different climatic preferences, the distribution and proportion of these two moths varies gradually in space and time. Only 25 % of pest attacks were caused by E. ambiguella in 1996, the start of this long-term survey. But its importance increased steadily, hitting a peak of 97 % of damage caused in the end of 2001. After 2003, the hottest summer ever recorded, E. ambiguella was hardly detectable in Western Switzerland and even today its relative frequency is low. In this paper we aim to identify factors that explain the population dynamics of the two grapevine moths. The most important factor seems to be the population size of the previous generation, followed by climatic factors such as humidity and temperature that explain growth conditions during larval development. Besides, wind and the length of the flight period have a significant effect on the magnitude of insect flight. In conclusion, the population dynamic of grapevine moths is strongly affected by climatic factors and extreme weather conditions such as droughts. A profound knowledge of grapevine moths' population dynamics is indispensable for the success of sophisticated and environmental friendly insect pest management schemes such as mating disruption.

Key words: viticulture, Lobesia botrana, Eupoecilia ambiguella

Introduction

The European grapevine moth, Lobesia botrana (Den. & Schiff.) (Lep., Tortricidae), and the European grape berry moth, Eupoecilia ambiguella Hübner (Lep., Phaloniidae), are the two main insect pests in Swiss vineyards. Their frequency and damage increased significantly between 1996 and 2000 (Fig. 1). In 2000, damage exceeded for the first and only time the level of 10 %. Ever since pest pressure clearly decreased, with the exception of 2003, the hottest summer ever recorded in Switzerland (Charmillot et al., 2005). However, it has to be noted that the population density of L. botrana remained nearly constant over the period of this survey (Fig. 1b). Differences in grape damage are therefore mainly explained by fluctua- ting E. ambiguella populations. These differences are also reflected by the proportion of the two moth species to each other. In 1996, only 25 % of grape damage was caused by E. ambiguella (Fig. 1b). However, its importance steadily increased over the following years, hitting a peak of 97 % of attacks caused in the end of 2001. However, after 2003 damage by E. ambiguella was hardly detectable and even today it is still low. Whereas L. botrana favours hot and dry ecotopes, E. ambiguella shows a marked preference for cool and humid weather conditions (Charmillot et al., 2005). These preferences are probably explained by the biogeo- graphic history of the two moths. E. ambiguella is known as a local grape pest since the Middle Ages, whereas L. botrana has not been observed in Central Europe before the end of the 19th century. In summary, the distribution and proportion of these two moths varies gradually in space and time and there. Moreover, there is only a poor correlation between the flight intensity of grapevine moths and actual grape damage measured in adjacent vineyards (Fig. 1).

321 322

Fig. 1. Population dynamics of grapevine moths in Western Switzerland over the past decade: a) flight intensity of males and b) grape damage by E. ambiguella and L. botrana.

In this paper we aim to identify and quantify the effect of the most important factors explaining the population dynamics of the two grapevine moths. We made use of data gained in a long-term survey. By stepwise regression we analysed the impact of several biological and climatic factors on the population cycle of the two moths. Here we outline first results of our preliminary analysis.

323

Material and methods

Agroscope Changins-Wädenswil has followed the flight of E. ambiguella and L. botrana males in Western Switzerland since 1992. Pheromone traps were installed at a total of 19 locations. Traps were checked at least once a week. Besides, grape damage was assessed in more than 50 vineyards at the end of the first and second generation of grapevine moths since 1996. A subsample of larvae was identified and the ratio of the two species to each other was calculated. Data were subsumed per species, location, year and generation. The values received were related to biological and climatic factors. Population density in the previous generation was used as reference. Climatic factors comprised temperature, humidity, precipitation and wind speed. The length of each generation adumbrated the biology of grapevine moths. The dependence of E. ambiguella and L. botrana abundance upon these biological and climatic factors was analysed by multilinear stepwise regressions.

Results and discussion

A first analyse over the first two generations of the two species pooled indicated that the starting population size, the length of the flight period, temperature, humidity and wind speed had a major effect on the number of E. ambiguella and L. botrana males caught (R2 = 0.61). Whereas population density in the previous generation together with the length of the flight period and temperature during larval development had a positive effect, humidity and wind speed decreased the number of flying moths. However, such an analysis over the whole dataset is only of limited value because differences between the two species and between the spring and summer generation are ignored. Focusing on the 1st generation of E. ambiguella we see that the size of the starting population as well as weather, in particular rain and wind, affect flight intensity (R2 = 0.75). Similar is true for the 2nd generation of E. ambiguella (R2 = 0.75). However, hibernating pupae of L. botrana seem to be more sensitive to frost and pathogens, but its flight is nevertheless determined by the population density in the previous generation and the length of the flight period (R2 = 0.74). However, humidity has a positive effect on larval development in summer (R2 = 0.68). Our results correspond quite well with the biology of the insects. Moreover, our preliminary analysis does not only identify well-known key factors, it also quantifies their impact on male flight. Quite surprising was the goodness of fit of our rudimentary analysis. Upcoming analyses are hopefully also shedding light on the key factors determining the rate of damage caused by E. ambiguella and L. botrana. Overall, a better knowledge on grapevine moths' population dynamics may further boost environmental friendly insect pest management schemes such as mating disruption.

References

Charmillot, P.J., Pasquier, D. & Degen, T. 2005: Climat et population respectives des vers de la grappe eudémis et cochylis. – Rev. Suis. Vitic. Arboric. Hortic. 37: 53-54. 324

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 325-329

Effects of grape downy mildew on interactions between fungicides and predatory mites on grapevines

Alberto Pozzebon, Carlo Duso and Paola Tirello Department of Environmental Agronomy and Crop Science - University of Padua - Viale dell’Università, 16 – 35020, Legnaro (PD) Italy. ([email protected])

Abstract: The biological control of phytophagous mites based on phytoseiid mites is a valuable alternative to chemical control. The success of bio-control strategies is favoured by agricultural practices that enhance predatory mites’ habitats, and by the choice of selective pesticides. Generalist phytoseiids can persist in the absence of herbivore mites by exploiting alternative food resources that include Grape downy mildew (GDM) mycelium and spores. We evaluated the effect of GDM on the interactions between fungicides and phytoseiids with a long term perspective.

Key words: Integrated pest management, Phytoseiidae, generalist predators, alternative food, Grape downy mildew, fungicides

Introduction

Biological control based on phytoseiid mites (Acari Phytoseiidae) is an effective alternative to the chemical control of phytophagous mites infesting vineyards. However, pesticides applied for the control of grape pests and diseases can exert pronounced effects on phytoseiid survival, development and reproduction. The use of pesticides detrimental to predatory mites was considered the main factor affecting tetranychid outbreaks in European vineyards (Ivancich Gambaro, 1973; Girolami, 1981). Therefore, in the last decades several studies have been conducted on the side-effects of pesticides on predatory mites (e.g.: Duso & Girolami, 1985; Englert & Maixner, 1988; Angeli et al., 1997). Among pesticides applied on grapevines the frequent application of non-selective fungicides (e.g. mancozeb) can determine a decline in predatory mite populations, while the use of copper derived fungicides and of folpet has been considered selective to predatory mites (Duso et al., 1983; Girolami & Duso, 1984; Zacharda & Hluchy, 1991; Kreiter et al., 1996; Blümel et al., 2000; Nicotina, 2003). In vineyards, fungicides are frequently applied to control Grape downy mildew (GDM) Plasmopara viticola (Berk. and Curtis ex de Bary) Berlese and De Toni that is also considered an alternative food source for predatory mites (Duso et al., 2003; Pozzebon, 2006). Field observations and experiments showed that phytoseiid population dynamics on plants is related to alternative food availability (e.g.: Eichhorn and Hoos, 1990; Engel and Ohnesorge, 1994b; Duso et al., 2003; Duso et al., 2004). Here we tested the effect of GDM availability on pesticide-predatory mite interactions, considering the short and long term effects of fungicide applications on predatory mite populations on grapevines.

Materials and methods

Experimental design The experiments were carried out during the 2003 and 2004 growing seasons in a vineyard located in the Veneto region, comprised of vines of the Merlot variety colonized by generalist phytoseiids (Amblyseius andersoni (Chant), Typhlodromus pyri Scheuten and Kampimo-

325 326

dromus aberrans (Oudemans)). The experiments were conducted by comparing phytoseiid mite densities on untreated (control) and fungicide treated plots: copper hydroxide (35% a.i., 200 g/hl) and Bordeaux mixture (20% a.i., 375 g/hl). Differential fungicide applications were made every 7-8 days from sprouting to blooming. After differential fungicide applications, these plots were sprayed with copper hydroxide (35% a.i., 200 g/hl) every 7-8 days. Each treatment consisted of 3 replicates of 10 plants each. Observations were performed approximately every 14 days from May 28th to October 1st in 2003 and from May 24th to July 20th in 2004, by randomly selecting 32 leaves per treatment. The leaves were taken to the laboratory and analyzed under a dissecting microscope in order to evaluate phytoseiid population densities. GDM leaf symptom extents were evaluated with a 1 cm2 scale printed transparent.

Data analysis Data on predatory mite densities and GDM leaf symptom extents were analyzed with the MIXED procedure of SAS® where treatment, time and their interaction were fixed effects. In the analyses, phytoseiid densities and GDM leaf symptom extents were considered as response variables with repeated measures taken at different times. Differences among treatments were evaluated applying a t-test (α = 0.05) to the least-square means calculated with the ESTIMATE option. Moreover, when considering the 2003 data, treatment effects were evaluated applying a t-test (α = 0.05) to the least-square means estimated across two separated periods: 1) the fungicide application period + one month (from May 28th to July 10th) as effects assessment period (Blümel et al., 2000); 2) the subsequent period (from July 11th to October 1st). Prior to the analysis data were log (x+1) transformed.

Results and discussion

During 2003, phytoseiid densities proved to be similar in all treatments during the fungicide application period + one month period (for all comparisons: p > 0.05), while in the second part of the season, lower predatory mite numbers were observed in Bordeaux mixture plots than in the control (p < 0.001) and in copper hydroxide plots (p = 0.003) (Figure 1). No differences emerged between the control and copper hydroxide (p > 0.05). GDM foliar symptoms were not observed in all treatments during the fungicide application period + 1 month period (for all comparisons: p > 0.05) (Figure 1). During the second part of the season GDM symptoms were more extensive in the control than in Bordeaux mixture plots (p = 0.002) (Figure 1). The incidence of GDM symptoms was not different between the control and copper hydroxide plots (p = 0.07) nor between copper hydroxide and Bordeaux mixture plots (p = 0.09) although the number of infected leaves was higher in the former treatment. The results that emerged during the 2003 fungicide effects assessment period, indicate that copper hydroxide and Bordeaux mixture had no detrimental effects on predatory mite populations. However, phytoseiid population increases late in the season reached the highest population level in the control, and proved to be more abundant in copper hydroxide than in Bordeaux mixture plots. Since the differences among treatments were highlighted after differential fungicide application period, effects other than direct ones must be involved. In fact, the differences in phytoseiid populations reflected the GDM incidence among treatments. GDM mycelium and spores are alternative food sources that can affect predatory mite abundance on vines (Duso et al, 2003; Pozzebon, 2006). The presence of GDM induced an increase in predatory mite densities favouring the recovery of phytoseiid populations. Observations carried out during 2004 provided further data on the importance of GDM for 327

predatory mite populations. During the fungicide application period, higher population densities were observed in the control plots than on treated plots. Both fungicides appeared to have a detrimental effect on phytoseiids. However, copper-derived products are well-known as a selective fungicide for predatory mites (e.g. Girolami & Duso, 1984; Nicotina, 2003) as was seen during 2003; therefore, the effects reported are not related to direct toxicity but more probably to indirect effects mediated by GDM availability.

Fig. 1. Predatory mite abundance and GDM symptoms observed during 2003.

Fig. 2. Predatory mite abundance and GDM symptoms observed during 2004. 328

During 2004, the abundance of predatory mites proved to be higher in the control compared to other treatments (for all comparisons: p < 0.05) where phytoseiid population densities were similar among fungicide treated plots (p = 0.89) (Figure 2). Regarding GDM foliar symptoms, higher levels were recorded in the control than in fungicide treated plots (for all comparison: p < 0.05) (Figure 2). The data presented here highlight the importance of GDM in interactions between generalist predators and fungicides. Alternative food availability favoured the recovery of phytoseiid populations in late season and thus their persistence. Indirect effects mediated by alternative foods availability should also be considered when evaluating pesticide side-effects on predatory mites.

References

Angeli, G. & Ioriatti, C. 1994: Susceptibility of two strains of Amblyseius andersoni Chant (Acari: Phytoseiidae) to dithiocarbamate fungicides. – Exp. Appl. Acarol. 16: 669-679. Angeli, G., Forti, D. & Maines, R. 1997: Effetti collaterali di fitofarmaci di interesse fruttiviticolo verso gli acari fitoseidi. – Informatore agr. 53: 74-77. Blümel, S., Pertl, C. & Bakker, F. 2000: Comparative trials on the effects of two fungicides on a predatory mite in the laboratory and in the field. – Ent. Exp. Appl. 97: 321-330. Blümel, S., Aldershof, S., Bakker, F.M., Baier, B., Boller, E., Brown, K., Bylemans, D., Candolfi, M.P., Huber, b., Linder, C., Louis, F., Müther, J., Nienstedt, K.M., Oberwalder, C., Reber, B., Schirra, K.J., Ufer, A. & Vogt, H. 2000: Guidance document to detect side effects of plant protection products on predatory mites (Acari: Phytoseiidae) under field conditions: vineyards and orchards. – In: Guidelines to evaluate side-effects of plant protection products to non-target arthropods. Eds. Candolfi, M.P., Blümel, S., Forster R., et al.: 145-153. Duso, C. & Girolami, V. 1985: Strategie di controllo biologico degli acari della vite. – Atti XIV Congr. Naz. Ital. Entom.: 719-728. Duso, C., Girolami, V., Borgo, M. & Egger, E. 1983: Influenza di anticrittogamici diversi sulla sopravvivenza di predatori fitoseidi introdotti su vite. – Redia 66: 469-483. Duso, C., Malagnini, V., Paganelli, A., Aldegheri, L., Bottini, M. & Otto, S. 2004: Pollen availability and abundance of predatory phytoseiid mites on natural and secondary hedgerows. – Biocontrol 49: 397-415. Duso, C., Pozzebon, A., Capuzzo, C., Bisol, P.M., Malagnini, V. & Otto, S. 2003: Grape downy mildew spread and mite seasonal abundance in vineyards: Evidence for the predatory mites Amblyseius andersoni and Typhlodromus pyri. – Biol. Control 27: 229- 241. Eichhorn, K.W. & Hoos, D. 1990: Investigations in population dynamics of Typhlodromus pyri in vineyards of Palatina, F.R. Germany. – IOBC/WPRS Bull 13 (7): 120-123. Engel, R. & Ohnesorge, B. 1994: Die Rolle von Ersatznahrung und Mikroklima im System Typhlodromus pyri Scheuten (Acari, Phytoseidae) – Panonychus ulmi Koch (Acari, Tetranychidae) auf Weinreben II. Freilandversuche. – J. Appl. Entomol. / Z. Angew. Entomol. 118: 224-238. Englert, W.D. & Maixner, M. 1988: Laborzucht von Typhlodromus pyri und Auswirkungen von Pflanzenschutzmitteln auf Mortalität und Fekundität dieser Milbe. – Nachtrichtenbl. Deutsch. Pflanzenschutzd. 40: 121-124. Girolami, V. 1981: Danni, soglie di intervento, controllo degli acari della vite. – Atti "III incontro su la difesa integrata della vite" Latina, 3-4 dicembre 1981, Regione Lazio: 111- 143. 329

Girolami, V. & Duso, C. 1984: Ruolo positivo del rame nelle strategie di controllo biologico degli acari della vite. – Vignevini 9: 90-94. Ivancich Gambaro, P. 1973: Il ruolo del Typhlodromus aberrans Oudemans (Acarina Phyto- seiidae) nel controllo biologico degli Acari fitofagi del Veronese. – Boll. Zool. Agr. Bachic. 11: 151-165. Kreiter, S., Sentenac, G., Weber, M., Rinville, C.H., Barthes, D. & Auger, P.H. 1997: Effects non intentionnels de quelques produits phytopharmaceutiques sur Typhlodromus pyri, Kampimodromus aberrans et Phytoseius plumifer. – Phytoma. La défense des végétaux, 493: 51-58. Nicotina, M. 2003: Side effects of some fungicides on the population of Typhlodromus exhilaratus Ragusa and Phytoseius finitimus Ribaga sensu Denmark (1966) (Parasiti- formes, Phytoseiidae) on grapevines in Tuscany. – Adv. in Hort. Sci. 17(2): 72-76. Pozzebon, A., 2006. Relazioni tra acari predatori generalisti e funghi patogeni della vite. Aspetti ecologici ed applicativi. – PhD Thesis, University of Padua. Zacharda, M. & Hluchy, M. 1991: Long-term residual efficacy of commercial formulations of 16 pesticides to Typhlodromus pyri Scheuten (Acari: Phytoseiidae) inhabiting commercial vineyards. – Exp. Appl. Acarol. 13: 27-40. 330

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 331-336

Soil Pest Management with Herbicides?

Peter Schwappach Bavarian State Institute for Viticulture and Horticulture, Section of Grapevine Pathology and Physiology, Herrnstr. 8, D-97209 Veitshoechheim, Germany

Abstract: The herbicides glyphosate and triclopyr as well as four different application methods have been tested in a healthy vineyard planted with cv. Mueller-Thurgau near Wuerzburg in the Franconian wine growing region. The following applications techniques were compared: a. injection of herbicide inside of the stem, b. placing the herbicide at the cross-cut of the stem, c. superficial application along the stem and d. spraying at the leaves. Both the two agents and the four application methods showed significant differences. After five years almost every treated replication did show neither leaves nor shoots. Although there were still some plants with vital, active roots at the end of the trial. While it was best to inject glyphosate, triclopyr showed best results with total necrotizing of roots when painted on the surface of the stem. Also the effect of killing the roots was much faster with triclopyr including both small and structural roots. Over all triclopyr led to better and faster results compared to glyphosate. Also, triclopyr is preferred by the method of application, because it is much easier painting the herbicide along the stem or placing it at the stem-cut.

Key words: grapevine, phylloxera, nematodes, glyphosate, triclopyr, uprooting, fallow vineyards

Introduction

At present it is not possible to control root-attacking pests like grape phylloxera or nematodes in Germany. In addition no suitable systemically acting pesticide is expected to be registered within the next years. Since the abundance of grape phylloxera in German vineyards is increasing a special focus lies on efficient methods to control grapevine soil pests. Former experiments (Herrmann and Herrmann, 2000) showed, that gallicoles of grape phylloxera were able to infest freshly planted vines, even after uprooting the old vines following five years of fallow. Therefore a vineyard experiment was performed to show, if herbicides can destroy the entire rootstocks including every single root-tip and thus deprive all grape phylloxera, too, being obligatory dependant to vines (Herrmann, 2001, 2002).

Material and methods

Experimental design The field trial was conducted in a healthy vineyard because the effect from the herbicide is more easy to detect. Also root-necrosis is easier to monitor, coming only from the herbicide. The vineyard used had been planted in 1978 with cultivar Mueller-Thurgau on rootstock SO4. Plants were treated with the two systemic herbicides: glyphosate (Roundup) and triclopyr (Garlon 4) between 1999 and 2001. The following techniques of application were tested to optimize the effects: (Tab.1)

331 332

Table 1: Experimental scheme in Veitshoechheim Woelflein, field 1 (1999) treatments in field 1 glyphosate triclopyr control Injection (1 x 1ml) 21 3 16 Injection (2 x, 5ml/0,25ml) 29 3 3 Stem painting 4 3 1 Stem cut 3 3 2 Leaf application 6 - -

With a further experiment the most successful applications were tested again in 2001 after having analysed the experiment of 1999 (Herrmann, 2002). At this time the most successful treatments were repeated. The experimental plot is shown in table 2. The figures show the number of grapes per treatment.

Table 2: Experimental scheme in Veitshoechheim Woelflein, field 4 (2001) treatments in field 4 glyphosate triclopyr control Injection (1 x 1ml) 10 10 4 Injection (2 x 0,5ml) 10 10 3 Stem painting 10 10 7 Stem cut 12 10 6

In order to inject the herbicide into the stem one (or two respectively) injection holes were drilled 5-10 cm above ground, each with a diameter of 5 mm. In the hole the pure active agent was injected with 1,0 ml or twice 0,5 ml using a commercial disposable syringe. The holes were filled with vaseline afterwards to avoid fungal infections and other damages. For “painting the stem” 7 ml of the agent were spread along the stem at approximately 20 cm above ground, after having removed the loose bark. For the treatment “stem cut” 0,5 ml of the active agent was smeared at a cross-cut of the trunk about 10 cm above ground. Spraying the herbicide on the leaves was performed with a 10% dilution of the herbicide, using a hand sprayer. This application was only conducted with glyphosate. Every treatment was applied singularily between 1999 and 2001. Control plants were treated accordingly, using only water instead of an active agent. Beginning in 2002 the emergence of shoots, leaves and vitality of the root-system was monitored annually. For this the complete plant was rooted using a mini-dredger. Using simple visual, morphological characteristics at the rootstock, the structural and small lateral roots, respectively, the effects of the herbicide on the root-system were assessed (fig. 1). All roots branching directly from the rootstock showing a distinct bark were catagorized as main or structural roots. Roots smaller than 2mm in diameter were recognized as small or lateral roots. The vitality of roots was classified according to the following scheme: 1 = vital, 2 = damaged, 3 = necrotized/dead. Moreover the discoloration at the cross-section of the rootstock was monitored, being a measure for wood-degradation.

Results and discussion

Both herbicides as well as the various applications showed very different effects. As expected neither vital shoots nor leaves could be obtained on treated vines. On the contrast results were 333

quite unequal when monitoring the roots. All control plants showed vital shoots and leaves as a well as a healthy growth of the root system.

Fig. 1: Observation chart for assessment of vitality of plants

Glyphosate-treatment Glyphosate applicated on the leaves (var. d) showed no effects on the roots. The treated vines were hardly affected and developped shootings as ususal. The same was true when painting this agent along the trunk (var. c). Only few plants reacted with decreased growth. While typical symptoms of glyphosate treatment could be registered at leaves, the concentration of the infiltrated agent was not high enough to kill the vines. The same happened with the roots (fig. 2). While in both variants necrotized roots were registered, the majority of plants kept vital structural and lateral roots. Results were different when injecting glyphosate into the trunk (var. a) or smearing it onto the stem-cut (var. b). At the latest in the second year after application there were no shoots at all. When assessing rootstock and root-system one could see distinct differences even within these treatments. The treatments with twice 0,5 ml showed vital roots even in 2005, the last year of our trial. In contrast we could exclusively find damaged or dead main roots on plants after injection of 1 ml already in the second year of the trial. But four years after the application of glyphosate the treatment “injection” still had vital main roots (fig. 2). Smearing glyphosate at the horizontal cut showed similar effects as injecting the agent into the trunk. On the one hand no vital lateral and structural roots could be found, on the 334

other hand even in the last year of our investigation not all of the structural roots were completely necrotized. In contrast to the injection-treatment the rootstock of the stem-cut- vines were almost vital und hardly damaged until the last year of our experiments.

Fig. 2: Development of main roots between 2002 and 2005: glyphosate

Triclopyr-treatment The application of triclopyr also showed very different results. On the contrary to applications with glyphosate we could register effects already in the first year after applying triclopyr. Thus all variants did show neither shoots nor leaves. Beginning from the second year we monitored obvious necrosis at the cross-cut of the rootstocks. This led to complete discoloration and partial dead tissue proving strong degradation of wood (fig. 3). Examination of roots showed a different picture. The treatment “injection” and “stem- cut” until two and three years after application, respectively, had some vital or just damaged roots. Whereas “painting along the stem” already since 2002 only showed completely necrotized small roots und from 2003 onwards no vital structural roots at all. Four years after applying triclopyr roots of this treatment were mainly degraded. Even the rootstock was hardly to be discovered in the soil.

335

Fig. 3: Shoot emergence and rootstock vitality in 2004

Fig. 4: Vitality of structural roots 2004

336

Conclusion

The herbicides glyphosate and triclopyr are suitable to kill small lateral and main roots completely. The effect of the two agents is different. There is a clear difference between the two agents combined with the application technique. Using glyphosate best results were achieved when injecting 1 ml into the stem. Painting along the stem resulted in the most complete necrotizing of vines when using triclopyr. This result was confirmed with the second trial of field 4 (fig. 3 and 4). Proved over the whole period of investigations triclopyr had a faster and more complete effect on structural roots than glyphosate. The same is true for the effect on small lateral roots (results are not shown here). triclopyr is also preferred by the method of application, because it is much easier to paint along the stem or the cut. At the same time this will save time and money. This indirect method of soil pest control is suitable to destroy soil pests by destroying their host plant. In the meantime triclopyr has been registered in Germany for this special purpose. Vine- growers are allowed to treat vines with this agent either in fallow vineyards or before uprooting. This results will be continued under real infestation conditions in order to evaluate the results we obtained in this trial.

Acknowledgements

Many thanks to Josef V. Herrmann for the idea and the initiative to start this experiment. Special thanks for supporting the experiment when uprooting the vines even at adverse weather conditions to Heri Schmitt and Martin Pfeiffer from LWG, Veitshoechheim.

References

Herrmann, G., Herrmann, J.V. 2000: Untersuchungen und Beobachtungen zur Reblaus. – Rebe und Wein 6: 252-256. Herrmann, J.V. 2001: Poisoning grapevines to avoid the risk of grape phylloxera reinfection? – ISHS-workshop ´Rootstocks Performance in Phylloxera Infested Vineyards´, Geisenheim. Herrmann, J.V. 2002: Herbizide gegen die Reblaus? – Das Deutsche Weinmagazin 4: 20-24. Schwappach, P. 2006: Herbizide gegen Wurzelschädlinge? – Der Deutsche Weinbau 13: 26-29. Schwappach, P. 2006: Herbizide zur Bekämpfung von Bodenschädlingen? – Deutsches Wein- bau-Jahrbuch 2007 (58): 101-107.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 337-342

Spatio-temporal distribution of Lobesia botrana (Denis & Schiffermüller) male population in a central Italy agro-ecosystem

Andrea Sciarretta1, Antonio Zinni2, Angelo Mazzocchetti2, Pasquale Trematerra1 1 Department of Animal, Plant and Environmental Science, University of Molise, via De Sanctis, I-86100 Campobasso, Italy. ([email protected]) 2 Agenzia Regionale per i Servizi di Sviluppo Agricolo-Abruzzo, via Nazionale 38, I-65010 Villanova di Cepagatti, Pescara, Italy

Abstract: The spatial analysis of the pheromone trap catches of European grapevine moth Lobesia botrana (Denis & Schiffermüller) males was undertaken in Abruzzo, region situated in central Italy, during years 2005 and 2006. In the study area, vineyards are the most spread cultivation, surrounded by hedgerows, small woodlots and alternated with cereal and olive groves. The main purpose of the study was to investigate the spatio-temporal dynamics of pest, inside and outside vineyards and to evaluate the effect of the landscape elements on pest distribution. The activity of L. botrana adult males was monitored weekly using 40 sticky traps, baited with 1 mg of pheromone (E,Z)-7,9-dodecadienyl acetate. Geostatistical methods were used to characterize the spatial distribution of European grapevine moth. For the annual sum of catches and single flights, models fit from variogram analyses were used to interpolate insect catches by means of the kriging algorithm and contour maps were obtained. Our results showed that the presence of L. botrana was not limited to vineyards, particularly at the beginning of the season, when most of the males were cached inside olive groves. The adult distribution in the experimental area changed during the season: contour maps showed that hot spots of the I flight were positioned inside olive groves, during the II and the III flight they concentrated in vineyards. L. botrana males were observed also in uncultivated fields, but never as hot spot. Large proportion of European grapevine moth adults is in zones usually uncovered by pest management programs. In the area under study, the olive plant must be seen as the main reservoir of adult population outside vineyards. In many parts of central and southern Italy, vineyards and olive groves are two of the most spread cultivations and are frequently neighbouring. In these areas, L. botrana can disperse in the landscape and adults move between contiguous crops. In such a context, during monitoring and control operations, it is highly recommended to consider the whole landscape, with particular attention to olive crops.

Key words: Lobesia botrana, geostatistics, vineyard, olive grove.

Introduction

Little attention has been paid, in the past, to the importance of the mosaic of farmland habitats in Integrated Pest Management (IPM) projects. New approaches of IPM tend to recognize the influences of adjacent habitats to the entomocenosis of cultivated crops and consider insect- landscape interactions in the context of whole agro-ecosystem, including the role that different parts of the system play on the dynamics of pests (Ekbom, 2000; Koul and Cuperus, 2007). Such a knowledge is an important component when we intend to introduce an ecologically based management and a precision farming approach in the control strategies.

337 338

In this paper we report the results obtained from spatial analysis of pheromone trap catches of European grapevine moth Lobesia botrana (Denis & Schiffermüller) males. The research was undertaken for two years, from 2005 to 2006, in Abruzzo, region situated in central Italy. In this area, commercial vineyards are the most spread cultivation, surrounded by hedgerows and small woodlots, and alternated with cereal and olive groves. The main purpose of the study was to investigate the spatio-temporal dynamics of pest, inside and outside vineyards and to evaluate the effect of the landscape elements on pest distribution.

Material and methods

Study area The experimental area, about 50 ha, is located in a hilly landscape of Abruzzo, region situated in central Italy, at 100-200 m above sea. The landscape is characterized by the presence of alternated plots of cereals crops, olive groves, uncultivated fields, vineyards and woodlots; it is crossed by a ravine with hedgerows, that divides the area into a north-western and a south- eastern hillside (Figure 1). In the north-western side, there were vineyards V1 (about 1.5 ha in size), V2 (about 0.3 ha) and V3 (about 1 ha), with Montepulciano d’Abruzzo wine cultivar. The various plots were surrounded by cereal crops and uncultivated areas. In the south-eastern side, the following vineyards were located: V4 (about 0.4 ha, with Sangiovese and Trebbiano wine cultivars), V5 (about 2.5 ha, with Montepulciano d’Abruzzo wine cultivar) and V6 (about 1 ha, with Monte- pulciano d’Abruzzo wine cultivar). All plots were trained to pergola system, a horizontal training system, except V2 (simple curtain) and V6 (spurred cord). Vineyards were separated by olive and cereal crops. Some oaks and walnut woodlots were located between vineyards V2 and V4 and between vineyards V5 and V6.

Fig. 1. Experimental area, with indication of main landscape elements: cereal crops (C), olive grove (O), ravine (rv), vineyards (V1-V6), uncultivated fields (U), woodlots (W).

Data collection The activity of L. botrana adult males was monitored weekly using pheromone sticky traps of the delta type, baited with 1 mg of synthetic pheromone (E,Z)-7,9-dodecadienyl acetate (Nova- 339

pher, San Donato Milanese, Italy). The traps were placed in the grapevine canopy at 1.8 m above ground. Pheromone dispensers were replaced every 4 weeks and sticky boards of the traps were replaced every 2-4 weeks. Trapped adults were removed and counted weekly. The activity of males was monitored in 2005 and 2006 by using 40 pheromone traps (see Figure 3). During the first year (2005), field surveys were conducted from the beginning of April until the beginning of November. During the second year (2006), field surveys were conducted from the end of March until the end of October. The sampling points were positioned as previous year, with limited shift. Distances between any two traps varied from 51 m to 687 m.

Spatial analysis Spatial dependence among observations was examined calculating omnidirectional semivario- grams with maximum distance of 375 m, at five lag intervals of 72 m. Models obtained from variogram analyses were used to interpolate insect catches by means of the kriging algorithm. Spatial analyses were carried out using Surfer software Version 8.05 (Golden software, Golden, CO, USA) with x, y representing latitude and longitude expressed as Universal Transversal Mercator coordinates, and z the trap counts. The obtained interpolation grid was graphically represented by a contour map, which shows the configuration of the surface by means of isolines representing equal z-values; a base map showing the experimental area, with the same coordinate system, was placed on top of the contour map. Zones of the contour map with higher catches than in the surroundings are named in the text as “hot spots”. For a full description of utilized methods, refer to Sciarretta and Trematerra (2006).

Results and discussion

Captures of L. botrana males displayed three peak flights (Figure 2). In both years, the first peak was reached at the end of April and the second one at the end of June; the third peak was in the mid of August in 2005 and at the beginning of September in 2006. Later in the season, low level of catches continues until November (2005) and the end of October (2006).

Fig. 2. Weekly catches of L. botrana in 2005 and 2006. Average weekly data for temperature minima (Temp min), temperature maxima (Temp max) and rainfall were obtained from daily recordings of a meteorological station located in the experimental area.

340

2005 2006

4662700

4662600

4662500

4662400

(m) coordinate Y

4662300

4662200 467800 467950 468100 468250 468400 Insect X coordinate (m) counts 0 240 480 720 960 1200

Fig. 3. Contour maps of L. botrana male annual catches during 2005 and 2006. Triangles represent the trap position. Bold lines delimit vineyards.

In 2005, contour maps obtained from annual catches showed a hot spot in the north- western sector of the study area inside vineyard V3 and two main hot spot in the south-eastern sector: one located on the western border of V5, extending toward south into an olive grove, the other on the eastern border of V6 (Figure 3). Outside vineyards, density of L. botrana males was high throughout the experimental area, especially where olive plants were located. In 2006, in addition to hot spots in V3 and V5, a new one appeared in V1, while that in V6 decreased. The presence of individuals in olive groves was also high, while in other landscape elements such as cereal crops, uncultivated land and woodlots was limited (Figure 3). In Figure 4, contour maps of single flights observed in 2005 and 2006 were represented. During the I flight of both years, L. botrana males were distributed mainly on the south- eastern sector of the study area and hot spots were located in olive groves. Occurrence in vineyards was also consistent, especially in V5, where a hot spot was on the border of the field, near olive plants, and inside V1 in 2006. Also in uncultivated areas of north-western sector, was observed the presence of L. botrana. In the following flights, distribution changed and hot spots were concentrated inside vineyards. In particular, the localization of high density areas was very similar for the II and the III flight of the same year. In 2005 and 2006, hot spots were located in V1, V3, V4, V5 and V6, less evident during the II flight than the III flight; the presence of L. botrana beyond vineyards was limited in both years. Few information are available about spatial distribution of L. botrana. Aggregated distribution of larvae was observed by Badehnausser et al. (1999) and Ifoulis and Savopoulou-Soultani (2006). Studies on adult distribution were very scarce and directed to evaluate the behaviour of L. botrana in mating disruption applications (Feldhege et al., 1993; Charmillot et al., 1996). Peláez et al. (2006) utilized geostatistical maps to determine the spatial distribution of L. botrana inside vineyards. Our contour maps showed that spatial distribution of L. botrana was not limited to vineyards, but its presence is high particularly inside olive groves. The distribution changes radically during the season: at the beginning, hot spots were positioned inside olive groves, then during the II and the III flight they concentrate in vineyards.

341

2005 I flight II flight III flight

600

500

400

300

200

100

0 Insect counts 2006 I flight II flight III flight

4662650

4662500

(m) coordinateY 4662350

4662200 467800 467950 468100 468250 468400 X coordinate (m) Fig. 4. Contour maps of adult male L. botrana distribution obtained by kriging procedures applied to single flight trap counts in 2005 and 2006. Bold lines delimit vineyards.

The ability of L. botrana to develop on olive trees was well-known (Stavridis and Savopoulou-Soultani, 1998) and observed also in the experimental area (unpublished data). In our case during the I flight, all hot spots of male adult population are in or on the border of olive groves and catches remained high also in following flights; thus in the area under study this plant must be seen as the main reservoir of adult population outside vineyards. In many Mediterranean areas of central and southern Italy, vineyards and olive groves are two of the most spread cultivations and are frequently neighbouring. In these cases, L. botrana can disperse in the landscape and adults move between contiguous crops. The spreading in viticulture of monitoring by means of pheromone traps and of control techniques that modify insect behaviour, such as mating disruption, attract and kill and auto- confusion, requires a thorough evaluation in the time and space positioning of traps and dispensers, especially at the beginning of the season. In these cases, it is highly recommended to consider the whole landscape, with particular attention to olive crops in Mediterranean agro-ecosystems.

References

Badehnausser, I., Lecharpentier, P., Delbac, L. & Pracros, P. 1999: Contributions of Monte- Carlo test procedures for the study of the spatial distribution of the European Vine Moth, Lobesia botrana (Lepidoptera: Tortricidae) in European vineyards. – Eur. J. Entomol. 96: 375-380. Charmillot, P.J., Pasquier, D., Alipaz, N.J. & Scalco, A. 1996: Study of the vine moth Lobesia botrana Den. and Schiff. (Lep., Tortricidae) behaviour inside and outside of a dispenser belt. – J. Appl. Entomol. 120: 603-609. Ekbom, B. 2000: Interchanges of insects between agricultural and surrounding landscapes. – In: Interchanges of insects between agricultural and surrounding landscapes. Eds. Ekbom 342

B., Irwin M.E. and Robert Y., Kluwer Academic Publishers, Dordrecht (The Netherlands): 1-3. Feldhege, M., Eichhorn, K.W. & Louis, F. 1993: Mating disruption of the European grapevine moth Lobesia botrana Schiff. (Lepidoptera: Tortricidae). Investigations on the temporal and spatial distribution of populations. – Bulletin OILB/SROP 16 (10): 90-92. Ifoulis, A.A. & Savopoulou-Soultani, M.. 2006: Use of geostatistical analysis to characterize the spatial distribution of Lobesia botrana (Lepidoptera: Tortricidae) larvae in northern Greece. – Environ. Entomol. 35 (2): 497-506. Koul, O. & Cuperus, G.W. 2007: Ecologically based Integrated Pest Management. – CAB International, Wallingford, UK: 462 pp. Peláez, H., Maraña, R., Vasquez de Prada, P., Puras, A. & Santiago, Y. 2006: Local population behaviour of Lobesia botrana Denis & Schiffermüller (Lepidoptera: Tortric- idae). – Bol. San. Veg. Plagas 32 (2): 189-197. Sciarretta, A. & Trematerra, P. 2006: Geostatistical characterization of the spatial distribution of Grapholita molesta (Busck) and Anarsia lineatella (Zeller) males in an agricultural landscape. – J. Appl. Entomol. 130 (2): 73-83. Stavridis, D.G. & Savopoulou-Soultani, M. 1998: Larval performance on and ovideposition preference for known and potential hosts by Lobesia botrana (Lepidoptera: Tortricidae). – Eur. J. Entomol. 95: 55-63.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 343-349

Lutte biologique contre les cochenilles farineuses Heliococcus bohemicus Sulc et Phenacoccus aceris (Signoret) au moyen de lâchers de Chrysoperla lucasina (Lacroix)

Gilles Sentenac1, Thuy Pham2, Armel Salaun1, Julien Souvignet2 1 Institut Français de la Vigne et du Vin, 6 rue du 16 ème Chasseurs 21 200 Beaune 2 Station Régionale Bourgogne Centre Est, 6 rue du 16 ème Chasseurs 21 200 Beaune

Résumé : Les stratégies mises à la disposition des viticulteurs pour lutter contre les cochenilles farineuses, reposent sur l’emploi d’insecticides neurotoxiques. En l’absence d’alternative, nous nous proposons, en conditions de plein champ, d’évaluer l’activité biologique du prédateur Chrysoperla lucasina à l’égard d’une population de Pseudococcidae composée d’ Heliococcus bohemicus et de Phenacoccus aceris. Pour se faire un dispositif comprenant sept répétitions a été suivi tout au long de la campagne 2005, en l’absence de traitement et de différence statistiquement significative entre modalité, ce dernier a été validé. Nous avons constaté l’efficacité des deux lâchers de 15 C. lucasina (vieille L1/jeune L2) par cep, réalisés au printemps 2006 sur les formes hivernantes d’ H. bohemicus et de P. aceris en reprise d’activité. La régulation opérée sur la génération 2005 de cochenilles farineuses, se répercute sur la génération 2006 au sein de laquelle les différences entre « lâcher » et « témoin » sont encore significatives. Les populations dans la modalité « témoin » ont doublé entre le 12 septembre 2005 et le 18 septembre 2006, à cette date l’efficacité des « lâchers » est de 83%.

Mots clés : lutte biologique, Chrysoperla lucasina, cochenilles farineuses, Pseudococcidae, Helio- coccus bohemicus, Phenacoccus aceris.

Motivations et objectifs

La lutte contre les cochenilles farineuses en période végétative de la vigne repose sur l'emploi d'insecticides appartenant à la famille des organophosphorés. L'absence d'alternative à la lutte chimique nous conduit à évaluer, en conditions de plein champ, l'efficacité biologique du prédateur Chrysoperla lucasina à l'égard d’une population mixte de Pseudococccidae : Helio- coccus bohemicus et Phenacoccus aceris. Le choix du prédateur s'appuie sur les résultats d'un test de voracité réalisé en conditions contrôlées, la proie étant H. bohemicus. Les chrysopes sont introduites sous forme de larves dès la reprise d'activité des formes hivernantes d' H. bohemicus et P. aceris.

Matériels et méthodes

Le dispositif expérimental retenu est un essai en blocs éclatés à 7 répétitions. La dimension de chaque parcelle élémentaire est de 10 ceps (2 rangs x 5 ceps). La validation de ce dispositif repose sur l'analyse statistique des résultats, nombre de cochenilles par cep ou par lot de 10 feuilles, issus des différentes observations réalisées en 2005. Les lâchers de larves sont réalisés le 02 mai et le 11 mai 2006, à raison de 15 larves de C. lucasina (vieilles L1, jeunes L2) par cep. Les larves sont expédiées dans des boîtes remplies de causses de sarrasin, la présence d'œufs d'Ephestia kuehniella a pour objectif de limiter le cannibalisme. Ces mêmes boîtes d'expédition qui contiennent donc 15 larves, sont attachées sur la tête de souche de

343 344

chaque cep de la modalité Lâcher, une fois le couvercle enlevé. Au total 2 100 larves de C. lucasina sont lâchées en deux apports de 1 050 larves chacun. En début de campagne, du 11 avril (bourgeon dans le coton-03) au 12 juin (10 à 12 feuilles étalées- 17) les observations portent sur le bois d'un an ainsi que sur l'ensemble de la végétation (feuilles, rameaux) de chaque cep. Par la suite les notations ne concernent que 10 feuilles par cep, 5 sur chaque face, de rang 1 à 5. La variable observée est le nombre de cochenilles H. bohemicus ou P. aceris par cep ou pour 10 feuilles. Les résultats des différents comptages, nombre de cochenilles par cep ou lot de 10 feuilles, sont soumis à une analyse statistique du type comparaison des moyennes, méthodes des couples, test unilatéral (hypothèse alternative à l'hypothèse nulle : le lâcher est (>) plus efficace que le témoin), α = 5 %.

Résultats et discussion

Validation du dispositif En 2005, année précédant celle au cours de laquelle les lâchers sont effectués, un suivi des populations de cochenilles a été réalisé sur les quatorze parcelles élémentaires de 10 ceps chacune. Nous pouvons à partir de ces dernières, constituer les sept répétitions des deux modalités prochainement à l'étude. Avant traitement, 15 notations sont donc effectuées (voir Tableau 1 ), la variable observée est le nombre de cochenilles, H. bohemicus ou P. aceris, par cep ou lot de 10 feuilles. Les blocs sont formés, a posteriori, de parcelles élémentaires présentant des densités par cep équivalentes.

Tableau 1 : nombre de Pseudococcidae par cep ou par lot de 10 feuilles en 2005, moyenne des 7 répétitions – Lâchers de C. lucasina en 2006

futur futur

date lâcher témoin différence de moyennes

mod. 1 mod. 2 21 avril 05, bourgeon ds le coton, 0.04 0 n.s. 03 2004 Géné 28 avril, pointe verte, 05-06 0.31 0.17 n.s. 19 mai, 6 feuilles étalées, 12 1.59 0.93 n.s. 22 juin, début nouaison, 25 2.29 1.80 n.s. 04 juillet, nouaison petit pois, 31 1.89 1.64 n.s. 11 juillet, fermeture de la grappe, 1.36 1.36 n.s. 33

par lot de 10 25 juillet, fermeture de la grappe 0.91 1.41 p.s. 01 août, fermeture de la grappe 0.97 1.10 n.s. 08 aout, début véraison, 35 0.83 0.70 n.s. 16 août, 25 % véraison 0.70 0.71 n.s. 22 août, 75 % véraison 0.73 0.31 s. (mais témoin moins occupé) 29 août, 95 % véraison 0.44 0.80 n.s. feuilles, génération 2005 05 septembre, 100 % véraison 3.44 3.40 n.s. 12 septembre, début maturité 3.74 4.61 n.s. nbre de Pseudococcidae 19 septembre, récolte 2.93 3.11 n.s.

345

Du 21 avril 2005 au 19 septembre 2005, les résultats de chacune des 15 observations ont fait l'objet d'une analyse statistique (méthodes des couples) qui n'a pas mis en évidence de différence significative, hormis le 22 août mais à cette date le témoin est moins peuplé que la future modalité lâcher (voir Tableau I, graphique n°1). Le dispositif expérimental en blocs éclatés est validé.

Nantoux 2005 Validation du dispositif expérimental Lutte biologique contre H. bohemicus et P. aceris au moyen de C. lucasina 10 ITV Beaune 9 obs: baguettes, rameaux, feuilles Observation de 10 feuilles par cep

6 futur Lâcher futur Témoin 5

4 changement de génération changement 3 pas d'observation durant cette période pas d'observation

2 Nombre de cochenilles par cep ou lot 10 feuilles . Nombre 1

0 temps 2-avr 9-avr 2-juil 9-juil 7-mai 4-juin 5-févr 3-sept 1-janv 8-janv 6-août 16-avr 23-avr 30-avr 16-juil 23-juil 30-juil 5-mars 14-mai 21-mai 28-mai 11-juin 18-juin 25-juin 12-févr 19-févr 26-févr 10-sept 17-sept 24-sept 15-janv 22-janv 29-janv 13-août 20-août 27-août 12-mars 19-mars 26-mars femelles sous présence de cochenilles sous écorce écorce présence de cochenilles sous écorce

Graphique n° 1

Analyse des résultats : Une, voire deux visites par semaine, ont permis de détecter avec précision la reprise d'activités des formes hivernantes d'H. bohemicus et/ou P. aceris. Les premières Pseudococc- idae ont été observées le 18 avril 2006. A l'origine, en s'appuyant sur la dynamique des populations suivies lors de la campagne dernière, le premier lâcher de C. lucasina était programmé le 28 avril, le second le 10 ou 11 mai 2006. La précocité de la reprise d'activité ou bien la disponibilité en nombre et en stade souhaité des prédateurs peuvent amener à modifier ces dates. En fait, nous avons reçu le 26 avril 2006, 70 boîtes de 15 jeunes L1 de C. lucasina chacune (démarrage de l'élevage), elles ont été conservées dans leur état d'origine, à 20-21°C sur une paillasse sans éclairement naturel direct, jusqu'au mardi 02 mai afin de lâcher des L2. Le second lâcher a eu lieu quant à lui le 11 mai, jour de réception du second lot de vieilles L1/jeunes L2. 10 à 13 boîtes ont fait l'objet d'un dénombrement de prédateurs le jour même du lâcher, pour se faire le contenu d'une boîte est répandu le plus uniformément possible sur un plateau de couleur blanche aux dimensions suivantes l: 30 x L : 40 cm. Les larves, parfois très mobiles, sont prélevées délicatement au moyen d'un pinceau, l'observation du contenu d'une boîte dure 10 minutes. Les résultats (voir Tableau 2) obtenus laissent apparaître une certaine hétérogénéité. 346

Tableau 2 . Nombre de Larves de C. lucasina par boîte.

nombre de larves de Chrysope, C. lucasina n° boîte lâcher du 02 mai lâcher du 11 mai 1 16 18 2 5 54 3 6 31 4 13 15 5 11 23 6 17 38 7 5 22 8 16 22 9 15 52 10 16 34 11 11 12 2 13 16 moyenne 11.46 30.9

Le déplacement des larves de Chrysopes n'a pas pu être complètement contrôlé par le dispositif expérimental retenu, en effet leur présence remarquée dans les parcelles «lâcher» du 04 mai au 24 mai, a également été notée à plusieurs reprises dans les parcelles «témoin» (voir Tableau 3), certaines en train de vampiriser une cochenille. La distance minimale entre une parcelle «lâcher» et une parcelle «témoin» est comprise entre 3 et 9 mètres. Bien qu'elles aient été, après leur détection, retirées des parcelles «témoin», elles ont pu générer un biais quant à l'évaluation de leur efficacité biologique qui repose sur une comparaison du nombre de cochenilles présentes dans les parcelles «lâcher» et «témoin». Un autre biais peut trouver son origine dans le déplacement des néonates ou autre larves de la génération 2006, déplacement qui peut éventuellement les amener à coloniser des ceps autres que ceux qui les ont vues naître alors que les larves de chrysope ont achevé depuis plusieurs semaines leur développement et se sont transformées en adultes dont le régime alimentaire repose sur des substances sucrées (nectar, miellat).

Tableau 3. Parcelles «témoin» fréquentées par des larves de C. lucasina (nombre de larves présentes)

Date Parcelles témoin 10 mai 2BIII (1), 2BVII* (2) 15 mai 2BII (2), 2BIII (1), 2BV (2), 2BVI (1), 2BVII (1) 18 mai 2BII (1), 2BIII (1) 24 mai 2BI (1), 2BII (1) *: prédation observée in situ

Contrairement aux comptages de Pseudococcidae effectués en 2005 en absence de traitement, les dénombrements de cochenilles réalisés en mai 2006 mettent en évidence des différences entre les modalités «lâcher» et «témoin» qui sont statistiquement significatives (voir Tableau 4, graphique n° 2). Si deux jours seulement après le premier lâcher , le 04 mai les différences sont presque significatives (significatives au risque α = 10 %), elles sont 347

significatives au risque α = 5 %, le 10, le 15, le 18 et le 24 mai, le 05 et 12 juin soit pendant les 41 jours qui suivent le premier lâcher ou les 32 jours qui suivent le second. Les larves de C. lucasina lâchées en mai 2006 ont montré une réelle efficacité de prédation sur les stades hivernants de la génération 2005 d’H. bohemicus et P. aceris, des captures suivies de prises d’alimentation ont été observées au cours des différents comptages réalisés le 04 et 10 mai 2006.

Tableau 4 : Nombre de Pseudococcidae par cep ou par lot de 10 feuilles, moyenne des 7 répétitions – lâcher 2006 de C. lucasina sous forme de larve.

Différence de Date Lâcher Témoin Efficacité moyennes 11 avril 2006, bourgeons dans le coton, 03 0.00 0.00 - - 13 avril, bourgeons dans le coton, 03 0.00 0.00 - - 18 avril, bourgeons dans le cotons, 03 0.01 0.19 - n.s. 25 avril, bourgeons éclatés, 06 0.31 0.39 - n.s. premier lâcher de 15 larves (L1/L2) de C. 2 mai 2006, 1 à 2 feuilles étalées lucasina par cep 04 mai, 2-3 feuilles étalées, 09 1.26 2.89 56% p.s. 10 mai, 4-5 feuilles étalées/inflorescences 0.39 2.21 82% s. visibles, 09 second lâcher de 15 larves (L1/L2) de C. 11 mai 2006, 4 à 5 feuilles étalées lucasina par cep 15 mai, 5-6 feuilles étalées, 12 0.16 1.30 88% s. génération 2005 18/19 mai, 7-8 feuilles 0.04 0.61 93% s. 24 mai, 7-8 feuilles étalées/inflorescences 0.07 0.40 82% s. séparées

Nombre de Pseudococcidae par cep 05 juin, grappes séparées et allongement, 0.07 0.31 77% s. 15 12 juin, 10-12 feuilles étalées/boutons 0.10 0.41 76% s. floraux séparés, 17

19 juin, Pleine floraion, 23 0.34 1.69 80% p.s. 26 juin, Nouaison, 27 0.44 2.56 83% s. 10 juillet, fermeture de la grappe, 33 0.67 4.00 83% p.s. 24 juillet, fermeture de la grappe, 33 1.14 4.43 74% s. 07 août, premières baies vérées, 35 0.13 6.84 98% n.s. 21 août, 50% véraison, 36 0.69 2.46 72% s. 04 septembre, 98% véraison 1.19 6.89 83% p.s. 18 septembre, début maturité, 38 1.56 9.10 83% s. génération 2006

par lot de 10 feuilles 02 octobre, récolté 0.46 5.60 92% p.s. 16 octobre, chute des feuilles, 43 0.09 0.09 0% n.s. Nombre de Pseudococcidae 30 octobre, fin de la chute des feuilles,47 0.00 0.00 0% -

Dès le 19 juin, les observations portent sur les jeunes larves descendantes des formes hivernantes, principalement des néonates (taille inférieure ou égale à 0.5 mm), nouvelle génération d’H. bohemicus et/ou P. aceris. Le nombre de cochenilles par lot de 10 feuilles, hormis lors de la notation du 21 août, croît tout au long de l’été, du 19 juin au 18 septembre 348

2006, date à laquelle la présence de cochenilles farineuses est optimale avec plus de 9 Pseudococcidae par cep dans les parcelles « témoin ». Par la suite, en octobre, la baisse des populations dans cette modalité est due à la migration des individus vers les zones d’hivernation, sous les écorces. L’effet des lâchers perdure au niveau de la génération 2006 de cochenilles farineuses, en effet, statistiquement, des différences significatives au risque α = 5 ou 10 % sont mises en évidence au cours de toute la campagne estivale à l’exception de la notation du 07 août.

Essai de lutte biologique contre H. Bohemicus et P. aceris au moyen de C. lucasina

Nantoux 2006 ITV Beaune observation sur baguettes, rameaux et feuilles Observation sur 10 feuilles par cep 10 9 Lâcher Lâcher de larves de 8 Témoin chrysope, 02 et 11 mai 7

4 4 Changement de génération 3

1 Nombre de Pseudococcidae par cep ou lot 10 feuilles . Nombre 0 Temps 1-oct 8-oct 2-avr 9-avr 2-juil 9-juil 7-mai 4-juin 5-févr 3-sept 1-janv 8-janv 6-août 15-oct 22-oct 29-oct 16-avr 23-avr 30-avr 16-juil 23-juil 30-juil 5-mars 14-mai 21-mai 28-mai 11-juin 18-juin 25-juin 12-févr 19-févr 26-févr 10-sept 17-sept 24-sept 15-janv 22-janv 29-janv 13-août 20-août 27-août 12-mars 19-mars 26-mars

Présences de cochenille sous écorce Présences de cochenille sous écorce

Graphique n° 2

Conclusions

L’efficacité des deux lâchers de printemps de C. lucasina au stade L2, sur les formes hivernantes d’ H. bohemicus et P. aceris en reprise d’activité, a été démontrée. La régulation opérée sur la génération 2005 de cochenilles farineuses, se répercute sur la génération 2006 au sein de laquelle les différences entre « lâcher » et « témoin » sont encore significatives. Les populations dans la modalité « témoin » ont doublé entre mi-septembre 2005 et mi-septembre 2006 Cette étude sera reconduite en 2007 sur un autre site expérimental à l’évidence plus peuplé, la densité moyenne sur les 10 parcelles élémentaires était de 20 Pseudococcidae par lot de 10 feuilles, le 28 septembre 2006.

349

Summary: Biological control of Pseudococcidae Heliococcus bohemicus and Phenacoccus aceris with the generalist predator Chrysoperla lucasina. Practical control strategies proposed to the winers to manage mealybugs lead to application of neurotoxic insecticids. As alternative we propose to value the biocontrol activity of the predator Chrysoperla lucasina in field trials on a mixed mealybugs population composed by Heliococcus bohemicus and Phenacoccus aceris. For that purpose an experimental design with seven replications was surveyed all along the 2005 season: with no statistical differences between experimental plots before any treatment, the design was validated. We observed the efficiency of two release of each time 15 C. lucasina per vine (late first larval instar/early second larval instar), made in spring 2006 on overwintering outgoing stages of H. bohemicus and P. aceris. The regulation realised on the 2005 generation was noticed even on the 2006 generation, whose differences between "release" and "untreated" where still significant. The scale populations in "untreated" are twice in september 2006 than in september 2005. At the same dates, the efficiency of "release" is about 83% compared to "untreated".

Key words: biological control, Chrysoperla lucasina, mealybugs, Heliococcus bohemicus, Phena- coccus aceris.

Bibliographie

Boudon-Padieu, E. Esmenjaud, D. Kreiter, S. Roehrich, R.Sfora, R. Stockel, J. Van Helden, M. 2000 : Ravageurs de la vigne. – Editions Féret : 231 pp. Dagnélie, P. 1984 : Théories et méthodes statistiques, Volume 2. – Les presses agronomiques de Gembloux : 463 pp. Plant, C.W. 1997 : A key to adults of British lacewings and their allies. – The AIDGAP publications : 90 p. Semeria, Y.& Berland, L. 1988 : Atlas des Névroptères de France et d’Europe. – Société Nouvelle des Editions Boubée, 190 p.

350

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 351-353

Volatiles from grape drive the oviposition of Lobesia botrana at short distance

Marco Tasin1,2, Gianfranco Anfora1, Anna-Carin Bäckman1, Claudio Ioriatti2, Antonio De Cristofaro3, Elena Pozzolini4, Elisabetta Leonardelli2 and Andrea Lucchi4 1 SafeCrop Centre IASMA Research Center Via E.Mach 2, 38010 S.Michele a/A, Italy; 2 Plant Protection Department IASMA Research Center Via E.Mach 2, 38010 S.Michele a/A, Italy; 3 Dept. Scienze Animali, Vegetali e dell’Ambiente, University of Molise,Via De Sanctis, 86100 Campobasso Italy; 4 Dept C.D.S.L., Sect. Entomologia agraria, University of Pisa, Via San Michele degli Scalzi,2 – 56124 - Pisa, Italy.

Abstract: Females of phytophagous moths use multisensory modalities to select oviposition sites. In laboratory experiments we examined the effect of olfaction on the oviposition behaviour of the grapevine moth Lobesia botrana, a pest of grapevine. In a dual choice oviposition test where insects could choose only using volatile and visual cues, females did not discriminate between the odours released by green bunches of two different varieties (Trebbiano and Sangiovese). Chemical and electrophysiological analysis of bunch odours showed only quantitative differences which did not account for a variety preference in oviposition in our experimental conditions. An artificial odour of the antennally active compounds was subsequently used in further oviposition tests. Females deposited their eggs onto the substrate holding the mimic with a dose-response pattern. Results of this work showed that volatiles released by green bunches as well as synthetic blends mimicking the bunch odour bouquet provided a signal for oviposition at a short range for grapevine moth females.

Key words: Lobesia botrana, behaviour, oviposition, plant volatiles, short range attraction.

Introduction

Females of phytophagous moths use multisensory modalities to select oviposition sites. In this study we carried out laboratory experiments to evaluate the effect of olfaction on the oviposition behaviour of the grapevine moth Lobesia botrana (Den. & Schiff.). The grapevine moth is a polyphagous herbivore distributed across the paleartic area, inhabiting cultivated vineyards. In Europe, gravid females lay their eggs on flower buds, green berries and mature bunch during first, second and third generation of the year. L. botrana oviposit on several genotypes of cultivated grape Vitis vinifera L. (Thiéry & Moreau, 2006). We hypothesised here that L. botrana females use olfactory cues beside contact chemoreception to choose the site of oviposition at a short distance. We also investigated the possible role of volatile compounds from grapevine bunch in encoding genotype attraction to the grapevine moth.

Material and methods

Insects Moths were selected from a colony of L. botrana reared on artificial diet at IASMA Research Centre. Insects were reared under a 17:1:6 photoperiod and 22°C. Cardboard strips with pupae were placed in cages where emerging adults were allow to mate. Oviposition assays were carried out with 48-96 h old mated females.

351 352

Volatiles collection Green bunches belonging to the varieties Sangiovese and Trebbiano were collected from a vineyard at University of Pisa. Grapevine plants were sprayed with fungicides to control powdery and downy mildew. No insecticides were used. Twenty clusters of either variety were collected from 10 different plants and immediately stored in a portable fridge at +4°C for transportation to the laboratory. Bunches were then placed in a polyacetate bag for volatiles collection. Air from the bag was drawn through a sorbent cartridge during 6 hs. Volatiles were desorbed from the cartridge by solvent at room temperature. Volatiles collections were stored in vials at -18°C until use. Three to four collections were carried out in each phenological stage.

Electrophysiological and chemical analysis Collection of plant compounds were analysed by a Hewlett Packard (Palo Alto, CA) 5890 gas chromatograph (GC) coupled to an electroantennodetector (EAD, Synthec, Hilversum, The Netherlands). A female antennae was mounted on the EAD and placed into the air stream carrying the volatiles compounds eluting from the GC column. Volatiles collected in headspace entrapments were identified by gas-chromatography coupled to mass-spectrometry (GC-MS, Perkin-Elmer, AutoSystem XL Turbomass Gold ).

Oviposition bioassay Behavioural choice assays were conducted in ventilated cylindrical cages (25 cm ID x 45 cm; metallic net, mesh 2mm wide; Tasin et al., in press). One bunch of each variety (Trebbiano and Sangiovese) was placed into a plastic conical pierced glass. In this way the choice of female moth for preferred oviposition site was based exclusively on visual and olfactory cues. In order to evaluate the role of odours on oviposition, we formulated a synthetic mimic of grape headspace with GC-EAD active compounds. Synthetic blends were loaded on rubber septa. In oviposition assays, females could choose between a pierced glass containing a septum releasing the blend and another septum loaded with lone solvent. The mimic was tested at 15, 150 and 1500 µg/septum. Three mated females were released into the centre of each cage. After 72 hs at 25°C and 16L:1T:6D photoperiod, the moths were removed and the eggs counted. Fifteen to twenty replicates were conducted. Deviation from equal preference in choice assay was tested by Chi Square test.

Results and discussion

Chemical and electrophysiological analysis Mated females L. botrana responded to 10 compounds released by bunches during phenological development. These compounds were released by both vine varieties. Six GC-EAD active compounds from green bunch headspace were used to formulate a synthetic bunch mimic.

Oviposition bioassay The oviposition preference of L. botrana for volatiles emitted by the green berries of the two varieties was assessed through a dual choice oviposition assay. As a result, the number of eggs was equally distributed among the two varieties. In a second test with synthetic blends, the response of females was dose-dependent. While females showed no preference for the lowest dose (P>0.05), a significant preference for the blend was found for the intermediate dose (P<0.05). At the highest load females were significantly repelled by the blend (P<0.05). Volatiles released by green bunches as well as synthetic blends mimicking the grapevine bunch odour bouquet, provided a signal for oviposition at a short range for grapevine moth females. However, females did not discriminate between the odours of the two tested 353

grapevine genotypes. Behavioral, electrophysiological and chemical analysis of the corresponding headspace confirmed that differences among released volatiles were only quantitative and probably did not account for a variety preference in our experimental conditions. Further investigations on the ovipositional response of females to grapevine volatiles will provide base knowledge for the development of behavioural manipulation methods for a safe control of L. botrana.

Acknowledgements

This study was funded by the Italian Ministry of University and Research (PRIN 2006) and by the Government of Trento Province (SafeCrop Center Project). Our thanks are due to Damiano Giotti (University of Pisa) for his help in some of the experiments.

References

Tasin, M., Anfora, G., Leonardelli, E., Ioriatti, C., Lucchi, A., De Cristofaro, A. and Pertot, I. 2008: An improved oviposition assay for the grapevine moth. – IOBC/wprs Bull.: In press. Thiéry, D. & Moreau, J. 2006. Grape cultivar affects larval and female fitness of the European grapevine moth, Lobesia botrana (Lepidoptera: Tortricidae). – IOBC/wprs Bull. 29(11): 131-138.

*This study was partially supported by MIUR (PRIN 2006 funds). 354

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 355-361

Grapevine Pests in Sicily

Haralabos Tsolakis, Ernesto Ragusa Dipartimetno Senfimizo, Sezione Entomologia, Acarologia E Zoologia - Università Di Palermo, Viale Delle Scienze, 13 -90128 Palermo (Italy) - ([email protected])

Abstract: More than twenty phytophagous species are reported associated with Sicilian vineyards. Only a quarter of them have to be considered of economic importance. Among moths, the grapevine moth Lobesia botrana (Denis & Schiffermüller) is certainly the most frequent species. It is considered the key pest on both table and wine grapes, while the incidence of the grape berry moth Eupoecilia ambiguella (Hübner) is scarce and linked to particular biotopes. Three to four flights are reported for the grapevine moth from February-March to October-November, while no data are available for the grape berry moth. Damages caused by L. botrana are strictly linked to the climatic trends during the summer period, to the microclimatic conditions in the island biotopes, as well as to the micro climate created by the canopy, or to the different grape-clusters of the various cultivars. The cicadellid Jacobiasca lybica (Bergevin) was reported for the first time in 1962, but it is considered an old inhabitant of Sicilian vineyards. Until 1990s it was present mainly on the western part of the island and it was considered of no economic importance for the autochthonous cultivars. Today the cicadellid is distributed on the whole island and often causes severe damages especially on allochthonous cultivars. The mealybug Planococcus ficus (Signoret) has to be considered a serious problem in Sicilian vineyards as in the past two decades an increase of infestations was registered, particularly on new plantings. No data are available up to now on the biology of the species in Sicily, while its presence causes serious concerns for both damages to yields and virus transmission. Calepitrimerus vitis (Nalepa), recorded in Sicily at the beginning of the XX century, is now considered another emergent problem in Sicilian vineyards, on both autochthonous and allochthonous cultivars, especially on newly planted vineyards.

Key words: grape moths, Jacobiasca lybica, Planococcus ficus, mites

Introduction

Viticulture in Sicily is of ancient origin. The first peoples inhabiting the island (Siculi, Sicani and Elimi) knew grapevine, but ancient Greeks are responsible for cultivation and diffusion on the whole island. For them wine was so important that they had a specific God called Dionysos. Although information was available about the extension of vineyards during dominations of Sicily, as far as pests are concerned, the first information affecting grapes dates to 1880, when phylloxera, Dactulosphaira vitifoliae (Fitch), arrived in Sicily. This aphid, as in other parts of the world, reduced Sicilian vineyards from 320.000 Hectares to 175.000 Hectares in few years. Once the hurdle of phylloxera was cleared using the American rootstocks, Sicilian viticulture remained a marginal culture, although the success of “Marsala”, the popular fortified aged wine, allowed a consistent presence of vineyards mainly in the western part of Sicily (where viticulture is still considered an important crop more although for historical and social reasons than for economic ones).

355 356

Diffusion of the French cultivars (Cabernet, Syrah, Chardonnay etc), after the second half of ’80s, gave the Sicilian viticulture the necessary push. Today this sector is very important from a social and economic point of view and as a consequence also problems related to vine protection have become important. As far as the phytophagous species are concerned, more than twenty species are reported associated with Sicilian vineyards (table 1), but only a quarter of them are to be considered of economic importance in Sicily. In this paper we report some of them in order of economic importance, according to literature and personal knowledge.

Grape moths Among grape moths, both Eupoecilia ambiguella (Hübner) and Lobesia botrana (Denis & Schiffermüller) are present in Sicily (Genduso, 1985), even if the presence of the first one is scarce and limited to particular biotopes. Up to now, this species has been found in vineyards up to 400 m a.s.l., but no particular damages were registered for this pest. On the other hand, the grapevine moth L. botrana, as in other Mediterranean countries, is certainly the most frequent species and is considered the key pest in Sicilian vineyards. It should be mentioned that damages caused by this pest are strictly related to the climatic trends during the summer period, to the microclimatic conditions in the island biotopes, as well as to the micro clime created by the canopy, or to the different grape-clusters of the various cultivars.

Fig. 1. Lobesia botrana flights in different years and Sicilian biotopes

357

Three and often four flights and the same number of generations of grape moth are reported in Sicily (fig. 1). The first flight starts on the second half of March, even if in some years male catches are registered in the last week of February (Genduso, 1985). This flight lasts until the second half of May, and, as it was shown, has no economic importance (Roehrich & Schmid, 1979; Delrio et al., 1985). The second flight usually starts during the first week of June and lasts 3-5 weeks. For most Sicilian Cultivars also this generation has to be considered almost not harmful, as Brix values until the end of July are maintained below 15.0. For early-harvesting Cultivars (Chardonnay, Viogner etc), for which the harvest occurs within the 1st half of August, both the 2nd and 3rd generation have to be considered harmful. The latter starts with the 3rd flight that occurs in Sicily from the 1st to the last week in July, and lasts until the end of August. This generation has to be considered harmful as the grey mould or the sour bunch rot or ochratoxinogenic fungus establish on damaged bunches. The 4th flight occurs in Sicily from the last week of August to the second week of September and lasts until half October (sometimes until the first week of November). It should be mentioned that larvae of the 4th generation complete their development only on abandoned vineyards or on some cultivars where harvest is very late (October); harvesting takes place in the majority of Sicilian vineyards within the first half of September.

Fig. 2. Trend of male captures and infested bunches on two adjacent vineyards: Cv Grecanico and Cv Catarratto (Tsolakis, unpublished)

Many natural enemies have been found associated with L. botrana (Moleas, 1979; Marchesini e Dalla Montà, 1994; Thiery et al., 2001) and some of them were reported for Sicilian vineyards (Parasitoids, Chrysopids, Coccinellids, spiders etc). However, because of the particular conditions of Sicilian viticulture (that is, absence of hedges and any kind of spontaneous plants in the surroundings of vineyards), the natural biological control is not sufficient to reduce infestations of the grapevine moth under the economic threshold. The vine cultivar is another important factor influencing the infestations by L. botrana. Thiéry and Moreau (2006) demonstrated that some cultivars have an influence on the 358

developmental time of the grape moth larvae, while Baldacchino and Moleas (2000) showed a different receptivity among some cultivars. In two organically managed Sicilian vineyards, we found significant differences between two endemic cultivars, Grecanico and Catarratto (Fig. 2). This different receptivity is probably due both to the micro clime created by the canopy of the different cultivars (summer pruning and vegetation conditioning are absent in traditional Sicilian vineyards), and to the different grape-clusters. Both factors play an important role, not only for the grape moth infestations, but also for the damages derived by mould infection (Fig. 3). As we can note in figure 3, both sour bunch rot and grey mould are significantly higher in no conditioned plants in comparison to the canopy managed.

Fig. 3. Bunches damaged at the grape harvest in the two different canopy managed vineyard (Cv Grecanico) (Tsolakis, unpublished)

Among the abiotic factors influencing the infestation trends of grape moth, the most important one is “Scirocco” a hot wind blowing from Africa in summer. This wind usually lasts 2-3 days. During this period temperatures can exceed 40°C, while relative humidity is reduced below 30%. These conditions have lethal effects on the majority of insects associated with Sicilian vineyards.

Leafhoppers Among leafhoppers, the cotton jassid Jacobiasca lybica (Bergevin) is considered another important pest in Sicilian vineyards on some red or allochthonous cultivars that have widely spread in Sicily in the last two decades, while it is considered of low or no economic importance on the main white Sicilian cultivars. This species was recorded for the first time in Sicily in 1962 (Vidano, 1962), even if it is considered an old inhabitant of Sicilian vineyards. J. lybica is an important pest in most parts of Northern Africa, various Mediterranean countries (Portugal, Spain, Italy and Cyprus) and in the Midle East. It is reported as a polyphagous species, but in Sicily damages are recorded only on vines, even if specimens were collected on various spontaneous plants near vineyards. 359

From information collected from old vine growers, we can say that the leafhopper was present in Belice Valley before its first report in 1962. After 1995, occasional infestations of the leafhopper have been reported, while in 2001 very strong infestations were registered in most parts of western Sicilian vineyards. Damage caused by J. lybica on the leaf is similar to the one provoked by Empoasca vitis Gothe. The damage, called hopperburn, is mechanically caused by plugging the xylem and destroying the phloem, but mainly injecting toxins during feeding (Vidano, 1963; DeLong, 1971). Different cultivars, then, react in different ways to jassid’s attack. This is the principal cause of different damage levels among the various cultivars located in the same territory, rather than the repellence of the cultivar towards J. lybica (Tsolakis, 2003). Fortunately, “scirocco” very often inhibits early infestations which are very dangerous even at low population level: 0.5-1 nymph/leaf during July-August, causes a significant reduction of must soluble solids (Lentini et al., 2000). On the other hand Zygina rhamni Ferrari is very common in all Sicilian vineyards, but no damages have been reported until now, for both autochthonous and allochthonous cultivars.

Planococcus ficus (Signoret) The vine mealybug Planococcus ficus Signoret is another economically important pest. In the last decades, infestations by P. ficus have become more frequent, especially in newly planted vineyards. No data are available in literature about its biology in our region. From our observations in an organically managed vineyard, this species overwinters under the bark or on the main roots of the plant mainly as female. The population increases in spring and trunks are usually wet by honeydew produced by the vine mealybug. Colonies of crawlers were observed, during June, at the base of leaf stem before their migration along the ribs of the leaf. In July a great number of crawlers migrate to bunch causing severe damages. Franco et al. (2004) report three generations/year in the Mediterranean vineyards, but we do not have data confirming this statement for Sicily.

Mites As far as mite species are concerned, three tetranychid species are common in Sicilian vineyards: Tetranychus urticae Koch, Panonychus ulmi (Koch) and Eotetranychus carpini (Oudemans). Sporadic but severe infestations of T. urticae were registered during ’80s in various Sicilian vineyards, most probably, related to the abuses of Carbaryl or other insecticides and the consequent elimination of phytoseiid fauna, more than to climatic or cultivar factors (Ragusa, unpublished). P. ulmi shows a constant presence in Sicilian vineyards but it is considered of economic importance only on table vines (Buonocore et al., 1992). No data are available for E. carpini in Sicilian vineyards. As far as Eriophyid mites are concerned, both the grape rust mite Calepitrimerus vitis (Nalepa) and the grape erineum mite Colomerus vitis (Pagenstecher) are present in Sicily. Both species are considered obligate pests of Vitis vinifera L. (Jeppson et al., 1975) and their diffusion coincide with vine diffusion. First recorded in Sicily in 1911 (Pantanelli, 1911), the grape rust mite causes leaf and shoot distortions and retarded shoot growth during springtime (Duso & De Lillo, 1996; Duffner et al., 2001). Despite its long-time presence in Sicily, no damages were reported for this eriophyid until the last decade, when in different new planted vineyards severe damages were registered (Tsolakis & Ragusa, unpublished). Finally, the grape erineum mite Col. vitis is to be considered of no economic importance in Sicilian vineyards even if the leaf symptoms (erinea) sometimes worry Sicilian growers.

360

Considerations

Entomofauna of Sicilian vineyards is very similar to that occurred in other Meditarranean countries but the different microclimatic conditions of the biotopes present in the island strongly influence the specific composition of vine pests. The absence, up to now, of Scapho- ideus titanus Ball or other phytoplasmas vectors in Sicily subjected to mandatory insecticide sprayings, allow a drastic reduction of the number of sprayings. As a matter of fact, in some environments no pesticides are needed while inmost parts of vineyards one or two sprayings are sufficient to guarantee an efficient control of key pests.

Acknowledgements Authors are deeply indepted to Prof. E. Chiavetta who checked the English text. This work carried out with funds by “Progetto per lo sviluppo dell’Agricoltura biologica in Sicilia” Assessorato Agricoltura e Foreste-Servizio IV.

References

Baldacchino, F. & Moleas, T. 2000: Suscettibilità di alcune cultivar di vite agli attacchi di Lobesia botrana (Denis et Schiffermüller) (Lepidoptera Tortricidae). – Atti Giorn. Fitopatol. 1: 293-298 Buonocore, E., Cataldo, V. & Tropea Garzia, G. 1992: Note sulla difesa dell’uva Italia nel territorio di Chiaramonte Gulfi. – Tecnica agricola 4: 3-23. DeLong, D.M. 1971: The bionomics of leafhoppers. – Ann. Rev. Entomol. 16: 179-210. Delrio, G., Luciano, P. & Prota, R. 1985: Researches on grape-vine moths in Sardinia. – Proc. Meeting on Integrated Pest Control in Viticulture, Portoferraio 26-28 sep 1985: 57-67. Duffner, K., Schruft, G. & Guggenheim, R. 2001: Passive dispersal of the grape rust mite Calepitrimerus vitis Nalepa 1905: (Acari, Eriophyoidea) in vineyards. – Anzeiger Schädlingskunde 74: 1-6. Duso, C. & De Lillo, E. 1996: Damage and control of eriophyoid mites in crops: 3.2.5 Grape. – In: Eriophyoid Mites. - Their Biology, Natural Enemies and Control. Eds. Lindquist, Sabelis and Bruin, Elsevier Science Publishers, Amsterdam, Netherlands: 571-582. Franco, J.C., Russo, A., Suma, P., Neto, E., Zada, A. & Mendel, Z. 2004: Comparative biology of the citrus mealybug and the vineyard Mealybug (Hemiptera: Pseudococcidae). – Proc. X Int. Symp. Scale insects stud. 19th-23rd April 2004: 232 Genduso, P. 1985: The grape-vine moths in theframework of IPM in Sicily. – In: Integrated pest control in viticulture. Ed. Cavalloro, Proc. Meeting EC Experts group, Portoferraio 26-28 sep 1985: 69-82. Jeppson, L.R., Keifer, H.H. & Baker, E.W. 1975: Mites injurious to economic plants. – University of California Press, Berkeley, California, USA: 1-614. Lentini, A., Delrio, G. & Serra G. 2000: Observations on infestation of Jacobiasca lybica on grapevine in Sardinia. – IOBC/wprs Bull. 23 (4): 127-130. Marchesini, E. & Dalla Montà, L. 1994 Observations on natural enemies of Lobesia botrana (Den. & Schiff) (Lepidoptera: Tortricide) in Venetian vineyards. – Boll. Zool. Agr. Bachic. Serie II, 26(2): 201-230. Moleas, T. 1979: Essais de lutte dirigee contre la Lobesia botrana Schiff. dans les Pouilles (Italie). – Proc. Int. Symp. IOBC/WRPS on Integrated Control in Agriculture and Forestry, Wien 8-12 Oct. 1978: 542-551 Pantanelli, E. 1911: L’acariosi della vite. – Marcellia 10: 133-160. 361

Roehrich, R. & Schmid, A. 1979: Lutte intégrée en viticulture. Tordeuses de la grappe: Evaluation du risque, determination des periodes d’intervention et recherche de methodes de lutte biologique. – Proc. Int. Symp. IOBC/WRPS on Integrated Control in Agriculture and Forestry, Wien 8-12 Oct. 1978: 245-253. Thiery D., Xuereb A., Villemant C., Sentenac G., Deblac L., Kuntzman P. 2001: Les parasites larvaires de tordeuses de vignobles: apercu de quelques espèces présentes dans 3 régions viticoles françaises. – IOBC/WPRS Bull. 24 (7):135-142. Thiéry, D. & Moreau, J. 2006: Grape cultivar affects larval and female fitness of the European grapevine moth, Lobesia botrana (Lepidoptera: Tortricidae). – IOBC/wprs Bull. 29(11): 131-138. Tsolakis, H. 2003: La cicalina africana Jacobiasca Bergevin (Homoptera, Cicadellidae) ricompare nei vigneti siciliani. – Informatore fitopatol. 1: 34-40. Vidano, C. 1962: Massiccio attacco alle viti insulari dell’Empoasca lybica. – Giornale di Agricoltura 45: 1-8. Vidano, C. 1963: Alterazioni provocate da insetti in Vitis osservate, sperimentate e comparate. – Ann.Fac. Sc. Agr. Univ. Torino 1: 513-644. 362

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 363-368

Incidence of grapevine moth Lobesia botrana (Den. & Schiff.) on occurrence of ochratoxin A in grapes

Haralabos Tsolakis1, Onofrio Corona2, Antonio S. Pulizzi3, Francesca Grippi4, Vincenzo Mondello5 1 Dipartimetno Senfimizo, Sezione Entomologia, Acarologia E Zoologia - Università di Palermo, Viale Delle Scienze, 13 - 90128 Palermo (Italy); 2 Dipartimento Itaf - Università di Palermo, Viale delle Scienze, 13 - 90128 Palermo (Italy); 3 via Gazzarella,19 -Petrosino (Tp) (Italy); 4 Consorzio Co.Ri.Bi.A., c/o ZS via Marinuzzi, Palermo (Italy); 5 Dipartimento Senfimizo, Sez. di Patologia vegetale e Microbiologia agraria, Università di Palermo, Viale delle Scienze, 90128Palermo (Italy) - ([email protected])

Abstract: observations were carried out in an organic vineyard (cv malvasia di candia), at salemi (sicily), during 2006-2007, in order to verify the influence of grapevine moth lobesia botrana (den. & schif.) on presence of ochratoxin a (ota) in bunches at harvest time. a percentage of 12.98% of bunches were attacked by grape moth during 2006, while 8.65% of them were infected by grey mould and sour bunch rot (5.77% and 2.88% respectivelly). analyses carried out on 12.98% infested bunches, showed a level of 20 µg/kg of ota, while on intact samples 0.04 µg/kg of toxin was registered. During 2007, infestation of bunches by grape moth was very low (2.05%), due to the particular climatic conditions, and no infections by grey mould and sour bunch rot were registered at harvest time. analyses on 2.05% infested bunches by l. botrana showed a presence of 0.055 µg/kg of ota, while no detectable presence of toxin was registered on intact bunches (<0.014 µg/kg). two aspergillus species were recorded in 2007: the most representative was aspergillus fumaricus wehm (86%) which is not considered the producer of ota, while 14% was aspergillus fonsecaeus (thom & raper) considered as the probably producer of the toxin.

Key words: Lobesia botrana, Ochratoxin A, Aspergillus spp.

Introduction

Infestations by Lobesia botrana (Den. & Schiff.) represent the principal cause of berries infections by grey mould and sour bunch rot in field. These diseases reduce the organoleptic quality of wines as far as the grape yield. Sometimes the indirect damages caused by L. botrana are overestimated because infections of grapes by moulds are not directly proportio- nal to the grape moth infestation, depending on other abiotic and biotic factors. Another indirect problem caused by L. botrana infestations was, until few years ago, unknown. This is the case of some fungi species belonging to the genera Aspergillus and Penicillium that produce a powerful mycotoxin named ochratoxin A (OTA) with nephrotoxic, carcinogenic, immunosupressive and teratogenic properties towards homeothermic animals (Sweeny and Dobson, 1998). These fungi are mainly saprophytes and invade the berries tissue through wounds caused on the berries by different abiotic or biotic agents. These infections seem to be another economically important consequence of grapevine moth infestations (Minguez et al., 2004; Cozzi et al. 2006). The present work aimed at verifying the influence of grapevine moth infestations and of some abiotic factors on mould infections of grapes as well as on the presence of ochratoxinogenic fungi and OTA in bunches at harvest time.

363 364

Materials and methods

Observations were made in an organic vineyard (Cv Malvasia di Candia), at Salemi (Sicily), during 2006 and 2007. The traditional vertical trellis with cane pruned system (Guyot) is present in the vineyard. 13 blocks of 15 plants each were taken into account. One bunch per plant was weekly sampled in order to survey grape moth larvae and bunches damaged by grey mould or sour bunch rot. Meteorological data were recorded by a digital thermohygrograph (Tinytag Ultra). No sprayings were applied in the vineyard for the entire observation period. To evaluate the presence in grapes of ochratoxinogenic Aspergillia and Penicillia, samples of intact and infested grapes have been collected in July and at harvest time. Grapes were manually pressed and diluted with equal part of sterile water-peptone at 1‰; this suspension was progressively diluted to obtain suspensions with concentration from 10-2 to 10-6. 100 µl of each suspension was inoculated in Petri dishes with selective medium DYSG (Dichloran Yeast Extract 18% Glycerol Agar), kept at 25±1 °C and periodically observed. The developed colonies of Aspergillus and Penicillium were first counted with a stereomicroscope, then singly transferred to pure culture for identification. To verify the OTA presence in berries at harvest time, berry samples were prepared according to the method described by Serra et al. (2004), while immunoaffinity clean-up was carried out as described by Visconti et al. (1999). Analyses were carried out on samples reflecting the percentage of infested bunches recorded at the last sampling. Intact bunches were used as control. Proportions of infested bunches were transformed using an arcsine-square-root equation prior to one-way analysis of variance. Differences were considered significant when 95% fiducial limits did not overlap. Analyses were computed using the software “Statistica” (StatSoft Inc., 2003).

Results and discussion

During 2006 an increasing trend of infested bunches was registered after the 2nd week of July (Fig. 1). Infestation rate decreased in the 2nd half of August, probably because of high temperatures and the low relative humidity values recorded in this period (Fig. 2, 3). At harvest time 12.98±1.96% of bunches showed attack symptoms by L. botrana, while 5.76±1.31% and 2.88±0.89% of them were infected by Botrytis and sour bunch rot respectively (Fig 1). It should be mentioned that larvae of L. botrana were present in all bunches damaged by moulds. The first infection data were registered on August 9, when sugar contents in berries was up to 16°Brix, despite the infestation of bunches floated from 5% to 12% during July (Fig. 1). On the other hand the number of infected bunches by moulds was statistically lower than those damaged by L. botrana (p<0.05). This fact has to be correlated to the high mean temperatures and low percentages of relative humidity registered during the second half of August (Fig. 2, 3). Very different climatic conditions were registered during 2007. As it can be shown in figures 2 and 3, high temperatures were registered during the whole period, accompanied by low relative humidity values. This climatic trend had a negative influence on both L. botrana infestation and grey mould and sour bunch rot infections (Fig. 4). As a matter of fact the grape moth infestation remained at very low levels during August (about 2% of infested bunches), while no bunches damaged by grey mould or sour bunch rot were recorded at harvest time (Fig. 4). On infested bunches 0.055 µg/kg of OTA were present, while no detectable presence of toxin was registered on intact bunches (<0.014 µg/kg) (Table 1). 365

Fig. 1. Trend of infested bunches by L. botrana larvae, infected bunches by grey mould and sour bunch rot and ° Brix of grapes during summer 2006 (mean±S.E.M.).

A lower value of OTA, in both intact and infested bunches was registered in 2007, in comparison to that obtained in 2006 (Tab. 1), confirming the strict correlation of fungi presence with climatic trend. No surveys for ochratoxinogenic fungi were carried out during 2006. However, during harvest time a 0.04 µg/kg of OTA was present in intact bunches, while this value was very high, 20.0 µg/kg, in 12.98% infested bunches (Table 1). Data on OTA presence during 2006, led us to investigate on presence of ochratoxinogenic fungi on bunches during 2007. Analyses for Aspergillia and Penicillia were carried out on bunches at the beginning of surveys and at harvest time. These analyses registered the absence of both fungi in July, while at the harvest time 7,3x106 CFU/g of Aspergillus and 8,5x105 CFU/g of Penicillium were present on infested grapes. No Aspergillia were found in intact grapes while 7,05x105 CFU/g of Penicillia was present. It should be mentioned that 100% of Penicillia was Penicillium griseofulvum Dierckx, that is not considered an ochratoxinogenic species. On the other hand, two species of Aspergilia were found: 86% belonging to A. fumaricus Wehm of no ochratoxinogenic properties, while 14% was A. fonsecaeus (Thom et Raper) that is considered an ochratoxinogenic species (van der Merwe et al., 1965). As a matter of fact, no Aspergillus carbonarius (Bainer) or Aspergillus section nigri, that are considered the main OTA producers on grapes, were found during our surveys.

366

20 Temperature °C Temperature

16 2006 2007 14

10 22-Giu 2-Lug 12-Lug 22-Lug 1-Ago 11-Ago 21-Ago 31-Ago 10-Set

Fig. 2. Trend of temperatures during the two observation years

100 2006 2007 80

60 %

0 22-Giu 2-Lug 12-Lug 22-Lug 1-Ago 11-Ago 21-Ago 31-Ago 10-Set

Fig. 3. Trend of relative humidity trend during the two observation years

367

Table 1. Ochratoxin A presence in grapes at harvest during 2006 and 2007

Ochratoxin A (µg/kg) 2006 2007 Intact bunches 0.04 < 0.014 Infested bunches by L. botrana (12.98% of infestation) 20.0 ( 2.05% of infestation) 0.055

Fig. 4. Trend of infested bunches by L. botrana larvae, infected bunches by grey mould and sour bunch rot and °Brix of grapes during summer 2007 (mean±S.E.M.).

This fact deserves further investigation in order to survey the ochratoxinogenic mycoflora on various Sicilian cultivars and to improve our knowledge on the system: grapevine moth infestation-macro and micro climatic conditions-grey mould and sour bunch rot infections-ochratoxinogenic fungi.

Acknowledgements

Authors are deeply indepted to Prof. E. Chiavetta who checked the English text. This work carried out with funds by “Progetto per lo sviluppo dell’Agricoltura biologica in Sicilia” Assessorato Agricoltura e Foreste-Servizio IV.

368

References

Cozzi, G. Pascale, M. Perrone, G. Visconti, A. & Logrieco A. 2006: Effect of Lobesia botrana damages on black aspergilli rot and ochratoxin A content in grapes. – Int. J. Food Microbiol. 111: S88-S92 Minguez, S., Cantus, J.M., Pons, A., Margot, P., Cabanes, F.J., Masque, C., Accensi, F., Elorduy, X., Giralt, L.L., Vilavella, M., Rico, S., Domingo, C., Blasco, M. & Capdevila, J. 2004: Influence of the fungus control strategy in the vineyard on the presence of Ochratoxin A in the wine. – Bull.OIV 77: 821-831. Serra, R., Delgadillo, I., Mendonça, C., Ducauze, C. J., Abrunhosa, L., Rutledge, D. N., Townshend, A., Pietri, A. & Venâncio, A. 2004: Determination of ochratoxin A in wine grapes: comparision of extraction procedures and method validation. – Analyt. Chim. Acta 513: 41-47 StatSoft Inc. 2003: Statistica, users manual 2003. – ed. Stat Soft Inc., Tusla, OK. Sweeny, M.J. & Dobson, A.D. 1998: Review: mycotoxin production by Aspergillus, Fusarium and Penicillium species. – Int. J. Food Microbiol. 43: 141-158. van der Merwe, K.J., Steyn, P.S. & Fourie, L. 1965: Mycotoxins. Part II. The constitution of ocrhratoxins A, B and C, metabolites of Aspergillus ochraceus Wilh. – J. Chem. Soc. 5: 7083-7088. Visconti, A., Pascale, M. & Centone, G. 1999: Determination of ochratoxin A in wine by means of immunoaffinity column clean-up and high-performance liquid chromatography. – J. Chromatogr. 864(1): 89-101.

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 369-373

Landscape characteristics influencing pest populations in viticulture

Maarten Van Helden1, Guillaume Pain2, Josephine Pithon2 1 UMR INRA/ENITA Santé Végétale 1065. 1 cours Gal de Gaulle 33175 Bordeaux France, [email protected]; 2 Groupe ESA, 55 rue Rabelais, BP 30748, 49007 Angers cedex 01

Abstract. Landscape-scale characteristics can influence pest insects directly, for instance by providing hibernation sites or by creating barriers for migration. We developed a new insect trap (Tri-∆nglué® trap), able to monitor the adult flight periods of all major vine insect pests. Pest insect distribution was sampled at the landscape scale over three years in four French wine growing regions ranging from 60- 200 km2 (Pessac-Léognan, Buzet, Sauternes, Saumur-Champigny ). Between 40 and 80 vineyard plots were monitored in each region. The spatial distributions of the insects varied little between years. Lobesia botrana and Scaphoideus titanus) had clustered distributions at this scale but spatial structure was less pronounced for Empoasca vitis and Eupoecilia ambiguella. A geographical information system was used to describe and quantify land cover characteristics at a variety of local and landscape scales, using buffers of increasing diameter (250 to 1000m). Insect abundance was significantly correlated with a number of both local and landscape-scale land cover variables. The two most abundant pest species, L. botrana and E. vitis, showed very dissimilar distributions. L. botrana was more abundant in large continuous monocultures while E. vitis was more abundant in heterogeneous landscapes including woodlands. The strength of these correlations increased with increasing buffer size, up to 750 m, revealing that population levels are indeed influenced by landscape characteristics at this scale. Longer term, landscape-scale monitoring will be continued to try to determine how landscape configuration may influence pest insect movements, so as to better explain underlying mechanisms. Tri-∆nglué trap networks like this one are being adopted by professional organisations, for example to monitor Scaphoideus titanus, vector of the quarantine disease ‘flavescence dorée’ in areas under imposed sprayings

Key words: viticulture, Empoasca vitis, Lobesia botrana, Scaphoideus titanus, Eupoecilia ambiguella, landscape, geographical information system, France.

Introduction

Certain landscape features outside vine plots are considered to be ecological compensation areas able to enhance the beneficial effects of natural enemies (ECA, Boller et al. 2004, van Helden et al., 2004). However such features also directly influence pest insects, providing habitat supplementation (alternative host plants) or complementation (hibernation sites) or by acting as physical barriers to dispersion (Decante & Van Helden, 2006). Vine plots vary in their vulnerability to the pest insects L. botrana, Eu. ambiguella, E. vitis and S. titanus. These species have a range of different ecological traits: mono to tri- voltine, pure specialist to generalist, sedentary to highly mobile (Stockel, 2000). To investigate the relationships between pest insect relative abundance and local and landscape characteristics we conducted a 3-year study to compare insect distributions in several “appellations” (areas of origin).

369 370

Material and methods

Study sites Over three years, between 40 and 70 vine plots were sampled per “appellation” (Pessac- Léognan (PL, 2005-6), Saumur-Champigny (SC, 2005-7), Buzet (Bz, 2007) and Sauternes (Sa, 2006). Plot size was > 1 ha and minimum spacing between traps was 500m. Management was entirely left to the owners but was rather homogeneous within each “appellation”. Insect monitoring Adult insects were trap monitored and larvae were counted three weeks after peak captures as described earlier (Van Helden et al., 2006) using Tri-∆nglué® traps (a yellow delta 2 µg pheromone trap). Second larval generations were not monitored because of insecticide applications. Geographical Information System Land use was defined and digitised, using high-resolution ortho-rectified aerial photographs (BDORTHO, IGN) and GIS software (ARCGIS – ESRI). Two different land cover classifica- tions were used, composed of 3 or 12 habitats. Only the results based on the simple habitat classification are presented here. This first step enabled us to calculate the amount of each land cover type (vine, forest, others) around each trap, in a set of buffers of increasing radius (250, 500, 750, 1000 m). The total continuous area of vineyards (CaV) to which each trap belonged was also determined.

Insect Trappings / Week Saumur Champigny 2006 120 40 L. botrana E. vitis Eu. ambiguella S. titanus 100 30 80

60 20

L.botrana and E.vitis and L.botrana 10

20 S.titanus and Eu.ambiguella

0 0 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 39 Week Number Fig. 1: Example of weekly insect captures on Tri-∆nglué® trap in the Saumur-Champigny area in 2006. Means of 36 traps NB. Lobesia botrana and Empoasca vitis: left axis, Eupoecilia ambiguella and Scaphoideus titanus: right axis.

Data analysis Insect abundances were summed within each generation (Lobesia botrana spring adult Generation = LbaG0, first larval generation = LblG1 etc.) and for each year (Lba2005 etc.). 371

These were compared with the plot and landscape variables using Spearman rank correlation coefficients (CC).

Results and discussion

Insect dynamics and distribution Insect trap samples showed ‘classic’ population dynamics (Fig 1). Insecticide treatments on second generation larvae (tortricids, E. vitis) sometimes interfered with our observations. Overall, trapping levels and flight periods varied between both “appellations” and years.

Variations between generations within a single year As during 2005 (Van Helden et al., 2006) strong correlations appeared between successive generations and stages of L. botrana for all “appellations” and years (Spearman r ≅ 0.8). E. vitis often showed significant correlations between immigrating adults (EvaG0) and subsequent larvae (EvlG1) and between G1 and G2 adults. CCs of G1 larvae and adults were nonexistent confirming the hypothesis that many G1 adults migrate (Decante & Van Helden, 2006). E. ambiguella, (SC), showed no significant CCs between G0 and G1 but trapping levels were generally low (Fig 1).

Table 1: Spearman rank corr. coefficients (rs) of total insect capture (data from 29 traps) between three years in the Saumur-Champigny area. Significant values in BOLD (α=0.05, bilateral test).

Insect L. botrana E. ambiguella E. vitis S. titanus Year 2005 2006 2007 2005 2006 2007 2005 2006 2007 2005 2006 2007 2005 1 1 1 1 2006 0.90 1 0.64 1 0.50 1 0.75 1 2007 0.91 0.94 1 0.54 0.64 1 0.41 0.51 1 0.49 0.55 1

Between-year comparisons Strong to very strong correlations were found when comparing L.botrana, Eu.ambiguella or S.titanus plot total population levels for 29 plots among years (Table 1). For E. vitis these correlations are slightly weaker but still significant, in spite of its hibernation outside the plot. Each species presents a rather comparable spatial distribution between years, in spite of differences in insecticide applications among plots. This distribution therefore seems to be related to some perennial factor of the plot or its surroundings

Insects and landscape characteristics for different buffer sizes As in Van Helden et al. (2006), abundance of L. botrana was always positively correlated with the % surface area of the buffer planted with vines (though not always significantly, Table 2). The size of the vine patch (CaV) correlated more strongly. This may relate to direct attraction of pests or to natural enemy exclusion. One management solution to be tested would be fragmentation of vineyards through hedgerow planting. Grape load (food density) is clearly correlated with Lobesia population density. Immigrating E. vitis (G0) and first generation (G1) adults correlate negatively with these same variables (Table 2). For spring immigrants (G0), this is probably due to the proximity of hibernation sites (winter hosts) hosts in the nearby vegetation (Decante & van Helden 2006). For summer adults (G1) we were unable to identify alternative summer hosts (Van Helden & Decante 2001, 2002). Long distance passive migration, resulting in homogeneous deposition and 372

subsequent dispersion towards vine plots can also explain higher population levels in plots surrounded by non-habitat. Damage (known as hopperburn) was often observed at plot borders, which could represent a barrier for migrating individuals, reluctant to leave the plot. For E. ambiguella and S. titanus, there were no significant results. Increasing buffer size generally increased correlation strength (with % of vine) up to 750 m (Table 2), indicating that landscape composition has an influence at this scale. This diameter is larger than expected considering the adult mobility of L. botrana (Torres-Vila et al. 1997).

Table 2: Spearman rank correlations with % of vine (for different buffer sizes), total continuous area of vines around the trap (CaV) and grape load of the plot. – Significant values in BOLD (a=0,050, bilateral test).

% vine in buffer of X m CaV Grape Insect 250 500 750 1000 Load EvaG0 -0,12 -0,26 -0,20 -0,35 -0,10 0,24 EvaG1 -0,29 -0,34 -0,42 -0,35 -0,40 -0,09 LbaG0 0,25 0,25 0,25 0,14 0,41 0,33 LbaG1 0,15 0,15 0,21 0,18 0,33 0,35 LbaG2 0,14 0,13 0,24 0,20 0,24 0,38 EvlG1 -0,19 -0,20 -0,25 -0,15 -0,21 0,23

Future observations From this preliminary work we can identify the major landscape-scale factors influencing pest abundance and develop hypotheses with regard to underlying mechanisms. Longer term, landscape-scale monitoring will be continued to try to determine how landscape configuration may influence pest insect movements or natural enemy impact. In addition, the apparently opposite responses of major pest insects (L. botrana versus E. vitis) to some parameters (CaV) need to be taken into account. In the Buzet region we will try to include disease monitoring in this landscape ecology study. A new application for the trap network The new Tri-∆nglué® trap has been adopted by professional organisations to monitor insects for different surveillance networks (Fulchin & Van Helden, 2007). This trapping system is now also accepted by the French plant protection services (SRPV) for the monitoring of the ‘flavescence dorée’ vector S. titanus in areas under imposed sprayings (quarantine disease). Trap networks have made it possible to demonstrate the near absence of adults after a single larvicide application, thus avoiding the usual subsequent adult treatment. This can lead to a substantial reduction in pesticide use.

References

Boller, E.F., Häni, F. & Poehling, H.M. (Eds.). 2004: Ecological infrastructures; ideabook on functional biodiversity at the farm level - temperate zones of Europe. – IOBC/WPRS Commission on Integrated Production Guidelines and Endorsement: 212 pp. Decante, D. & Van Helden, M. 2006. Population ecology of Empoasca vitis (Goethe) and Scaphoideus titanus (Ball) in Bordeaux vineyards: Influence of migration and landscape. – Crop Protection 25 (7): 696-704. 373

Decante, D. & Van Helden, M. 2007. Spatial and temporal distribution of Empoasca vitis (Göthe) inside a vineyard plot (in press for Agricultural and Forest Entomology) Fulchin, E. & Van Helden, M. 2007. Development of monitoring schedules for grape diseases at regional scale. –IOBC/WPRS Bulletin 36: 95-99. Stockel, J. (Ed.) 2000. Les ravageurs de la vigne. – Editions Féret, Bordeaux: 121-129. Torres-Vila, L.M.; Stockel J., Roehrich R. & Rodriguez-Molina M.C. 1997: The relation between dispersal and survival of Lobesia botrana and their density in vine inflorescence. – Entomologia Experimentalis et Applicata 84: 109-114. Van Helden, M. & Decante, D. 2001. The possibilities for conservation biocontrol as a management strategy against Empoasca vitis. – IOBC/WPRS Bulletin 24 (7): 291-297. Van Helden, M. & Decante, D. 2002. Les zones écologiques réservoirs (ZER): un moyen pour gérer les ravageurs ? – 6ième Conférence Internationale sur les ravageurs en Agri- culture AFPP Montpellier: 53-61. Van Helden, M.; Fargeas E., Fronzes M., Maurice O., Thibaud M., Gil F. & Pain G. 2006. The influence of local and landscape characteristics on insect pest population levels in viticulture. – IOBC/WPRS Bulletin 29(6): 145-149. Van Helden, M.; Roland A., Merignac J.B., Rodriguez San Martin M., Valles Jimenez M.D. 2004. Lutte biologique par conservation en vignoble, le rôle des haies et des zones enherbées. – Progrès Agricole et Viticole 122(5): 113-118. 374

Integrated Protection in Viticulture IOBC/wprs Bulletin Vol. 36, 2008 pp. 375-377

Occurrence and spread of Scaphoideus titanus in Austria

Norbert Zeisner AGES, Austrian Agency for Health and Food Safety, Institute for Plant Health, Spargelfeldstr. 191, A-1226 Vienna (Austria), ([email protected])

Abstract: Grapevine yellows are widespread in many viticultural areas of the world. Flavescence dorée (FD) is one of them. It causes crop losses, a decrease in the lifespan of grapevines and finally the death of the affected plant. FD is transmitted specifically by the cicadellid leafhopper Scaphoideus titanus. This species was first introduced to Southern France in the 1950s from North America and it is now expanding its range to the north. In the summer of 2004 the first S. titanus-specimen reached the vineyards around Bad Radkersburg near the Slovenian border in Styria. Since then the Austrian Agency for Health and Food Safety carried out intense monitoring surveys with yellow sticky traps in selected Austrian vineyards along the Hungarian and Slovenian border. During the last 3 years, in the middle of August, north-westerly winds brought hundreds of specimen of S. titanus over the border to Austria by passive spread along the valleys of rivers from Balkan States. Further inland the abundance of the vector was much lower. The results of the monitoring surveys show, that the distribution of this vector in Austria is limited to a restricted area in southern Styria and so far the distribution of S. titanus does not present a phytosanitary problem.

Key words: Scaphoideus titanus, Flavescence dorée, vector, monitoring, grapevine.

Introduction

Grapevine yellows caused by phtytoplasmas are widespread in many viticultural areas of the world. Grapevine flavescence dorée phytoplasma (FD) is listed in Annex II A2 of the EU Directive 2000/29 /EC and is also classified as an A2 pest in the EPPO region. FD can cause epidemic disease in Vitis vinifera and is considered to be the most important phytoplasmosis in Europe (Maixner and Holz, 2003). Its natural vector is the cicadellid leafhopper S. titanus which is originating from North-America. This insect was first introduced into Southern France in the 1950s. Today S. titanus is widespread in Northern Spain, Northern Portugal, in Italy, Switzerland (Höhn et al. 2007; Clerc et. al., 2007) Croatia, Serbia, Slovenia (Seljak & Petrovic, 2001) and also in Hungary (Dèr, pers. com., September 14, 2006). Maybe due to the climatic change this pest is now expanding its range to the north. In Austria the first specimen were detected in September 2004 in vineyards around Bad Radkersburg in Styria (Zeisner, 2005).

Material and methods

The intense monitoring surveys were started in 2005 at 5 experimental sites in south-eastern Styria (Sicheldorf, Straden, Klöch, St. Anna am Aigen, Kapfenstein), south-western Styria (Ratsch a. d. Weinstraße; only in 2007) and at 3 experimental sites in Burgenland (Halbturn, Lutzmannsburg, Eisenberg). All sites were located along the Hungarian and Slovenian border. Yellow sticky traps were used during the vegetation period of 2005, 2006 and 2007 in order to capture S. titanus. The samplings were selected and identified in the laboratory of the Austrian Agency for Health and Food Safety (AGES) at the Institute for Plant Health.

375 376

Results and discussion

In 2005 only two larval instars were found. The climatic conditions during this summer were relatively cool and wet (Anonymous, 2007). By contrast in the warm and dry summer of 2006 (Anonymous, 2007) all 5 larval instars were detected within one month. This was the first proof that S. titanus is able to complete its whole life-cycle under Austrian climatic conditions. The range of S. titanus is limited by climatic parameters, because it requires cold winters to break diapause of eggs and warm summers to complete the life-cycle (Boudon-Padieu, 1999). It can be confirmed, that these insects need warm summers to accomplish its life-cycle, however the past mild winter conditions of 2006/07 had the effect that S. titanus larvae occurred almost one month earlier than in the years before. It is assumed, that cold winters are not necessarily essential for the accomplishment of the life-cycle of S. titanus. The results of the monitoring-surveys show that the highest numbers of S. titanus were collected at the experimental site in Sicheldorf close to the Slovenian border in August 2006 and 2007 (Figure 1 and 2), while at the other experimental sites in Styria (Straden, Klöch, St. Anna am Aigen, Kapfenstein), only very few specimen of S. titanus were found (data not shown). In Burgenland no S. titanus specimen were detected. As S. titanus is widespread in Croatia, Slovenia and Serbia (Seljak and Petrovic, 2001), it is assumed, that the high numbers of S. titanus in August are probably due to a passive spread by north-westerly winds along the valleys of the river Mur in Slovenia, the river Drau in Croatia and the river Danube in Serbia.

500

450

400

350

300

S. titanus Larvae 250 Adults 200

150 Numbers of of Numbers 100

06 06 06 06 06 8. 8. 0 0 5. 1.09. 30.05.06 29.06.06 13.07.06 03. 2 2 10.10. 17.10.

Fig. 1. Numbers of S. titanus found at the monitoring site of Sicheldorf in 2006.

In south-western Styria, which is also a vine producing region, no evidence of the occurrence of S. titanus was found, although it is known from Seljak (2002) that S. titanus is also present around Maribor, which is only about 20 kilometres away from the experimental site. The reason is probably the fact that between Maribor and the vine producing area in south-western Styria forested mountains prevent the spread northwards, because they cannot be crossed over by S. titanus. 377

From the monitoring survey it can be concluded that S. titanus is not a phytosanitary problem so far, since the distribution of S. titanus is restricted to a small area in south-eastern Styria (Sicheldorf, Straden, Klöch, St. Anna am Aigen, Kapfenstein) and Flavescence dorée phytoplasma was not found yet, neither in grapevine nor in the vector (Reisenzein, pers. com., November 28, 2007).

500

450

400

350

300

S. titanus Larvae 250 Adults 200

150 Numbers of 100

07 07 07 07 07 5. 6.07 6.07 6.07 7. 7. 7.07 7.07 8.07 0. 0 0 0 0 0 0 0 0 1 4. . . 5. . 30. 1 19 27 04.07.07 12. 18. 2 31 30.0 18.09. 17.

Fig. 2. Numbers of S. titanus found at the monitoring site of Sicheldorf in 2007.

References

Anonymous, 2007: Landesstatistik Steiermark - Klimadaten. – Retrieved November 12, 2007 from http://www.verwaltung.steiermark.at/cms/ziel/7954241/DE/ Boudon-Padieu, E. 1999: Grapevine phytoplasmas. – Proceedings of the First Internet conference on Phytopathogenic Mollicutes, 57-62. Retrieved September 12, 2007 from http://web.uniud.it/phytoplasma//intconf.pdf. Clerc L., Linder C. & Günthart H. 1997: First detection in western Switzerland of the leafhopper Scaphoideus titanus Ball (Homoptera, Jassidae), vector of the grapevine flavescence dorée. – Revue Suisse de Viticulture d’Arboriculture et d’Horticulture 29: 245-247. Höhn, E., Linder, C. & Schaub, L. 2007: Goldgelbe Vergilbung der Rebe – Informationen zum Vektor. – Schweizer Zeitschrift für Obst-Weinbau 15: 4-6. Maixner, M. & Holz, B. 2003: Risiken für den Weinbau durch gebietsfremde Schaderreger. – BMLEV-Forschungsreport 2: 19-23. Seljak, G. & Petrovic, N. 2001: An overview of the presence and research of the phytoplasma diseases on grapevine and fruit trees in Slovenia. – Sodobno Kmetijstvo 34: 466-471. Seljak, G. 2002: Non-European Auchenorrhyncha (Hemiptera) and their geographical distribution in Slovenia. – Acta Entomologica Slovenica 10: 97-101. Zeisner, N. 2005: Amerikanische Zikaden im Anflug. – Der Winzer 5: 20-21.