Integrated Irrigation and Canopy Management Strategies for Vitis Vinifera Cv

INTEGRATED IRRIGATION AND CANOPY MANAGEMENT STRATEGIES FOR VITIS VINIFERA CV. SHIRAZ.

A thesis submitted in fulfillment of the requirements for the

Degree of Doctor of Philosophy at The University of Adelaide

RACHEL MARGARET ASHLEY

September 2004

This work contains no material which has been accepted for the award of any other degree or diploma in any university or other tertiary institution and, to the best of my knowledge and belief, contains no material previously published or written by another person, except where due reference has been made in the text.

I give consent to this copy of my thesis, when deposited in the University Library, being available for loan or photocopying.

______Rachel Margaret Ashley 15 September 2004

Dedicated to the memory of Eve Cottral.

iii

SUMMARY

Modern canopy management practices and irrigation strategies have improved the economic and environmental sustainability of Australia’s wine industry, in terms of increased production and improved wine quality for minimal production cost and environmental impact. This study tested the hypothesis that partial rootzone drying (PRD) integrated with low input, minimal pruning practices can improve sustainability of winegrape production in warm-climate, irrigated vineyards. The bi-factorial experiment investigated three conventional pruning practices; hand spur pruning (SPUR), mechanical hedging (MECH) and minimal pruning (MIN) integrated with standard drip (SD) and PRD irrigation strategies. The sustainability of winegrape production of field-grown cv. Shiraz grapevines was determined by examining yield, fruit composition, wine composition and quality, vine physiology and susceptibility of bunches to Botrytis bunch rot.

Winegrape production was strongly influenced by pruning level and the resultant bunch number per vine. Increased node retention at pruning of minimal pruned vines resulted in 4- fold more bunches per vine than spur pruned vines. Mechanical hedged vines had an intermediate number of bunches per vine. Yield generally reflected the trend in bunch number per vine. However, minimally pruned and mechanically hedged vines compensated for greater carbohydrate partitioning between reproductive sinks by producing smaller bunches with fewer berries per bunch. Partial drying of the grapevine rootzone had a detrimental effect on yield relative to SD irrigation (18%). The additive effect of SD combined with light pruning treatments resulted in few statistically significant interactions for the measured yield components. Berry weight was the only parameter influenced by the interaction between irrigation and pruning during the three experimental seasons; PRD + MIN reduced berry weight by 36% compared to SD + SPUR, in response to lower irrigation inputs and higher bunch number. A 2-fold increase in water use efficiency (tonnes per megalitre) was found by the reduced irrigation inputs of PRD combined with the high crop levels of MIN vines compared to SD + SPUR vines.

Fruit and wine composition was also largely unaffected by combined irrigation and pruning treatments, as a result of the additive effect of PRD and MIN. However, light pruning levels (MIN and MECH) and their associated small berry size and high bunch exposure, reduced

iv pH and increased titratable acidity, and anthocyanin and phenolic concentrations of berry juice compared to SPUR. Minor pruning level effects on wine composition can be directly correlated with those observed on fruit composition. PRD had minimal effect on basic fruit composition but strong effects on wine spectral parameters: density, hue, total anthocyanin and phenolic concentration and ionised anthocyanin concentration, possibly as a result of co- pigmentation of anthocyanin compounds with exocarp tannins. Berry size was strongly correlated with fruit and wine quality. Small berries (i.e. from PRD and MIN) had lower pH and higher anthocyanin and phenolic concentrations in the juice and produced wine that was more acidic, brighter and had higher colour density and anthocyanin (total and ionised) and phenolic concentrations than all other treatments.

Midday and diurnal leaf gas exchange were manipulated by partially drying the rootzone. PRD reduced midday stomatal conductance, photosynthesis and transpiration compared to SD. Stomatal limitation on photosynthesis and transpiration was probable, given the strong positive relationship with stomatal conductance and reduced carbon isotope discrimination by PRD. Transpiration efficiency was improved for PRD irrigated vines compared to SD irrigated vines. Leaf water potential and osmotic potential were measured diurnally, in conjunction with leaf gas exchange to investigate the response of PRD irrigated vines to increasing vapour pressure deficit. Diurnally, stomatal conductance was reduced by PRD compared to SD, which maintained leaf water potential, while no osmotic adjustment occurred. Therefore, PRD irrigation maintained hydraulic water status by hydrating half of the rootzone, whilst dehydration of the other half of the rootzone resulted in the partial closure of stomata. Pruning treatment effects on vine physiology were less pronounced. Minor gas exchange effects showed that pruning level influenced carboxylation efficiency and not stomatal limitations, as photosynthesis was not directly correlated with stomatal conductance.

Bunches were least resistant to infection by Botrytis when fully developed and at maximum maturity. The development of bunches into tighter clusters as berry size increased from veraison to harvest and the increase in sugar content may have encouraged development of Botrytis. The distinct bunch architecture resulting from the combined pruning and irrigation treatments influenced the incidence and severity of Botrytis bunch rot. Light pruning combined with PRD irrigation produced small, loose bunches in season 2001-02, which

v were less susceptible to Botrytis bunch rot development compared to the large, compact bunches produced on SD + SPUR vines. However, low bunch numbers and high fruit-set on MIN and MECH vines in season 2002-03 led to a significant change in bunch architecture. As a consequence of the increased compactness of bunches in season 2002-03, no pruning effects on Botrytis development were observed.

Long term economic and environmental sustainability of winegrape production is dependent on continual improvement in fruit and wine quality, preservation of yield, reduced water and chemical usage. This study has shown partial drying of the rootzone combined with light pruning techniques improved yield, fruit and wine composition, water use efficiency and transpiration efficiency and reduced the incidence and severity of Botrytis bunch rot compared to SD and severe pruning levels. Therefore, over the three experimental seasons, PRD combined with minimal pruning was determined as the preferred strategy to enhance the sustainability of winegrape production of Shiraz cv. in warm-climate, irrigated vineyards.

ACKNOWLEDGEMENTS

I am grateful to the Co-operative Research Centre for Viticulture and its associated funding and research bodies, in particular CSIRO Plant Industry, DPI Victoria, MDBC, The University of Adelaide and Riverlink, for financial and research support throughout the duration of my PhD candidature. Also, I am greatly appreciative to my co-supervisors Mr Peter Clingeleffer, CSIRO PI, Assoc. Prof. Peter Dry, The University of Adelaide and Dr Bob Emmett, DPI. Special thanks to Peter C for sharing his vast knowledge on practical viticulture, to Bob E for exposing me to the fascinating world of pathology and teaching me important mycology techniques and to Peter D for proof-reading my manuscript and providing important feedback.

I am indebted to my colleagues and friends at CSIRO PI, Horticulture Unit, Merbein for their continual research support, especially Dr Rob Walker, Dr Nicola Cooley for assistance with diurnal measurements, Dr Mark Gibberd for help with plant physiology, Dr Paul Petrie for viticultural advice, Peter Lo Iacono and Sonja Needs for sharing their small-lot wine making expertise and producing wine each season. Thanks are given to the viticultural team at DPI; Dr Mark Krstic, Yasmin Chalmers, Glenda Kelly and Lisa Mitchell, for their continual technical and harvest support. Thankyou to my fellow CRCV PhD students for sharing their experience and knowledge, particularly Keren Bindon for proofreading my final draft. A special mention is given to Deakin Estate, Wingara Wine Group Ltd., especially Craig Thornton, Assistant Vineyard Manager, Jeff Milne, Vineyard Manager and Will Davies, Technical Assistant. The experimental site was located in this busy commercial vineyard and it was with Craig’s continued co-operation, that all experiments and harvests were successful.

I am truly indebted to my family and friends for their enduring encouragement, support and love throughout the past 4 years. Without the frequent reassurance and guidance of Mum and Dad, it is highly unlikely I would have commenced let alone finished my PhD. Likewise, I am truly thankful for the support and companionship of my dear friends in Mildura, especially Angelica, Ang, Chelsea, Nicky, Sonja, Nicole and Fiona. Finally, a giant thank you to my fiancé, Damian, who has encouraged and loved me through the difficult times of writing-up my thesis.

vii

LIST OF ABBREVIATIONS

Abbreviation Explanation and Units

PRD Partial rootzone drying SD Standard commercial practice drip irrigation SPUR Hand spur pruning MECH Mechanical hedging MIN Minimal pruning PAR Photosynthetically active radiation (µmol.s-1.m-2) DAB Days after budburst DAF Days after flowering DAI Days after irrigation AEDT Australian Eastern Daylight Savings Time LSD Least significant difference

θv Volumetric soil-water content (%) T Temperature (°C) RH Relative humidity (%) VPD Vapour pressure deficit (kPa) SGR Shoot growth rate (cm.day-1) LA Leaf area per vine (m2) LA:F Leaf area to fruit ratio (m2.g.vine-1) WUE Water use efficiency (t.ML-1) TSS Total soluble solids (°Brix) TA Titratable acidity (g.L-1) α Percentage of anthocyanin ionisation (%) -2 -1 gs Stomatal conductance (m mol.m .s ) A Net leaf photosynthesis (µmol.m-2.s-1) T Net leaf transpiration (µmol.m-2.s-1) A/T Transpiration efficiency (µmol.mmol-1) Ci Internal leaf carbon dioxide concentration (µmol.mol-1) Ca Atomospheric carbon dioxide concentration (µmol.mol-1) ∆ Carbon isotope discrimination

ΨL Leaf water potential (MPa)

ΨS Osmotic potential (MPa) SS Surface sterilisation NS Non-surface sterilisation

viii

TABLE OF CONTENTS

1 Introduction and literature review...... 1 1.1 Introduction ...... 1 1.2 Canopy management ...... 1 1.2.1 Winter pruning ...... 3 1.2.2 Pruning systems...... 5 Hand spur pruning...... 5 Mechanical hedging...... 6 Minimal pruning...... 7 1.3 Irrigation management ...... 9 1.3.1 Partial rootzone drying...... 10 1.4 Integration of canopy and irrigation management...... 12 1.5 Conclusion...... 13 2 Methodology ...... 15 2.1 Experimental site...... 15 2.1.1 Vineyard characteristics ...... 15 2.1.2 Soil characteristics...... 16 2.2 Experimental field trial...... 17 2.2.1 Irrigation and pruning treatments...... 17 2.2.2 Experimental design...... 17 2.2.3 Irrigation scheduling...... 19 2.3 Climatic conditions for seasons 2000-2003...... 24 2.4 Phenological growth dates for seasons 2001-2003 ...... 25 3 Irrigation and pruning effects on yield...... 27 3.1 Introduction and experimental aims...... 27 3.2 Methodology ...... 28 3.2.1 Canopy development and morphology ...... 28 3.2.2 Berry development...... 29 3.2.3 Harvest components ...... 30 3.2.4 Statistical analysis ...... 30 3.3 Results ...... 31 3.3.1 Canopy development and morphology ...... 31 3.3.2 Leaf area development...... 32 3.3.3 Pruning weights...... 34 3.3.4 Berry development...... 35 3.3.5 Berry maturation ...... 35 3.3.6 Bunch number and bunch weight ...... 42 3.3.7 Yield...... 44 3.3.8 Berry weight and berry number ...... 46

3.3.9 Vine balance...... 48 3.4 Discussion...... 49 3.4.1 Pruning and irrigation effects on canopy development...... 49 3.4.2 Treatment effects on berry development...... 51 3.4.3 Pruning effect on yield components ...... 52 3.4.4 PRD influence on yield components ...... 53 3.4.5 Treatment effects on vine balance...... 54 3.4.6 Improvements in water use efficiency caused by PRD ...... 55 3.5 Conclusions ...... 56 4 Irrigation and pruning effects on fruit and wine composition...... 58 4.1 Introduction and experimental aims...... 58 4.2 Methodology ...... 60 4.2.1 Determination of fruit total soluble solids, pH and titratable acidity...... 60 4.2.2 Determination of fruit anthocyanin and phenolic concentration...... 61 4.2.3 Small-scale wine production ...... 61 4.2.4 Determination of wine composition...... 62 4.2.5 Statistical analysis ...... 63 4.3 Results ...... 63 4.3.1 Fruit total soluble solid, pH and titratable acidity results...... 63 4.3.2 Fruit anthocyanin and phenolics concentrations and content ...... 66 4.3.3 Relationship between anthocyanin concentration and temperature ...... 68 4.3.4 Wine pH and titratable acidity results ...... 69 4.3.5 Wine spectral evaluation...... 71 4.4 Yield and fruit composition correlations ...... 76 4.5 Yield, fruit and wine composition correlations ...... 79 4.6 Discussion...... 82 4.7.1 Pruning effects on fruit and wine composition...... 82 4.7.2 Irrigation effects on fruit and wine composition ...... 83 4.7.3 Irrigation and pruning effects on fruit and wine composition ...... 85 4.7.4 Influence of yield on fruit and wine composition...... 87 4.7.5 Influence of berry size and TSS on fruit and wine composition ...... 88 4.7 Conclusions ...... 89 5 Physiological response to irrigation and pruning treatments ...... 91 5.1 Introduction and experimental aims...... 91 5.2 Methodology ...... 93 5.2.1 Midday leaf gas exchange ...... 93 5.2.2 Carbon isotope discrimination ...... 94 5.2.3 PRD vine response to increasing vapour pressure deficit...... 95 5.2.4 Statistical analysis ...... 95 5.3 Results of irrigation and pruning effects on midday leaf gas exchange ...... 96 5.3.1 Midday leaf gas exchange for 2001-02 and 2002-03 ...... 96

5.3.2 Vapour pressure deficit and soil water content...... 100 5.3.2 Phenological effect on midday leaf gas exchange...... 101 5.4 Irrigation and pruning effects on carbon isotope discrimination ...... 109 5.5 PRD vine physiological response to increasing vapour pressure deficit ...... 110 5.5.1 Diurnal increase in vapour pressure deficit...... 110 5.5.2 Diurnal response of PRD vines ...... 112 5.6 Discussion...... 126 5.6.1 Irrigation effects on leaf gas exchange...... 126 5.6.2 Pruning effects on leaf gas exchange ...... 127 5.6.3 Irrigation and pruning effect on leaf gas exchange...... 127 5.6.4 Leaf gas exchange at different phenological stages ...... 129 5.6.5 Irrigation and pruning effects on carbon isotope discrimination...... 131 5.6.6 Diurnal response of PRD vines to increasing vapour pressure deficit...... 132 5.6.7 Proposed mechanisms for stomatal closure by PRD...... 134 5.6.8 Single leaf gas exchange measurements...... 135 5.7 Conclusions ...... 135 6 Botrytis bunch rot development and bunch architecture...... 137 6.1 Introduction and experimental aims...... 137 6.2 Methodology ...... 139 6.2.1 Isolates of Botrytis cinerea...... 139 6.2.2 Spore concentration ...... 140 6.2.3 Botrytis field inoculations at flowering ...... 141 6.3.4 Late season Botrytis field inoculations...... 141 6.2.5 Bunch architecture...... 142 6.2.6 Statistical analysis ...... 142 6.3 Results ...... 143 6.3.1 Effect of spore concentration on incidence and severity of Botrytis ...... 143 6.3.2 Botrytis field inoculations at flowering ...... 143 6.3.3 Late season Botrytis field inoculations...... 150 6.3.4 Seasonal effects on bunch architecture...... 154 6.3.5 Integrated irrigation and pruning effects on bunch architecture ...... 155 6.3.6 Bunch architecture and Botrytis incidence and severity at harvest ...... 158 6.4 Discussion...... 159 6.4.1 Influence of spore concentration on inoculation experiments ...... 159 6.4.2 Botrytis incidence and severity at different bunch development stages...... 160 6.4.3 Susceptibility of bunches to late season Botrytis inoculations ...... 161 6.4.4 Influence of irrigation and pruning treatments on Botrytis bunch rot...... 162 6.4.5 Irrigation and pruning effects on bunch architecture and Botrytis infection...... 163

6.5 Conclusions ...... 165 7 General Discussion ...... 166 7.1 Introduction to the experiment ...... 166 7.2 Irrigation effects on sustainability of winegrape production ...... 167 7.3 Pruning effects on sustainability of winegrape production...... 169 7.4 Integrated irrigation and pruning effects on sustainability of winegrape production ..171 7.5 Recommendations for the Australian Wine Industry...... 173 8 References...... 174 APPENDIX A...... 194 APPENDIX B ...... 195 APPENDIX C ...... 196 APPENDIX D...... 197

xii

LIST OF FIGURES

Figure 1.1: Response of net photosynthesis of grapevine laminae to sunlight, measured as PAR (Smart 1985)...... 2

Figure 1.2: Stomatal conductance of Chardonnay (gs) for PRD vines (σ) and non-irrigated vines (□) expressed as a percentage of fully irrigated vines (Dry and Loveys 1999) ...... 11

Figure 1.3: Net Photosynthesis of Chardonnay (A) for PRD vines (σ) and non-irrigated vines (□) expressed as a percentage of fully irrigated vines (Dry and Loveys 1999)...... 11

Figure 1.4: Shoot growth rate of Chardonnay (SGR; cm.d-1) for PRD vines (σ) and non-irrigated vines (□) expressed as a percentage of fully irrigated vines (Dry and Loveys 1999)...... 12

Figure 2.1: Systematic diagram of alternating PRD irrigation cycles and infrastructure at Deakin Estate; represents non-irrigating drip line and dry rootzone and represents irrigating drip line and associated wet rootzone of grapevine...... 18

Figure 2.2: Schematic diagram of experimental field site at Deakin Estate; 2 contiguous latin squares of pruning plots (8 vines each) superimposed on to irrigation rows...... 18

Figure 2.2: Photograph of PRD irrigation infrastructure at Deakin Estate, 2 parallel drip lines irrigate half the rootzone of the vine at a time as indicated by the wet and dry rootzone...... 19

Figure 2.4: Figure 2.4: a. Standard drip (SD) volumetric soil-water content of surface (20 cm) and sub-surface soil (50 cm) b. PRD irrigation volumetric soil-water content of surface soil (20-40cm) of north and south side of vine and c. soil-water content of sub-surface soil (50-70cm) of north and south side of vine, as measured by TDR and corresponding irrigation and rainfall events for 2000-01 growing season...... 21

Figure 2.5: a. Standard drip (SD) volumetric soil-water content of surface (20 cm) and sub-surface soil (50 cm) b. PRD irrigation volumetric soil-water content of surface soil (20-40cm) of north and south side of vine and c. soil- water content of sub-surface soil (50-70cm) of north and south side of vine, as measured by TDR and corresponding irrigation and rainfall events for 2001-02 growing season...... 22

Figure 2.6: a. Standard drip (SD) volumetric soil-water content of surface (20 cm) and sub-surface soil (50 cm) b. PRD irrigation volumetric soil-water content of surface soil (20-40cm) of north and south side of vine and c. soil- water content of sub-surface soil (50-70cm) of north and south side of vine, as measured by TDR and corresponding irrigation and rainfall events for 2002-03 growing season...... 23

Figure3.1: Pruning treatment effects on shoot growth in season 2000-01; a. shoot length, b. shoots growth rate, c. number of nodes.shoot-1 and d. internode length from budburst to 89 days after budburst (DAB). Significant differences were calculated by Fisher’s least significant difference (LSD) at each time point. LSD is represented by bars and levels of significance are denoted by ***P<0.001, **P<0.01, *P<0.05...... 32

Figure 3.2: a Pruning and b. Irrigation effects on total leaf area development n season 2000-01. Significant differences were calculated by Fisher’s least significant difference (LSD) at each time point. Levels of significance are denoted by ***P<0.001, **P<0.01, *P<0.05, ns = not significant...... 33

Figure 3.3: Irrigation and pruning effects on berry weight during the maturation period in season 2000-01. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant...... 36

Figure 3.4: Irrigation and pruning effects on berry weight during the maturation period in season 2001-02. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant...... 36

xiii graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant...... 37

Figure 3.6: Integrated irrigation and pruning effects on sugar (TSS) accumulation during the maturation period in season 2000-01. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant...... 38

Figure 3.7: Integrated irrigation and pruning effects on sugar (TSS) accumulation during the maturation period in season 2001-02. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant...... 38

Figure 3.8: Integrated irrigation and pruning effects on sugar (TSS) accumulation during the maturation period in season 2002-03. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant...... 39

Figure 3.9: Parallel regression model of sugar accumulation in berries on vines with integrated irrigation and pruning treatments for season 2000-01. Refer to Table 3.3 for equations ...... 40

Figure 3.10: Parallel regression model of sugar accumulation in berries from vines with integrated irrigation and pruning treatments for season 2001-02. Refer to Table 3.3 for equations ...... 41

Figure 3.11: Common regression model of sugar accumulation in berries from vines with integrated irrigation and pruning treatments for season 2002-03. Refer to Table 3.3 for equations ...... 41

Figure 5.1: Curvilinear relationship between photosynthesis and stomatal conductance for midday measurements of all treatments (R2 = 0.748), P<0.001 ...... 99

Figure 5.2a: Midday stomatal conductance (gs) of SD (c) and PRD ( ) irrigation treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 102

Figure 5.2b: Midday stomatal conductance (gs) of SPUR (c), MECH (c) and MIN ( ) pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 103

Figure 5.2c: Midday stomatal conductance (gs) of integrated irrigation and pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 103

Figure 5.3a: Midday leaf photosynthesis (A) of SD (c) and PRD ( ) irrigation treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 104

Figure 5.3b: Midday leaf photosynthesis (A) of SPUR (c), MECH (c) and MIN ( ) pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 104

Figure 5.3c: Midday leaf photosynthesis (A) of integrated irrigation and pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 105

Figure 5.4a: Midday transpiration (T) of SD (c) and PRD ( ) irrigation treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 106

Figure 5.4b: Midday transpiration (T) of SPUR (c), MECH (c) and MIN ( ) pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant ...... 106

xiv

Figure 5.4c: Midday transpiration (T) of integrated irrigation and pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 106

Figure 5.5a: Midday internal CO2 concentration (Ci) of SD (c) and PRD ( ) irrigation treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 107

Figure 5.5b: Midday internal CO2 concentration (Ci) of SPUR (c), MECH (c) and MIN ( ) pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 107

Figure 5.5c: Midday internal CO2 concentration (Ci) of integrated irrigation and pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant ...... 108

Figure 5.6a: Midday transpiration efficiency (A/T) of SD (c) and PRD ( ) irrigation treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 108

Figure 5.6b: Midday transpiration efficiency (A/T) of SPUR (c), MECH (c) and MIN ( ) pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant...... 109

Figure 5.6c: Midday transpiration efficiency (A/T) of integrated irrigation and pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant ...... 109

Figure 5.7: Diurnal change in vapour pressure deficit (VPD, kPa) for four diurnal measurement days...... 111

Figure 5.8a: Diurnal response of PRD on stomatal conductance (gs) on 24 January 2002, ± SEM...... 113

Figure 5.8b: Diurnal response of PRD on photosynthesis (A) on 24 January 2002, ± SEM...... 113

Figure 5.8c: Diurnal response of PRD on transpiration (T) on 24 January 2002, ± SEM...... 114

Figure 5.8d: Diurnal response of PRD on transpiration efficiency (A/T) on 24 January 2002, ± SEM ...... 114

Figure 5.8e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 24 January 2002, ± SEM ....114

Figure 5.8f: Diurnal response of PRD on leaf water potential (ΨL) on 24 January 2002, ± SEM ...... 115

Figure 5.8g: Diurnal response of PRD on osmotic potential (ΨS) on 24 January 2002, ± SEM...... 115

Figure 5.8h: Diurnal response of PRD on turgor on 24 January 2002, ± SEM...... 115

Figure 5.9a: Diurnal response of PRD on stomatal conductance (gs) on 25 January 2002, ± SEM...... 116

Figure 5.9b: Diurnal response of PRD on photosynthesis (A) on 25 January 2002, ± SEM...... 117

Figure 5.9c: Diurnal response of PRD on transpiration (T) on 25 January 2002, ± SEM...... 117

Figure 5.9d: Diurnal response of PRD on transpiration efficiency (A/T) on 25 January 2002, ± SEM...... 117

Figure 5.9e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 25 January 2002, ± SEM ....118

Figure 5.9f: Diurnal response of PRD on leaf water potential (ΨL) on 25 January 2002, ± SEM ...... 118

Figure 5.9g: Diurnal response of PRD on osmotic potential (ΨS) on 25 January 2002, ± SEM...... 118

Figure 5.9h: Diurnal response of PRD on turgor on 25 January 2002, ± SEM...... 119

Figure 5.10a: Diurnal response of PRD on stomatal conductance (gs) on 26 February 2002, ± SEM...... 120

Figure 5.10b: Diurnal response of PRD on photosynthesis (A) on 26 February 2002, ± SEM...... 120

Figure 5.10c: Diurnal response of PRD on transpiration (T) on 26 February 2002, ± SEM...... 120

Figure 5.10d: Diurnal response of PRD on transpiration efficiency (A/T) on 26 February 2002, ± SEM...... 121

Figure 5.10e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 26 February 2002, ± SEM 121

Figure 5.10f: Diurnal response of PRD on leaf water potential (ΨL) on 26 February 2002, ± SEM ...... 121

Figure 5.10g: Diurnal response of PRD on osmotic potential (ΨS) on 26 February 2002, ± SEM...... 122

Figure 5.10h: Diurnal response of PRD on turgor on 26 February 2002, ± SEM...... 122

Figure 5.11a: Diurnal response of PRD on stomatal conductance (gs) on 14 February 2003, ± SEM...... 123

Figure 5.11b: Diurnal response of PRD on photosynthesis (A) on 14 February 2003, ± SEM...... 123

Figure 5.11c: Diurnal response of PRD on transpiration (T) on 14 February 2003, ± SEM...... 124

Figure 5.11d: Diurnal response of PRD on transpiration efficiency (A/T) on 14 February 2003, ± SEM ...... 124

Figure 5.11e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 14 February 2002, ± SEM 124

Figure 5.11f: Diurnal response of PRD on leaf water potential (ΨL) on 14 February 2003, ± SEM ...... 125

Figure 5.11g: Diurnal response of PRD on osmotic potential (ΨS) on 14 February 2003, ± SEM ...... 125

Figure 5.11h: Diurnal response of PRD on turgor on 14 February 2003, ± SEM ...... 125

Figure 6.1: Mean incidence and severity of Botrytis in seasons 2001-02 and 2002-03 at 4 stages of bunch development; flowering, fruit-set, veraison and harvest. Significant differences between treatment means denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) ...... 144

Figure 6.2: Incidence of Botrytis in season 2001-02 of surface (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 phenological stages; flowering, fruit-set, veraison and harvest. Significant differences between phenological stage*treatment*sterilisation were calculated by Fisher’s least significant difference (LSD 5% level). Significance levels are represented by ***P<0.001, **P<0.01, *P<0.05, ns = not significant...... 148

Figure 6.3: Incidence of Botrytis in season 2002-03 of surface (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 phenological stages; flowering, fruit-set, veraison and harvest. Significant differences between phenological stage*treatment*sterilisation were calculated by Fisher’s least significant difference (LSD 5% level). Significance levels are represented by ***P<0.001, **P<0.01, *P<0.05, ns = not significant...... 148

Figure 6.4: Severity of Botrytis in season 2001-02 of surface (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 phenological stages; flowering, fruit-set, veraison and harvest. Significant differences between phenological stage*treatment*sterilisation were calculated by Fisher’s least significant difference (LSD 5% level). Significance levels are represented by ***P<0.001, **P<0.01, *P<0.05, ns = not significant...... 149

xvi

Figure 6.5: Severity of Botrytis in season 2002-03 of surface (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 phenological stages; flowering, fruit-set, veraison and harvest. Significant differences between phenological stage*treatment*sterilisation were calculated by Fisher’s least significant difference (LSD 5% level). Significance levels are represented by ***P<0.001, **P<0.01, *P<0.05, ns = not significant...... 150

Figure 6.6: Photograph of representative bunch architecture for each of the combined irrigation and pruning treatments in season 2001-02...... 156

Figure 6.7: Effect of irrigation and pruning treatments at harvest 2001-02 () and 2002-03 () on bunch architectural parameters; a. mean bunch weight, b. berry number vine-1, c. bunch volume, d. maximum width, e. bunch length and f. compactness. Significant differences were calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant.157

Figure 6.8 Linear relationships between a. incidence (R2= 0.87, P<0.01) and b. severity (R2= 0.94, P<0.01) of Botrytis and mean bunch weight for integrated irrigation and pruning treatments in 2001-02...... 158

Figure 6.9: Linear relationships between a. incidence (R2= 0.014, P<0.01) and b. severity (R2= 0.009, P<0.01) of Botrytis and mean bunch weight for integrated irrigation and pruning treatments in 2002-03...... 159

xvii

LIST OF TABLES

Table 1.1: Influence of pruning level on mean photosynthetic rate (Pn), stomatal conductance (gs), leaf water potential (ψL), average leaf area and shoot length of Cabernet Sauvignon vines at Mildura, Victoria. Significant difference represented by different letters (P<0.05), adapted from Sommer et al. (1993) ...... 4

Table 1.2: Influence of pruning level on mean yield, bunch number, berry weight, total soluble solid (ºBrix), pH and titratable acidity of Cabernet Sauvignon vines at Mildura, Victoria. Significant difference represented by different letters (P<0.05), adapted from Sommer et al. (1993)...... 4

Table1.3: Comparison of incidence and severity of Botrytis bunch rot on bunches of Chardonnay at Coonawarra from minimally pruned, mechanically hedged and hand pruned vines, adapted from Emmett et al. (1995) ...... 8

Table 2.1: Average monthly climatic data from Mildura Airport: latitude 34°14′S, longitude 142°5′E, 54 years of records (Bureau of Meteorology 2003) ...... 15

Table 2.2: Soil characteristic of surface and subsurface soil of Shiraz vineyard, Deakin Estate, Iraak, Victoria: texture (LS= loamy sand, SL= sandy loam, SCL = sandy clay loam), ECe (dS/m), pH, TOC (%), N (%) and P (mg/kg) of surface soil (10-20 cm) and subsurface soil (50-70 cm)...... 17

Table 2.3: Total seasonal irrigation inputs (ML.ha-1) for the three experimental seasons for SD and PRD irrigation treatments at Deakin Estate, Iraak, Victoria...... 20

Table 2.4: Mean monthly climatic data from Nangiloc weather station for the 2000-01 growing season...... 24

Table 2.5: Mean monthly climatic data from Nangiloc weather station for the 2001-02 growing season...... 25

Table 2.6: Mean monthly climatic data from Nangiloc weather station for the 2002-03 growing season...... 25

Table 2.7: Calendar date for phenological stages of budburst, flowering, veraison and harvest for Shiraz grapevines at Deakin Estate for seasons 2000-01, 2001-02 and 2002-03...... 26

Table 3.1: Irrigation and pruning effect on total leaf area.vine-1 (LA, m2.vine-1) at maximum canopy development, 22 November 2000-02. * Indicates significance level; *** P<0.001, ** P<0.01, * P<0.05, ns = non significant .34

Table 3.2: Irrigation and pruning effects on pruning weight, cane number.vine-1 and mean cane weight at winter pruning in season 2002-03. * Indicates significance level; *** P<0.001, ** P<0.01, * P<0.05, ns = not significant, different letters denote significant differences between means in each column...... 34

Table 3.3: Equations and coefficient of determination (R2) for parallel regression model of sugar accumulation for season’s 2000-01 and 2001-02 and common regression model of sugar accumulation for season 2002-03. Regression model Equations are Total Soluble Solids = constant + slope*Days After Veraison...... 40

Table 3.4: Integrated treatment effects on sugar accumulation rate (°Brix.day-1) for seasons 2000-01, 2001-02 and 2002-03. Different letters denote significant differences between seasonal means. *** indicates P<0.001...... 42

TABLE 3.5: Bunch number.vine-1 of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 43

TABLE 3.6: Bunch weight (g) of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 44

xviii

Table 3.7: Coefficient of determination (R2) and statistical significance level (P value) of the negative linear relationship between bunch number vine-1 and bunch weight for all treatments in each season and over the 3 seasons. Significance level is denoted by ** P<0.01, * P<0.05 ...... 44

TABLE 3.8: Yield of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 45

TABLE 3.9: Berry weight of vines with integrated irrigation and pruning treatments and seasonal means for 2000- 01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 47

TABLE 3.10: Berry number.bunch-1 of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 47

TABLE 3.11: Leaf area: fruit ratio of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 48

TABLE 3.12: Water use efficiency of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 49

TABLE 4.1: The total soluble solid concentration of berries from vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant ...... 64

TABLE 4.2: The pH of berries from vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 65

TABLE 4.3: The titratable acidity of berries from vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 65

TABLE 4.4: The anthocyanin a. concentration and b. content of berries from vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant ...... 67

xix significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant ...... 68

Table 4.6: Correlation coefficients of mean anthocyanin concentration of all treatments with mean daily temperature for the months of December, January and February for the three experimental seasons, n=3. Bold script represents significance at 1% level, P<0.01...... 69

TABLE 4.7: The pH of wine from vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 70

TABLE 4.8: The titratable acidity of wine from vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 70

TABLE 4.9: The colour density of wine from integrated irrigation and pruning treatments for season 2000-01, 2001-02, 2002-03 and seasonal means. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = non significant...... 71

TABLE 4.10: The colour hue of wine from integrated irrigation and pruning treatments for season 2000-01, 2001- 02, 2002-03 and seasonal means. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = non significant...... 72

TABLE 4.11: The ionised anthocyanin concentration of wine from integrated irrigation and pruning treatments for season 2000-01, 2001-02, 2002-03 and seasonal means. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = non significant...... 73

TABLE 4.12: The total anthocyanin concentration of wine from integrated irrigation and pruning treatments for season 2000-01, 2001-02, 2002-03 and seasonal means. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = non significant ...... 74

TABLE 4.13: The degree of ionisation of anthocyanins of wine from integrated irrigation and pruning treatments for season 2000-01, 2001-02, 2002-03 and seasonal means. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = non significant...... 75

TABLE 4.14: The total phenolic concentration of wine from integrated irrigation and pruning treatments for season 2000-01, 2001-02, 2002-03 and seasonal means. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = non significant ...... 75

Table 4.16: Correlation coefficients of mean yield components [yield, bunch number.vine-1 (Bunch No.), bunch weight (Bunch Wt.), berry number.bunch-1 (Berry No.), leaf area: fruit (LA:F), water use efficiency (WUE)] and fruit composition parameters[total soluble solids (TSS), pH, titratable acidity (TA), anthocyanin concentration (Antho.) and phenolic concentration (Phenol)] for seasons 2000-03, n=18. Bold script represents significance at 5% level...... 78

Table 5.1: Mean midday leaf stomatal conductance (gs) of vines with integrated irrigation and pruning treatments and seasonal means for 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 97

Table 5.2: Mean midday photosynthesis (Α) of vines with integrated irrigation and pruning treatments and seasonal means for 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 97

Table 5.3: Mean midday Transpiration (Τ) of vines with integrated irrigation and pruning treatments and seasonal means for 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant ...... 98

Table 5.4: Mean midday intercellular CO2 concentration (Ci) of vines with integrated irrigation and pruning treatments and seasonal means for 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant ...... 98

Table 5.5: Mean midday transpiration efficiency (Α/Τ) of vines with integrated irrigation and pruning treatments and seasonal means for 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 99

Table 5.6: Midday climatic conditions: temperature (T), relative humidity (RH) and vapour pressure deficit (VPD), number of days after an irrigation event (DAI), corresponding amounts of irrigation water applied to SD and PRD treatments, and number of days after a rainfall event (DAR) and amount of rain (mm) for each of the midday gas exchange measurements...... 100

Table 5.7: Midday volumetric soil water content (θv %) of SD and PRD (north and south side of the vine) of surface (20-40 cm) and subsurface soil (50-70 cm) as measured hourly by TDR probes located adjacent to vines...... 101

xxi

Table 5.8: Mean Carbon Isotope Discrimination (∆) of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant...... 110

Table 5.9: Hourly climatic conditions (Temperature °C, Relative Humidity % and Vapour Pressure Deficit kPa) for each of the diurnal measurement days...... 111

Table 5.10: Average volumetric soil water contents (θv %) of diurnal measurement days for SD and PRD (north and south side of the vine) of surface (20-40 cm) and subsurface soil (50-70 cm) as measured hourly by TDR probes located adjacent to vines ...... 111

Table 6.1: Incidence and severity of B. cinerea on bunches from all treatments inoculated in the field at flowering in seasons 2001-02 and 2002-03 at spore suspension concentrations of 0, 102 or 104 spores.mL-1. Significant differences calculated by Fisher’s least significant difference (LSD 5% level) and denoted by ***P<0.001, **P<0.01, *P<0.05, ns = non significant ...... 143

Table 6.2: The mean incidence of Botrytis in season 2001-02 on inoculated and control bunches, surface sterilised (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 stages of bunch development. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 145

Table 6.3: The mean incidence of Botrytis in season 2002-03 on inoculated and control bunches, surface sterilised (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 stages of bunch development. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 146

Table 6.4: The mean severity of Botrytis in season 2001-02 on inoculated and control bunches, surface sterilised (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 stages of bunch development. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 146

Table 6.5: The mean severity of Botrytis in season 2002-03 on inoculated and control bunches, surface sterilised (SS) and non-surface sterilised (NS) bunches from integrated irrigation and pruning treatments at 4 stages of bunch development. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 147

Table 6.6: The mean incidence and severity of Botrytis in seasons 2001-02 and 2002-03 on bunches inoculated at flowering, veraison and pre-harvest. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 151

Table 6.7: The incidence of Botrytis in seasons 2001-02 and 2002-03 on bunches inoculated at flowering, veraison and pre-harvest. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 151

Table 6.8: Severity of Botrytis in seasons 2001-02 and 2002-03 on bunches inoculated at flowering, veraison and pre-harvest. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance level are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 152

Table 6.9: The incidence of Botrytis in seasons 2001-02 and 2002-03 on bunches from integrated irrigation and pruning treatments inoculated at flowering, veraison and pre-harvest. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 153

xxii

Table 6.10: Severity of Botrytis in seasons 2001-02 and 2002-03 on bunches from integrated irrigation and pruning treatments inoculated at flowering, veraison and pre-harvest. Significant differences between treatment means are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant ...... 154

Table 6.11: Bunch architectural parameters, pooled over all treatments, at harvest in seasons 2001-02 and 2002- 03. Significant differences between seasons are denoted by different letters as calculated by Fisher’s least significant difference (LSD 5% level) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant...... 155

Chapter 1: Introduction and Literature Review 1

1 LITERATURE REVIEW

1.1 INTRODUCTION

Modern viticultural practices have been employed by Australia’s wine industry to ensure sustainable winegrape production. Sustainable winegrape production encompasses viticultural management, economics and the environment and for this study, is defined as the consistent production of quantities of high quality grapes sufficient to return a net profit over an extended period of time with minimal inputs and environmental impact. The long-term economic and environmental sustainability of winegrape production is dependent on continual improvement in fruit and wine quality, preservation of yield and reduced water and chemical inputs. Mechanisation of vineyard processes at harvest and pruning has been widely accepted by Australia’s wine industry because of reductions in production cost. Similarly, drip irrigation strategies that trigger water stress responses in grapevines are being adopted by industry to improve fruit quality and reduce water inputs, particularly on red winegrape varieties. The adoption of mechanised light pruning practices or induced water stress irrigation strategies have generally improved production, wine quality and water use efficiency (Clingeleffer 1993; McCarthy 1996; Dry and Loveys 1998).

Nevertheless, further enhancement of the sustainability of Australia’s wine industry is essential because of the continual expansion of wine production and subsequent, increased competition at a global scale. This is particularly important for warm, irrigated regions, where the bulk of Australia’s winegrapes are sourced. Sustainability, in terms of production, quality, water use efficiency and disease control may be improved by adopting an integrated approach to viticulture. In this review, the effects of canopy management and induced water-stress irrigation strategies on winegrape production, fruit and wine quality, vine physiology and disease development and the potential of integrated viticultural management to improve sustainability of winegrape production are discussed.

1.2 CANOPY MANAGEMENT

The microclimate within and directly surrounding a canopy is dependent on the canopy density (Smart and Robinson 1991). Climatic factors most influenced by canopy density in

Chapter 1: Introduction and Literature Review 2 grapevines are photosynthetic photon flux density, light quality, wind speed, evaporation, temperature and humidity (Smart et al. 1985). Transmittance of photosynthetically active radiation through leaves is small (less than 10%) (Pearman 1966). As a result, the light levels within dense grapevine canopies are very low. Although, interior leaves of dense canopies have lower photosynthetic rates (Figure 1.1), they still transpire and reduce net water use efficiency of the grapevine (Escalona et al. 1998).

Consequences of excessive canopy density and vine vigour are reduced yield due to lower fruit initiation in buds (May 1965) and poor fruit set associated with early bunch stem necrosis (Jackson 1991). In addition, dense canopies may produce poor quality fruit and wine. Fruit may have lower sugar, tartaric acid, anthocyanin and phenolic concentrations and increased berry size, K+ concentrations, pH and malic acid (Morrison 1989; Dokoozlian and Kliewer 1995). Wines produced from Sauvignon blanc berries from dense canopies may have an undesirable “herbaceous” character (Allen et al. 1996).

The microclimate of high-density canopies has also been associated with increased disease development on grapevines because of restricted air movement and increased humidity (Thomas et al. 1988). High-density canopies may have increased the incidence of Botrytis bunch rot (Gubler et al. 1987). Disease control may also be difficult because of inefficient chemical spray penetration within the canopy to infected areas (ie. bunches).

Figure 1.1: The response of net photosynthesis of grapevine laminae to sunlight, measured as photosynthetically active radiation (PAR) (Smart 1985).

Chapter 1: Introduction and Literature Review 3

Canopy management involves a range of viticultural practices that alter the position and number of shoots, leaves and bunches. The aim of canopy management is to optimise sunlight interception, photosynthetic capacity and vine microclimate (Smart et al. 1990). The benefits of canopy management are improved yield and quality (Smart et al. 1990), reduced incidence of some diseases (Savage and Sall 1984; Emmett et al. 1995) and the facilitation of mechanisation (Clingeleffer 1983, 1992). Canopy management includes temporary practices such as winter and summer pruning, leaf removal in the fruiting zone, shoot thinning and permanent practices such as trellis-training systems. The level of winter pruning can control shoot number and spacing and thereby control canopy density. Leaf removal may increase sun exposure in the fruiting zone by reducing leaf area, while shoot thinning may be used to reduce shoot crowding. Trellis-training systems may increase grapevine growth and capacity for assimilation (May et al. 1973). Trellis-training systems, such as vertical shoot positioning (VSP) may also reduce shading within the canopy.

This project focuses on the comparison of different winter pruning techniques for canopy management, in particular hand spur pruning, mechanical hedging and minimal pruning. The impact of spur pruning, mechanical hedging and minimal pruning on canopy development, physiology, productivity, fruit composition and disease development will be discussed.

1.2.1 Winter Pruning

Pruning is the removal of vegetative parts of the vine including shoots, canes and leaves (Winkler et al. 1974) and is an important cultural practice of the vineyard. Pruning is also the most expensive and labour consuming aspect of vineyard management after harvest (Clingeleffer 1992). Pruning practices are chosen in response to vineyard vigour, grapevine variety, desired wine quality, production costs and climatic conditions of the region.

Pruning aims to select nodes that produce fruitful shoots, regulate shoot number, bunch number, bunch architecture and improve fruit quality (Tassie and Freeman 1992). Severe pruning results in vigorous growth to compensate for the reduced number of shoots per vine. Alternatively, light pruning reduces vegetative growth because of a greater number of shoots per vine. The level of pruning affects canopy architecture and, ultimately, canopy density. Light pruning practices result in large open, permanent canopies in contrast to dense canopies associated to severe pruning levels (Sommer et al. 1993). Although canopy architecture is

Chapter 1: Introduction and Literature Review 4 distinctly different between light and severe pruning levels, Downton and Grant (1992), Sommer et al. (1993) and Lasko et al. (1996) reported no differences in photosynthetic capacity when gas exchange and leaf water potential measurements were conducted on individual, fully sun-exposed leaves (Table 1.1). However, as photosynthesis, stomatal conductance and leaf water potential vary with leaf size, thickness, age (Syvertsen et al. 1995) and, in particular, position (Candolfi-Vasconcelos et al. 1994), there are problems associated with single leaf photosynthetic measurements. Harley and Baldocchi (1995) found that net photosynthesis was reduced by 31% in shaded leaves compared to fully sun exposed leaves in a deciduous forest. The rate of development of the canopy, degree of canopy density, ratio of shaded to exposed leaves and ultimately pruning level can affect net photosynthesis and stomatal conductance of the grapevine. Lasko et al. (1996) and Poni et al. (2000) showed canopy net CO2 exchange was significantly higher in minimally pruned vines compared to hand-pruned vines early in the growing season when their canopies were more advanced. The difference in canopy net CO2 exchange between light and severe pruned vines was reduced after flowering as canopy size equilibrated between treatments.

Table 1.1: Influence of pruning level on mean leaf photosynthetic rate (A), leaf stomatal conductance (gs), leaf water potential (ψL), average leaf area and shoot length of Cabernet Sauvignon vines at Mildura, Victoria. Significant difference in means represented by different letters (p<0.05), adapted from Sommer et al. (1993).

Pruning Level A gs ΨL Leaf Area Shoot Length (µmol m-2s-1) (mmol m-2s-1) (MPa) (cm-2) (cm) Light (minimally pruned) 10.6 234 -0.95 65b 39a Severe (cane pruned) 11.3 245 -0.95 76a 72b

Severe pruning reduces yield because of reduced crop load but bunch weight, berries per bunch and berry weight are increased (Table 1.2) (Sommer et al. 1993). Light pruning levels increase yield through increased crop load but bunches are smaller and less compact (Reynold et al. 1994). The reduced bunch and berry weight may be due to increased partitioning of carbohydrates between more reproductive sinks.

Table 1.2: Influence of pruning level on mean yield, bunch number, berry weight, total soluble solid (TSS), pH and titratable acidity (TA) of Cabernet Sauvignon at Mildura Victoria. Significant difference in means represented by different letters (p<0.05), adapted from Sommer et al. (1993).

Pruning Level Yield Bunch No. Berry Weight TSS pH TA (t ha-1) (vine-1) (g) (ºBrix) (g L-1) Light (minimally pruned) 37.7b 585b 0.85a 24.0a 3.42a 5.92 Severe (cane pruned) 24.0a 144a 1.04b 24.6b 3.49b 5.83

Chapter 1: Introduction and Literature Review 5

Sommer et al. (1993) found that the effects of pruning level on fruit composition were negligible (Table 1.2). However, Bravdo et al. (1985) found that berry maturation was generally delayed in lightly pruned vines due to the greater crop load. Also, colour may be enhanced by lighter pruning due to increased exocarp to juice ratio of the smaller berries (Sommer and Clingeleffer 1993).

1.2.2 Pruning Systems

Hand Spur Pruning

Spur pruned vines are normally bilaterally cordon trained and two-node spurs are retained as the bearers. Spur pruning suits most grapevine varieties, excluding those with low fruitfulness at basal nodes (eg. Sultana). At pruning the distal shoot is removed and the proximal shoot is retained as the new two-node spur. Most spur-pruned vines are classified as severely to moderately pruned when there are 50 – 180 nodes per vine. Vegetative growth of spur pruned vines is highly vigorous because of the low shoot density and stimulation of growth near the pruning cut (Clingeleffer 1992).

Shoots of spur-pruned vines are long with high node number, long internode length and large leaf size (Clingeleffer and Krake 1992). As a consequence of low shoot number per vine, crop load is also 1-2 fold lower than lighter pruning systems (Clingeleffer and Sommer 1995). However, bunch weight, berry weight and berry number per bunch are increased. Reduced crop load of spur pruned vines increases the rate and degree of sugar maturation in berries compared to lighter pruned vines (Poni et al. 1994). Other fruit quality parameters including pH, titratable acidity and K+ concentration are largely unaffected by decreased crop load. However, improved wine quality (sensory evaluation, colour density and hue) has been reported for spur-pruned vines (Clingeleffer and Sommer 1995).

The resulting bunch architecture of spur-pruned vines (large, compact bunches) can have serious implications on disease development (Phillips et al. 1990; Vail and Marois 1991). The presence of free water is required for the germination of Botrytis cinerea (Jarvis 1980). Compact bunches exposed to rain may take longer to dry than loose bunches and fully

Chapter 1: Introduction and Literature Review 6 hydrated berries in compact bunches are more likely to rupture (Sall et al. 1982). This produces a humid microclimate within and around the bunch that can encourage the development of Botrytis bunch rot (Thomas et al. 1988). Also, the cuticle and epicuticular wax layers of a grape berry are its main defence against infections by pathogens. Marois et al. (1985) showed that epicuticular wax plays an important role in the resistance of berries to Botrytis infection. Damage to the cuticle and epicuticular wax can occur by berry-to-berry contact within the bunch (Marois 1986). Therefore, compact bunches are expected to have greater berry contact and, consequently, less resistance to infection.

The facilitation of mechanisation in commercial vineyards, labour shortages and adoption of lighter pruning techniques for increased production has seen a reduction in hand spur pruning in recent years in Australia. However, spur pruning is generally still used in small privately- owned or cool climate vineyards in Australia, which do not use mechanical harvesters as low bunch density is better suited to hand picking (Clingeleffer and Sommer 1995).

Mechanical Hedging

Mechanical pruning by hedging is a widely accepted cost saving viticultural practice in Australia, particularly in large vineyards. Hedged vines can be trained to a single wire, narrow T or 2-vertical wire trellis and pruned to 1 to 6 nodes per bearer depending on the desired pruning level. Pruning occurs vertically and horizontally to produce a hedge shape, using circular saws or cutter bars mounted on tractors. Increased vigour can occur in mechanically hedged vines in cool regions of Australia, as a result of pruning cuts through one-year old wood (Clingeleffer 1992). Hedged vines have greater canopy density, node number, shoot number, bunch number and yield per vine compared to hand pruned vines (Smart et al. 1979). The increase in productivity of mechanically hedged vines generally has not compromised berry quality in warm climates. Clingeleffer (1993) found no significant differences in berry and wine quality between hedged, spur and minimally pruned vines but malate and K+ concentrations in the fruit and pH in the wine were slightly higher from hedged vines relative to the other pruning treatments. High K+ and pH levels in fruit and wine are often associated large canopies, as a result of excessive bunch shading (Allen et al. 1996).

Chapter 1: Introduction and Literature Review 7

The incidence and severity of Botrytis bunch rot may be increased by the adoption of mechanical hedging, as a result of increased source inoculum within the shoot zone (Emmett et al. 1994; Gubler et al. 1987). Large amounts of dead wood and mummified bunches that can support disease inoculum may be left within the shoot zone of vines after mechanical hedging of the vines. Also restricted air movement, greater humidity levels and decreased fungicide spray penetration in dense canopies of hedged vines may encourage disease development.

Mechanical hedging has been widely adopted in large-scale commercial vineyards as a major timesaving and cost-efficient canopy management technique. Although, productivity is increased without reduction to fruit quality compared to more intense pruning techniques, problems with increased disease incidence may be experienced due to denser canopy structure and greater levels of background inoculum. Therefore, chemical inputs to the vineyard may increase to aid disease control. The inherent disease risk associated with mechanical hedging may be minimised with removal of pruned material from the canopy and vineyard floor. However, the effectiveness of this method of disease control has not been evaluated in the literature.

Minimal Pruning

In the minimal pruning system of cordon-trained vines (MPCT) developed at CSIRO Merbein (Clingeleffer 1983), vines are trained on a high single or two-wire vertical trellis and left unpruned. Minimal pruning has been successful on a range of grapevine varieties grown on irrigated vineyards in warm regions (i.e. Sunraysia). Minimal pruning is particularly suitable for varieties with low basal fruitfulness (Sommer et al. 1995). However, in cool regions and areas of high shoot vigour (i.e. New Zealand), shoot skirting and summer bunch thinning may be necessary to reduce bunch number and encourage berry maturation on minimal pruned vines (Smart and Robinson 1991).

Minimally pruned vines have the capacity to “self-regulate” growth and maintain shape and productivity by abscission of non-lignified terminal growth during autumn (Possingham et al. 1990). Minimally pruned vines have a 3 to 10-fold increase in shoot number per vine compared to spur pruned vines. As a result node number per shoot, internode length and leaf size are reduced. Bunch number and yield are significantly increased but bunches are smaller,

Chapter 1: Introduction and Literature Review 8 less compact and consist of smaller berries compared to those on traditionally hand pruned vines (Clingeleffer 1983).

The fruiting zone of minimally pruned vines is predominantly on the outside of the canopy (Emmett et al. 1995) and bunches have greater exposure to the sun. The open canopy structure, bunch architecture and exposed fruiting zone of minimally pruned vines have been associated with improved disease control of Botrytis bunch rot and powdery mildew (Kidd 1989; Emmett et al. 1994, 1995). Emmett et al. (1995) found the incidence and severity of Botrytis bunch rot in Chardonnay from Coonawarra was reduced by 75% and 87% respectively, compared to spur pruned vines (Table 1.3). Small, loose bunches from minimally pruned vines may be less inclined to be infected by pathogens because berry cuticle and epicuticular wax damage from berry to berry contact is lower than in tight, compact bunches. Also, epicuticular wax production is greater on berries from exposed bunches where light intensity and temperature are higher (Marois et al. 1986; Percival et al. 1993).

Table 1.3: Comparison of incidence and severity of Botrytis bunch rot on bunches of Chardonnay from minimally pruned, mechanically hedged and hand pruned vines at Coonawarra, adapted from Emmett et al. (1995).

Botrytis bunch rot Minimally Pruned Mechanically Hedged Spur Pruned Incidence % 18.3 71.2 71.2 Severity % 2.40 18.8 16.5

Minimal pruning does not significantly alter winegrape composition compared to more severe pruning techniques (Poni et al. 2000). However, sugar maturation may be delayed by minimal pruning given greater crop level. In conjunction with lower sugar concentrations, titratable acidity may be slightly lower and pH may be slightly higher in winegrapes from minimal pruned vines compared to mechanical hedged or spur pruned vines at the same phenological stage (Clingeleffer 1992). Anthocyanin and phenolic concentrations in berries from minimal pruned vines may be higher due to an increased exocarp to juice ratio when compared to berries from spur pruned vines (Sommer and Clingeleffer 1993).

The substantial increase in yield, improved berry colour, disease control and low input costs of minimal pruned vines compared to mechanically hedged and spur pruned vines makes minimal pruning favourable to large mechanised vineyards. However, the unknown long-term repercussions of minimal pruning and delayed ripening of fruit has deterred industry from

Chapter 1: Introduction and Literature Review 9 mass adoption of this pruning technique. Research is being conducted into crop thinning treatments on minimal pruned vines to increase sugar accumulation, in both warm and cool winegrape growing regions of Australia (Clingeleffer et al. 2000).

1.3 IRRIGATION MANAGEMENT

Irrigation management of grapevines encompasses a range of different water application methods including overhead sprinklers, furrow irrigation, under-vine irrigation and drip irrigation. Irrigation is essential in regions of low rainfall for healthy grapevine development and increased productivity (Williams and Matthews 1990). However, in recent years the need to reduce water consumption in irrigated vineyards has risen considerably because of increased water costs and restrictions on allocated amounts of irrigation water. In addition, major environmental problems have long been associated with excessive irrigation in vineyards, such as water logging of the soil, rising ground water table levels and soil salinisation. Degradation of the soil by water logging and salinisation may lead to decreased vineyard productivity and wine quality.

Water availability affects vegetative growth indirectly as a result of physiological mechanisms including leaf water potential, turgor, photosynthesis and transpiration (Bravdo and Hepner 1987). As a result, irrigation management can successfully control vigorous growth in grapevines, which are known to be sensitive to water stress (Loveys et al. 1998). Grapevines respond to changes in soil water availability by regulating stomatal conductance to adjust the rate of transpiration from leaves. Root signals induced by plant growth hormones, such as abscisic acid (ABA) can influence stomatal conductance (Zhang and Davies 1990; Tardieu et al. 1992). ABA synthesised in roots exposed to a water deficit can be transported in the xylem to leaves (Davies and Zhang 1991). In response, stomata reduce their aperture and restrict water loss by transpiration. Also, ABA has been shown to have an inhibitory effect on leaf growth (van Volkenburgh and Davies 1983). Therefore, as the concentration of root-sourced ABA is increased in plants, stomatal conductance and leaf growth rate may be reduced (Loveys 1984; Zhang and Davies 1990; Tardieu et al. 1992).

Irrigation management of grapevines can involve partially drying the rootzone (PRD) to stimulate root signals, such as ABA to control shoot growth and transpiration. PRD has been successfully applied to a range of horticultural crops, such as maize, hot peppers and pear

Chapter 1: Introduction and Literature Review 10

(Kang et al. 1998, 2001, 2003). PRD reduces grapevine vigour while maintaining crop yield and improving fruit quality and water use efficiency (Dry and Loveys 1998). The potential of PRD to control vine vigour, improve winegrape composition and water use efficiency is reviewed below.

1.3.1 Partial Rootzone Drying

Partial rootzone drying (PRD) is an irrigation practice that enables half of the plant root system to dry out while the other half is irrigated. After a period of time the irrigation is alternated, so that the roots that were dry are watered and the roots that were hydrated dry out. The hydrated roots maintain a high plant-water status in the grapevine (shoot water potential near zero) while the dry roots induce a water-stressed plant response. The concentration of ABA is increased in the roots from the dry side and has been associated with reduced stomatal aperture in the leaves. Dry et al. (1996) and Dry and Loveys (1999) found that PRD significantly reduced stomatal conductance (Figure 1.2), photosynthesis (Figure 1.3), shoot growth rate (Figure 1.4), leaf area and pruning weights of several grapevine varieties in both split-pot and field experiments. As a result, grapevine vigour and canopy density are also reduced (Dry and Loveys 1998). This may have a positive effect on canopy microclimate and disease control. Although, no reported work has been conducted in this area. Decreased canopy density may increase air movement and evaporation, thus reduce humidity within the canopy. The resultant changes in microclimate may be less suitable for the development of Botrytis bunch rot. The effect of PRD combined with different pruning systems on incidence and severity of Botrytis bunch rot will be assessed as a component of this study.

Also, an improvement in fruit composition has been attributed to the reduction of canopy density and increase in berry exocarp to juice ratio by PRD. Cabernet Sauvignon berries had consistently higher levels of titratable acidity, lower pH and higher concentrations of anthocyanins, phenols and glycosyl-glucose when vines were irrigated by PRD (Dry et al. 1996; Dry 1997; Loveys et al. 1998). There have been no reported reductions in yield or berry weight associated with PRD in experimental situations (Dry et al. 1999). However, the amount of irrigation is halved and thus water use efficiency (WUE), in terms of yield produced per mega litre of water applied per hectare is improved 2-fold (Düring et al.1996; Loveys et al. 2000).

Chapter 1: Introduction and Literature Review 11

Improved fruit composition and WUE with no associated yield losses may make PRD more attractive to the wine industry than other modern irrigation practices which subject vines to water deficits, such as regulated deficit irrigation (RDI). PRD is also applied to vines throughout the entire growing season, thus may be less influenced by climatic conditions compared to deficit irrigation practices applied for short durations at critical phenological stages. However, PRD does require changes to standard drip irrigation infrastructure including additional drip lines and taps for swapping irrigations. Also, more research is required into the length of the drying cycle for sufficient production of root signals and physiological changes, as it will vary for each vineyard in response to climatic conditions, grapevine variety and soil type.

-2 -1 Fig. 1.2: Stomatal conductance of Chardonnay (gs; mmol m s ) for PRD vines (▲) and non-irrigated vines (□) expressed as a percentage of fully irrigated vines (Dry and Loveys 1999).

-2 -1 Fig. 1.3: Net Photosynthesis of Chardonnay (Pn; µmol m s ) for PRD vines (▲) and non-irrigated vines (□) expressed as a percentage of fully irrigated vines (Dry and Loveys 1999).

Chapter 1: Introduction and Literature Review 12

-1 Fig. 1.4: Shoot growth rate of Chardonnay (SGR; cm d ) for PRD vines (▲) and non-irrigated vines (□) expressed as a percentage of fully irrigated vines (Dry and Loveys 1999).

1.4 INTEGRATION OF CANOPY AND IRRIGATION MANAGEMENT

Investigations of integrated canopy and irrigation management are limited to the effects of crop control and irrigated or non-irrigated treatments on grapevine growth, production and quality (Freeman et al. 1979; Bravdo et al. 1985; Poni et al. 1994). Grapevine vegetative growth is increased by the combination of low cropping levels and irrigation (Poni et al. 1994). Crop control practices including thinning and pruning intensity have a greater influence on bunch number per vine, berry size and yield than increasing available water with irrigation (Freeman et al. 1979; Bravdo et al. 1985; Poni et al. 1994).

An interaction between crop production and irrigation exists for sugar accumulation in fruit. Total soluble solids are significantly lower in berries from irrigated vines with high crop load (Freeman et al. 1980; Bravdo et al. 1985). However, the effects of crop control and irrigation on other fruit quality parameters (pH, titratable acidity, colour and phenolics) were non- significant (Freeman et al. 1980; Bravdo et al. 1985).

The consequence of combined canopy and irrigation management on canopy development, bunch architecture and development of persistent grapevine diseases, such as Botrytis bunch rot, are as yet unknown.

Chapter 1: Introduction and Literature Review 13

1.5 CONCLUSION

The effects of various canopy or irrigation management techniques on grapevine development, winegrape production and quality and disease control have been established for both cool and warm climates. The control of grapevine vigour and reduction of canopy density is required for the production of high yields and quality winegrapes. To date, this has been achieved in vineyards by either canopy or irrigation management. Improved canopy microclimate and productivity has been associated with minimal pruning and partial drying of the rootzone has improved winegrape composition and wine quality. However, more research is required into the integration of low-input pruning systems (mechanical hedging and minimal pruning) with modern irrigation strategies (PRD). Assessment of grapevine development, physiology, productivity, quality and disease development is essential to develop holistic vineyard management strategies for improved sustainability and quality.

I have developed the following hypotheses from gaps within the literature that will be addressed by this project:

• Partial drying of the rootzone integrated with light pruning techniques can improve the sustainability of winegrape production in warm, irrigated vineyards, in terms of fruit and wine quality, yield, water-use efficiency and Botrytis bunch rot development.

• Light pruning practices (i.e. minimal pruning) integrated with PRD will increase grape production by increasing crop level and improve water use efficiency by applying half the amount of water of standard drip irrigation.

• PRD combined with minimal pruning will improve berry anthocyanin and phenolics concentration due to an increase in exocarp to juice ratio and reduction in berry size.

• Grapevine vigour and single leaf stomatal conductance will be reduced by the application of PRD combined with minimal pruning.

• Light pruning practices integrated with PRD irrigation will produce high numbers of small, loose exposed bunches that will reduce bunch infection and the expression of Botrytis bunch rot.

Chapter 1: Introduction and Literature Review 14

The objectives of this project were:

• To determine the effects of light pruning practices and PRD irrigation on canopy development, physiology, productivity, fruit and wine quality, water use efficiency, bunch architecture and control of Botrytis disease of cv. Shiraz (Vitis vinifera L.). • To investigate the relationship between yield, fruit composition and wine quality for cv. Shiraz grown in a warm climate. • To assess the effects of canopy and bunch architecture on incidence and severity of Botrytis bunch rot. • To identify the optimal integrated irrigation and pruning strategy for sustainable winegrape production in warm-climate, irrigated vineyards.

Chapter 2: Methodology 15

2 METHODOLOGY

2.1 EXPERIMENTAL SITE

2.1.1 Vineyard Characteristics

The experimental site was established on a 5-hectare commercial Shiraz vineyard at Deakin Estate, Wingara Wine Group Ltd at Iraak in the Sunraysia region of southeast Australia, latitude 34°25′S, longitude 142°21′E. The climate of the Sunraysia region is semi-arid with average annual rainfall of 272 mm and mean January temperature of 23.9 °C (Coombe and Dry 1988). Average monthly climatic data from the Mildura Airport (Table 2.1) show a mean maximum daily temperature range of 15.3°C to 32°C and mean minimum daily temperature range of 4.3 °C to 16.6 °C. Mean daily relative humidity ranges from 51% to 87% at 0900 AEST and 28% to 57% at 1500 AEST. Median monthly rainfall (mm) ranges from 10.4 mm to 27.7 mm, with the majority of rainfall occurring during the winter months. The topographical position of the vineyard is the swale of east-west running sand dunes. The soil- type is Nookamka sandy loam; consisting of a red-brown to brown sandy loam (0-36 cm), reddish brown to light brown sandy clay loam with slight lime (36 – 66 cm) and light brown sandy clay to light clay (66 – 150 cm) (Hubble and Crocker 1941).

Table 2.1: Average monthly climatic data from Mildura Airport: latitude 34°14′S, longitude 142°5′E, 54 years of records (Bureau of Meteorology 2003).

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Year Daily Mean Temperatures °C Maximum 32.0 31.5 28.2 23.3 18.9 15.9 15.3 17.1 20.2 23.7 27.2 30.0 23.6 Minimum 16.6 16.4 13.8 10.1 7.5 5.2 4.3 5.3 7.3 9.8 12.3 14.7 10.3 Relative Humidity % 0900 EST 53 57 61 70 83 88 87 80 69 58 54 51 67 1500 EST 28 31 34 41 51 57 55 48 41 36 31 28 40 Rainfall mm Mean 21.8 21.5 18.8 19.3 27.3 23.1 26.9 27.5 28.8 31.5 24.5 22.2 293 Median 13.0 10.6 10.4 13.2 22.9 16.3 27.7 26.1 25.5 22.4 18.1 12.7 287

Chapter 2: Methodology 16

Vitis vinifera cv Shiraz grafted on Schwarzmann (V. riparia x V. rupestris) rootstock was planted in 1994, at a vine spacing of 2.4 m and row spacing of 3.0 m, in an east-west orientation. Vine planting density is 1389 vines per hectare. The vines were trained to 4 cordons on a two-wire vertical trellis, mechanically hedged and drip irrigated. Partial rootzone drying irrigation was applied to every third row in 1997. Six rows each of drip irrigated and PRD irrigated vines were randomly selected from the 46 available rows for this field experiment in 2000. Three conventional canopy management practices were applied to vines in the selected irrigation rows in July 2000 to assess the effects of integrated viticulture on grape production, fruit composition, disease control and water use efficiency.

2.1.2 Soil Characteristics

Soil physical and chemical parameters were measured on surface (10-20 cm) and sub-surface (50-70cm) samples collected 20 cm from the base of a vine, every 10 metres in experimental rows. Soil samples were air-dried, 2mm sieved and the following parameters were measured: soil texture, electrical conductivity, pH, total organic carbon, total nitrogen and phosphorus. Soil texture was determined by hand feel and classified as loamy sand, sandy loam, light sandy clay loam, sandy clay loam, clay loam sand or light clay. Electrical conductivity (ECe) of a soil suspension was used to estimate the concentration of soluble salts in the soil and pH indicated the intensity of acidity of the soil. ECe and pH were determined on a 1:5 soil/H2O extract as described by Rayment and Higgins (1992). Total organic carbon (TOC) and total Nitrogen (N) in the soil was determined by combustion in a high frequency induction furnace and detection by mass spectrometry. Soil phosphorus (P) content was determined using a

1:100 extraction with 0.5M sodium bicarbonate (NaHCO3) followed by colorimetric determination.

Soil texture of the surface soil (10-20 cm) ranged from loamy sand to sandy loam and light sandy clay loam to sandy loam clay in the subsurface soil (50-70 cm). Electrical conductivity increased significantly with depth by 0.32 dS/m (Table 2.2). However, levels fell below the recommended salinity threshold for grapes of 1.5 dS/m, at which level growth and yield declines may be expected. An increase (3%) in soil alkalinity and total organic carbon content (41%) also occurred with depth. Total nitrogen and phosphorus content were greater in the surface soil than the subsurface soil, by 2-fold and 11-fold, respectively. The higher levels of

Chapter 2: Methodology 17 phosphorus content in the surface soil can be attributed to regular granular phosphate fertilizer applications along the irrigation rows.

Table 2.2: Soil characteristics of surface and subsurface soil of Shiraz vineyard, Deakin Estate, Iraak, Victoria: texture (LS= loamy sand, SL= sandy loam, SCL = sandy clay loam), ECe (dS/m), pH, TOC (%), N (%) and P (mg/kg) of surface soil (10-20 cm) and subsurface soil (50-70 cm).

Depth cm Texture ECe pH TOC % N % P mg/kg 10-20 LS-SL 0.43a 7.67a 0.33a 0.02b 17.1 50-70 LSCL-SLC 0.75b 7.90b 0.56b 0.01a 1.5 LSD (5% level) 0.111 0.095 0.080 0.002 15.3

2.2 EXPERIMENTAL FIELD TRIAL

2.2.1 Irrigation and Pruning Treatments

The standard drip (SD) irrigation treatment consisted of a single drip-line; 19 mm diameter, drip spacing of 700 mm and average seasonal output of 5.4 ML/ha. The partial rootzone drying (PRD) irrigation treatment applied irrigation water to half of the vine rootzone at a time by using two parallel drip lines of 19 mm diameter with 4 L hr-1 emitters, spaced at 450 mm on alternate (east or west) sides of the vine (Figure 2.1 and 2.3). The average seasonal output of PRD was 3.0 ML/ha. Three conventional pruning techniques of different pruning levels were superimposed on the irrigation treatments. The pruning treatments were hand spur pruning (SPUR), mechanical hedging (MECH) and minimal pruning (MIN). Hand spur pruned vines had 30 spurs (ten spurs on upper cordons and five spurs on the lower cordons) with two buds per spur. Mechanical hedged vines were pruned by tractor with vertical and horizontal cutter bars at a distance of 200 mm from the trellis wire, resulting in an approximate bud number vine-1 of 150. Minimally pruned vines were skirted at a height of 1 m from ground level to reduce disease incidence by cane contact with soil, approximate bud number vine-1 was >300.

2.2.2 Experimental Design

The experimental design was a 6 x 6 row-column design. Six irrigation rows (3 SD + 3 PRD) were randomly selected and divided into six pruning plots, aligned in columns. The three pruning treatments were assigned to the SD and PRD irrigation rows in two contiguous Latin

Chapter 2: Methodology 18

Squares (Figure 2.2). Thus the combined irrigation and pruning treatment plots were replicated six times in the vineyard. The pruning plots consisted of eight vines each and n=288. The two outer vines of each plot served as buffer vines, the two central vines were chosen for all harvest measurements and remaining vines were used for physiology and pathology experiments. The Latin Square design was maintained to enable testing (and elimination) of any column effects along the rows, as well as the orthogonal estimation of the irrigation effects, pruning effects and also treatment interactions.

Irrigation Cycle 1 Grapevine Dry rootzone North drip line

South drip line Wet rootzone

Irrigation Cycle 2

Grapevine Dry rootzone North drip line

South drip line Wet rootzone

Row Irrigation Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6

42 SD Mech Min Spur Spur Min Mech 63 SD Spur Mech Min Mech Spur Min 75 SD Min Spur Mech Min Mech Spur

49 PRD Spur Min Mech Mech Min Spur 61 PRD Mech Spur Min Spur Mech Min 73 PRD Min Mech Spur Min Spur Mech

Figure 2.2: Schematic diagram of experimental field site at Deakin Estate; 2 contiguous latin squares of pruning plots (8 vines each) superimposed on to irrigation rows.

Chapter 2: Methodology 19

Northern Drip line Wet Rootzone

Southern Drip line Dry Rootzone

Figure 2.3: Photograph of PRD irrigation infrastructure at Deakin Estate, 2 parallel drip lines irrigate half the rootzone of the vine at a time as indicated by the wet and dry rootzone.

2.2.3 Irrigation Scheduling

Irrigation scheduling was monitored by time domain reflectometry (TDR) using CS615 water content reflectometers (Campbell Scientific Inc., Logan, Utah). The reflectometers measured volumetric soil-water content (θv). The volumetric soil-water content was derived from the effect of changing dielectric constant on electromagnetic waves propagating along a wave- guide. The water content reflectometers consisted of two parallel 30 cm steel rods (wave- guide) and were buried at the base of SD and PRD irrigated vines at two soil depth ranges (20-

40 cm and 50-70 cm) to encompass the rooting depth of 20 to 70cm. θv was logged every hour by CR10X data loggers (Campbell Scientific Inc., Logan, Utah). Irrigation water was applied when θv reached 23%. Frequency of irrigation events was dependent on plant-water use and evapotranspiration. PRD was applied to the vines on an alternate cycle; thus each irrigation

Chapter 2: Methodology 20 event was swapped between the north and south drip lines. This cycle was chosen to ensure a dry rootzone on one side of the vine without having detrimental effects on root health, given the low water-holding capacity of the sandy loam soil and high mean summer temperatures of the region.

Soil-water content, irrigation water input and rainfall input for SD and PRD treatments for each growing seasons are presented in Figure 2.4a, b and c (2000-01), Figure 2.5a, b and c (2001-02) and Figure 2.6a, b and c (2002-03). Irrigations began in October or early November, depending on the growing season and winter rainfall. Irrigation began on the 9 November in 2000, as spring rains were relatively high in October. However, irrigations started a month earlier in 2001 and 2002, on 8 and 9 October respectively, because of the lack of winter and spring rainfall. Irrigation frequency was four days when water demands were greatest (January and February) and approximately 18-21 mm of water per irrigation event was applied for SD and 11-13.8 mm for PRD treatments, depending on daily plant water use.

The total amount of irrigation water applied for SD and PRD was monitored by inline flow meters, located at the lateral valve in two of the three experimental rows for each treatment. Total seasonal irrigation amounts per hectare were calculated from average flow meter readings for the three experimental growing seasons. Seasonal irrigation inputs ranged from 5.0 to 5.4 ML ha-1 for SD and 2.5 to 3.1 ML ha-1 for PRD irrigation and are presented in (Table 2.3).

Table 2.3: Total seasonal irrigation inputs (ML ha-1) for the three experimental seasons for SD and PRD irrigation treatments at Deakin Estate, Iraak, Victoria.

Irrigation Treatment 2000-01 2001-02 2002-03 Standard Drip 5.44 5.07 5.21 Partial Rootzone Drying 3.13 2.50 2.93

Chapter 2: Methodology 21 a. Rainfall 0.80 Irrigation 70.0 SD 20 cm 0.70 60.0 SD 50 cm 0.60

50.0 Input Water mm

0.50 40.0 0.40 30.0 0.30

Soil Water Content 20.0 0.20 10.0 0.10 0.00 0.0 18/10 1/11 15/11 29/11 13/12 27/12 10/1 24/1 7/2 21/2 7/3 Date b. Rainfall PRD irrig north 0.80 70.0 PRD irrig south 0.70 PRD 20cm north 60.0 PRD 20cm south 0.60

50.0 Input Water mm 0.50 40.0 0.40 30.0 0.30

Soil Water Content 20.0 0.20 0.10 10.0

0.00 0.0 18/10 1/11 15/11 29/11 13/12 27/12 10/1 24/1 7/2 21/2 7/3 Date c. Rainfall PRD irrig north 0.80 70.0 PRD irrig south 0.70 PRD 50cm north 60.0 PRD 50cm south 0.60

50.0 Input Water mm 0.50 40.0 0.40 30.0 0.30

Soil Water Content 20.0 0.20

0.10 10.0

0.00 0.0 18/10 1/11 15/11 29/11 13/12 27/12 10/1 24/1 7/2 21/2 7/3 Date

Figure 2.4: a. Standard drip (SD) volumetric soil-water content of surface (20 cm) and sub-surface soil (50 cm) b. PRD irrigation volumetric soil-water content of surface soil (20-40cm) of north and south side of vine and c. soil- water content of sub-surface soil (50-70cm) of north and south side of vine, as measured by TDR and corresponding irrigation and rainfall events for 2000-01 growing season.

Chapter 2: Methodology 22 a. Rain 0.80 Irrig SD 70.0

SD 20cm 0.70 60.0 SD 50 cm 0.60

50.0 Input Water mm 0.50 40.0 0.40 30.0 0.30

Soil Water Content 20.0 0.20

0.10 10.0

0.00 0.0 1/10 15/10 29/10 12/11 26/11 10/12 24/12 7/1 21/1 4/2 18/2 4/3 Date b. Rain Irrig N 0.80 70.0 Irrig S

0.70 PRD 20cm north 60.0 PRD 20cm south 0.60

50.0 Input Water mm 0.50 40.0 0.40 30.0 0.30

Soil Water Content 20.0 0.20 0.10 10.0

0.00 0.0 1/10 15/10 29/10 12/11 26/11 10/12 24/12 7/1 21/1 4/2 18/2 4/3 Date c. Rain

Irrig N 0.80 70.0 Irrig S

0.70 PRD 50cm north 60.0 PRD 50cm south 0.60 50.0 Input Water mm 0.50 40.0 0.40 30.0 0.30

Soil Water Content 20.0 0.20 0.10 10.0

0.00 0.0 1/10 15/10 29/10 12/11 26/11 10/12 24/12 7/1 21/1 4/2 18/2 4/3 Date

Chapter 2: Methodology 23 a. Rain 0.80 70.0 Irrig SD 0.70 SD 20cm 60.0 SD 50 cm t 0.60 n 50.0 Input W

e t

n 0.50 40.0 r Co 0.40 a ter m te 30.0 0.30 il Wa m

So 20.0 0.20

0.10 10.0

0.00 0.0 1/10 15/10 29/10 12/11 26/11 10/12 24/12 7/1 21/1 4/2 18/2 4/3

Date b. Rain Irrig N 0.80 70.0 Irrig S

0.70 PRD 20cm north 60.0 PRD 20cm south 0.60

50.0 Input Water mm 0.50 40.0 0.40 30.0 0.30

Soil Water Content 20.0 0.20

0.10 10.0

0.00 0.0 1/10 15/10 29/10 12/11 26/11 10/12 24/12 7/1 21/1 4/2 18/2 4/3

Date c. Rain Irrig N 0.80 70.0 Irrig S 0.70 PRD 50cm north 60.0 PRD 50cm south 0.60

50.0 Input Water mm

0.50 40.0

0.40

30.0 0.30

Soil Water Content 20.0 0.20

0.10 10.0

0.00 0.0 1/10 15/10 29/10 12/11 26/11 10/12 24/12 7/1 21/1 4/2 18/2 4/3 Date

Chapter 2: Methodology 24

2.3 CLIMATIC CONDITIONS FOR SEASONS 2000-2003

Climatic data was collected daily (0900 AEST) from a weather station located 5 km southeast of the vineyard at Nangiloc. Climatic parameters included mean daily temperature (minimum and maximum), relative humidity (minimum and maximum), rainfall (previous 24 hours), mean daily evaporation and mean daily sunshine hours. Monthly means are presented in Tables 2.4, 2.5 and 2.6 for the three grape-growing seasons when experiments were conducted. The three experimental growing seasons deviated from 50-year season averages with higher summer temperatures in 2000-01, lower summer temperatures in 2001-02 and below average annual rainfall in 2001-02 and 2002-03. Mean maximum temperature for January 2001 (36°C) was 14% greater than the 50 year mean maximum January temperature (32°C) for the region (Bureau of Meteorology 2003). The following season was milder than the 50-year average with lower mean maximum and minimum temperatures from October 2001 to February 2002. Drought conditions prevailed in the second and third season; the total annual rainfall was 44% and 36% lower than the 50-year annual rainfall average in 2001-02 and 2002-03, respectively. The majority of rainfall occurred during the late winter to early spring months in 2000-01 and 2001-02, however two large rainfall events in December and February dominated the annual rainfall in 2002-03. Relative humidity was consistent throughout the three growing seasons with maximal values recorded throughout the winter months. Daily sunshine hours were greatest during the summer months and minimal variation was recorded between growing seasons from 2000-2003.

Table 2.4: Mean monthly climatic data from Nangiloc weather station for the 2000-01 growing season.

2000-01 May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr Daily Mean Temperatures °C Maximum 17.6 15.6 15.3 17.0 21.3 23.0 29.6 32.2 36.3 33.9 27.8 23.8 Minimum 5.3 4.9 4.5 4.7 8.4 8.6 14.9 15.1 19.1 19.2 11.8 8.7 Relative Humidity % Maximum 98.5 97.5 98.6 97.2 96.5 91.0 88.4 73.3 71.2 74.6 84.2 88.5 Minimum 50.5 55.8 56.6 45.5 41.4 34.8 31.9 19.9 19.5 27.7 28.4 33.5 Rainfall mm Total 13.4 8.6 23.2 22.4 35.2 40.6 14.2 15.8 1.4 30.0 19.2 3.4 Evaporation mm (24 hrs) Mean 1.5 1.2 1.3 1.8 2.7 3.7 4.8 6.2 6.6 5.2 4.1 2.8 Daily Sun Hours Mean 6.9 5.2 6.2 6.6 7.5 8.9 10.1 9.8 11.4 9.7 10.0 8.4

Chapter 2: Methodology 25

Table 2.5: Mean monthly climatic data from Nangiloc weather station for the 2001-02 growing season

2001-02 May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr Daily Mean Temperatures °C Maximum 19.7 17.0 14.6 18.0 22.2 22.8 26.3 28.2 31.9 30.6 28.5 26.3 Minimum 6.5 5.4 4.3 6.1 9.3 8.5 10.6 11.9 14.7 14.2 12.4 10.4 Relative Humidity % Maximum 95.9 77.5 99.8 90.7 89.9 88.9 88.0 84.9 67.1 73.8 85.5 91.7 Minimum 46.1 60.0 65.6 43.9 37.1 32.8 25.6 23.2 16.0 22.2 30.0 32.7 Rainfall mm Total 0.0 2.2 6.6 35.8 42.8 22.8 1.6 3.2 0.2 8.2 22.2 18.6 Evaporation mm (24 hrs) Mean 1.7 1.2 1.1 1.4 2.7 3.5 4.6 4.9 6.2 5.4 4.3 2.9 Daily Sun Hours Mean 6.4 6.0 5.4 7.1 7.4 8.9 10.1 9.8 11.1 11.4 10.6 9.1

Table 2.6: Mean monthly climatic data from Nangiloc weather station for the 2002-03 growing season.

2002-03 May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr Daily Mean Temperatures °C Maximum 20.8 16.9 17.4 18.0 22.0 25.0 30.2 31.7 33.7 33.2 24.5 25.9 Minimum 7.0 5.1 4.1 3.6 6.9 8.0 12.8 14.6 16.6 16.2 10.9 9.5 Relative Humidity % Maximum 91.8 99.8 94.8 91.4 88.0 88.2 78.7 73.1 69.7 79.5 89.0 84.8 Minimum 39.6 50.1 44.2 31.9 30.8 26.0 21.0 19.5 20.0 23.6 28.3 36.4 Rainfall mm Total 16.2 13.0 4.8 15.6 7.0 7.0 20.0 58.2 0.2 45.0 0.0 1.2 Evaporation mm (24 hrs) Mean 1.8 1.2 1.4 2.3 3.1 4.5 5.2 6.0 6.7 5.6 4.2 2.9 Daily Sun Hours Mean 6.9 6.2 7.3 8.6 9.5 10.0 9.9 10.6 11.8 9.6 9.9 8.7

2.4 PHENOLOGICAL GROWTH DATES FOR SEASONS 2001-2003

Important phenological dates of grapevine development were recorded for the three experimental seasons and are presented in Table 2.7. Budburst occurred in the month of September of each experimental season, however it became progressively later in the month over the three seasons. Similarly, flowering, as determined by 80% capfall, became later in the calendar year over the duration of the study. Veraison was early in the first experimental year compared to the other two seasons, which reflects the early flowering date for that season, as a

Chapter 2: Methodology 26 result of relatively high temperatures at the end of October and beginning of November in 2000. Harvest occurred in the 2001 vintage on March 6 for all irrigation and pruning treatments. However, irrigation and pruning treatments were harvested by berry maturity level in subsequent seasons. The range in harvest date of the six treatments was approximately 11 to 14 days in the latter two seasons.

Table 2.7: Calendar date for phenological stages of budburst, flowering, veraison and harvest for Shiraz grapevines at Deakin Estate for seasons 2000-01, 2001-02 and 2002-03.

Phenological Stage Season 2000-01 Season 2001-02 Season 2002-03 Budburst 12/9/00 15/9/01 23/9/02 Flowering 4/11/00 15/11/01 18/11/02 Veraison 12/1/01 30/1/01 21/1/03 Harvest 6/3/01 28/2-14/3/01 27/2-10/3/03

Chapter 3: Irrigation and pruning effects on yield 27

3 IRRIGATION AND PRUNING EFFECTS ON YIELD

3.1 INTRODUCTION AND EXPERIMENTAL AIMS

Sustainable winegrape production is dependent on the carbon balance between vegetative and reproductive development. Vine balance is achieved when “vegetative vigour and fruit load are in equilibrium and consistent with high fruit quality” (Gladstones 1992). Excessive shoot vigour and foliage growth alters the “source-sink” relationship. This can have detrimental effects on yield and fruit quality because of increased canopy density and fruit shading (May 1965; Morrison 1989; Reynolds and Wardle 1989; Dokoolizan and Kliewer 1995; Allen et al. 1996). On the other hand, excessively heavy cropping is known to limit canopy growth and fruit quality (Bravdo et al. 1985; Smart and Robinson 1991). The vegetative to reproductive ratio is used to measure vine balance and varies with canopy management (pruning, trellising, shoot positioning) and irrigation (water supply, timing, scheduling).

The level of winter pruning and number of nodes retained per vine affects the source-sink ratio and carbohydrate partitioning. Severe levels of pruning, such as spur pruning can increase shoot vigour (Smart et al. 1990; Downton and Grant 1992). Shoot vigour is increased by severe pruning, as a result of fewer shoots per vine and subsequently, reduced competition for carbohydrates by shoot apices. Also, fewer nodes retained by severe pruning leads to lower crop levels because of reduced bunch numbers per vine (Clingeleffer 1993; Sommer et al. 1993; Lasko et al. 1996; Poni et al. 2000). In contrast, light pruning levels, such as mechanical hedging and minimal pruning retain a greater node number at pruning than hand spur pruning. Carbohydrate competition and partitioning is increased for both vegetative and reproductive growth in mechanically hedged and minimally pruned vines. As a result shoot vigour, bunch weight and berry weight is reduced but yield is maintained by increased crop level (Clingeleffer 1984).

Irrigation is used to successfully increase productivity of grape vines in areas of low rainfall, however irrigation can also be used as a management tool to control shoot vigour. The imposition of regulated water deficits at crucial growth stages can inhibit shoot growth (van

Chapter 3: Irrigation and pruning effects on yield 28

Zyl 1984; Williams and Grimes 1987) and leaf area (Williams and Matthews 1990), as well as manipulate berry size (Goodwin and Jerie 1992; Poni et al. 1993; McCarthy 1997a). An alternative to deficit irrigation is partial rootzone drying (PRD), where a drought response is induced in the grapevine by drying half the rootzone, while plant water status is maintained by irrigating the other half of the rootzone. PRD has been shown to reduce grapevine shoot vigour and subsequently canopy density but crop load is relatively unaffected (Loveys et al. 2000). PRD irrigation also improves water use efficiency by applying approximately half the amount of irrigation water to the vines compared to standard drip irrigation practices.

This study aims to:

1. Evaluate the effects of irrigation and pruning treatments on canopy development.

2. Assess the influence of integrated irrigation and pruning treatments on berry development and maturation.

3. Determine the effects of integrated irrigation and pruning treatments on yield components at harvest.

The following hypothesis was tested:

The combination of light pruning levels of minimal pruning with PRD irrigation will improve the “source-sink” carbohydrate relationship. Retaining greater number of nodes at pruning will increase yield, whilst inducing a water-stress response in the grapevine by PRD will control vegetative vigour.

3.2 METHODOLOGY

3.2.1 Canopy development and morphology

Vine morphology, canopy size and pruning weights were assessed on vines from each irrigation and pruning treatment. However, vegetative growth was investigated for pruning treatment effects only, as measurements were conducted monthly from October to December 2000 before irrigations were applied for the 2000-01 growing season.

Chapter 3: Irrigation and pruning effects on yield 29

Primary shoot measurements were done on four representative shoots of eight replicate vines from each pruning treatment. Shoot growth rate, length, internode number and internode length were assessed. The length of the tagged shoot was measured from the basal bud to the shoot tip and shoot growth rate was calculated as centimetres of growth per day. Internode length was determined between the leaf four and five from the basal leaf on each tagged shoot from each of the pruning treatments. Leaf area development of vines was estimated monthly through the growing season in 2000-01 on 12 replicate vines using an indirect optical method. The central vine from each plot of each row was measured using the LAI-2000 canopy analyser (LI-COR Inc., Lincoln, Nebraska, USA) according to Sommer et al. (1995).

Canopy size (total leaf area.vine-1) was estimated on fully developed canopies of six replicate vines per treatment when primary shoot growth had ceased in subsequent seasons using the LAI-2000 canopy analyser. Total leaf area.vine-1 was calculated using the relationship between direct and indirect measurements of leaf area index from Sommer and Lang’s (1994) LAI-2000 validation study. Maximum leaf area.vine-1 was measured on 22 November in three consecutive growing seasons (2000-02) on all treatments just prior to summer trimming. Destructive sampling was used to assess mean pruning weight.vine-1 and cane number.vine-1 and mean cane weight was calculated (pruning weight.vine-1 /cane number.vine-1) on six replicate vines on all treatments at winter pruning, 2003.

3.2.2 Berry Development

Mean berry weight and total soluble solids (TSS) were monitored at weekly intervals from veraison to harvest as an indication of berry maturation for the three consecutive seasons. Three vines per treatment were randomly selected each week and five bunches were sampled in a diagonal cross-section from the bottom north-east corner, over the top of the vine to the south-west corner. All berries were removed from the five bunches and 100 berries were randomly selected and weighed (g). The 100-berry sub-sample was crushed by hand using a mortar and pestle; grape juice was then drained through a 2 mm gauze sieve. Grape juice samples were left to settle for 1 hour and 1 mL extracted to measure TSS

Chapter 3: Irrigation and pruning effects on yield 30

(°Brix) using a temperature compensating PR-101 digital refractometer (Atago Co. Ltd., Tokyo, Japan).

3.2.3 Harvest Components

In the first season (2000-01) all six treatments were picked on the same day (124 days after flowering, DAF) but in subsequent seasons (2001-02 and 2002-03) treatments were picked at a maturity level of 24.0 ± 1.0 °Brix. The two central vines (Vines 4 and 5) of each pruning plot in the six experimental irrigation rows were tagged as harvest vines: 12 replicate vines represented each treatment. At harvest, 1 m transects were assigned to either the west, central or east region of the vine. Transect ladders (1 m wide) were used and all bunches within the 1 m zone were counted, picked and weighed. Yield was calculated as kg.vine-1 and t.ha-1 and mean bunch weight was calculated (yield kg.m-1/bunch number.m-1). Sub-samples of 10 bunches were collected from each of the harvest vines and all berries were plucked from the rachis. One hundred berries were randomly selected and weighed; mean berry weight was calculated (100 berry weight g/100). A further 250 g of berries was sampled and stored at –20°C for analysis of anthocyanin and phenolic concentration.

Leaf area to fruit ratio (LA:F) was calculated as the ratio of total leaf area per vine (cm2), measured using the LAI-canopy analyser (as described in section 3.2.1) to yield per vine (g). Water use efficiency (WUE) was calculated as the amount of yield produced (t.ha-1) per mega-litre of irrigation water applied for each irrigation treatment over the growing season.

3.2.4 Statistical Analysis

Harvest data was analysed by Analysis of Variance (ANOVA) using GenStat® 6th Edition (Lawes Agricultural Trust, Rothamsted Experimental Station). All interactions were balanced and analysed by general ANOVA. The treatment structure was Irrigation*Pruning and the blocking structure was Row*Plot (Appendix B for GenStat output). Treatment effects at each sample collection date were analysed using ANOVA on maturation data.

Chapter 3: Irrigation and pruning effects on yield 31

3.3 RESULTS

3.3.1 Canopy Development and Morphology Shoot Development Imposed pruning treatments influenced all aspects of shoot development; shoot length, growth rate, node number.shoot-1 and internode length, as shoot growth measurements occurred before irrigation was applied in season 2000-01.

Shoot length Mean shoot length was greater for SPUR vines at the start of shoot growth measurements compared to MECH or MIN vines (Figure 3.1a). By 61 days after budburst (DAB), SPUR and MECH vines had significantly longer shoots than MIN vines. Shoot elongation had ceased by 89 DAB; SPUR shoots were 23% longer than MECH shoots, which in turn were 54% longer than MIN shoots.

Shoot Growth Rate Shoot growth rate between 27 and 61 DAB was less on shoots from MIN vines compared to SPUR and MECH vines (Figure 3.1b). The shoot growth rate of SPUR vines was greater than MECH and MIN vines at both 61 and 89 DAB. The rate of shoot growth rapidly declined from 2.2 cm.day-1 to 1.6 cm.day-1 for SPUR vines between 61 and 89 DAB, whereas the decrease in shoot growth rate was less obvious on MIN vines.

Node Number.Shoot-1 The node number.shoot-1 was significantly less for SPUR vines compared to MECH and MIN vines at 27 DAB (Figure 3.1c). However, by 61 DAB SPUR had the highest node number.shoot-1 compared to the other pruning treatments. This trend continued with SPUR having the highest number, MECH having an intermediate number and MIN vines having the lowest node number.shoot-1 by the end of shoot elongation (89 DAB)

Internode Length MIN vines had significantly shorter internodes than MECH or SPUR vines on all measurement days (Figure 3.1d). The difference in internode length between pruning treatments was much greater at 89 DAB than at 27 or 61 DAB.

Chapter 3: Irrigation and pruning effects on yield 32

180 2.6 b. *** a. *** 2.4 160 *** 2.2

140 -1 y

a 2.0 ***

d 120 m 1.8 cm c te 1.6 100 a

r th length 1.4 80 w o

r 1.2 g *** t Shoot 60 o 1.0 40 *** Sho 0.8

20 0.6

0 0.4 20 30 40 50 60 70 80 90 100 20 30 40 50 60 70 80 90 100

Days after budburst Days after budburst

22 70 c. *** d. *** 20 *** 60

16 c

-1 t h t 40

14 ng e

l e 12

od 30 n r Nodes Shoo 10 te

In 20 ** *** 8 *** 10 6

4 0 20 30 40 50 60 70 80 90 100 20 30 40 50 60 70 80 90 100

Days after budburst Days after budburst

Spur Mech Min

Figure 3.1: Pruning treatment effects on shoot growth in season 2000-01; a. shoot length, b. shoot growth rate, c. node number.shoot-1 and d. internode length from budburst to 89 days after budburst (DAB). Significant differences were calculated by Fisher’s least significant difference (LSD) at each time point. LSD is represented by bars and levels of significance are denoted by ***P<0.001, **P<0.01, *P<0.05.

3.3.2 Leaf Area Development

Total Leaf Area development Total leaf area development of vines during the 2000-01 growing season showed leaf expansion from budburst to 22 November (42 DAB) was greater for MIN compared to

Chapter 3: Irrigation and pruning effects on yield 33

MECH and SPUR vines (Figure 3.2a). Between 42 DAB and 68 DAB, summer trimming of all vines occurred and total leaf area.vine-1 of MIN vines was similar to both SPUR and MECH vines. No significant pruning effects were found for the remainder of the growing season. However, the rate of leaf area development was higher for SPUR vines later in the growing season due to the increase in lateral shoot growth. Irrigation influenced leaf area development towards the end of the growing season (Figure 3.2b). SD increased the rate of leaf area development between 18 December (68 DAB) and 13 February (124 DAB) compared to PRD irrigation. Total leaf area.vine-1 of SD irrigated vines was 17% larger than PRD irrigated vines prior to harvest.

35 35 a. *** b. ns

30 30

*** ns 25 25 * -1 -1 ns e e n n * ns vi vi 2 2 m 20 *** ea 20 Ar

area m

eaf

tal leaf tal L o o T

T 15 15 ns

10 10

Spur Mech SD Min PRD 5 5 0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140 Days after budburst Days after budburst

Maximum leaf area.vine-1 The integration of pruning and irrigation or irrigation alone had no significant effect on maximum leaf area.vine-1 over the three seasons. However, differences in maximum leaf area.vine-1 between the three pruning treatments were observed; MIN produced a higher total leaf area.vine-1 than SPUR (22%) or MECH (16%) (Table 3.1).

Chapter 3: Irrigation and pruning effects on yield 34

Table 3.1: Irrigation and pruning effect on total leaf area.vine-1 (LA m2.vine-1) at maximum canopy development, 22 November 2000-02. * Indicates significance level; *** P<0.001, ** P<0.01, * P<0.05, ns = non significant.

Irrigation LA m2.vine-1 Pruning LA m2.vine-1 SD 23.1 SPUR 20.7a PRD 22.1 MECH 21.9a MIN 25.3b Irrigation P value ns Pruning P value ***

3.3.3 Pruning Weights

The application of PRD irrigation alone or the integration between irrigation and pruning treatments did not affect pruning weights, cane number and mean cane weight (Table 3.2). However, there was a pruning response for all parameters. SPUR had a higher mean pruning weight than MECH and MIN by 2.5-fold and 3.2-fold respectively. MIN increased cane number.vine-1 compared to MECH and SPUR by 1.8-fold and 2.2-fold respectively. Mean cane weight was greater for SPUR relative to MECH and MIN by 1.5-fold and 6.5-fold. Hence, severe pruning to 60 nodes produced more one-year old wood and enhanced shoot vigour more than lighter pruning treatments.

Treatment Pruning Wt. (kg) Cane Number.Vine-1 Cane Wt. (g) Grand Mean 1.3 111.9 14.9 SD 1.4 114.3 15.4 PRD 1.3 109.6 14.4 Irrigation ns ns ns SPUR 1.9c 77.0a 24.8c MECH 1.5b 93.2b 16.1b MIN 0.6a 165.6c 3.8a Pruning *** *** *** SD + SPUR 2.0 78.0 25.6 PRD + SPUR 1.8 76.0 24.0 SD + MECH 1.6 94.5 16.9 PRD + MECH 1.4 91.8 15.3 SD + MIN 0.6 170.3 3.6 PRD + MIN 0.6 160.8 4.0 Irrigation*Pruning ns ns ns

Chapter 3: Irrigation and pruning effects on yield 35

3.3.4 Berry Development

Berry Weight Two stages of berry weight change were observed each season between veraison and harvest. The first period (Period 1) involved a gradual increase in berry weight between veraison and 20 to 34 days after veraison, depending on the imposed treatment. In the second period (Period 2) there was a slow decline in berry weight from 34 days after veraison to harvest because of berry shrivel. The mean rate of berry weight increase across all treatments in Period 1 was consistent over the 3 seasons (0.011 g.day-1). However, the rate of berry weight loss in Period 2 was season-dependent, ranging from -0.007 g.day-1 in season 2002-03 to -0.010 g.day-1 in season 2001-02.

The treatment trends for berry weight accumulation were comparable for all three experimental seasons (Figure 3.3, 3.4, 3.5). However, the final berry weight differences between treatments were greater in seasons 2001-02 and 2002-03 than 2000-01, when treatments had been imposed for more than 2 or more seasons. Initial berry weight at the onset of veraison was influenced by the imposed irrigation and pruning treatments. Initial berry weight and berry weight throughout the maturation period was greatest for SD + SPUR for all 3 seasons. Variation was also large for SD + SPUR compared to treatments that produced smaller berries. SD + MECH, PRD + SPUR, SD + MIN and PRD + MIN had progressively smaller initial berry weights and this trend continued through the maturation period. The lowest berry weight throughout the maturation period for each season was produced by PRD + MIN.

3.3.5 Berry Maturation

Sugar Accumulation TSS was measured weekly to give an indication of sugar accumulation in the berries. Initial sugar concentration varied across treatments in the three experimental seasons. This suggests delayed veraison, particularly for lightly pruned treatments with high bunch numbers per vine (MIN and MECH).

Chapter 3: Irrigation and pruning effects on yield 36

2.0 *** **

1.8 *** *

** ** *** 1.6 ***

(g) 1.4 ight e W

1.2

Mean Berry

1.0

SD+Spur PRD+Spur 0.8 SD+Mech PRD+Mech SD+Min PRD+Min 0.6 10 20 30 40 50 Days after veraison

SD+Min ns PRD+Min ** 1.8 * * **

1.6 (g) *** ight

e W

y 1.4 r r

e B an 1.2 Me

1.0

0.8

0.6 10 20 30 40 50 Days after veraison

Chapter 3: Irrigation and pruning effects on yield 37 graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant. 2.0

*** ** 1.8 ns ** **

1.6 ns *** ***

(g) 1.4 ight

Berry W 1.2 n a e M

1.0

SD+Spur 0.8 PRD+Spur SD+Mech PRD+Mech SD+Min PRD+Min 0.6 10 20 30 40 Days after veraison

Figure 3.5: Irrigation and pruning effects on berry weight during the maturation period in season 2002-03. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant.

Sugar Accumulation, season 2000-01

In season 2000-01, the pruning treatments had an overriding influence on sugar accumulation, as a response to crop load (Figure 3.6). SPUR treatments reached maturity first, followed by MECH and MIN treatments. SPUR pruning had a lower crop load than MECH or MIN pruning in 2001. Thus the number of potential sink sites (bunches) for carbohydrates produced by SPUR pruned vines was lower than higher yielding treatments.

Sugar Accumulation, season 2001-02

PRD irrigation influenced sugar accumulation in season 2001-02 (Figure 3.7). Berries from vines irrigated by PRD had a greater initial sugar concentration than SD irrigated vines for all pruning treatments. PRD + SPUR and PRD + MECH reached 24 °Brix 13 days earlier than other treatments. SD + MIN had the lowest TSS levels throughout the ripening period between veraison and harvest.

Chapter 3: Irrigation and pruning effects on yield 38

26 ** **

** 24

* ns 22

) *

Brix o (

s 20 d

i *** l

e S bl

lu 18

So l

Tota * 16

SD+Spur PRD+Spur 14 SD+Mech PRD+Mech SD+Min PRD+Min 12 10 20 30 40 50 Days after veraison Figure 3.6: Irrigation and pruning effects on sugar (TSS) accumulation during the maturation period in season 2000-01. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant. 28

** * ns * 26

*** 24

)

x **

Bri 22

o (

s d i l o 20 e S bl

lu ns o l S a

t 18 o T

16 SD+Spur PRD+Spur 14 SD+Mech PRD+Mech SD+Min PRD+Min 12 10 20 30 40 50 Days after veraison

Figure 3.7: Irrigation and pruning effects on sugar (TSS) accumulation during the maturation period in season 2001-02. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant.

Chapter 3: Irrigation and pruning effects on yield 39

Sugar Accumulation, season 2002-03

The treatment differences in sugar accumulation were less apparent in 2002-03 (Figure 3.8). PRD + MECH and SD + MIN reached 24 °Brix 10 days earlier than the other treatments. This was the result of a rapid increase in sugar concentration in berries after a major rainfall event 34 days after veraison.

ns ns ns 26 ns ns

ns 24

) x i

Br 22 o (

* 20

tal Soluble Solids

To 18

SD+Spur 16 PRD+Spur SD+Mech PRD+Mech SD+Min PRD+Min 14 10 20 30 40 Days after veraison

Figure 3.8: Irrigation and pruning effects on sugar (TSS) accumulation during the maturation period in season 2002-03. Significance of irrigation and pruning interactions were calculated by Fisher’s LSD and are represented on the graph by LSD bars. Levels of significance are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = not significant.

Sugar Accumulation Regression Models

Three linear regression models were fitted to the rate of TSS increase for each season to test which model provided best representation of the treatment effects. The three linear regression models tested were distinct regression lines for each treatment, parallel regression lines for each treatment and a common regression line of all treatments. Equations of the regression models and coefficients of determination (R2) for each regression line are presented in Table 3.3. The rates of sugar accumulation for season 2000-01 and 2001-02 were best fitted to parallel regression lines for each of the six treatments (Figure 3.9 and

Chapter 3: Irrigation and pruning effects on yield 40

3.10, R2 values presented in Table 3.3). Therefore, the rates of sugar accumulation in berries from all six treatments were consistent for each season but the start of sugar accumulation (veraison) varied between treatments. The rate of sugar accumulation was greater in season 2001-02 than the previous season, since regression line slopes were 0.245 and 0.196 respectively. Sugar accumulation in season 2002-03 was best fitted to a common regression line (R2 = 0.89), which suggests the rate of maturation and the start of veraison was similar for all six treatments (Figure 3.11).

Table 3.3: Equations and coefficient of determination (R2) for parallel regression model of sugar accumulation for seasons 2000-01 and 2001-02 and common regression model of sugar accumulation for season 2002-03. Regression model Equations are Total Soluble Solids = constant + slope*Days After Veraison.

Treatment Rate of Sugar Accumulation 2001 Parallel Regression Constant Slope R2 SD + SPUR 14.30 0.196 0.96 PRD + SPUR 13.45 0.196 0.95 SD + MECH 12.91 0.196 0.95 PRD + MECH 14.17 0.196 0.95 SD + MIN 13.57 0.196 0.93 PRD + MIN 12.21 0.196 0.93 2002 Parallel Regression Constant Slope R2 SD + SPUR 12.85 0.246 0.93 PRD + SPUR 15.21 0.246 0.94 SD + MECH 12.55 0.246 0.87 PRD + MECH 14.20 0.246 0.98 SD + MIN 10.81 0.246 0.92 PRD + MIN 12.12 0.246 0.96 2003 Common Regression Constant Slope R2 All treatments (6) 15.412 0.218 0.89

24.00

22.00

Brix) o 20.00

18.00

SD+Spur 16.00 PRD+Spur Total Soluble Solids ( SD+Mech PRD+Mech 14.00 SD+Min PRD+Min 12.00 74 79 84 89 94 99 104 109 114 119 DAF Figure 3.9: Parallel regression model of sugar accumulation in berries on vines with integrated irrigation and pruning treatments for season 2000-01. Refer to Table 3.3 for equations.

Chapter 3: Irrigation and pruning effects on yield 41

26.00

24.00

Brix) 22.00 o

20.00

18.00 SD+Spur PRD+Spur Total Soluble Solids ( 16.00 SD+Mech

PRD+Mech 14.00 SD+Min PRD+Min

12.00

76 81 86 91 96 101 106 111 116

DAF Figure 3.10: Parallel regression model of sugar accumulation in berries from vines with integrated irrigation and pruning treatments for season 2001-02. Refer to Table 3.3 for equations.

26.0

24.0

Brix) 22.0 o

20.0

18.0

SD+Spur PRD+Spur Total Soluble Solids ( 16.0 SD+Mech PRD+Mech 14.0 SD+Min PRD+Min 12.0 76 81 86 91 96 101 106 111 116 DAF

Figure 3.11: Common regression model of sugar accumulation in berries from vines with integrated irrigation and pruning treatments for season 2002-03. Refer to Table 3.3 for equations.

Sugar Accumulation Rate The rate of sugar accumulation (°Brix.day-1) varied between seasons (Table 3.4). The rate was greater in seasons 2001-02 and 2002-03 compared to 2000-01. No difference in the rate of sugar accumulation was observed between treatments in each season but PRD + SPUR

Chapter 3: Irrigation and pruning effects on yield 42 and PRD + MECH had a more rapid rate of sugar accumulation compared to other treatments in season 2001-02.

Treatment Rate of Sugar Accumulation (°Brix.day-1) 2001 2002 2003 SD + SPUR 0.20 0.25 0.23 PRD + SPUR 0.20 0.29 0.24 SD + MECH 0.21 0.24 0.22 PRD + MECH 0.19 0.29 0.25 SD + MIN 0.18 0.24 0.27 PRD + MIN 0.21 0.23 0.25 Season Mean *** 0.20a 0.26b 0.24b

3.3.6 Bunch Number and Bunch Weight

Bunch Number.Vine-1 The level of pruning had a significant effect on bunch number.vine-1 but irrigation treatments had no effect for the three experimental seasons (Table 3.5). Since irrigation did not influence bunch number.vine-1, there were no significant interactions between irrigation and pruning treatments. Seasonal variation had a large influence on bunch number.vine-1. Bunch number.vine-1 for the two irrigation treatments and light pruning treatments (MECH and MIN) were heavily reduced in season 2002-03 compared to the two previous seasons due to poor bunch initiation.

The pruning effect on bunch number.vine-1 was large in seasons 2000-01 and 2001-02. A 2.0-fold increase in bunch number.vine-1 from SPUR to MECH and 4.2-fold from SPUR to MIN was recorded for season 2000-01. Bunch number.vine-1 increased 2.5-fold from SPUR vines to MECH vines and 4.6-fold from SPUR to MIN vines in season 2001-02. However, the difference in bunch number.vine-1 between pruning treatments was significantly less in 2002-03. A 1.3-fold increased in bunch number.vine-1 was recorded from SPUR to MECH and a 2.3-fold increase from SPUR to MIN. Light pruning treatments, MIN and MECH were greatly affected by seasonal variation. Large reductions in bunch number.vine-1 were found for both MIN and MECH in season 2002-03 compared to the previous seasons.

Chapter 3: Irrigation and pruning effects on yield 43

However, no year-to-year variation was recorded for SPUR because of the increased crop control associated with severe pruning.

Bunch Number.Vine-1 Grand Treatment 2000-01 2001-02 2002-03 Mean Season LSD Seasonal Mean 330 379 159 290 *** 22.2 SD 349 348 150 280 PRD 312 370 169 283 Irrigation ns ns ns ns ** 63.9 SPUR 137a 140a 102a 126a MECH 275b 345b 137b 252b MIN 580c 641c 239c 478c Pruning *** *** *** *** *** 52.1 SD + SPUR 128 147 96 124 PRD + SPUR 144 132 108 127 SD + MECH 280 368 141 263 PRD + MECH 269 322 132 241 SD + MIN 637 618 212 473 PRD + MIN 523 657 267 482 Irrigation*Pruning ns ns ns ns ** 82.4

Mean Bunch Weight Mean Bunch weight was also strongly affected by imposed pruning treatments and seasonal variation (Table 3.6). The influence of pruning treatments on bunch weight was consistent for the three experimental seasons; SPUR bunches > MECH bunches > MIN bunches. Bunch weight was reduced by MECH by 35% compared to SPUR and a further 42% reduction was recorded from MECH to MIN. Irrigation treatments only affected bunch weight in season 2002-03 with PRD reducing bunch weight by 28% compared to SD irrigation. As a consequence of the limited influence irrigation treatments imposed on mean bunch weight, no significant interactions between irrigation and pruning were found for the three experimental seasons.

A significant negative linear relationship was found between bunch number.vine-1 and mean bunch weight across all treatments in each season. The greater the bunch number.vine-1, the smaller the resultant bunch weights. The coefficient of determination and level of significance for each season are presented in Table 3.7.

Chapter 3: Irrigation and pruning effects on yield 44

Bunch Weight g Grand Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season Mean 59.8 58.9 71.2 63.3 *** 4.1 SD 64.0 69.8 82.6b 72.2b PRD 55.7 52.2 59.7a 55.8a Irrigation ns ns ** * ** 12.9 SPUR 88.0c 100.3c 94.2c 94.1c MECH 59.1b 48.9b 74.4b 60.8b MIN 32.6a 26.6a 44.9a 35.1a Pruning *** *** *** *** *** 9.2 SD + SPUR 96.0 111.6 112.2 106.6 PRD + SPUR 80.1 89.0 76.2 81.6 SD + MECH 61.5 55.7 83.9 67.0 PRD + MECH 56.7 42.1 64.8 54.5 SD + MIN 34.7 28.3 51.7 39.5 PRD + MIN 30.4 25.4 38.1 31.3 Irrigation*Pruning ns ns ns ns ns 15.6

Table 3.7: Coefficient of determination (R2) and statistical significance level (P value) of the negative linear relationship between bunch number.vine-1 and bunch weight for all treatments in each season and over the 3 seasons. Significance level is denoted by ** P<0.01, * P<0.05.

Season R2 P value 2000-01 0.86 ** 2001-02 0.82 ** 2002-03 0.75 * Mean 0.83 *

3.3.7 Yield

Yield was influenced by seasonal variation. A large decrease in yield was found in association with the decrease in bunch number.vine-1 in season 2002-03 relative to the preceding seasons. Irrigation influenced yield in seasons 2001-02 and 2002-03 and pruning practices influenced yield in seasons 2000-01 and 2001-02 (Table 3.8). Combined irrigation and pruning treatments did not have a significant effect on yield in any experimental season.

PRD irrigation reduced yield relative to SD irrigation in seasons 2001-02 and 2002-03 by 23% and 18% respectively. The lower yield in response to PRD corresponded to reductions in berry weight (Refer to section 3.5.4). Pruning levels strongly influenced yield in the first

Chapter 3: Irrigation and pruning effects on yield 45 two experimental seasons, when bunch numbers were very different between treatments. Yield was higher on lightly pruned treatments compared to SPUR pruning. The high yield produced by MIN vines in season 2000-01 may be attributed to the retention of a high number of nodes at winter pruning. Pruning level did not influence yield in season 2002-03, when bunch numbers were relatively consistent between treatments.

There was no significant interactive effect of irrigation and pruning treatments on yield because of the additive effects of SD irrigation and light pruning treatments. However, interesting trends were observed in each season. PRD reduced yield by 28% when combined with MIN in season 2000-01, whereas reductions in the following two seasons were only 7% and 5%. However, the seasonal trend in yield reductions by PRD was reversed for both SPUR and MECH vines. A small decrease by PRD was observed in season 2000-01 for SPUR and MECH vines and large decreases were observed in the subsequent seasons.

Yield kg.vine-1 Grand Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season mean 15.6 16.1 10.1 14.0 *** 1.13 SD 17.1 17.9b 11.1b 15.2b PRD 14.2 13.8a 9.1a 12.4a Irrigation ns * * * * 3.08 SPUR 11.7a 13.9a 9.5 11.7a MECH 16.1b 16.7b 10.1 14.3b MIN 19.2c 16.7b 10.7 15.4b Pruning *** ** ns *** *** 2.40 SD + SPUR 12.2 16.1 10.7 13.0 PRD + SPUR 11.3 11.6 8.4 10.4 SD + MECH 16.9 19.9 11.6 16.1 PRD + MECH 15.3 13.6 8.6 12.5 SD + MIN 22.3 17.4 10.9 16.8 PRD + MIN 16.1 16.1 10.4 14.2 Irrigation*Pruning ns ns ns ns * 3.91

Chapter 3: Irrigation and pruning effects on yield 46

3.3.8 Berry Weight and Berry Number

Berry Weight Both PRD irrigation and MIN pruning significantly reduced mean berry weight in each season (Table 3.9). An interaction effect was found between irrigation and pruning treatments for mean berry weight in season 2001-02. Also, seasonal variation influenced berry weight; the largest berries developed in the first experimental season (2000-01) and the smallest berries developed in the following season (2001-02).

Irrigation had a significant effect on berry weight for each season; PRD reduced berry weight by 19% compared to SD. Pruning also had a significant effect on berry weight in all seasons. SPUR produced the largest berries and MIN produced the smallest, possibly as a response to bunch number.vine-1. No difference in berry weight was found for SPUR between the three seasons. However, berry weight from MECH and MIN treatments were influenced by seasonal variation.

There was a significant interaction between irrigation and pruning treatments for mean berry weight in 2001-02. A large decrease (42%) was observed from the treatment with the largest berries (SD + SPUR) to that with the smallest berries (PRD + MIN). PRD irrigation resulted in reductions in berry weight when vines were SPUR (24%) or MECH (20%) pruned. However, berry weight of MIN vines was not significantly reduced by PRD. There was also a significant interaction for berry weight between irrigation and pruning treatments in the three seasons. Berry size was reduced by PRD compared to SD when combined with all pruning treatments but the magnitude of reduction by PRD was greater on SPUR and MECH vines than MIN vines.

Berry Number.Bunch-1 Mean berry number.bunch-1 progressively increased over the experimental period, which relates to flower number and fruit-set (Table 3.10). Pruning level had a significant effect on berry number.bunch-1. However, neither irrigation nor the combination of irrigation and pruning had any effect on berry number.bunch-1.

Chapter 3: Irrigation and pruning effects on yield 47

Berry number.bunch-1 was reduced from SPUR to MECH (31%) and from MECH to MIN (30%) in the three seasons. The number of berries per bunch increased in 2002-03 for MECH and MIN but decreased for SPUR compared to the previous two seasons. Therefore a change in bunch architecture occurred in 2002-03 for MECH and MIN, i.e. more compact bunches with more berries per bunch (Refer to section 6.3.5).

TABLE 3.9: Berry weight of vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant.

Berry Weight (g) Grand Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season Mean 1.21 1.01 1.14 1.12 *** 0.03 SD 1.32b 1.14b 1.26b 1.24b PRD 1.11a 0.91a 1.03a 1.01a Irrigation * ** ** ** ns 0.14 SPUR 1.30c 1.19c 1.25b 1.25b MECH 1.22b 1.01b 1.23b 1.15b MIN 1.12a 0.81a 0.95a 0.97a Pruning *** *** *** *** *** 0.06 SD + SPUR 1.43 1.35e 1.41 1.40c PRD + SPUR 1.17 1.02c 1.09 1.09b SD + MECH 1.33 1.12d 1.30 1.25c PRD + MECH 1.10 0.90b 1.15 1.05b SD + MIN 1.19 0.84a 1.05 1.05b PRD + MIN 1.05 0.79a 0.84 0.89a Irrigation*Pruning ns ** ns ** * 0.15

Berry Number.Bunch-1 Grand Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season Mean 48.8 55.2 60.6 54.9 *** 3.2 SD 48.0 58.0 64.3 56.6 PRD 49.9 55.2 57.2 54.0 Irrigation ns ns ns ns * 6.7 SPUR 68.3c 84.5c 75.0c 75.8c MECH 49.3b 48.1b 60.1b 52.5b MIN 29.1a 32.9a 46.8a 36.5a Pruning *** *** *** *** *** 6.9 SD + SPUR 67.8 82.5 79.3 76.6 PRD + SPUR 68.9 86.6 70.7 75.1 SD + MECH 46.9 49.6 63.8 53.4 PRD + MECH 51.7 46.6 56.3 51.6 SD + MIN 29.3 33.9 49.0 37.8 PRD + MIN 29.0 32.2 44.7 35.3 Irrigation*Pruning ns ns ns ns ns 10.1

Chapter 3: Irrigation and pruning effects on yield 48

3.3.9 Vine Balance

Leaf Area: Fruit Ratio The ratio of total leaf area.vine-1 and yield (LA:F) was used as an indication of vine balance between vegetative and reproductive growth. No significant interaction or irrigation effects were found. However, in general, LA:F was greater for PRD, as a response to lower yield. A significant pruning effect was observed in 2000-01: MECH and MIN reduced the LA:F ratio compared to SPUR (30% and 25% respectively) (Table 3.11). Therefore, less leaf area was required to mature 1 g of fruit mass on MECH and MIN vines than on SPUR vines.

Leaf Area (cm2): Fruit (g) Ratio Grand Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season Mean 18.8 16.4 19.2 18.1 ** 1.87 SD 17.9 14.8 16.8 16.3 PRD 19.2 18.3 22.3 19.9 Irrigation ns ns ns ns * 4.30 SPUR 21.6b 17.5 20.8 20.0b MECH 16.5a 15.0 17.4 16.3a MIN 17.6a 17.5 19.2 18.1ab Pruning * ns ns * ns 4.10 SD + SPUR 23.0 15.8 15.8 18.2 PRD + SPUR 20.1 19.3 25.9 21.9 SD + MECH 15.7 12.9 14.7 14.4 PRD + MECH 17.2 17.1 20.1 18.1 SD + MIN 15.0 16.1 17.5 16.2 PRD + MIN 20.2 18.3 21.0 19.8 Irrigation*Pruning ns ns ns ns ns 6.11

Water Use Efficiency WUE (t.ML-1) was greater for each treatment in 2001-02 than in any other season because of the higher yields recorded during this season. WUE was improved by PRD relative to SD by 1.4, 1.6, 1.3-fold in seasons 2000-01, 2001-02 and 2002-03, respectively. MIN and MECH significantly improved WUE compared to SPUR in 2000-01 and 2001-02 but no pruning effect was found in 2002-03 due to reduced yield. No significant interactions were observed between irrigation and pruning treatments (Table 3.12). However, a 2-fold improvement in

Chapter 3: Irrigation and pruning effects on yield 49

WUE by PRD + MIN compared to SD + SPUR was recorded in 2000-01 and 2001-02, as a response to lower irrigation inputs and higher yield.

Water Use Efficiency (t.ML-1) Grand Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season Mean 5.32 6.36 3.66 5.11 *** 0.44 SD 4.4a 4.9a 3.0a 4.0a PRD 6.3b 7.7b 4.3b 6.1b Irrigation * ** *** ** * 0.91 SPUR 4.1a 5.4a 3.3 4.3a MECH 5.5b 6.5b 3.4 5.2b MIN 6.4c 7.3b 3.6 5.8b Pruning *** ** ns *** * 0.91 SD + SPUR 3.1 4.4 2.8 3.5 PRD + SPUR 5.0 6.5 4.0 5.1 SD + MECH 4.3 5.5 3.1 4.3 PRD + MECH 6.8 7.5 4.1 6.1 SD + MIN 5.7 4.8 2.9 4.4 PRD + MIN 7.2 9.0 4.9 7.0 Irrigation*Pruning ns ns ns ns ns 1.34

3.4 DISCUSSION

3.4.1 Pruning and irrigation effects on canopy development

The level of pruning influenced canopy development, morphology, shoot growth, maximum leaf area and consequently, the balance between reproductive and vegetative growth. MIN changed the morphology of the Shiraz grapevine framework to produce large permanent woody structures. In terms of total leaf area, pruning treatments had a significant effect on canopy development from budburst to fruit-set, as a direct result of higher bud numbers retained after winter pruning. MIN and MECH vines developed full canopies earlier in the growing season than SPUR vines, as previously reported by Downton and Grant (1992) and Sommer et al. (1993). Therefore net photosynthesis and subsequently, cumulative carbon production may have been greater for MIN and MECH vines, having a beneficial effect on growth and fruit development. Shoot growth and subsequently, canopy morphology was altered by light pruning levels. MIN and MECH produced large numbers of smaller shoots with shorter internode lengths and reduced node number, as a result of increased vegetative

Chapter 3: Irrigation and pruning effects on yield 50 sink competition and partitioning of carbohydrates. These results agree with reports by Clingeleffer (1984) and McCarthy and Cirami (1990). Vegetative “vigour”, as determined by pruning weights at the end of the growing seasons was increased by severe pruning treatments compared to light pruning treatment because of rapid growth of 1-year old wood on SPUR vines. As a consequence of shoot growth dominance of SPUR vines, the balance between vegetative and reproductive growth was altered compared to lightly pruned treatments. Summer pruning was imposed in December to achieve vine balance and allow machinery access to the vineyard.

Canopy development was largely unaffected by irrigation treatments, since the majority of vegetative growth occurred before irrigation was applied each season. However, total leaf area of the vine was reduced by the application of PRD between late November and March. The reduced total leaf area of PRD vines was consistent with reduced canopy biomass reported for Cabernet Sauvignon in response to PRD (Loveys et al. 1998). The reduction in canopy biomass caused by PRD may have been associated with reduced leaf size or restricted development of lateral shoots, as a result of ABA synthesis in the dry rootzones. A reduction in total leaf area after veraison may have negative repercussions on fruit development and maturation, if net photosynthesis and subsequently carbon production is sufficiently reduced. Alternatively, increased bunch exposure due to fewer leaves per vine or smaller leaves may enhance fruit quality, in particular anthocyanin and phenolic concentrations.

Surprisingly, PRD irrigation had no influence on pruning weight in this study, given results from previous work (Dry and Loveys 1998; Loveys et al. 1998) and its effect on total leaf area. PRD field and split-pot experiments conducted on Shiraz, Cabernet Sauvignon and Riesling have demonstrated a reduction in pruning weight, as well as shoot growth rate and leaf area in relation to the biochemical responses produced by PRD (Dry et al. 1996; Dry and Loveys 1998, 1999). Since total leaf area was reduced but pruning weight was unaffected in this study, this suggests that PRD had a greater affect on canopy biomass and lateral growth than on primary shoot growth. The lack of influence of PRD on pruning weight may be a consequence of the integration with pruning systems, particularly in the first experimental season when pruning treatments had an overriding effect on canopy development.

Chapter 3: Irrigation and pruning effects on yield 51

3.4.2 Treatment effects on berry development

Pruning and irrigation effects on berry weight were apparent at the onset of veraison in all three seasons; MIN and PRD reduced berry weight prior to veraison compared to other treatment combinations. Partitioning of carbon between more sinks per vine combined with a PRD-induced hormonal (ABA) response during the initial phase of berry development between flowering and veraison, may have caused the low initial berry weight of PRD + MIN vines. It is well established that berry size is most influenced by stress factors between flowering and veraison, which coincides with the phase of pericarp cell division (Hardie and Considine 1976; Matthews et al. 1987; McCarthy 1997). It has previously been accepted that early water deficits reduce the rate of cell division of the pericarp, which only occurs during the initial stages of berry growth. However, Ojeda et al. (2001) used DNA extraction profiles to show water deficits do not affect cell division of Syrah berries but rather cause a reduction in pericarp volume. This suggests that water deficits or water deficit responses induced by PRD may modify the structural properties of the cell components, such as flexibility, thereby limiting cell enlargement.

The rate of berry weight increase between veraison and harvest occurred in two stages and was constant between all pruning and irrigation treatments. Period 1 involved an initial rapid increase in berry weight with cell expansion and water accumulation in the berry. This was followed by a slow decline of berry weight (Period 2) due to differential water loss, a phenomenon commonly associated with Shiraz (McCarthy 1999). These results suggest a pre-veraison restriction in cell division and solvent accumulation (predominantly water) in berries but cell expansion and solute accumulation (predominantly diffusible sugars) after veraison was not affected by pruning or irrigation treatments. The restrictions on berry expansion by MIN and PRD before veraison may be explained by increased carbohydrate partitioning between greater numbers of sinks and decreased cell wall flexibility respectively. The constancy in the rate of sugar accumulation after veraison is well documented for various growing conditions and berry sizes (Coombe 1980, 1984; Coombe and Phillips 1982; Coombe et al. 1987). Brown and Coombe (1984) have suggested that sugar accumulation in grape berries is an increase in diffusible sugars and suggested the primary control was unloading from the phloem into the apoplast. However, if phloem sap in the berry has the same composition before and after veraison, the question remains as to the factors that trigger the onset of veraison. ABA has been previously reported to play an

Chapter 3: Irrigation and pruning effects on yield 52 important role in triggering veraison (DaPeng et al. 1997); ABA is translocated pre-veraison from grapevine leaves and accumulates in berries (Antolin et al. 2003). This suggests PRD may encourage the onset of veraison by the induced synthesis of ABA in the dry rootzone.

Sugar maturation was delayed by MIN in 2000-01 probably because of strong sink competition due to high crop levels. Other researchers have reported delays in fruit maturity due to light pruning levels (Sommer and Clingeleffer 1993; Poni et al. 1994; Reynold and Wardle 2001; Naor et al. 2002). Delays in veraison and maturation were also observed for SD in 2001-02, which may be attributed to intensive vegetative growth of shoot apices induced by high irrigation inputs, as was shown by Freeman et al. (1980), Bravdo et al. (1985) and Dry et al. (1999). No treatment effects on sugar accumulation were observed in 2002-03 when yield was not significantly different between pruning treatments. Sugar accumulation increased at a faster rate in seasons 2001-02 and 2002-03 compared to 2000- 01. This may be attributed to the reduced total leaf area in the first season, as a result of very high temperatures in December 2001 causing severe leaf senescence at fruit-set. The large leaf area reduction, which occurred prior to veraison, may have resulted in a source: sink imbalance. It has been shown that a reduction in the source of assimilation (i.e. leaves) can result in a delay of the onset of veraison and the rate of maturation (Howell 2001).

3.4.3 Pruning effect on yield components

The change in vine balance and increased sink size of light pruning levels had an important role on determining yield, bunch weight, berry weight and berry number. The conversion of pruning treatments from MECH to MIN and SPUR during winter dormancy in 2000 affected bunch numbers on vines in the subsequent season. MIN increased the number of nodes by retaining previous season’s canes and therefore increased number of bunches per vine. A two-fold difference in bunch number.vine-1 was observed in season 2000-01 for MECH relative to SPUR and a further two-fold difference for MIN to MECH. These results agree with those reported for manually pruned, mechanically hedged and minimally pruned Chancellor vines in British Colombia by Reynolds and Wardle (2001). The increased reproductive sink size of MIN and MECH vines was counter-balanced by increased total leaf area.vine-1 and reductions in bunch weight, berry weight and berry number.bunch-1. Bunch weight and berry weight were significantly reduced by MIN and MECH pruning

Chapter 3: Irrigation and pruning effects on yield 53 treatments compared SPUR in response to high bunch numbers. As a consequence of the reduced bunch and berry weights by MIN, there was no significant difference in the mean yield with MECH and mean yield was only 24% lower for SPUR, despite a 74% reduction in bunch number per vine. The reduction in reproductive growth of light pruning treatments may be attributed to increased competition for photo-assimilates among reproductive sinks (berries). Also reductions in flower numbers and/or fruit set may have contributed to the lower bunch weights and berry number.bunch-1 found on MECH and MIN vines, however these measurements were beyond the scope of this project.

Large seasonal variation in the number of bunches per vine was observed over the three experimental seasons, which had direct repercussions on other yield components. Bunch number.vine-1 was reduced for all treatments in 2002-03 compared to the first two seasons. As a result of the low bunch number.vine-1 and consequently reduced carbohydrate partitioning in the final experimental season, bunch weight and berry weight increased for MIN and MECH vines relative to the previous seasons. Berry number.bunch-1 was also increased in the final season. This suggests flower number and/or fruit set were greater, which corresponds to favourable weather conditions at the appropriate reproductive phenological stages (i.e. flowering and fruit set). Freeman et al. (1979) also reported greater seasonal variation in fruitfulness (as determined by bunches per shoot) than treatment response in irrigated and non-irrigated Shiraz vines of four different pruning levels.

3.4.4 PRD influence on yield components

The application of PRD irrigation restricted berry size and as a consequence, yield penalties were incurred over the three experimental seasons. Previous studies on PRD have generally shown no yield loss for both potted and field-grown vines compared to control irrigation treatments, as vine water status is maintained by irrigating the wet side of the vine but a biochemical response to water stress occurs (Dry et al. 1996, 1999; Loveys et al. 1998). This would suggest PRD imposed a limitation in water availability during the phase of berry expansion, particularly when combined with pruning treatments with high bunch numbers (i.e. MECH and MIN). As a result, the biochemical synthesis of ABA triggered by PRD, in response to water-stress would not have been separated from the physical effects of reduced water availability. However, physiological data presented in Chapter 5 clearly shows leaf

Chapter 3: Irrigation and pruning effects on yield 54 water potential was maintained by PRD throughout the season. Therefore, PRD produced a water-stress response whilst maintaining plant-water status compared to SD irrigation. Alternatively, PRD could have induced a hormonal or chemical response to a pre-veraison water deficit, which may have reduced pericarp volume and restricted berry expansion by modifying the structural properties of cell components and limiting subsequent enlargement of pericarp cells, as shown by Ojeda et al. (1999, 2001).

Over the three seasons, the only interactive effect between irrigation and pruning treatments was on berry weight. The combination of PRD and MIN restricted berry size in each season compared to the other combined irrigation and pruning treatments. This indicates the combination of increased sink size, change in vine balance by early canopy development and the water-stress response of PRD all culminated to result in severe berry size reductions. However, SPUR vines were more sensitive to the application of PRD, as shown by the large differences in berry size and final yield compared to SPUR vines irrigated by SD. The sensitivity of SPUR vines to PRD may be related to the relatively large initial berry size and high water demands for rapid vegetative and reproductive growth at flowering and fruit set (refer to section 3.3.1). If PRD indeed reduces pericarp volume and modifies cell structural properties, it would be expected that a more significant response would occur on rapidly expanding berries with a greater surface area, as was the case for berries produced by SPUR vines.

3.4.5 Treatment effects on vine balance

The balance between vegetative and reproductive growth of vines was assessed as a ratio of total leaf area to fruit produced per vine. Total leaf area required to mature one gram of fruit mass varied between pruning levels, as determined by leaf area index. MECH significantly lowered LA:F compared to SPUR over the three seasons. Therefore, SPUR vines had greater vegetative growth compared to reproductive growth, which may produce within- canopy shading of fruit and consequently undesirable “vegetative” characteristics in the wine (Allen et al. 1996). However, the low crop loads of SPUR vines may have also influenced the higher LA:F than simply differences in canopy size between the pruning systems. Interestingly, MIN vines produced an intermediate LA:F ratio. Thus, vegetative and reproductive growth was greater on MIN vines compared to MECH vines. This may be

Chapter 3: Irrigation and pruning effects on yield 55 the result of maintaining wood at pruning on MIN vines to create a large permanent canopy structure. Also, more vegetative growth and potentially greater photosynthetic capacity may have been required to sustain and ripen the larger crop produced on MIN vines. Given that fruit produced on MECH and MIN vines reached optimal sugar maturity (24 °Brix), the smaller total leaf area was sufficient to produce sugar-ripe fruit. This suggests the reduced total leaf area of MECH and MIN pruning systems was more efficient in carbohydrate production for sugar accumulation in berries than the SPUR vines with larger total leaf area.

There were no irrigation effects on the ratio of vegetative growth to reproductive growth but PRD generally increased LA:F relative to SD. This was an indication of the reduced yield on PRD vines since total leaf area was less for PRD vines than SD vines. Hence the balance between canopy and fruit of both PRD and SD was sufficient to maintain and ripen the crop to an acceptable standard. No significant interactions were found for LA:F, and all treatments produced grapes that met industry-recognised sugar concentrations despite LA:F values ranging from 14.4 cm2.g (SD + MECH) to 21.9 cm2.g (PRD + SPUR). The large range of LA:F for the different irrigation and pruning combinations suggests cv. Shiraz has the capacity to synthesise sufficient carbohydrates or mobilise carbohydrates stored in wood to ripen large cropping levels.

3.4.6 Improvements in water use efficiency caused by PRD

Water use efficiency was influenced by treatments that reduced water inputs (PRD) and increased yield (MIN and MECH). WUE was significantly improved by PRD each season, which agrees with previous studies (Dry et al. 1996; Düring et al. 1996). Loveys et al. (2000) reported the WUE of PRD irrigated Cabernet Sauvignon, Shiraz and Riesling varieties were doubled, in conjunction with no yield penalties relative to control treatments. The large increase in WUE by PRD irrigation has major environmental and economic benefits. Australia’s water resources are limited and currently restrictions on water use from the River Murray for irrigation purposes are enforced to maintain water flow and reduce environmental impact of soil salinisation. As a result, average vineyard irrigation costs have increased and expansion capacity has reduced. Therefore the application of 50% of irrigation water by PRD can potentially reduce production costs, reduce the impact of soil salinisation

Chapter 3: Irrigation and pruning effects on yield 56 by decreasing the amount of water percolating into the ground water table and allow for further vineyard expansion.

3.5 CONCLUSIONS

a) The combination of PRD + MIN reduced the “source-sink” carbohydrate relationship by increasing the number of potential sinks (bunches per vine) whilst maintaining the same level of vegetative biomass as the other combined irrigation and pruning treatments.

b) Pruning level had an overriding effect on canopy development, shoot growth and maximum leaf area. Canopies of MIN vines developed faster and reached maximum size sooner than SPUR and MECH due to the higher bud numbers retained at winter pruning.

c) PRD irrigation did not affect primary shoot growth but total leaf area.vine-1 was reduced later in the growing season. This differs from previous PRD studies (Dry and Loveys 1998; Loveys et al. 1998) and suggests PRD had a transient effect rather than a permanent effect on canopy development.

d) The combined treatments influenced cell division and water accumulation in berries from fruit set through to veraison. PRD + MIN produced the smallest berries with the lowest sugar concentrations on the same calendar day. However, treatments did not have an effect on cell expansion and solute accumulation (predominantly diffusible sugars) after veraison, since berry weight increments and rates of sugar accumulation were consistent between treatments.

e) The yield response of Shiraz grapevines was dominated by bunch number.vine-1, which was predetermined by pruning level and climatic conditions during bunch initiation. The high node number retained on MIN vines at the pruning resulted in bunch numbers 4-fold greater than SPUR vines.

Chapter 3: Irrigation and pruning effects on yield 57 f) PRD reduced berry size, as determined prior to veraison. Since plant-water status was maintained by PRD (refer to Chapter 5), the restriction in berry size was not due to a limitation in water availability during cell division but possibly a hormonally-induced reduction in pericarp volume and pericarp cell expansion.

g) Water use efficiency (t.ML-1) was doubled by the reduced water inputs of PRD and increased yields of MIN and MECH relative to the low yielding, high water input treatments (SD + SPUR).

h) MIN increased yield, whilst PRD induced a water-stress response that reduced total leaf area. Therefore the hypothesis that “The combination of light pruning levels of minimal pruning with PRD irrigation will improve the “source-sink” carbohydrate relationship may be accepted.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 58

4 IRRIGATION AND PRUNING EFFECTS ON FRUIT AND WINE COMPOSITION

4.1 INTRODUCTION AND EXPERIMENTAL AIMS

Optimisation of fruit and wine quality is important to sustain the competitiveness of the Australian wine industry, but to maximise profitability quality must be balanced with yield and production costs. The relationship between yield and quality is complex, since it depends on the interactions between external climatic and cultural factors (Jackson and Lombard 1993). The effects of viticultural management strategies on yield and fruit composition have been extensively studied with diverse results, given the large variation in climate, cultivars, irrigation regimes and canopy management (Freeman et al. 1983; Bravdo et al. 1985a,b; Bravdo and Hepner 1987; Rojas-Lara and Morrison 1989; Sommer and Clingeleffer 1993; Reynolds et al. 1994; Dry et al. 1996; Naor et al. 2002). Some viticultural strategies employed to improve fruit and wine composition are associated with yield reductions, e.g. regulated deficit irrigation (RDI) and bunch thinning. Yet, other management strategies may increase or maintain yield with either no effect or improvements on fruit specification, e.g. minimal pruning and partial rootzone drying (PRD). The influence of viticultural management on resultant fruit composition is dependent on changes in grapevine photosynthetic capacity (Smart and Coombe 1983; Matthews and Anderson 1988; Williams and Matthews 1990), berry size (Clingeleffer 1983; Bravdo et al. 1985b), canopy microclimate (Smart and Robinson 1991; Allen et al. 1996) and/or vine vigour (Dry 1997).

The benefits of irrigation for fruit and wine quality have been questioned in the literature. This has resulted in the lack of a uniform hypothesis because of the disparity in irrigation regimes, irrigation application method, environment and cultivar (Hardie and Considine 1976; Freeman et al. 1983; van Zyl 1984). Irrigation primarily alters the water status and photosynthetic capacity of vines, which may result in greater sugar accumulation in berries (Matthews and Anderson 1988; Poni et al. 2000). Increased plant water status can stimulate shoot vigour and alter canopy microclimate that in turn, leads to excessive shading of bunches. This may have a negative influence on quality by lowering total soluble solids, anthocyanin and phenolic concentrations in berries and produce ‘herbaceous’ characteristics in wine (Smart and Robinson 1991; Allen et al. 1996; Esteban et al. 2001). However,

Chapter 4: Irrigation and pruning effects on fruit and wine composition 59 grapevine vigour has been controlled through the stimulus of a biochemical water stress response by partially drying the rootzone (PRD) (Dry et al. 2000). Previous PRD pot and field experiments on grapevines have shown no yield penalties because plant water status is maintained by irrigating one half of the vine root system, while the other half dries out (Dry et al. 1996; Dry 1997; Loveys et al. 1998). In addition, PRD may enhance anthocyanin concentration by a reduction in bunch shading and/or by increasing the exocarp to juice ratio (Dry 1997). The effect of PRD on wine quality requires evaluation, as it does not appear to have been documented to date.

Similarly, the effects of canopy management on fruit and wine composition are varied in the literature, given the range of pruning techniques, trellising, crop level, climate and cultivar (Freeman and Kliewer 1983; Morris and Cawthon 1982; Poni et al. 2000; Sommer and Clingeleffer 1993). Pruning level has a large effect on crop level and subsequently, fruit and wine quality (Reynolds et al. 1994). Light pruning levels, as achieved by mechanical hedging or minimal pruning, produce higher crop levels than severe pruning treatments, such as spur pruning, since a greater number of nodes are retained at winter pruning. Reynolds and Wardle (2001) showed light pruning levels reduced fruit quality by increasing yield and pH and lowering sugar concentration and titratable acidity when compared to severe pruning levels of Chancellor grapes. In contrast, light pruning by mechanical hedging and minimal pruning has been found to improve fruit quality (tartrate to malate ratio, phenolic and anthocyanin concentration) and wine spectral parameters by increased bunch exposure and reduced berry size, when adequate sugar levels were reached in both hot and cool climates (Clingeleffer 1992, 2001; Schultz et al. 2001).

Several researchers have investigated the effects of both irrigation and cropping level on fruit and wine quality by using bi-factorial experiments (Neja et al. 1977; Freeman et al. 1980; Morris and Cawthon 1982; Freeman and Kliewer 1983; Kliewer et al. 1983; Bravdo et al. 1985a,b;). Neja et al. (1977), Freeman et al. (1980) and Kliewer et al. (1983) reported crop level had no influence on fruit and wine quality of Vitis vinifera but delays in sugar accumulation were observed for high crop levels by Bravdo et al. (1985a,b). Water stress or alternatively, excessive irrigation reduced sugar concentration and pH in the must (Neja et al. 1977; Freeman et al. 1980). Morris and Cawthon (1982) also showed that irrigation

Chapter 4: Irrigation and pruning effects on fruit and wine composition 60 delayed sugar maturation of cv. Carignane but increased pH and colour compared to no irrigation.

The aim of this study is to:

1. Determine irrigation and pruning effects on fruit composition, in respect to total soluble solids, pH, titratable acidity, anthocyanin and phenolic concentration.

2. Evaluate effects of irrigation and pruning treatments on wine composition.

3. Assess the relationship between fruit and wine composition.

4. Assess the relationship between yield components, as discussed in the previous chapter (refer to section 3.4.3), fruit and wine composition.

Given the potential for irrigation management and pruning techniques to improve fruit and wine composition irrespective of yield, this study will test the following hypotheses:

a) The effects of irrigation and pruning treatments on fruit and wine composition of cv. Shiraz will be minimal because vegetative and reproductive growth is balanced. b) Small berry size will improve fruit and wine composition parameters, as a result of an increase in exocarp to juice ratio and concentration of solutes in the mesocarp.

4.2 METHODOLOGY

4.2.1 Determination of fruit total soluble solids, pH and titratable acidity

Fruit composition was assessed at harvest on 10 bunches randomly sampled from each of the harvest vines (see Chapter 3.2.3 for harvest details). All berries were plucked from the rachis and one hundred berries were randomly selected and weighed; mean berry weight was calculated. A further 100 berries were sampled and stored at –20°C for analysis of anthocyanin and phenolic concentrations.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 61

The 100-berry sub-sample was homogenised by hand using a mortar and pestle. Grape juice was extracted through a 2mm gauge sieve, left to settle for 1 hour and a 1 mL sub-sample was used for measurement of total soluble solids (TSS, °Brix) using a temperature compensating PR-101 digital refractometer (Atago Co. Ltd., Tokyo, Japan). A further 30 mL aliquot was centrifuged for 10 minutes at 3500 rpm and 10 mL of the remaining juice was used for the measurement of juice pH and titratable acidity (TA) using VIT90 autotitrator (Radiometer Co. Ltd., Brønshøj, Copenhagen). The pH electrode was calibrated against buffers of pH 3 and 7 and titratable acidity was determined by titration to an end point of pH 8.3 with 0.100 M NaOH.

4.2.2 Determination of fruit anthocyanin and phenolic concentration

Anthocyanin and phenolic concentrations were determined using the methodology of Iland (1996). At harvest, a 100-berry sub-sample was collected, its fresh weight recorded and it was stored at –20°C until analysis. Berry samples were partially thawed at room temperature 1 hour prior to maceration. Berries were macerated to a consistent mixture of exocarp, mesocarp and seeds using a Barmix® handheld food blender. A sub-sample of homogenate (ca. 1.0 g) was weighed, suspended in aqueous ethanol at pH 2 (50% v/v). The supernatant was mixed by inversion four times during a 1-hour period and centrifuged at 3500 rpm for 10 minutes. An aliquot of supernatant (1.0 mL) was added to 10 mL of 1.0M HCl, mixed thoroughly and incubated at room temperature for 3 hours. The volume of remaining supernatant was recorded and added to the total amount.

Absorbance was measured through a cell with a path length of 50 mm at 520 nm (Abs520) for anthocyanin concentration and 280 nm (Abs280) for phenolic concentration, employing a

Cintra 10e UV-visible spectrophometer (GBC Scientific Equipment Pty. Ltd., Victoria, Australia). Formulae for the calculation and anthocyanin and phenolic concentrations are presented in Appendix C.

4.2.3 Small-scale Wine Production

Two replicates of approximately 4 kg of fruit pooled from across the experimental site for the six treatments were fermented at harvest, using a similar method to Becker and Kerridge

Chapter 4: Irrigation and pruning effects on fruit and wine composition 62

(1972). Grapes were cooled overnight at 4°C and returned to room temperature (20°C) before being processed through a de-stemmer/crusher (Amos Ltd., Heilbronn, Germany).

The grapes were crushed into 10L polyethylene containers, and 0.1 g/L SO2 (10% w/v sodium metabisulphite solution) and 0.75 g/L diammonium hydrogen orthophosphate (10% w/v solution) were added. Tartaric acid was added as required to lower the pH of must to between 3.40 - 3.55. Thiamine hydrochloride and re-hydrated Lalvin 254D yeast were also added at a rate of 1 g/L and 0.2 g/L, respectively.

The must was fermented at 25°C for 3 days, plunged twice daily and then pressed by a 30 L Willmes air-bag press (Willmes Ltd., Bensheim, Germany). The must was transferred to

CO2 filled glass bottles to continue fermenting at 25°C until less than 0.25% sugar remained.

The wine was racked, adjusted to 20 mg/L SO2 (using 10% w/v sodium metabisulphite) and stored for 14 days at 15°C, after which a second racking occurred and SO2 re-adjustment was made if required. The wine was cold stabilized for +30 days at 4°C and sterile-filtered into 375 mL bottles.

4.2.4 Determination of Wine Composition

Wine pH and TA were assessed at bottling, approximately 6-8 weeks after pressing. A 25 mL sample of wine was degassed using a sonicator for 30 minutes from which a 10 mL sub- sample was diluted with 40 mL distilled water. The pH and TA were measured using a 702 SM Titrino titrator (Metrohom Ltd. Herisau, Switzerland). The pH electrode was calibrated against buffers of pH 3 and 7 and titratable acidity was determined by titration to an end- point of pH 8.3 with 0.100 M NaOH. Spectral evaluation of the wine samples was conducted according to Somers and Evans (1977) (refer to Appendix D) to measure colour density, colour hue, total anthocyanin concentration, ionised anthocyanin, degree of ionisation of anthocyanins (α) and total phenolic concentration. Ionised anthocyanins represent the fraction of anthocyanins present in the coloured form and the degree of ionised anthocyanins is the percentage of total anthocyanins present in the coloured form.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 63

4.2.5 Statistical Analysis

Fruit and wine composition data were analysed by Analysis of Variance (ANOVA) using GenStat® 6th Edition (Lawes Agricultural Trust, Rothamsted Experimental Station). All interactions were balanced and analysed by general ANOVA (Refer to Appendix B for an example of the ANOVA analysis). The treatment structure was analysed as Irrigation*Pruning and the blocking structure was Row*Plot. Yield and fruit composition, yield and wine composition and yield, fruit and wine composition were correlated using the correlation coefficient (r) to find interrelationships at the 5% significance level. Correlations were done on treatment means for each season and across all seasons.

4.3 RESULTS

4.3.1 Fruit Total Soluble Solid, pH and Titratable Acidity Results

Berry composition with respect to TSS, pH and TA was predominantly affected by pruning treatments. As a consequence, the interactions between irrigation and pruning for fruit analysis were generally insignificant.

Total Soluble Solids Pruning level affected TSS concentration in seasons 2000-01 and 2002-03 (Table 4.1). MECH and MIN had lower TSS concentrations overall compared to SPUR (3%). No irrigation effects were observed. The interaction between irrigation and pruning was significant for TSS in 2002-03, however the range of TSS concentrations was only 1 °Brix. The variation between seasons was not a true representation because of the change in harvest strategy from calendar date in 2000-01 to harvest by maturity level in 2001-02 and 2002-03.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 64

Total Soluble Solids (°Brix) Treatment 2000-01 2001-02 2002-03 Grand Mean Year LSD Season Mean 22.9 24.4 24.5 23.9 *** 0.24 SD 23.1 24.6 24.5 24.0b PRD 22.8 24.1 24.5 23.8a Irrigation ns ns ns ** ns 0.13 SPUR 23.7b 24.5 25.0b 24.4b MECH 22.7a 24.2 24.4a 23.7a MIN 22.4a 24.4 24.1a 23.7a Pruning *** ns *** *** *** 0.23 SD + SPUR 24.0 24.6 25.0bc 24.5e PRD + SPUR 23.5 24.3 25.1c 24.3d SD + MECH 23.0 24.4 24.7bc 24.0c PRD + MECH 22.4 24.0 24.0a 23.5a SD + MIN 22.3 24.7 23.8bc 23.6b PRD + MIN 22.6 24.1 24.4ab 23.7b Irrigation*Pruning ns ns ** * ns 0.55 pH

The difference in pH of berry juice between treatments was low. However, these differences were significant because of the low level of variation. SPUR increased the pH of berry juice relative to the other pruning treatments in each experimental season (Table 4.2). Irrigation had a minor effect in season 2002-03; PRD increased pH relative to SD by 1%. The interaction between irrigation and pruning had no significant effect on pH.

Titratable Acidity

The irrigation and pruning effects on juice TA were relatively small and the difference in mean TA between combined treatments was only ± 0.18 g.L-1 (Table 4.3). However, MIN increased TA compared to MECH and SPUR in seasons 2000-01 and 2002-03. The effect of irrigation on TA was negligible. A significant interaction between irrigation and pruning occurred in season 2001-02. PRD + MECH increased TA compared to all other treatment combinations.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 65

pH Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 4.00 3.61 3.65 3.75 *** 0.03 SD 3.98 3.64 3.62a 3.75 PRD 4.02 3.59 3.67b 3.76 Irrigation ns ns * ns *** 0.06 SPUR 4.09c 3.62b 3.70c 3.80b MECH 4.00b 3.57a 3.65bc 3.74a MIN 3.91a 3.66b 3.60ab 3.72a Pruning *** * ** ** *** 0.05 SD + SPUR 4.10 3.67 3.67 3.81 PRD + SPUR 4.07 3.56 3.72 3.79 SD + MECH 3.99 3.58 3.62 3.73 PRD + MECH 4.02 3.55 3.68 3.75 SD + MIN 3.84 3.67 3.57 3.69 PRD + MIN 3.96 3.66 3.62 3.75 Irrigation*Pruning ns ns ns ns ns 0.08

Titratable Acidity (g.L-1) Treatment 2000-01 2001-02 2002-03 Grand Mean Year LSD Mean 4.72 4.31 4.02 3.96 *** 0.10 SD 4.71 4.32 4.06 3.96 PRD 4.72 4.34 3.98 3.95 Irrigation ns ns ns ns ns 0.16 SPUR 4.63a 4.23a 3.83a 3.85a MECH 4.63a 4.64b 4.04b 4.05b MIN 4.90b 4.07a 4.18b 3.97b Pruning ** *** *** *** *** 0.17 SD + SPUR 4.63 4.37bc 3.90 3.92 PRD + SPUR 4.63 4.09a 3.76 3.78 SD + MECH 4.62 4.49c 4.08 4.01 PRD + MECH 4.63 4.78d 4.00 4.09 SD + MIN 4.89 3.98a 4.20 3.94 PRD + MIN 4.91 4.14ab 4.17 4.00 Irrigation*Pruning ns * ns ns ns 0.26

Chapter 4: Irrigation and pruning effects on fruit and wine composition 66

4.3.2 Fruit Anthocyanin and Phenolic Concentrations and Content

Anthocyanin and phenolic concentrations were improved by light pruning treatments but the content of these compounds per berry was influenced by berry size. The combined irrigation and pruning effects on anthocyanin and phenolic concentrations varied from season to season.

Anthocyanin Concentration Anthocyanin concentrations were much higher in 2001-02 than 2000-01 (105%) and 2002- 03 (51%) (Table 4.4 a). A significant difference between pruning technique occurred in season 2001-02; MIN increased anthocyanin concentration by 8% compared to SPUR. Irrigation method had no significant effect on anthocyanin concentration, nor did the combination of irrigation and pruning.

Anthocyanin Content Anthocyanin content per berry was influenced by the large difference in berry weight between combined irrigation and pruning treatments (Table 4.4 b). Anthocyanin content was highest on combined treatments with large berries (e.g. SD + SPUR) and lowest on treatments with small berries (e.g. PRD + MIN). The reductions in anthocyanin content associated with PRD were more pronounced when combined with SPUR and MECH.

Phenolic Concentration Phenolic concentrations were also high in the cool season of 2001-02 (Table 4.5 a). Differences between pruning treatments were significant in seasons 2001-02 and 2002-03. MIN increased phenolic concentration relative to MECH (10%) and SPUR (15%) in season 2001-02. MIN and MECH both increased phenolic concentration relative to SPUR in season 2002-03 by 11% and 6%, respectively. PRD increased phenolic concentration by 14% compared to SD in 2002-03 only. Combined irrigation and pruning treatments had a significant effect on phenolic concentration in 2003 and over the three seasons. PRD + MIN and PRD + MECH increased phenolic concentrations, while PRD + SPUR decreased phenolic concentration.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 67

a. Anthocyanin concentration (mg Anthocyanin.berry mass-1 g) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 0.94 1.93 1.42 1.43 *** 0.06 SD 0.93 1.85 1.39 1.39 PRD 0.97 2.00 1.46 1.47 Irrigation ns ns ns ns ns 0.18 SPUR 0.95 1.86a 1.39 1.40 MECH 0.94 1.93ab 1.42 1.43 MIN 0.95 2.01b 1.47 1.47 Pruning ns * ns ns ns 0.11 SD + SPUR 0.92 1.82 1.38 1.37 PRD + SPUR 0.98 1.89 1.41 1.42 SD + MECH 0.93 1.80 1.39 1.37 PRD + MECH 0.94 2.06 1.45 1.48 SD + MIN 0.93 1.96 1.41 1.43 PRD + MIN 0.97 2.04 1.52 1.51 Irrigation* ns ns ns ns ns 0.20 Pruning

b. Anthocyanin Content (mg Anthocyanin.berry-1) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 1.14 1.92 1.62 1.56 *** 0.07 SD 1.22 2.07 1.75b 1.66b PRD 1.05 1.79 1.48a 1.45a Irrigation ns ns * * ns 0.16 SPUR 1.22 2.18c 1.73b 1.72c MECH 1.14 1.92b 1.73b 1.60b MIN 1.06 1.62a 1.38a 1.34a Pruning ns *** *** *** *** 0.12 SD + SPUR 1.31 2.43d 1.94 1.89d PRD + SPUR 1.12 1.92bc 1.52 1.53bc SD + MECH 1.24 2.00c 1.80 1.68c PRD + MECH 1.03 1.85bc 1.66 1.51b SD + MIN 1.11 1.63a 1.49 1.38ab PRD + MIN 1.02 1.61ab 1.28 1.30a Irrigation* ns *** ns ** ns 0.20 Pruning

Phenolic Content Phenolic content was dominated by berry size: large berries produced on vines from the combined treatment of SPUR and SD had the highest content of phenolics per berry. Phenolic content was significantly affected by the combination of irrigation and pruning treatments in 2001-02 and 2002-03 (Table 4.5 b). Treatments which produced the largest berries had the highest level of phenolics per berry, namely SD + SPUR and SD + MECH.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 68

TABLE 4.5: The a. concentration and b. content of berries from vines with integrated irrigation and pruning treatments and seasonal means for 2000-01, 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant.

a. Phenolic Concentration (mg Phenolics.berry mass-1 g) Treatment 2000-01 201-02 2002-03 Grand Mean Season LSD Season Mean 1.04 1.60 1.36 1.33 *** 0.05 SD 1.01 1.54 1.27a 1.28 PRD 1.07 1.64 1.45b 1.38 Irrigation ns ns * ns * 0.19 SPUR 1.06 1.50a 1.21a 1.26a MECH 1.03 1.58a 1.41b 1.34b MIN 1.03 1.73b 1.46b 1.40b Pruning ns ** *** *** *** 0.09 SD + SPUR 1.06 1.53 1.23a 1.28a PRD + SPUR 1.06 1.47 1.19a 1.24a SD + MECH 1.03 1.51 1.29a 1.28a PRD + MECH 1.04 1.65 1.53b 1.41ab SD + MIN 0.96 1.61 1.29a 1.29a PRD + MIN 1.09 1.80 1.62b 1.51b Irrigation* ns ns ** ** ns 0.20 Pruning

b. Phenolic Content (mg Phenolics.berry-1) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 1.26 1.59 1.53 1.46 *** 0.06 SD 1.34 1.73 1.59 1.55 PRD 1.17 1.47 1.46 1.37 Irrigation ns ns ns ns ns 0.18 SPUR 1.37b 1.77c 1.51b 1.55b MECH 1.26ab 1.58b 1.72c 1.52b MIN 1.14a 1.39a 1.36a 1.29a Pruning ** *** *** *** *** 0.11 SD + SPUR 1.51 2.05c 1.74b 1.76c PRD + SPUR 1.22 1.50ab 1.28a 1.33a SD + MECH 1.37 1.68b 1.67b 1.57b PRD + MECH 1.14 1.49ab 1.76b 1.46ab SD + MIN 1.14 1.35a 1.37a 1.28a PRD + MIN 1.14 1.42a 1.35a 1.31a Irrigation* ns *** *** *** ** 0.20 Pruning

4.3.3 Relationship between Fruit Anthocyanin Concentration and Temperature

The relationship between mean anthocyanin concentration of all treatments at harvest and mean daily temperature for the months of December, January and February were examined over the three experimental seasons. Significant negative correlations between anthocyanin

Chapter 4: Irrigation and pruning effects on fruit and wine composition 69 concentration and January and February mean daily temperatures were observed (Table 4.6). When daily temperatures were lower than average during maturation period (January to February), colour concentration in berries was higher.

December January February R2 Mean Temp (ºC) Mean Temp (ºC) Mean Temp (ºC) Anthocyanin Concentration -0.930 -0.994 -0.999 mg/g

4.3.4 Wine pH and Titratable Acidity Results The pH and TA of the wine produced was largely unaffected by either irrigation or pruning. Large variation in pH and TA was observed in season 2001-02; wine was significantly higher in TA and lower in pH than the other two experimental seasons.

Wine pH A minor pruning effect on wine pH occurred in the first experimental season (Table 4.7). Wine pH was lower for MIN compared to SPUR (4%) and MECH (4%). SD increased the pH of the wine in 2001-02 by 2% but no other irrigation effects were recorded. No interactions between irrigation and pruning treatments were observed.

Titratable Acidity SD irrigation raised TA in season 2000-01 compared to PRD but no other significant effects were recorded (Table 4.8) The lack of treatment difference between the wine TA was expected, since the must was acid adjusted during the wine making process.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 70

Wine pH Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 3.50 3.45 3.58 3.51 *** 0.05 SD 3.49 3.49b 3.58 3.52 PRD 3.52 3.41a 3.59 3.51 Irrigation ns * ns ns ns 0.07 SPUR 3.54b 3.49 3.57 3.53b MECH 3.55b 3.45 3.58 3.52ab MIN 3.41a 3.42 3.61 3.48a Pruning * ns ns * * 0.08 SD + SPUR 3.55 3.56 3.59 3.49 PRD + SPUR 3.54 3.42 3.56 3.52 SD + MECH 3.54 3.47 3.56 3.49 PRD + MECH 3.56 3.43 3.59 3.41 SD + MIN 3.37 3.44 3.59 3.58 PRD + MIN 3.46 3.40 3.63 3.59 Irrigation* Pruning ns ns ns ns ns 0.11

Wine Titratable Acidity (g.L-1) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 7.25 8.26 7.77 7.76 *** 0.29 SD 7.61b 8.01 7.70 7.77 PRD 6.89a 8.50 7.83 7.74 Irrigation * ns ns ns ** 0.41 SPUR 7.48 8.17 7.63 7.76 MECH 7.01 8.01 7.86 7.63 MIN 7.25 8.60 7.81 7.89 Pruning ns ns ns ns ns 0.50 SD + SPUR 8.02 7.84 7.49 7.78 PRD + SPUR 6.94 8.50 7.77 7.74 SD + MECH 7.36 8.01 7.84 7.74 PRD + MECH 6.67 8.02 7.88 7.52 SD + MIN 7.45 8.19 7.78 7.80 PRD + MIN 7.06 9.00 7.85 7.97 Irrigation* Pruning ns ns ns ns ns 0.71

Chapter 4: Irrigation and pruning effects on fruit and wine composition 71

4.3.5 Wine Spectral Evaluation

Wine spectral parameters were strongly influenced by PRD irrigation. Improvements in wine colour density, hue, colour and phenolics were generally associated with PRD. Large seasonal variation was also observed for wine density and anthocyanin and phenolic concentrations.

Wine Colour Density Wine colour was significantly improved by PRD compared to SD, by 22% and by 31% in seasons 2000-01 and 2001-02, respectively (Table 4.9). MECH significantly improved colour density in 2002-03 relative to SPUR (17%) and MIN (15%). Interestingly, trends in wine colour density for both irrigation and pruning were reversed in 2002-03 compared to previous seasons as SD and MECH both significantly improved density. No interaction effects on colour density were observed for combined irrigation and pruning treatments.

Wine Colour Density (AU) Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season Mean 8.75 10.36 8.07 9.06 *** 0.72 SD 7.90a 9.03a 8.60b 8.51a PRD 9.60b 11.70b 7.54a 9.61b Irrigation * ** * *** *** 1.02 SPUR 8.75 10.50 7.59a 8.95 MECH 8.05 9.34 8.90b 8.76 MIN 9.46 11.25 7.72a 9.48 Pruning ns ns * ns * 1.25 SD + SPUR 7.75 8.38 8.45 8.19 PRD + SPUR 9.74 12.62 6.74 9.70 SD + MECH 7.08 7.97 9.91 8.32 PRD + MECH 9.01 10.71 7.89 9.20 SD + MIN 8.88 10.73 7.44 9.02 PRD + MIN 10.05 11.77 7.99 9.94 Irrigation* Pruning ns ns ns ns * 1.77

Chapter 4: Irrigation and pruning effects on fruit and wine composition 72

Wine Hue Wine hue is an indication of the red-brown colouration of the wine. Brighter and more red coloured wines have lower hue values. On average, wines were brighter and redder in seasons 2000-01 and 2001-02 compared to 2002-03, particularly for SPUR and MIN treatments (Table 4.10). PRD irrigation reduced hue by 8% in 2001-02 and by 13% over the three seasons compared to SD. A significant interaction between irrigation and pruning was found in 2001-02: PRD reduced hue in wines produced from SPUR and MECH vines but not from MIN vines.

Wine Colour Hue (AU) Treatment 2000-01 2001-02 2002-03 Mean Season LSD Season Mean 0.52 0.51 0.54 0.52 *** 0.01 SD 0.52 0.53b 0.55 0.53b PRD 0.52 0.49a 0.54 0.51a Irrigation ns ** ns *** ** 0.02 SPUR 0.53 0.51 0.56b 0.53 MECH 0.53 0.51 0.52a 0.52 MIN 0.50 0.51 0.55b 0.52 Pruning ns ns ** ns * 0.02 SD + SPUR 0.53 0.55b 0.57 0.55d PRD + SPUR 0.52 0.48a 0.54 0.51ab SD + MECH 0.53 0.54b 0.53 0.53cd PRD + MECH 0.52 0.48a 0.51 0.50a SD + MIN 0.49 0.51ab 0.55 0.51abc PRD + MIN 0.51 0.51ab 0.55 0.52bc Irrigation* Pruning ns * ns ** ns 0.03

Ionised Anthocyanins The fraction of anthocyanins present in the coloured form was significantly higher in 2001- 02 and 2002-03 relative to 2001-02 (Table 4.11). Ionised anthocyanin content was not affected by the interaction between irrigation and pruning. However, PRD increased it compared to SD in 2000-01, 2001-02 and 2002-03 by 22%, 41% and 18% respectively. MECH and MIN significantly improved ionised anthocyanin concentration in 2002-03

Chapter 4: Irrigation and pruning effects on fruit and wine composition 73 compared to SPUR and MIN had the highest level of ionised anthocyanins over the three seasons.

Ionised Anthocyanins (mg.L-1) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 83 100 95 93 ** 8.7 SD 75a 83a 87a 81a PRD 91b 117b 103b 104b Irrigation * ** ** *** ns 12.3 SPUR 80 100 78a 86a MECH 74 91 101b 89a MIN 95 109 106b 103b Pruning ns ns ** ** * 15.1 SD + SPUR 70 74 68 71 PRD + SPUR 90 126 89 101 SD + MECH 64 73 92 76 PRD + MECH 84 110 111 102 SD + MIN 90 102 101 98 PRD + MIN 100 115 111 109 Irrigation* Pruning ns ns ns ns ns 21.3

Total Anthocyanins Total anthocyanin content of wine increased progressively over the three experimental seasons (Table 4.12). The seasonal differences in total anthocyanin content of the wine were greater than the irrigation and pruning effects. PRD consistently increased total anthocyanin content of the wine compared to SD in 2000-01, 2001-02 and 2002-03 by 13%, 15% and 19%, respectively. A pruning effect was only observed in 2000-01: SPUR improved total anthocyanin content compared to MECH and MIN. An irrigation and pruning interaction was also observed in 2000-01: PRD significantly increased total anthocyanin content of SPUR (14%) and MIN (24 %) vines but not of MECH vines.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 74

b. Total Anthocyanins (mg.L-1) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 458 731 966 719 *** 43.1 SD 430a 679a 883a 664a PRD 487b 784b 1049b 773b Irrigation *** * ** *** * 60.9 SPUR 490b 754 966 737 MECH 447a 704 960 704 MIN 438a 736 972 715 Pruning ** ns ns ns ns 74.6 SD + SPUR 458bc 675 840 658 PRD + SPUR 521d 833 1091 815 SD + MECH 441b 626 876 647 PRD + MECH 454bc 781 1044 760 SD + MIN 391a 736 934 687 PRD + MIN 485cd 737 1010 744 Irrigation* Pruning * ns ns ns ns 105.5

Degree of Ionisation of Anthocyanins (α) The percentage of total anthocyanins present in the coloured form decreased progressively over the three seasons as total anthocyanin concentration increased (Table 4.13). A pruning effect was observed in 2000-01: MIN increased the degree of ionisation of anthocyanins compared to SPUR and MECH. An irrigation effect occurred in 2001-02 as PRD increased α compared to SD. No significant interactions between irrigation and pruning were found.

Total Phenolics A progressive increase in mean total phenolic concentration occurred with each season (Table 4.14). Irrigation and pruning treatments both had an effect on total phenolics of the wine in seasons 2000-01 and 2002-03. PRD improved total phenolic concentration in the wine compared to SD irrigation in both seasons. Whilst, total phenolics were increased by SPUR relative to MIN and MECH in season 2000-01 and by MIN relative to SPUR and MECH in season 2002-03. A significant interaction between irrigation and pruning was observed in 2002-03: PRD increased phenolic concentrations in wine from all three pruning treatments but to a lesser degree for MIN than SPUR or MECH.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 75

Degree of ionisation of Anthocyanins (%) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 18.2 13.5 9.9 13.9 *** 1.69 SD 17.6 12.1a 9.8 13.2a PRD 18.8 15.0b 9.9 14.6b Irrigation ns ** ns * ns 2.39 SPUR 16.3a 13.1 8.1 12.5a MECH 16.5a 12.8 10.6 13.3a MIN 21.8b 14.8 10.9 15.8b Pruning * ns ns ** ns 2.92 SD + SPUR 15.3 11.0 8.1 11.5 PRD + SPUR 17.2 15.1 8.0 13.5 SD + MECH 14.5 11.5 10.5 12.2 PRD + MECH 18.6 14.0 10.8 14.5 SD + MIN 23.1 13.9 10.9 15.9 PRD + MIN 20.5 15.8 11.0 15.8 Irrigation* Pruning ns ns ns ns ns 4.14

TABLE 4.14: total phenolic concentration of wine from integrated irrigation and pruning treatments for season 2000-01, 2001-02, 2002-03 and seasonal means. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments denoted by *** P<0.001, ** P<0.01, * P<0.05, ns = non significant.

Total Phenolics (mg.L-1) Treatment 2000-01 2001-02 2002-03 Grand Mean Season LSD Season Mean 44 61 78 61 *** 2.74 SD 42a 58 71a 57a PRD 46b 64 84b 65b Irrigation ** ns *** *** * 3.87 SPUR 47b 62 75a 62ab MECH 42a 57 78ab 59a MIN 43a 65 80b 63b Pruning * ns * * * 4.74 SD + SPUR 44 57 68a 56 PRD + SPUR 50 67 82c 66 SD + MECH 41 53 69a 54 PRD + MECH 43 61 86c 63 SD + MIN 40 64 77b 60 PRD + MIN 46 66 82c 65 Irrigation* Pruning ns ns ** ns ns 6.71

Chapter 4: Irrigation and pruning effects on fruit and wine composition 76

4.4 YIELD AND FRUIT COMPOSITION CORRELATIONS

Season 2000-01 Yield was strongly correlated with the number of bunches per vine in season 2000-01 (Table 4.15a). MIN produced high yield because of a high number of small bunches with less berries of smaller mean size. The leaf area: fruit ratio (LA:F) was lower for MIN treatments and water use efficiency (WUE) was greater because yield was higher. Berries were less mature, TA was greater and pH was lower at harvest on high yielding treatments. Anthocyanin concentration was negatively correlated with berry weight: smaller berries had higher anthocyanin levels.

Season 2001-02 The correlation between bunch number and yield was weaker in 2001-02 than the previous season, but bunch number was still negatively correlated with bunch weight and berry number (Table 4.15b). Berry number and berry weight positively correlated with bunch weight. As expected, LA:F was negatively correlated with yield, since yield was the denominator in the leaf area to fruit ratio. The degree of maturation of berries was related to WUE: low yielding treatments, such as SPUR had lower WUE and more mature berries at harvest. Fruit TA and pH were not correlated with yield components. However, anthocyanin concentration was greater in smaller berries. Anthocyanin and phenolic concentrations were positively correlated, thus smaller berries should also have higher phenolic concentrations.

Season 2002-03 Similarly, bunch number did not correlate with yield but negatively correlated with bunch weight, berry number and berry weight in season 2002-03 (Table 4.15c). Also, berry number and berry weight dominated bunch weight and there was a positive correlation between berry number and berry weight. Pruning and irrigation treatments that produced more bunches with fewer berries per bunch had higher levels of TA and lower pH levels. Anthocyanin concentrations were higher in smaller berries and anthocyanins and phenolics were interrelated.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 77

Table 4.15: Correlation coefficients of mean yield components [yield, bunch number.vine-1 (Bunch No.), bunch weight (Bunch Wt.), berry number.bunch-1 (Berry No.), leaf area:fruit (LA:F), water use efficiency (WUE)] and fruit composition parameters[total soluble solids (TSS), pH, titratable acidity (TA), anthocyanin concentration (Antho.) and phenolic concentration (Phenol)] for seasons a. 2000-01, b. 2001-02, c. 2002-03, n=6. Bold script represents significance at 5% level. a. Yield and Fruit Composition Parameters 2000-01 Yield Bunch Bunch Berry Berry LA:F WUE TSS pH TA Antho. Phenol. r (kg) No. Wt. (g) No. Wt. (g) (cm2/g) (t/ML) (°Brix) (g/L) (mg/g) (mg/g) Yield 1.00 0.89 -0.79 -0.86 -0.22 -0.81 0.36 -0.78 -0.97 0.63 -0.14 -0.69 Bunch No. 1.00 -0.93 -0.96 -0.50 -0.57 0.60 -0.79 -0.95 0.91 0.24 -0.40 Bunch Wt. 1.00 0.96 0.71 0.61 -0.81 0.91 0.87 -0.82 -0.47 0.21 Berry No. 1.00 0.51 0.59 -0.65 0.83 0.91 -0.84 -0.28 0.25 Berry Wt. 1.00 0.23 -0.96 0.72 0.39 -0.56 -0.86 -0.14 LA:F 1.00 -0.34 0.77 0.77 -0.20 0.10 0.72 WUE 1.00 -0.83 -0.49 0.58 0.75 0.13 TSS 1.00 0.82 -0.57 -0.34 0.36 pH 1.00 -0.76 -0.08 0.63 TA 1.00 0.45 -0.14 Antho. 1.00 0.45 Phenolics 1.00

b. Yield and Fruit Composition Parameters 2001-02 Yield Bunch Bunch Berry Berry LA:F WUE TSS pH TA Antho. Phenol. r (kg) No. Wt. (g) No. Wt. (g) (cm2/g) (t/ML) (°Brix) (g/L) (mg/g) (mg/g) Yield 1.00 0.47 -0.29 -0.49 0.12 -0.85 -0.36 0.43 0.42 0.00 -0.38 0.07 Bunch No. 1.00 -0.90 -0.94 -0.77 -0.02 0.36 0.01 0.48 -0.35 0.54 0.77 Bunch Wt. 1.00 0.96 0.90 -0.02 -0.51 0.28 -0.08 0.08 0.08 -0.69 Berry No. 1.00 0.76 0.17 -0.39 0.16 -0.25 0.02 -0.60 -0.75 Berry Wt. 1.00 -0.45 -0.65 0.37 0.02 0.27 -0.82 -0.69 LA:F 1.00 0.55 -0.35 -0.05 -0.38 0.58 0.31 WUE 1.00 -0.87 -0.26 0.13 0.76 0.73 TSS 1.00 0.62 -0.56 -0.59 -0.45 pH 1.00 -0.56 -0.03 0.36 TA 1.00 0.07 -0.03 Antho. 1.00 0.82 Phenolics 1.00

c. Yield and Fruit Composition Parameters 2002-03 Yield Bunch Bunch Berry Berry LA:F WUE TSS pH TA Antho. Phenol. r (kg) No. Wt. (g) No. Wt. (g) (cm2/g) (t/ML) (°Brix) (g/L) (mg/g) (mg/g) Yield 1.00 0.32 0.10 -0.11 0.24 -0.83 -0.52 -0.16 -0.77 0.62 -0.22 -0.11 Bunch No. 1.00 -0.87 -0.91 -0.82 0.02 0.47 -0.58 -0.72 0.82 0.76 0.66 Bunch Wt. 1.00 0.96 0.94 -0.36 -0.64 0.65 0.47 -0.63 -0.79 -0.68 Berry No. 1.00 0.82 -0.09 -0.49 0.79 0.66 -0.82 -0.73 -0.73 Berry Wt. 1.00 -0.58 -0.77 0.38 0.26 -0.41 -0.85 -0.61 LA:F 1.00 0.68 0.25 0.58 -0.49 0.41 0.09 WUE 1.00 -0.01 0.25 0.03 0.91 0.73 TSS 1.00 0.70 -0.79 -0.30 -0.51 pH 1.00 -0.94 -0.13 -0.23 TA 1.00 0.41 0.55 Antho. 1.00 0.88 Phenolics 1.00

Chapter 4: Irrigation and pruning effects on fruit and wine composition 78

Seasons 2001-2003 Treatment means were also pooled for the three seasons to assess the relationships between yield components and fruit composition parameters (Table 4.16). Yield was strongly correlated with bunch number and berry number. Treatments with highest bunch numbers with fewer berries per bunch, such as MIN, produced the highest yield. High bunch number was associated with low average bunch weight, low berry number per bunch and low average berry weight. Improved WUE was associated with high bunch numbers and low bunch and berry weights. Treatments that produced smaller berries as a result of higher crop level (MIN) or reduced irrigation input (PRD) had lower pH levels and higher anthocyanin and phenolic concentrations in berry juice. TSS content was correlated with the other fruit composition parameters. Berries with high TSS invariably had low TA and high pH, high anthocyanin concentrations and high phenolic concentrations.

Table 4.16: Correlation coefficients of mean yield components [yield, bunch number.vine-1 (Bunch No.), bunch weight (Bunch Wt.), berry number.bunch-1 (Berry No.), leaf area:fruit (LA:F), water use efficiency (WUE)] and fruit composition parameters[total soluble solids (TSS), pH, titratable acidity (TA), anthocyanin concentration (Antho.) and phenolic concentration (Phenol)] for seasons 2000-03, n=18. Bold script represents significance at 5% level.

Yield and Fruit Composition Parameters 2000-03 Yield Bunch Bunch Berry Berry LA:F WUE TSS pH TA Antho. Phenol. r (kg) No. Wt. (g) No. Wt. (g) (cm2/g) (t/ML) (°Brix) (g/L) (mg/g) (mg/g) Yield (kg) 1.00 0.75 -0.43 -0.56 -0.06 -0.64 0.42 -0.48 0.14 0.01 0.00 -0.07 Bunch No. 1.00 -0.83 -0.91 -0.61 -0.34 0.48 -0.19 -0.22 0.08 0.17 0.23 Bunch Wt. (g) 1.00 0.95 0.71 0.27 -0.60 0.31 0.15 -0.06 -0.11 -0.16 Berry No. 1.00 0.47 0.27 -0.41 0.42 -0.04 0.08 0.10 0.02 Berry Wt. (g) 1.00 0.11 -0.77 -0.21 0.56 -0.43 -0.62 -0.65 LA:F 1.00 -0.06 -0.04 0.50 -0.44 -0.32 -0.25 WUE 1.00 -0.16 -0.40 0.36 0.38 0.43 TSS (°Brix) 1.00 -0.64 0.63 0.80 0.79 pH 1.00 -0.91 -0.94 -0.87 TA 1.00 0.88 0.84 Antho. 1.00 0.98 (mg/g) Phenolics 1.00 (mg/g)

Chapter 4: Irrigation and pruning effects on fruit and wine composition 79

4.5 YIELD, FRUIT AND WINE COMPOSITION CORRELATIONS

Season 2000-01 During 2000-01, treatments with high yield and bunch numbers produced more acidic wine (i.e. wine with a lower pH) as a response to more acidic juice. As a result, less tartaric acid adjustment was required during the winemaking phase (Table 4.17a). The amount of tartaric acid added increased with bunch weight, berry weight and ripeness of the berries (TSS). Small berries produced wine with higher density and ionised anthocyanins. Reductions in yield by PRD were associated with increased total anthocyanins in the wine.

Season 2001-02 The ratio of LA:F was highly correlated with wine quality in season 2001-02 (Table 4.17b). High LA:F was associated with high wine density, total anthocyanins, ionised anthocyanins, phenolics and degree of ionisation of anthocyanins. Total anthocyanin content was negatively correlated with yield. Again, the reduction in yield by PRD was positively correlated with colour and flavour components of the wine produced. Larger berries were associated with higher wine pH.

Season 2002-03 Yield correlated with the amount of tartaric acid added to the wine in season 2002-03. The higher the yield, the less tartaric acid that was required to increase TA of the wine (Table 4.17c). Berry anthocyanin and phenolics concentrations were positively correlated with wine pH, probably as a response to bunch number and berry size. Pruning treatments that produced more bunches consisting of smaller berries had higher anthocyanin and phenolic concentrations in the berries and also higher wine pH. Wine total anthocyanins and phenolics were greater on low yielding treatments, e.g. PRD. PRD decreased berry size and ultimately yield in 2002-03 with all pruning systems. As a result, colour and phenolics of the fruit was intensified. The reduced volume of water applied by PRD also improved WUE. Subsequently, WUE was positively correlated with total anthocyanins, ionised anthocyanins and total phenolics of the wine. Wine ionised anthocyanin concentration was correlated with berry anthocyanin and total phenolic concentrations. High degrees of ionisation of anthocyanins were associated with low berry numbers per bunch, low TSS and high juice TA, as found for MIN vines.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 80

Table 4.17: Correlation coefficients of mean yield components [yield, bunch number.vine-1 (Bunch No.), bunch weight (Bunch Wt.), berry number.bunch-1 (Berry No.), leaf area:fruit (LA:F), water use efficiency (WUE) and fruit parameters [total soluble solids (TSS), pH, titratable acidity (TA) anthocyanin concentration (Antho.) and phenolic concentration (Phenol.) and wine parameters [tartaric acid addition (TA added), wine pH, wine titratable acidity (TA), colour density, hue, total anthocyanin concentration (Total Antho.), ionised anthocyanin (Ionised Antho.), phenolic concentration (Phenolics), degree of ionisation (alpha)] for season a. 2000-01, b. 2001-02 and 3. 2002-03, n=6. Bold script represents significance at 5% level. r Yield and Fruit Composition Parameters 2000-01 Wine Yield Bunch Bunch Berry Berry LA:F WUE TSS pH TA Antho. Phenol. Parameters kg No. Wt. g No. Wt. g cm2/g t/ML °Brix g/L mg/g mg/g TA added -0.67 -0.78 0.91 0.78 0.86 0.65 -0.91 0.96 0.76 -0.65 -0.54 0.27 Wine pH -0.82 -0.92 0.74 0.79 0.35 0.44 -0.37 0.58 0.90 -0.90 -0.14 0.55 Wine TA 0.03 -0.09 0.41 0.16 0.84 0.31 -0.78 0.57 0.09 -0.05 -0.74 -0.15 Density -0.07 0.32 -0.43 -0.24 -0.86 0.19 0.73 -0.33 -0.14 0.56 0.87 0.26 Hue -0.69 -0.80 0.67 0.63 0.53 0.43 -0.49 0.62 0.80 -0.80 -0.26 0.60 Total -0.85 -0.58 0.39 0.53 -0.22 0.67 0.04 0.48 0.71 -0.26 0.62 0.75 Antho. Ionised 0.16 0.54 -0.60 -0.45 -0.87 0.05 0.78 -0.48 -0.37 0.74 0.81 0.12 Antho. Phenolics -0.76 -0.44 0.32 0.47 -0.29 0.66 0.07 0.44 0.58 -0.08 0.67 0.60 Alpha 0.67 0.86 -0.79 -0.73 -0.68 -0.37 0.69 -0.73 -0.79 0.86 0.38 -0.39 r b. Yield and Fruit Composition Parameters 2001-02 Wine Yield Bunch Bunch Berry Berry LA:F WUE TSS pH TA Antho. Phenol. Parameters kg No. Wt. g No. Wt. g cm2/g t/ML °Brix g/L mg/g mg/g TA added -0.41 -0.61 0.65 0.71 0.43 0.20 -0.68 0.59 -0.01 -0.34 -0.45 -0.75 Wine pH 0.29 -0.53 0.72 0.51 0.91 -0.53 -0.74 0.52 0.31 0.26 -0.67 -0.48 Wine TA -0.24 0.51 -0.45 -0.32 -0.66 0.65 0.76 -0.38 0.14 -0.54 0.51 0.61 Density -0.73 0.19 -0.29 -0.06 -0.69 0.93 0.60 -0.37 -0.16 -0.45 0.64 0.33 Hue 0.80 0.02 0.28 0.03 0.62 -0.76 -0.54 0.54 0.55 0.02 -0.65 -0.15 Total -0.92 -0.14 -0.05 0.19 -0.47 0.91 0.43 -0.38 -0.36 -0.21 0.56 0.09 Antho. Ionised -0.76 0.17 -0.32 -0.08 -0.71 0.91 0.65 -0.46 -0.26 -0.33 0.69 0.34 Antho. Phenolics -0.65 0.27 -0.27 -0.08 -0.65 0.92 0.46 -0.14 0.12 -0.62 0.61 0.37 Alpha -0.56 0.41 -0.52 -0.30 -0.83 0.84 0.77 -0.49 -0.13 -0.39 0.76 0.54 r c. Yield and Fruit Composition Parameters 2002-03 Wine Yield Bunch Bunch Berry Berry LA:F WUE TSS pH TA Antho. Phenol. Parameters kg No. Wt. g No. Wt. g cm2/g t/ML °Brix g/L mg/g mg/g TA added -0.84 -0.42 -0.01 0.09 -0.01 0.52 0.44 0.06 0.70 -0.48 0.14 0.25 Wine pH 0.11 0.76 -0.57 -0.64 -0.59 -0.01 0.57 -0.42 -0.36 0.57 0.84 0.84 Wine TA -0.21 0.46 -0.75 -0.72 -0.61 0.30 0.55 -0.47 -0.21 0.41 0.51 0.55 Density 0.50 -0.19 0.41 0.25 0.51 -0.63 -0.16 0.27 -0.05 0.11 -0.15 0.10 Hue 0.10 -0.19 0.49 0.55 0.21 0.04 0.06 0.75 0.37 -0.44 0.00 -0.20 Total -0.84 -0.08 -0.27 -0.05 -0.43 0.85 0.82 0.30 0.70 -0.48 0.50 0.35 Antho. Ionised -0.33 0.55 -0.77 -0.75 -0.69 0.35 0.81 -0.46 -0.09 0.39 0.81 0.86 Antho. Phenolics -0.82 0.06 -0.39 -0.22 -0.50 0.76 0.88 0.08 0.58 -0.30 0.65 0.57 Alpha 0.30 0.74 -0.75 -0.90 -0.50 -0.27 0.26 -0.85 -0.73 0.90 0.53 0.72

Chapter 4: Irrigation and pruning effects on fruit and wine composition 81

Seasons 2001-2003 Yield was not significantly correlated with wine quality parameters over the three seasons. However, berry weight was strongly correlated with most of the measured wine quality parameters (Table 4.18). Small berries resulted in wine with lower pH and hue, higher wine density, high total and ionised anthocyanins and higher total phenolic concentration. Treatments with higher WUE improved wine colour (ionised anthocyanins), density, hue and reduced pH. Fruit composition was strongly associated with wine quality over the three seasons. The TSS of berry juice was positively correlated with total anthocyanins and phenolics and negatively correlated with α. Thus more mature berries at harvest were correlated with higher total anthocyanin and phenolic concentrations in the wine. Fruit pH correlated with all wine components with the exception of wine hue. A negative relationship was found between fruit pH and wine TA, total anthocyanins, ionised anthocyanins and phenolics. In comparison, a positive relationship was found between fruit TA and wine TA, total anthocyanins and phenolics. Anthocyanin and phenolics concentration in the berries was positively correlated with wine TA and negatively correlated with the amount of TA adjustment required by the must. Berry anthocyanin and total phenolic concentration also correlated highly with total anthocyanins, ionised anthocyanins and total phenolics in the wine.

Table 4.18: Correlation coefficients of mean yield components [yield, bunch number.vine-1 (Bunch No.), bunch weight (Bunch Wt.), berry number.bunch-1 (Berry No.), leaf area:fruit (LA:F), water use efficiency (WUE) and fruit parameters [total soluble solids (TSS), pH, titratable acidity (TA) anthocyanin concentration (Antho.) and phenolic concentration (Phenol.) and wine parameters [tartaric acid addition (TA added), wine pH, wine titratable acidity (TA), colour density, hue, total anthocyanin concentration (Total Antho.), ionised anthocyanin (Ionised Antho.), phenolic concentration (Phenolics), degree of ionisation (alpha)] for seasons 2000-03, n=18. Bold script represents significance at 5% level. r Yield and Fruit Composition Parameters Wine Yield Bunch Bunch Berry Berry LA: F WUE °Brix pH TA Antho. Phenol. Parameters No. Wt. g No. Wt. g mg/g mg/g TA added -0.30 -0.48 0.46 0.32 0.70 0.48 -0.67 -0.43 0.79 -0.80 -0.81 -0.85 Wine pH -0.40 -0.71 0.65 0.50 0.68 0.22 -0.58 -0.03 0.58 -0.38 -0.44 -0.40 Wine TA -0.03 0.22 -0.06 0.07 -0.43 -0.02 0.23 0.75 -0.71 0.58 0.79 0.81 Density -0.33 0.25 -0.30 -0.02 -0.81 0.25 0.68 0.28 -0.51 0.33 0.57 0.55 Hue 0.14 -0.27 0.39 0.17 0.58 -0.14 -0.53 0.12 0.35 -0.20 -0.25 -0.17 Total -0.28 -0.01 0.01 0.26 -0.63 -0.08 0.38 0.79 -0.86 0.78 0.95 0.91 Antho. g/L Ionised -0.27 0.31 -0.37 -0.09 -0.81 0.23 0.72 0.19 -0.52 0.35 0.54 0.52 Antho. g/L Phenolics -0.27 0.10 -0.05 0.20 -0.68 -0.03 0.40 0.79 -0.80 0.69 0.93 0.91 g/L Alpha 0.20 0.39 -0.43 -0.43 -0.04 0.25 0.27 -0.85 0.51 -0.59 -0.64 -0.63

Chapter 4: Irrigation and pruning effects on fruit and wine composition 82

4.6 DISCUSSION

4.7.1 Pruning Effects on Fruit and Wine Composition

Winter pruning affected fruit yield, as a result of the number of nodes retained per vine at pruning, but fruit and wine quality were largely unaffected. A significant 3.8-fold increase in bunch number per vine and 1.3-fold increase in yield were observed through the retention of nodes by light pruning (MIN) compared to severe pruning (SPUR). However, the effects of pruning level on fruit composition (i.e. sugar concentration, pH, TA, colour and flavour concentrations) and wine composition (i.e. pH, TA, colour density, hue, anthocyanin and phenolic concentrations) were minimal. Pruning had a minor effect on sugar concentration (TSS) in the first experimental season when vines were harvested by calendar date; berries from lightly pruned treatments, MIN and MECH, had slightly lower sugar concentrations than berries from SPUR. Since the rate of sugar accumulation was not significantly different between pruning treatments (refer to section 3.4.2), this suggests that the initiation of veraison was delayed by MIN and MECH. Also, the partitioning of carbohydrates between sinks (berries) was greater for MIN and MECH compared to SPUR since crop levels were higher. Reynolds and Wardle (2001) also observed delayed fruit maturity and reduced °Brix in berries from minimally pruned Chancellor vines and attributed this to greater crop stress due to higher crop levels.

Severe pruning and its associated larger berry size and higher canopy density influenced the pH and TA of berry juice: SPUR produced berries with higher pH and lower TA compared to MIN and MECH. This suggests that larger berries had a relatively greater solvent (water) to solute (sugars and organic acids) ratio, thus greater dilution of solutes per berry. Fruit from dense, shaded canopies, as found on SPUR vines, have relatively higher K+ juice concentrations and pH (Botting et al. 1996; Smart 1992). The degree of increase in berry pH by SPUR compared to MIN (0.08 units) and MECH (0.06 units) was comparable to studies on ‘Sunbelt” grapes (Striegler et al. 2002) but was unlikely to be of viticultural significance. The pruning effects on anthocyanin and phenolic concentrations and content were also associated with berry size. The small berries produced on MIN vines had higher concentrations of secondary metabolites (anthocyanins and phenolics) than those produced on SPUR and MECH vines, which may be attributed to an increased surface area to volume ratio. Increased anthocyanin concentration of berries on MIN vines may also be linked to

Chapter 4: Irrigation and pruning effects on fruit and wine composition 83 improved microclimate and subsequently, greater bunch exposure (Sommer and Clingeleffer 1993). Levels of anthocyanin and phenolics have been positively correlated with the degree of bunch exposure for grapevines, particularly on field-grown cv. Cabernet Sauvignon (Carbonneau 1985; Crippen and Morrison 1986; Rojas-Lara and Morrison 1989; Gao and Cahoon 1994). The results from this study agree with research on cv. Shiraz that showed high levels of colour and phenolics were associated with small bunches consisting of smaller berries and higher levels of ripeness (Holzapfel et al. 1999). However, Reynolds and Wardle (2001) found minimally pruned Chancellor grapevines produced berries with lower anthocyanin concentration than hand-pruned vines. When secondary metabolite content was determined on a per berry basis, SPUR had the highest levels because of the larger berry size. Iland et al. (1996) showed some viticultural management treatments that promote anthocyanin and phenolic accumulation per berry often increase berry size, thus lower the surface area to volume ratio and concentration of secondary metabolites.

As expected, the minor pruning effect on wine composition can be directly associated with those observed on fruit composition. Wine pH and TA were largely unaffected by the level of pruning and wine spectral parameters reflected the berry anthocyanin and phenolic concentrations. Wine colour density and colour hue results are comparable to studies on cv. Shiraz and cv. Cabernet Sauvignon grown in Coonawarra, South Australia (Clingeleffer 1992). However, wine ionised anthocyanin concentration and the percentage of anthocyanins present in the coloured form (α) were slightly improved by MIN and can be attributed to the higher anthocyanin concentration in the smaller berries on MIN vines. Therefore, light-pruning levels produced higher yields and improved wine spectral parameters (colour and brightness).

4.7.2 Irrigation Effects on Fruit and Wine Composition

Interestingly, PRD had a minor effect on fruit composition but strong effect on wine colour and flavour parameters. Fruit pH, TA, anthocyanin and phenolic concentration, and phenolic content were not significantly affected by the irrigation method applied. However, anthocyanin concentration in the fruit tended to increase with PRD irrigation. This increase in anthocyanin concentration may have been due to the increase in exocarp to juice ratio of smaller berry size associated with PRD. This explanation is supported by the fact that

Chapter 4: Irrigation and pruning effects on fruit and wine composition 84 anthocyanin content was lower in berries from PRD irrigated vines, thus it was a concentrating effect rather than an actual increase in anthocyanins per berry. Bravdo et al. (1985a), Hepner et al. (1985), Matthews et al. (1990) and Iland (2000) have shown induced water stress results in decreased berry size, increased berry colour and phenolic concentration and/or wine colour density and total phenolics. Also, the higher concentration of anthocyanins in the berries on PRD irrigated vines may be partly associated with improved bunch exposure, as a result of reduced leaf area (refer to section 3.3.2).

PRD positively affected wine quality, in particular the spectral parameters associated with colour and flavour. Wine density, total anthocyanin concentration, ionised anthocyanin and total phenolic concentration of the wine were all improved by PRD. The slightly raised anthocyanin levels in the smaller berries produced by PRD and increased exocarp tissue concentration in fermenting must may partially account for the higher concentration in the wine. However, similar trends in spectral parameters were not observed in wine produced from the small berries of MIN vines. This suggest small berry size alone, is not a precursor to improved wine spectral parameters. The question arises as to why large differences in spectral properties of wine produced from PRD grapes were observed when no significant irrigation differences in fruit analysis occurred. A previous study on cv. Cabernet Sauvignon has shown irrigation treatment effects on concentration of anthocyanin and proanthocyanin in wines were about 50% of those in berries (Kennedy et al. 2002). The difference between wine spectral parameters and fruit colour and phenolic concentrations may be due to the poor extraction of anthocyanins and phenolics from berries, co-pigmentation of anthocyanins or synthesis of anthocyanin compounds in response to PRD. The extraction and consequently, determination of anthocyanin and phenolic concentration in berry exocarp and mesocarp can be influenced by the degree and method of maceration of berries, storage time, defrosting temperature and exposure to oxygen (P. Petrie and N. Cooley, pers. comm.). Singleton (1972) found extraction efficiency of phenolic compounds from grape exocarp decreases as juice volume of berries decreases (i.e. small berries). However, no differences between fruit analysis and wine spectral properties were found for pruning treatments that also produced small berries (MIN and MECH). This suggest colour and phenolic extraction from berry exocarp was reflective of concentrations found in the wine. Alternatively, co- pigmentation of anthocyanins in wine produced from PRD vines may have induced greater absorbance of coloured pigments (hyperchromic shift) and possibly, a shift in the wavelength of maximum absorbance (bathochromic shift), producing blue-purple tones in

Chapter 4: Irrigation and pruning effects on fruit and wine composition 85 the wine. Co-pigmentation is due to molecular associations between colour pigments and co-pigment cofactors (usually non-coloured organic compounds) in solution. Co- pigmentation accounts for 30% to 50% of the colour of young red wines (Neri and Boulton 1996) and can be influenced by variety and cultural practices (Boulton 2001). Wines made from different cultivars have specific patterns of cofactors and cofactor concentrations (Vaadia 1997). Levengood (1996) also found substantial variation in co-pigmentation levels in Cabernet Sauvignon wine from different vineyard sites subjected to various cultural practices. This study shows the potential of cultural practices that influence berry size and flavonoid synthesis, such as irrigation and crop level, to encourage co-pigmentation of the wine. Further comprehensive studies are required to investigate the effects of cultural practices on the concentration of co-pigmentation in fruit and wine.

Unfortunately, spectrophotometric assays of anthocyanin content in red wine do not account for co-pigmentation (Boulton 2001). Colour measurements of wine were conducted using the methodology of Somers and Evans (1977) in this experiment, which involved bleaching and ionisation of anthocyanins to estimate total anthocyanin content, degree of ionisation of anthocyanins and chemical age (ratio between monomeric anthocyanins and polymeric pigments). Bleaching in the assay for free sulphur dioxide may result in an overestimated anthocyanin content due to enhanced colour released from the co-pigmented forms compared to other colour determination techniques (HPLC) (Bakker et al. 1986). Somers and Evans (1977) acknowledged the existence of co-pigmentation, however its contribution is omitted from anthocyanin readings in several equations. Further investigation into anthocyanin separation, identification and quantification is required using HPLC, NMR or circular dichroism (CD) to determine the contribution of co-pigmentation to total anthocyanin colour and concentrations and ratios of the major anthocyanin pigments (i.e. delphinidin, cyanidin, petunidin, peonidin and malvidin), as described by Esteban et al. (2001).

4.7.3 Irrigation and Pruning Effects on Fruit and Wine Composition

The combination of irrigation technique with pruning level had minor effects on fruit and wine analysis. SPUR and MECH pruning treatments were more sensitive to PRD irrigation than MIN, in terms of fruit sugar and secondary metabolite concentrations, and wine spectral

Chapter 4: Irrigation and pruning effects on fruit and wine composition 86 parameters. This was associated with the greater reduction in berry size by PRD on SPUR and MECH vines relative to MIN vines. Fruit and wine pH and TA levels were not significantly altered by the interaction between irrigation and pruning treatments, which agrees with previous bi-factorial studies (Freeman et al. 1980).

Fruit maturity, as determined by TSS concentration chronologically, was slightly reduced by high crop levels and PRD irrigation, which may have been due to increased partitioning of carbohydrates between more sinks (bunches) and an insufficient rate of carbon fixation by photosynthesis over time to allow fruit to ripen. Bravdo et al. (1985a, 1986) have shown high crop levels can delay ripening of fruit as a result of greater sink demands. However, previous irrigation studies of vines have also shown a reduction in TSS concentration by irrigation when berry growth is faster than the increase in sugar accumulation in the berry and dilution occurs (Freeman et al. 1980; McCarthy et al. 1983; Smart and Coombe 1983; Bravdo et al. 1985b; van Zyl 1984; Bravdo and Hepner 1987).

Anthocyanin and phenolic concentrations in the berries were much greater in 2001-02 for all combined treatments than other seasons. This may be due to slow accumulation under mild summer climatic conditions, as shown by the strong negative correlation with mean daily December to February temperatures. Kliewer (1970) and Hale and Buttrose (1974) showed similar negative correlations between anthocyanin concentration in grapes and temperature. No significant treatment effects were found for anthocyanin concentration of the berries but phenolic concentration was improved by the combination of PRD and MIN. The increased exocarp to juice ratio of small berries may explain the increased levels of phenolics. Anthocyanin and phenolic content were dependent on berry weight: treatments that reduced berry size (i.e. PRD and MIN) had less secondary metabolites per berry. However, light- pruning treatments (MIN and MECH) produced more bunches and consequently, more berries per vine than SPUR. Thus secondary metabolite content was greater for light pruning treatments when calculated on a per vine basis (data not shown).

When the effects of irrigation and pruning on wine quality were evaluated, it was apparent that like fruit quality, the treatments had very minimal effect. Wine pH, TA, colour density, ionised anthocyanin concentration and percentage of anthocyanins present in the coloured form (α) were not significantly influenced by the interactions between irrigation and

Chapter 4: Irrigation and pruning effects on fruit and wine composition 87 pruning, which agrees with previous findings (Freeman et al. 1980; Bravdo et al. 1985b). The lack of significant interactions may be attributed to the additive effect of PRD combined with MIN to improve wine quality, in terms of colour and flavour properties. Wine hue and total anthocyanin concentrations of the PRD + MIN were better than the other combined treatments. However, PRD improved both the brightness and total anthocyanin concentration of SPUR and MECH treatments compared to SD irrigated vines. This may be the result of co-pigmentation in the wine induced by PRD, as explained in section 4.7.2. Improvement to canopy microclimate due to decreased shoot vigour by PRD (Dry 1997), in conjunction with increased bunch exposure by MIN (Sommer and Clingeleffer 1993), may have also contributed to greater colour and brightness in these wines.

Sensory analysis of the wine produced from each treatment was not included because of the difficultly involved with quantification due to the subjective nature of the definition of wine quality. An objective measure of wine quality is required before quantification can occur.

4.7.4 Influence of Yield on Fruit and Wine Composition

Yield has the potential to affect fruit and wine quality directly by crop level, vine balance and berry size, and indirectly, by rate of ripening and bunch exposure, as reviewed by Jackson and Lombard (1993). Bunch number was positively correlated with yield and negatively correlated with each of its components (i.e. bunch weight, berry weight and berry number). However, bunch number or yield was not significantly correlated with fruit or wine quality. These findings support other studies conducted on cv. Shiraz, which have shown poor correlation between yield and fruit quality (Clingeleffer 1992, 1993; Johnstone et al. 1996; Gray et al. 1997; Holzapfel et al. 1999). Gray et al. (1997) found no relationship between yield and the value of the commercial wine, however berry colour and phenolic concentration were negatively correlated with berry size. This suggests the photosynthetic capacity and subsequent carbohydrate production of cv. Shiraz is sufficient to adequately ripen and maintain fruit quality at high crop levels, as achieved on minimal pruned vines. This has important economic implications for grapegrowers, since cv. Shiraz could sustain greater tonnages of fruit with no apparent wine quality penalties and, in the case of PRD + MIN treatments, possibly improve colour concentration.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 88

4.7.5 Influence of Berry Size and TSS on Fruit and Wine Composition

The correlation between berry size and both fruit and wine composition was more significant than that found for yield. Berry size alters the surface to volume ratio, which consequently affects the balance of extractable exocarp contents (i.e. secondary metabolites) to mesocarp contents (i.e. water and sugars) (Dry et al. 1999). The size of a berry also determines the degree of dilution of solutes (sugars and organic acids) to solvent (water) in the mesocarp of the grape. Therefore, small berries produced by PRD and MIN were expected to contain higher concentrations of sugar and organic acids. Berry size also correlated positively with wine quality: small berries were associated with wine that was more acidic, brighter and had higher colour density and greater anthocyanin (total and ionised) and phenolic concentration. However, the integrated results showed that PRD had a greater influence on wine spectral properties than light pruning levels, which suggests small berry size is an associated response with improved spectral characteristics of wine rather than a causative effect, as previously reported (Clingeleffer, 1983). As an alternative, PRD may have induced a biochemical reaction in the wine, such as co-pigmentation, to increase absorbance of colour pigments and shift maximum absorbance, as discussed in section 4.7.2.

The concentration of sugar (TSS) in berries was strongly correlated with other fruit compositional parameters: TA, pH and anthocyanin and phenolic concentrations. This suggests sugar maturity is important to the production of high quality fruit. Holzapfel et al. (1999) showed similar interactions between TSS and anthocyanin concentrations for cv. Shiraz grown in a hot climate. Gladstones (1992) suggested that the duration of ripening might also influence the compositional balance of grapes through the loss of aromatics. Botting et al. (1996) found lower concentrations of anthocyanins and glycosyl-glucose [G- G] in berries from vineyards with delayed maturation. Treatments that increase the rate of ripening may have a greater photosynthetic capacity to synthesise more carbohydrates and secondary metabolites. In addition, berry anthocyanin and phenolic concentrations correlated well with total anthocyanins, ionised anthocyanins and total phenolics in the wine. This is important to Australia’s wine industry, as several leading wine companies have recently included berry anthocyanin concentration as a payment parameter, as well as TSS and yield.

Chapter 4: Irrigation and pruning effects on fruit and wine composition 89

4.7 CONCLUSIONS

a) The combined effects of irrigation and pruning treatments on fruit and wine composition of cv. Shiraz were unlikely to be of viticultural significance. SPUR and MECH vines were more responsive to PRD than MIN, in terms of fruit sugar and secondary metabolite concentrations, and wine spectral parameters. This was associated with the relative berry size reduction by PRD on SPUR and MECH compared to MIN.

b) Light pruning (MIN and MECH) resulted in minor increases in anthocyanin and phenolic concentrations of fruit and wine compared to severe pruning (SPUR).

c) PRD did not significantly affect fruit composition, however PRD had a strong effect on wine spectral properties: density, total anthocyanin concentration, ionised anthocyanin and total phenolic concentration relative to SD. This may be attributed to co- pigmentation in the wine.

d) Bunch number or yield did not significantly correlate with fruit or wine quality, however berry size was strongly correlated with fruit and wine quality.

e) Small berries were associated with low pH and high anthocyanin and phenolic concentrations and wine that was more acidic, brighter and had high colour density and anthocyanin (total and ionised) and phenolic concentrations.

f) High berry TSS concentration was also correlated with improved fruit TA, pH and anthocyanin and phenolic concentrations. This indicates high sugar maturity of berries is important in the production of high quality fruit.

g) The hypothesis that “irrigation and pruning effects on fruit and wine composition will be minimal because vegetative and reproductive growth was balanced” is accepted. All treatments reached optimal sugar maturity, which suggests vine photosynthetic capacity

Chapter 4: Irrigation and pruning effects on fruit and wine composition 90

was sufficient to ripen fruit, despite large differences in bunch number and yield between treatments.

h) The hypothesis that “small berry size will improve fruit and wine composition parameters, as a result of an increase in exocarp to juice ratio and concentration of solutes in the mesocarp” is rejected. Integrated data clearly showed that PRD produced wine spectral changes in association with small berries (possibly due to co- pigmentation). However small berries produced on MIN vines did not replicate these improvements in wine spectral properties.

Chapter 5: Physiological response to irrigation and pruning treatments 91

5 PHYSIOLOGICAL RESPONSE TO IRRIGATION AND PRUNING TREATMENTS

5.1 INTRODUCTION AND EXPERIMENTAL AIMS

Sugar accumulation in the berry is an important determinant of fruit quality in cv. Shiraz and is affected by the photosynthetic capacity of the grapevine. Photosynthesis is controlled by the degree of opening of stomatal aperture and subsequent uptake of CO2. Photosynthesis and other gas exchange parameters (stomatal conductance, transpiration and internal leaf

CO2 concentration) are affected by environmental and internal factors, such as light intensity, vapour pressure deficit, plant water status, cultivar, carboxylation efficiency, crop load and water deficits (Chaves and Rodrigues 1987; Schultz 2003). Stomatal conductance is sensitive to changes in plant water status, as determined by leaf or xylem water potential (Lui et al. 1978; Naor et al. 1994). Osmoregulation of grapevine leaves is an alternate mechanism to maintain high water status and high rates of leaf gas exchange, associated with decreasing plant available water (PAW) (Düring 1984; Rodrigues et al. 1993; Schultz and Matthews 1993). The degree of solute accumulation in leaves can be dependent on climatic conditions and grapevine variety (Düring and Loveys 1982) and may affect the stomatal conductance-leaf water potential relationship.

Irrigation can influence both soil water content and plant water status and can result in drought conditions if insufficient. As a drought response, water-stressed vines close stomata to regulate transpiration so that sufficient carbon is gained while plant water status is maintained (Kriedemann and Smart 1971; Naor et al. 1994; Schultz 2003). Long-term or intensive water deficits can reduce photosynthetic capacity of grapevines, as a consequence of decreased ribulose biphosphate synthesis, Rubsico activity and carboxylation efficiency (Escalona et al. 1999). Several studies have indicated an association between abscisic acid (ABA) and the regulation of stomatal conductance in response to decreasing soil water content (Zhang and Davies 1989a, 1989b, 1990, 1991; Gowing et al. 1990). Reduction of water-stress by stomatal closure has the potential to affect carbohydrate production and sugar accumulation in berries. Partial rootzone drying (PRD) can induce a drought response by reducing stomatal conductance in a variety of crops, by drying half the rootzone but maintains plant water status by irrigating the alternate half of the rootzone (Dry and Loveys 1998; Kang et al. 2001; Liu et al. 2001). Since plant water status is maintained by PRD,

Chapter 5: Physiological response to irrigation and pruning treatments 92 typical physical drought stress responses, such as reduced photosynthesis, crop load loss and delayed sugar accumulation, should be minimised. Recent studies by dos Santos et al. (2003) and de Souza et al. (2003) showed PRD irrigation (50% of crop evapotranspiration) of field-grown grapevines generally maintained plant water status close to a fully irrigated control treatment (100% crop evapotranspiration) but increased water use efficiency, due to decreased stomatal conductance and maintained carbon assimilation and crop production.

The presence of fruit activates photosynthesis and stomatal conductance in grapevines (Kriedemann 1971; Chaves 1984; Downton et al. 1987). Manipulation of crop level by pruning (Sommer and Clingeleffer 1993; Poni et al. 2000; Intrieri et al. 2001) and fruit thinning (Wünsche et al. 2000; Petrie 2002) can lead to an increase in photosynthetic rate as source:sink ratios decrease. However the photosynthetic response of grapevine leaves to crop management does not appear to be consistent in the literature. Poni et al. (1994) found no significant effects of crop level on gas exchange or plant water use on field-grown Concord vines and Chaumont et al. (1994) reported no photosynthetic response to changes in crop load on potted vines. Crop manipulation by winter pruning affects canopy structure and total leaf area, subsequently altering the source:sink ratio (Sommer and Clingeleffer 1993). If grapevines are near the critical source:sink ratio because of high crop levels, then the vines may be more susceptible to external stress, such as drought, which can reduce leaf and vine photosynthesis (Düring and Loveys 1982; van Zyl 1987; Downton et al. 1988). Alternatively, when crop level is low due to fruit removal or severe pruning, photosynthesis can be increased by the demands of other sinks, such as rapidly growing shoot tips (Edson and Howell 1993).

Grapevine photosynthetic capacity can be determined on a single-leaf or whole-canopy scale. Single-leaf photosynthetic responses do not always reflect those of the whole canopy, as leaf responses vary with age, position and orientation (Intrieri et al. 1997). However, assessments of single-leaf photosynthesis have been used extensively for comparative studies of treatment effects (Naor and Wample 1994; Poni et al. 1994; Gomez-del-Campo et al. 2002). The environmental and internal factors that affect grapevine stomatal conductance and photosynthetic capacity also affect the ratio of intercellular CO2 (Ci) and atmospheric CO2 (Ca) partial pressures. In C3 plants, the ratio between Ci and Ca is related to discrimination against naturally occurring stable isotope 13C (Farquhar et al. 1982).

Chapter 5: Physiological response to irrigation and pruning treatments 93

Carbon isotope discrimination has been shown to vary between genotypes and water- stressed grapevines in pot and field trials (Iacono et al. 1993; Gibberd et al. 2002). Differences in carbon isotope discrimination can arise from changes in the balance between stomatal conductance and photosynthetic capacity. If a change in carbon isotope discrimination were due to a variation in photosynthetic capacity per unit area of leaf, a negative relationship with yield would be expected, as shown in peanuts by Wright et al. (1994). However, positive relationships between carbon isotope discrimination and yield have been found for cereal crops (Condon et al. 1987). This indicates that a variation in stomatal conductance may influence carbon isotope discrimination. As a result, WUE is increased because the stomatal limitation on transpiration is higher than on photosynthesis. If carbon isotope discrimination is lowered by increasing photosynthetic capacity whilst maintaining stomatal conductance, no loss in yield would be expected and WUE would be improved, as has been also been shown with peanuts (Nageswara Rao and Wright 1994).

The aims of this study are to:

1. Assess the effects of PRD integrated with three pruning levels on midday leaf gas exchange and carbon isotope discrimination of cv. Shiraz.

2. Investigate the diurnal response of field grown cv. Shiraz vines irrigated by PRD to increasing vapour pressure deficit (VPD) under warm climate conditions.

The physiological effects of combined irrigation and pruning on cv. Shiraz will be evaluated in this study to test the following hypothesis: a) The physiological response of cv. Shiraz will be negatively affected by the interaction between PRD and light pruning levels, as it nears the critical ‘source:sink’ relationship.

5.2 METHODOLOGY

5.2.1 Midday Leaf Gas Exchange

Ambient leaf gas exchange measurements were conducted on all irrigation by pruning combinations at midday (1100-1400 h) on cloudless days when air temperature was ≥25°C. Weather conditions (temperature, °C, relative humidity, % and VPD, kPa) and volumetric

Chapter 5: Physiological response to irrigation and pruning treatments 94

soil water content (θv, %) were determined at midday (12:00 AEDT) for each of the six midday gas exchange experimental days in 2002 and 2003.

Net photosynthesis (Α), stomatal conductance (gs), transpiration (Τ), internal leaf carbon dioxide (Ci) and atmospheric CO2 (Ca) were measured using a portable open infrared gas analyser (CIRAS; PP System, Hitchin, Herts, UK) and fitted to a Parkinson leaf cuvette operated at 300 mL.min-1. Readings were taken on a leaf area of 2.5 cm2 when steady state conditions in gas exchange were achieved (approximately 2 minutes). The CO2 concentration (400 µmol.mol-1), solar radiation, temperature and evaporative demand inside the leaf cuvette were controlled by the CIRAS at ambient conditions. Carbon dioxide and water vapour concentration differences between inlet and outlet gas circulating through the leaf chamber and leaf temperature obtained from energy balance equations were used to calculate leaf area based rates of Α, gs and Τ using the equation of von Caemmerer and Farquhar (1981). Transpiration efficiency (Α/Τ) was calculated as the ratio of net A to T. Vapour Pressure Deficit (VPD) was calculated by the Penman equation (Penman 1955), using relative humidity and air temperature recorded by the weather station.

Eight fully expanded (i.e. the fifth leaf from the shoot tip), sun exposed leaves of close proximity and similar age, size and thickness were randomly selected on four replicate vines per treatment, n=32 vines. Midday leaf gas exchange measurements were conducted on 18 and 23 February 2002 and 13, 14 and 16 January and 17 February 2003 and converted to days after flowering (DAF). Carbon isotope discrimination analysis was determined on leaves selected for gas exchange measurements, as described in section 5.2.2.

5.2.2 Carbon Isotope Discrimination

The discrimination against 13C (∆) is an indicator of the ratio between intercellular and atmospheric partial pressure of CO2 (Ci/Ca) and water-use efficiency (WUE) if VPD between the leaf and surrounding air is constant (Farquhar et al. 1982, 1989; Hubick and Farquhar 1989). The ∆ of oven dried (80°C for at least 4 days), finely ground (<100 µm) leaf laminae samples was determined by mass spectrometry. A standard of known carbon isotope composition was used relative to Pee Dee Belemnite (PDB). ∆ was calculated as

Chapter 5: Physiological response to irrigation and pruning treatments 95 described by Farquhar and Richards (1984), assuming an isotopic composition of air of 7.6% relative to PDB.

5.2.3 PRD vine response to increasing vapour pressure deficit

PRD vine response to increasing VPD throughout the day was assessed on 4 cloudless days between veraison and harvest when resource (water and carbohydrate) demands were greatest due to crop production and high VPD. Diurnal gas exchange (Α, gs, T, Ci, Ca, A/T), leaf water potential (Ψl) and osmotic potential (Ψs) measurements were taken hourly between 6:00 and 14:00 AEDT on 24 and 25 January and 26 February 2002 and between 6:00 and 18:00 AEDT on 14 February 2003. Gas exchange measurements were conducted using the CIRAS-1 under ambient conditions as described in section 5.2.1, on four randomly selected sun exposed, fully expanded leaves from MIN vines irrigated by PRD or SD. MIN vines were chosen as they had shown the greatest photosynthetic sensitivity to PRD in initial experiments in season 2000-01.

Immediately following gas exchange measurements, the selected leaf blade was sealed in a small polyethylene bag to reduce transpiration and the petioles were excised with a razor blade. Leaf water potential (ΨL) was determined by placing the leaf blade in a pressure chamber (Soil Moisture Equipment Corp., Santa Barbara, California, USA) and the petiole was sealed in a silicone gasket, as described by Turner (1988). After ΨL was measured, leaf blades were wrapped in aluminium foil and stored on dry ice. Leaf osmotic potential (Ψs) was determined using a vapour pressure Osmometer 5500 (Wescor, Logan, Utah, USA), as described by Düring (1984).

5.2.4 Statistical Analysis

The main effects and treatment interactions between irrigation and pruning for midday gas exchange measurements were analysed using generalised analysis of variance (ANOVA) in the GENSTAT® statistical package. Pruning effects on leaf development were analysed over the growing period using repeated measures, analysis of variance (ANOVA) in the GENSTAT® statistical package.

Chapter 5: Physiological response to irrigation and pruning treatments 96

5.3 RESULTS OF IRRIGATION AND PRUNING EFFECTS ON MIDDAY LEAF GAS EXCHANGE

5.3.1 Midday Leaf Gas Exchange for 2002 and 2003 Growing Seasons

Midday gas exchange data was averaged over each season (2 days in 2001-02 and 4 days in

2002-03) to evaluate the effects of the treatments on gs (Table 5.1), Α (Table 5.2), Τ (Table

5.3) and Α/Τ (Table 5.5). Gas exchange was dominated by the reduction in gs by PRD in both seasons (Table 5.1). MIN increased gs slightly in the first season and, as a result, significant interactions were found between irrigation and pruning. The rates of gas exchange were higher in 2002-03, which may be attributed to the timing of measurements and soil water content.

Vine physiology at midday was strongly influenced by the application of PRD in both 2001-

02 and 2002-03. PRD reduced gs by 17% and 9 %, Α by 12% and 6% and Τ by 16% and

15% in 2001-02 and 2002-03, respectively. As Ci was not significantly affected by PRD in either growing season, the sub-stomatal concentration of CO2 may not have been altered by the reduction in stomatal aperture. Α/Τ was improved by PRD as a direct response to a greater reduction in Τ rate relative to Α.

Pruning had a minor effect on midday leaf gas exchange. Stomatal conductance (gs), Α and Τ were higher for MIN compared to MECH or SPUR in 2001-02. No significant differences were found between pruning treatments in 2002-03 for the measured gas exchange parameters. Pruning had no effect on Ci or Α/Τ in either season.

The interaction between irrigation and pruning had a significant effect on gs, Α and Τ.

SPUR and MECH treatments showed a large response to PRD with reductions in gs, Α and

Τ. Similarly, MECH was influenced by PRD with large reductions in gs and Α in season 2002-03. However, no gas exchange differences were found between SD + SPUR and PRD

+ SPUR in 2002-03. Ci was not affected by imposed treatments for either season, although the mean results showed a significant reduction by PRD irrigated vines when they were spur pruned (Table 5.4). Α/Τ was not affected by the interaction between irrigation and pruning

Chapter 5: Physiological response to irrigation and pruning treatments 97 in 2001-02 but it was improved by PRD compared to SD, particularly combined with SPUR in 2002-03.

-2 -1 gs (mmol.m .s ) Treatment 2001-02 2002-03 Mean Season LSD Season Mean 226 351 320 *** 15.7 SD 247b 368b 338b PRD 204a 334a 301a Irrigation *** *** *** ns 22.2 SPUR 215a 364 327 MECH 220a 346 314 MIN 242b 344 319 Pruning * ns ns ns 27.2 SD + SPUR 255b 375 345 PRD + SPUR 176a 352 308 SD + MECH 237b 377 342 PRD + MECH 203a 314 286 SD + MIN 251b 353 327 PRD + MIN 234b 335 310 Irrigation*Pruning ** ns ns ns 38.5

Α (µmol.m-2.s-1) Treatment 2001-02 2002-03 Mean Season LSD Season Mean 13.5 17.3 16.4 *** 0.47 SD 14.4b 17.8b 17.0b PRD 12.7a 16.8a 15.8a Irrigation *** *** *** ns 0.66 SPUR 13.1ab 17.5 16.4 MECH 13.1a 17.2 16.2 MIN 14.4c 17.2 16.5 Pruning ** ns ns * 0.81 SD + SPUR 14.6b 17.5b 16.8cd PRD + SPUR 11.5a 17.5b 16.0b SD + MECH 13.9b 18.5c 17.4d PRD + MECH 12.3a 15.8a 15.0a SD + MIN 14.7b 17.4b 16.7cd PRD + MIN 14.2b 17.0b 16.3bc Irrigation*Pruning ** *** *** *** 1.15

Chapter 5: Physiological response to irrigation and pruning treatments 98

Τ (mmol.m-2.s-1) Treatment 2001-02 2002-03 Mean Season LSD Season Mean 5.70 7.62 7.14 *** 0.26 SD 6.19b 8.24b 7.72b PRD 5.21a 7.01a 6.56a Irrigation *** *** *** ns 0.36 SPUR 5.49a 7.64 7.11 MECH 5.61a 7.56 7.07 MIN 5.99b 7.67 7.25 Pruning * ns ns ns 0.44 SD + SPUR 6.31d 8.23 7.75 PRD + SPUR 4.68a 7.05 6.46 SD + MECH 5.99cd 8.34 7.75 PRD + MECH 5.23b 6.78 6.39 SD + MIN 6.27d 8.14 7.67 PRD + MIN 5.71c 7.19 6.82 Irrigation*Pruning *** ns ns ns 0.63

Table 5.4: Mean midday intercellular CO2 concentration (CI) of vines with integrated irrigation and pruning treatments and seasonal means for 2001-02, 2002-03. Different letters denote significant differences between treatment means for each season (column) and across seasons (rows), as calculated by Fisher’s least significant difference (LSD 5% level). Significant differences between treatments are denoted by *** P<0.001, ** P<0.01, * P<0.05, ns= not significant.

-1 CI (mmol.mol ) Treatment 2001-02 2002-03 Mean Season LSD Season Mean 228 241 238 *** 2.47 SD 230 242 239 PRD 226 241 237 Irrigation ns ns ns ns 3.50 SPUR 226 243 239 MECH 228 241 237 MIN 229 240 237 Pruning ns ns ns ns 4.28 SD + SPUR 231 245 241b PRD + SPUR 222 240 236a SD + MECH 229 241 238ab PRD + MECH 226 241 237a SD + MIN 229 239 237a PRD + MIN 228 240 237a Irrigation*Pruning ns ns * ns 6.05

Chapter 5: Physiological response to irrigation and pruning treatments 99

Α/Τ (µmol.mmol-1) Treatment 2001-02 2002-03 Mean Season LSD Season Mean 2.41 2.33 2.35 * 0.06 SD 2.34a 2.21a 2.24a PRD 2.48b 2.45b 2.46b Irrigation * *** *** ns 0.09 SPUR 2.42 2.35 2.37 MECH 2.38 2.33 2.34 MIN 2.43 2.32 2.35 Pruning ns ns ns ns 0.11 SD + SPUR 2.33 2.17a 2.21a PRD + SPUR 2.50 2.53c 2.52c SD + MECH 2.34 2.26a 2.28a PRD + MECH 2.42 2.39b 2.40b SD + MIN 2.35 2.21a 2.24a PRD + MIN 2.50 2.43bc 2.45bc Irrigation*Pruning ns * * ns 0.16

2 A strong curvilinear relationship (R = 0.7553, P<0.01) was found between midday Α and gs for all treatments. Therefore, leaf photosynthetic rate is likely to be a direct response to gs and thus stomatal aperture size in this experiment (Figure 5.1).

-1

s 20 2 m ol

m u

s 15

otosynthe 10 Ph

5 y = 7.8881Ln(x) - 28.625 R2 = 0.7553

0 100 200 300 400 500 600 700

Stomatal Conductance mmol m-2s-1

Figure 5.1: Curvilinear relationship between photosynthesis and stomatal conductance for midday measurements of all treatments (R2 = 0.748), P<0.001

Chapter 5: Physiological response to irrigation and pruning treatments 100

5.3.2 Vapour Pressure Deficit and Soil Water Content

Midday temperature of the gas exchange experimental days ranged from 27.8°C (57 DAF) to 34.6°C (56 DAF) and relative humidity ranged from 27% (56 DAF) to 40% (100 DAF) (Table 5.6). As a result of high temperatures and low relative humidity, the VPD at 56 DAF was very high (4 kPa). VPD was calculated to be between 2.5 and 2.6 kPa on all other days except at 59 DAF when midday temperature was slightly higher. Days after irrigation (DAI) and days after rainfall (DAR) were calculated for each scheduled experimental day, as these parameters may have affected θv. Irrigation occurred between three and six days prior to gas exchange measurements, and irrigation amounts ranged between 18 and 21 mm for SD and 10 and 12 mm for PRD through either the northern or southern drip line. Rainfall was expected to have a large effect on θv, particularly of the subsoil on 56, 57 and 59 DAF, as 58 mm of rain fell at 43 DAF, which completely refilled the soil profile. However, no more rain was recorded until late February. Thus, measurements at 91 DAF would not have been influenced by rain.

DAF Midday Irrigation Rainfall T °C %RH VPD DAI SD mm PRD mm DAR mm 56 34.6 27.2 4.0 3 18.0 10.3N 13 58.0 57 27.8 33.0 2.5 4 18.0 10.3N 14 58.0 59 31.9 31.4 3.2 6 18.0 10.3N 16 58.0 91 28.3 35.8 2.5 5 21.0 12.1N 47 58.0 95 29.8 37.4 2.6 6 21.0 12.1S 1 0.2 100 29.4 40.0 2.5 4 21 12.1N 6 0.2

The θv of the surface soil was not influenced by irrigation treatments, prior irrigation or rainfall events. This may have been due to higher levels of soil evaporation, plant water use in the upper 20 cm of soil and the predominantly sandy loam soil type. There was little difference between the surface soil θv of the SD and PRD irrigation treatments, which ranged from 19 to 20% for SD and 19 to 21% for PRD. Previous irrigation and rainfall events did not influence the θv of the surface soil (Table 5.7). Water uptake by vine roots in the surface soil, soil evaporation and infiltration to subsurface soil was high after an

Chapter 5: Physiological response to irrigation and pruning treatments 101

irrigation event, as was demonstrated by the rapid decrease in surface soil θv on the TDR graphs for both SD and PRD after an irrigation event (refer to section 2.2.3).

The θv of the subsurface soil (50-70 cm) was affected by irrigation treatment and prior rainfall events. The subsurface θv was greater for SD irrigation than PRD irrigation. The range of θv for the subsoil of the SD irrigated rows was 31 to 40%, whereas PRD reduced θv to a range between 26 to 34%. The subsurface soil was affected by the large rainfall event (58 mm in 24 hours) on the 31 December 2002, which completely refilled the soil profile.

As a consequence, θv of SD subsurface soil increased to 40% on 56 DAF and proceeded to decrease until the next irrigation event. The next rainfall event occurred in late February

2002, thus the remaining experimental days were not influenced by rain. The θv of PRD subsurface soil was also influenced by the large rainfall event on 43 DAF. However, the increase was not as substantial because the soil profile under PRD irrigation was drier prior to the rain event.

DAF Soil Depth SD PRD cm North South 56 20-40 cm 19.8 21.2 20.4 50-70 cm 40.2 33.8 29.7 57 20-40 cm 19.2 20.7 20.4 50-70 cm 37.9 30.8 29.7 59 20-40 cm 18.9 20.2 19.9 50-70 cm 34.4 27.3 28.5 91 20-40 cm 18.9 20.0 19.7 50-70 cm 37.8 26.1 29.8 95 20-40 cm 19.0 19.7 19.4 50-70 cm 30.5 26.3 30.4 100 20-40 cm 18.9 19.7 19.0 50-70 cm 30.5 28.0 27.6

5.3.2 Phenological Effect on Midday Leaf Gas Exchange

The midday photosynthetic capacity (gs, Α, and Τ) of vines was higher at post-veraison (56, 57 and 59 DAF) than a month later at pre-harvest.

Chapter 5: Physiological response to irrigation and pruning treatments 102

Stomatal Conductance

The rate of gs was greater at post-veraison, particularly 59 DAF, than pre-harvest. No significant differences between irrigation treatments were recorded early in the season, however PRD decreased gs significantly from 59 DAF through to harvest (Fig 5.2a). Pruning had minimal effect on gs through out the ripening phase of berry development. However,

MIN reduced gs at 57 and 91 DAF (Figure 5.2b). The interaction between irrigation and pruning had a significant effect on gs at 57, 91, 95 DAF (Fig 5.2c). At 57 DAF, SD +

MECH had the highest rates of gs, which indicates this treatment was functioning at a higher rate than all other treatments. Integrated irrigation and pruning treatments did not influence gs at 59 DAF, since all SD irrigated treatments had higher rates of gs compared to PRD and pruning had no effect. A large reduction in gs was measured 91 DAF and SD + SPUR and

SD + MECH had a significantly greater rate of gs compared to all other treatments. The combination of PRD + SPUR and PRD + MECH lowered the gs of compared to all other treatments at pre-harvest (95 DAF). A similar physiological response was observed 100 days after flowering but the interaction between irrigation and pruning did not significantly affect gs.

550 Post-veraison *** Pre-harvest 500 450 /s) 2 400 ns 350 ns ***

(mmol/m 300 s ** *** g 250

200 150

56 57 59 91 95 100

DAF

Chapter 5: Physiological response to irrigation and pruning treatments 103

550 Post-veraison Pre-harvest 500 ns

450 * /s) 2 400

ns 350 ***

(mmol/m 300 s

g ns ns 250 200

150 56 57 59 91 95 100 DAF

550 ns Post-veraison Pre-Harvest 500 SD+Spur 450 PRD+Spur **

/s) SD+Mech 2 400 ns PRD+Mech 350 * SD+Min PRD+Min * (mmol/m 300

s ns g 250

200

150

56 57 59 91 95 100 DAF

Photosynthesis

Net leaf photosynthesis (A) steadily decreased from the maximum rate at 59 DAF to harvest. No irrigation effects were found at 56 or 57 DAF but PRD decreased A from 59 DAF onwards relative to SD (Figure 5.3a). The photosynthetic response to PRD reflects the decrease in gs. However, the degree of difference in A between PRD and SD was only small relative to that of gs. Significant pruning effects occurred at 56, 59 and 95 DAF (Figure 5.3b). MIN had a higher rate of A at 56 and 95 DAF compared to SPUR and MECH. Yet

Chapter 5: Physiological response to irrigation and pruning treatments 104

SPUR had the highest level of A at 59 DAF. The interaction between irrigation and pruning significantly affected Α after veraison to 91 DAF (Figure 5.3c). The rate of A was generally reduced by PRD + MECH and PRD + MIN at post-veraison, relative to the other treatments. SD + SPUR and SD + MECH had amongst the highest rates of Α, which indicates these treatment were functioning at a higher rate than all other treatments. No significant treatment effects were observed at 95 and 100 DAF (i.e. pre-harvest).

25 Post-veraison Pre-Harvest

ns 20 ** ns /s) 2 *** 15 *** ** mol / m µ A ( 10

5 56 57 59 91 95 100 DAF

25 Post-veraison Pre-harvest

* 20 * ns /s) /s) 2

2 ns

m m

/ ** ns 15 mol mol /

µ µ (

A A ( 10

56 57 59 91 95 100

DAF

Chapter 5: Physiological response to irrigation and pruning treatments 105

25 Post-Veraison Pre-Harvest

*** * 20 ** * /s) /s) 2 2

m ns m / ns 15 mol mol /

µ µ SD+Spur

A ( A ( PRD+Spur 10 SD+Mech PRD+Mech SD+Min PRD+Min 5 56 57 59 91 95 100 DAF

Transpiration

Transpiration decreased progressively from veraison to harvest, with the exception of 59 DAF. PRD reduced T compared to SD from 57 DAF to 100 DAF, with the greatest reduction occurring at 59 DAF when T rates were highest (Figure 5.4a). By reducing stomatal aperture and the conductance of CO2 into the leaf and H2O and O2 out of the leaf, the rate of Α and Τ was reduced by PRD in conjunction with gs. Pruning treatments had little affect on T from 56 to 91 DAF but MIN increased T slightly relative to SPUR and MECH at 95 and 100 DAF (Figure 5.4b).

The rate of T was not significantly altered by the integration of irrigation and pruning treatments at 56 or 59 DAF (Figure 5.4c). The strongest combined treatment effect occurred at 57 DAF. SD + MECH had the highest rates of Τ, which indicates this treatment was functioning at a higher rate than all other treatments. No treatment effect was found on 59 DAF because of the dominant irrigation effect (i.e. T was reduced by PRD). A large reduction in T was measured at 91 DAF for all treatments, particularly those irrigated by PRD. The interaction between irrigation and pruning lowered the T of PRD + SPUR and PRD + MECH compared to all other treatments at pre-harvest (95 and 100 DAF). The reduction in gs had a direct effect on Τ.

Chapter 5: Physiological response to irrigation and pruning treatments 106

12 Post-veraison *** Pre-harvest

10 ns ***

/s) 8 2 *** *** / m *** 6

4 T (mmol

0 56 57 59 91 95 100 DAF Figure 5.4a: Midday transpiration (T) of SD (c) and PRD ( ) irrigation treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant

12 Post-veraison Pre-harvest ns 10 ns ns

/s) 8 2 * ns / m 6 *

4 T (mmol

0 56 57 59 91 95 100

DAF

12 ns Post-veraison Pre-harvest

10 ns ***

/s) 8

2 * ***

/ m * 6 SD+Spur 4 PRD+Spur T (mmol SD+Mech PRD+Mech 2 SD+Min PRD+Min 0

56 57 59 91 95 100

DAF

Chapter 5: Physiological response to irrigation and pruning treatments 107

Internal CO2 Partial Pressure

The Ci simulated the trend in gs from post veraison to harvest. Irrigation had no significant affect on Ci levels from 56 to 59 DAF but PRD reduced Ci at 91 and 100 DAF relative to SD

(Figure 5.5a). Pruning level had an influence on Ci at 56, 57 and 91 DAF (Figure 5.5b).

SPUR had the highest Ci at 56 and 57 DAF and MECH had the highest level of Ci at 91

DAF. MIN generally had the lowest Ci relative to the other pruning treatments. The combination of irrigation treatment and pruning level had a significant affect on Ci early in the ripening phase of berry development but no affect on Ci in the latter part of berry ripening (Figure 5.5c). SD + MIN lowered Ci at 56 and 57 DAF compared to SD + SPUR and PRD + SPUR lowered Ci at 59 DAF compared to PRD + MECH.

300 Post-veraison Pre-harvest 290 280 270 /s)

2 ns 260 /m ns 250 ns 240 *** * ***

Ci (mmol 230 220 210 200 56 57 59 91 95 100

DAF

300 Post-veraison Pre-harvest 290 280 270

/s) ns 2 260 /m *** 250 ** 240 ** ns ns

Ci (mmol 230 220

210

200 56 57 59 91 95 100 DAF

Chapter 5: Physiological response to irrigation and pruning treatments 108

300 Post-veraison Pre-harvest 290 SD+Spur 280 PRD+Spur SD+Mech 270 * PRD+Mech /s) 2 260 * SD+Min /m PRD+Min 250 ** ns ns 240 ns

Ci (mmol 230 220 210 200 56 57 59 91 95 100 DAF

Transpiration Efficiency

The phenological effect on transpiration efficiency (A/T) was minimal throughout the ripening phase of berry development and did not reflect the negative trend with time of the other physiological parameters. A/T ranged between 1.6 and 3.1 µmol.mol-1. PRD increased A/T at 57, 59, 91 and 100 DAF because of the reduction in transpiration by restriction of stomatal conductance (Figure 5.6a). Pruning effects on A/T were shown after veraison, although no consistent trends were found (Figure 5.6b). MIN, MECH and SPUR each increased A/T at 56, 57 and 59 DAF, respectively. The integration of irrigation and pruning also affected A/T after veraison (Figure 5.6c). SD irrigation combined with SPUR, MECH and MIN each reduced A/T relative to the other treatments between 56 and 59 DAF.

4 Post-veraison Pre-harvest 3.5 *** 3 *** ) *** 2.5 ns ns *** 2 mol /mmol 1.5

A/T ( 1 0.5 0 56 57 59 91 95 100

DA F

Chapter 5: Physiological response to irrigation and pruning treatments 109 4 Post-veraison Pre-harvest 3.5 ns 3 ns * 2.5 ** ns *** 2 mol /mmol)

µ 1.5

A/T ( 1

0.5 0

56 57 59 91 95 100

DAF

Figure 5.6b: Midday internal CO2 concentration (Ci) of SPUR (c), MECH (c) and MIN ( ) pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant.

4 Post-veraison Pre-harvest 3.5 ns ns 3 ** ) ** ns 2.5 ***

2 mol /mmol SD+Spur 1.5 PRD+Spur

A/T ( SD+Mech 1 PRD+Mec h SD+Min 0.5 PRD+Min 0 56 57 59 91 95 100 DA F Figure 5.6c: Midday transpiration efficiency (A/T) of integrated irrigation and pruning treatments measured on six days between veraison and harvest in seasons 2001-02 and 2002-03. Significant differences at each measurement day are represented by ***P>0.001, **P>0.01, *P>0.05, ns = non significant.

5. 4 IRRIGATION AND PRUNING EFFECTS ON CARBON ISOTOPE DISCRIMINATION

The determination of ∆ was influenced by seasonal variation, PRD irrigation and MIN pruning as a response to stomatal limitation (Table 5.8). PRD irrigation reduced ∆ in 2000- 01, 2001-02 and 2002-3 by 2%, 3% and 3% respectively, presumably as a result of increases in stomatal limitation. Pruning had no direct effect on ∆ in 2000-01. However, MIN reduced ∆ in the two following seasons. The interaction between irrigation and pruning had an effect on ∆ in 2003. Carbon isotope discrimination was reduced by PRD when combined

Chapter 5: Physiological response to irrigation and pruning treatments 110 with SPUR or MECH. However, PRD had no effect on the ∆ of MIN leaves. This interaction response was similar to the midday transpiration response in 2002-03.

∆ Treatment 2000-01 2001-02 2002-03 Mean Year LSD Seasonal Mean 21.1 19.4 19.7 19.6 ** 0.20 SD 21.3b 19.7b 20.0b 19.9b PRD 20.8a 19.2a 19.5a 19.4a Irrigation *** *** *** *** ns 0.28 SPUR 21.2 19.5b 20.1b 19.8b MECH 21.0 19.6b 19.6a 19.6b MIN 21.0 19.2a 19.5a 19.4a Pruning ns ** ** *** ns 0.34 SD + SPUR 21.4 19.7 20.5c 20.2 PRD + SPUR 20.9 19.3 19.6ab 19.5 SD + MECH 21.2 19.9 19.9b 19.9 PRD + MECH 20.9 19.4 19.4a 19.4 SD + MIN 21.2 19.5 19.6ab 19.5 PRD + MIN 20.8 18.8 19.5ab 19.2 Irrigation*Pruning ns ns * ns ns 0.48

The relationship between ∆ and yield was evaluated to determine if stomatal limitation or increased photosynthetic capacity influenced treatment physiological responses. When all data was pooled over the three seasons, no significant relationship was found between yield and ∆ (R2 = 0.07).

5.5 PRD VINE PHYSIOLOGICAL RESPONSE TO INCREASING VAPOUR PRESSURE DEFICIT

5.5.1 Diurnal Increase in Vapour Pressure Deficit

The diurnal physiological response of PRD irrigated vines to increasing VPD was measured between veraison and harvest in seasons 2001-02 and 2002-03. Climatic data for diurnal measurement days are presented in Table 5.9 and volumetric soil water content is presented in Table 5.10. VPD was plotted against time for the four measurement days (Figure 5.7).

Chapter 5: Physiological response to irrigation and pruning treatments 111

Table 5.9: Hourly climatic conditions (Temperature °C, Relative Humidity % and Vapour Pressure Deficit kPa) for each of the diurnal measurement days.

Time 24/1/02 25/1/02 26/2/02 14/2/03 AEDT T °C %RH VPD T °C %RH VPD T °C %RH VPD T °C %RH VPD 6:00 15.9 55.7 0.8 10.4 79.8 0.3 15.3 95.0 0.1 15.9 61.8 0.7 7:00 14.5 62.1 0.6 10.2 81.3 0.2 19.2 89.1 0.2 14.4 65.1 0.6 8:00 16.8 57.9 0.8 14.5 83.9 0.3 23.0 73.8 0.7 18.1 59.3 0.8 9:00 19.1 53.2 1.0 21.9 58.8 1.1 25.9 63.5 1.2 22.1 48.1 1.4 10:00 22.6 47.1 1.4 27.3 44.0 2.0 28.1 56.9 1.6 25.8 40.9 2.0 11:00 26.1 39.8 2.0 31.6 32.7 3.1 29.4 50.2 2.0 29.7 33.6 2.8 12:00 29.3 34.5 2.7 34.5 26.3 4.0 31.0 42.3 2.6 33.3 27.5 3.7 13:00 32.0 29.0 3.4 36.9 19.6 5.0 32.4 35.7 3.1 35.2 23.4 4.4 14:00 33.5 25.3 3.9 38.1 17.4 5.5 33.6 33.2 3.5 35.8 21.4 4.6 15:00 ------35.5 21.1 4.6 16:00 ------35.5 21.9 4.5 17:00 ------33.8 22.2 4.1 18:00 ------31.7 24.0 3.6

Diurnal Soil Depth SD PRD cm North South 24/1/02 20-40 cm 20.6 22.4 19.1 50-70 cm 36.5 36.7 29.3 25/1/02 20-40 cm 20.1 20.4 20.8 50-70 cm 34.0 31.8 28.7 26/2/02 20-40 cm 21.9 19.7 25.0 50-70 cm 32.2 26.9 36.3 14/2//03 20-40 cm 32.6 20.0 34.9 50-70 cm 38.3 26.7 34.1

6.0

5.5 5.0 4.5

4.0

3.5

3.0 24/01/2002

VPD KPa 25/01/2002 2.5 26/02/2002 2.0 14/02/2003 1.5 1.0 0.5 0.0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00

Time AEDT Figure 5.7: Diurnal change in vapour pressure deficit (VPD, kPa) for four diurnal measurement days.

Chapter 5: Physiological response to irrigation and pruning treatments 112

5.5.2 Diurnal Response of PRD Vines

The photosynthetic capacity of the vines increased rapidly from dawn to between 8:00 and 9:00 AEDT when photosynthetically active radiation (PAR) was >1200 µmol.s-1.m-2. A rapid increase in Α and gs occurred in the morning and then stabilised (as VPD increased between 11:00 and 2:00 AEDT) with a tendency to decline slightly in the late afternoon. The increase in Τ was more gradual throughout the day. However, the rate declined in the afternoon. The rates of gs, Α and Τ were all reduced by the application of PRD. Ci decreased rapidly as plants began to transpire in the early morning and stabilised at approximately -1 10:00 AEDT at a CO2 concentration of 200-250 µmol.mol . PRD also lowered the initial Ci within the leaf but reached equilibrium at the same concentration as the SD vines after 10:00

AEDT. ΨL was measured in conjunction with gas exchange measurements. No irrigation treatment differences in ΨL were measured throughout the day, although ΨL decreased (became more negative) as VPD increased during the day. Interestingly, A was maintained at a maximal rate despite the large reduction of ΨL to values of <-1.5 Mpa during the diurnal measurement period. Ψs and cell turgor also decreased (became more negative) as VPD increased during the day. Ψs and turgor were dependent on soil water content. Ψs was lower and turgidity was greater on PRD vines when soil water content was high (recently irrigated).

Diurnal measurements conducted on 24 January 2002 Gas exchange measurements were conducted on an hourly basis between 6:00 and 14:00

AEDT on the 24 January 2002. PRD significantly reduced gs throughout the day, with the exception of measurements taken at 9:00, 10:00 and 13:00 AEDT (Figure 5.8a). Net A and Τ were also reduced by PRD throughout the day (Figures 5.8b and c). Minimal variability was recorded for Τ measurements. Thus, treatment differences were observed. Irrigation strategy did not influence Α, except at 13:00 AEDT when PRD was slightly greater than SD. Α/Τ was improved by PRD during the morning when VPD was low, as a response to reduced Τ by PRD (Figure 5.8d). Ci was slightly lower for PRD compared to SD at several time points throughout the day (Figure 5. 8e).

Chapter 5: Physiological response to irrigation and pruning treatments 113

Pre-dawn ΨL was slightly lower for PRD irrigated vines (-0.75 MPa) compared to SD irrigated vines (-0.60 MPa) (Figure 5.8f). ΨL declined steadily throughout the day for both treatments as VPD increased. The imposition of PRD did not affect ΨL, except at 14:00

AEDT when SD vines demonstrated earlier recovery. PRD lowered ΨS but the treatment difference was predominantly insignificant (Figure 5.8g). The ΨS of the vine leaves decreased (i.e. became more negative) as VPD increased. As a result of maintained ΨL and decreased ΨS by PRD, an increase in cell turgor was observed from 8:00 AEDT onwards (Figure 5.8h).

400

350

300

250

1 - .s 2 -

m SD 200 PRD

mmol. s g 15 0

10 0

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00

Time TimAEDeT AEDT

Figure 5.8a: Diurnal response of PRD on stomatal conductance (gs) on 24 January 2002, ± SEM.

-1 15 s

-2 SD -1 .s

-2 PRD m

A mmol m 10 mol. µ

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00

Time AEDT

Figure 5.8b: Diurnal response of PRD on photosynthesis (A) on 24 January 2002, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 114 12

-1

.s -2 SD l.m 6 PRD mo

T 4

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00

Time ATimEDTe AEDT

Figure 5.8c: Diurnal response of PRD on transpiration (T) on 24 January 2002, ± SEM.

-1

ol 6 m m

l. SD 5

mo PRD µ

/T 4 A 3

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AEDT Time AEDT Figure 5.8d: Diurnal response of PRD on transpiration efficiency (A/T) on 24 January 2002, ± SEM.

400

350

300

250

-1 1 - l mo . l 200 mo mmol mol i µ i C

C SD 150 PRD

100

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AETimDTe AEDT

Figure 5.8e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 24 January 2002, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 115 Time AEDT

6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 0.00

-0.20

-0.40

SD -0.60 PRD -0.80

MP -1.00 L Ψ ψ -1.20

-1.40

-1.60

-1.80

-2.00

Figure 5.8f: Diurnal response of PRD on leaf water potential (ΨL) on 24 January 2002, ± SEM.

TimTime AEeD TAEDT

6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 0.00

-0.20

-0.40

-0.60

SD -0.80 PRD

MPa -1.00 s

Ψ ψ -1.20

-1.40

-1.60

-1.80

-2.00

Figure 5.8g: Diurnal response of PRD on osmotic potential (ΨS) on 24 January 2002, ± SEM.

1. 2 0

1. 0 0

0.80

0.60 SD MPa PRD

0.40

0.20

0.00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AEDT Time AEDT Figure 5.8h: Diurnal response of PRD on turgor on 24 January 2002, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 116

Diurnal measurements conducted on 25 January 2002

VPD was low between 6:00 and 8:00 AEDT on the 25 January 2002. A decrease in gs was observed for both treatments between 8:00 and 9:00 AEDT. gs then increased to a maximum rate at 12:00 AEDT, followed by a steady decline as VPD increased (Figure 5.9a). Photosynthesis and Τ were delayed (Figure 5.9b and 5.9c) compared to 24 January 2002, particularly for PRD irrigated vines. The reduction in Α and Τ rates was associated with lower gs. Α/Τ was relatively low throughout the day with no treatment differences due to the reduction in Τ by PRD (Figure 5.9d). Ci decreased as gs, Α and Τ increased.

Equilibrium was reached at 10:00 AEDT at a Ci concentration of approximately 230 µmol.mol-1 (Figure 5.9e).

Pre-dawn ΨL was slightly higher than for the previous day (-0.5 to –0.6 MPa) due to the higher relative humidity and lower temperatures at this time (Figure 5.9f). ΨL decreased for both irrigation treatments as VPD increased during the day. PRD was lower (more negative) at 11:00 AEDT but no other treatment differences were recorded throughout the day.

Similarly ΨS decreased as VPD increased, with no treatment differences recorded (Figure

5.9g). Turgor increased at 7:00 AEDT due to an increase in ΨL, with PRD irrigated vines having slightly more turgidity than SD (Figure 5.9h).

400

350

300

250 1

.s SD 2 - 200 m PRD

mmol. 15 0 s g

10 0

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00

TimeTim AEDeT AEDT

Figure 5.9a: Diurnal response of PRD on stomatal conductance (gs) on 25 January 2002, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 117

-1 15

.s -2

m SD

mol. PRD µ µ

A 10

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time ATimEDeT AEDT Figure 5.9b: Diurnal response of PRD on photosynthesis (A) on 25 January 2002, ± SEM. 12

8 -1

.s SD -2 m

. l 6 PRD

mo µ

T 4

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AEDT Time AEDT Figure 5.9c: Diurnal response of PRD on transpiration (T) on 25 January 2002, ± SEM.

-1

ol 6 m m

l. SD 5 mo PRD µ

A 4

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AETimDT e AEDT Figure 5.9d: Diurnal response of PRD on transpiration efficiency (A/T) on 25 January 2002, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 118 400

350

300

250

1 - l mo . l 200 mo

µ SD

µ i

C PRD 15 0

10 0

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AETimDT e AEDT

Figure 5.9e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 25 January 2002, ± SEM. Time AETimDT e AEDT

6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00

0.00

-0.20

-0.40

-0.60

-0.80 a

L -1.00

ψ Ψ -1.20 SD -1.40 PRD

-1.60 -1.80

-2.00

Figure 5.9f: Diurnal response of PRD on leaf water potential (ΨL) on 25 January 2002, ± SEM. Time ATimEDTe AEDT 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 0.00

-0.20

-0.40

-0.60 -0.80

MPa

s -1.00 MPa S ψ SD Ψ -1.20 PRD

-1.40

-1.60

-1.80

-2.00

Figure 5.9g: Diurnal response of PRD on osmotic potential (ΨS) on 25 January 2002, ± SEM

Chapter 5: Physiological response to irrigation and pruning treatments 119

1. 2 0

1. 0 0

0.80

SD 0.60 MPa PRD

0.40

0.20

0.00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 TimeTim AEDeT AEDT

Figure 5.9h: Diurnal response of PRD on turgor on 25 January 2002, ± SEM

Diurnal measurements conducted on 26 February 2002 The volumetric soil water content of the subsurface soil was slightly lower on the 26 February 2002 compared to previous measurements, as the preceding irrigation event occurred 96 hours prior to measurements (refer to Figures 2.5a, b and c). However, VPD was also lower than the previous diurnal (refer to Figures 5.7). The rate of gs was lower than for previous diurnals (Figure 5.10a). PRD had lowered gs between 10:00 and 13:00 AEDT and this lowered Α and Τ (Figures 5.10b and c). The rate of gs increased more for PRD than SD as VPD increased. Thus, at 13:00 the gs of PRD irrigated vines surpassed SD irrigated vines and Α and Τ were greater for PRD in the afternoon. SD had a greater Α/Τ in the morning due to higher A (Figure 5.10d). However by mid-morning, no treatment differences were observed. No significant differences in Ci were observed between PRD and SD (Figure 5.10e).

Pre-dawn ΨL was higher (less negative) than diurnals conducted in January. No treatment differences were found in the morning for ΨL until VPD reached 1.6 kPa, when PRD had lowered ΨL (Figure 5.10f). ΨS was not influenced by the imposition of PRD but it was lower than for previous diurnals (Figure 5.10g). However, cell turgor in the leaves was reduced by PRD irrigation (Figure 5.10h).

Chapter 5: Physiological response to irrigation and pruning treatments 120 400

350

300

250 1 - .s 2

- SD

m 200 PRD mmol.

s 15 0 g

10 0

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AETimDT e AEDT

Figure 5.10a: Diurnal response of PRD on stomatal conductance (gs) on 26 February 2002, ± SEM. 25

-1 15

.s -2

m SD

mol. PRD µ µ

A 10

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AEDT Time AEDT Figure 5.10b: Diurnal response of PRD on photosynthesis (A) on 26 February 2002, ± SEM. 12

8 -1

.s -2 m . l 6

mo µ SD T PRD 4

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00

Time AETimDTe AEDT Figure 5.10c: Diurnal response of PRD on transpiration (T) on 26 February 2002, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 121 10

7 -1

ol 6 m m l. 5 SD mo

µ µ PRD

A/T 4

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AEDTimT e AEDT Figure 5.10d: Diurnal response of PRD on transpiration efficiency (A/T) on 26 February 2002, ± SEM. 400

350

300

250 -1 mol 200 mol. µ µ i SD

C 15 0 PRD

10 0

0 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 Time AETimDT e AEDT

Figure 5.10e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 26 February 2002, ± SEM.

Time AEDTimT e AEDT 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 0.00

-0.20

-0.40 -0.60

-0.80 SD

a PRD

MP -1.00 L Ψ ψ -1.20 -1.40

-1.60

-1.80

-2.00

Figure 5.10f: Diurnal response of PRD on leaf water potential (ΨL) on 26 February 2002, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 122

TimeTim AEDeT AEDT

6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 0.00

-0.20

-0.40

-0.60

-0.80

MPa -1.00

Ψ SD ψ -1.20 PRD

-1.40

-1.60

-1.80

-2.00

Figure 5.10g: Diurnal response of PRD on osmotic potential (ΨS) on 26 February 2002, ± SEM.

1.2 0

1.0 0

0.80

0.60 SD

MPa PRD

0.40

0.20

0.00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00

Time AETimDT e AEDT Figure 5.10h: Diurnal response of PRD on turgor on 26 February 2002, ± SEM.

Diurnal measurements conducted on 14 February 2003

A 13 hour diurnal (dawn to dusk) was conducted on 14 February 2003 to assess the physiological response of PRD irrigated vines as VPD increased during the day and decreased during late afternoon. Stomatal conductance (gs) followed a similar trend to VPD throughout the day. It increased in the morning, stabilised at midday and decreased in the afternoon (Figure 5.11a). Α and Τ responded to the decrease in gs by PRD (Figures 5.11b and c). However Α decreased at 16:00 AEDT for SD vines which, may have been due to photoinhibition as no corresponding decrease in gs was observed. Α/Τ was relatively improved in the morning by PRD, although no treatment effects were found after 11:00

Chapter 5: Physiological response to irrigation and pruning treatments 123

AEDT (Figure 5.11d). A large difference in Ci was observed at pre-dawn between PRD and -1 SD vines (Figure 5.11e). Ci stabilised at a concentration of approximately 230 µmol.mol .

Pre-dawn ΨL was –0.42 MPa for both treatments and ΨL decreased as VPD increased

(Figure 5.11f). ΨS followed a similar trend and decreased as VPD increased (Figure 5.11g).

In the absence of treatment differences in ΨL and ΨS, cell turgidity was unaffected by the treatments (Figure 5.11h).

400

350

300

250

1 -

.s SD 2 - 200 m PRD

mmol.

s 15 0 g

10 0

0 6.00 8.00 10.00 12.00 14.00 16.00 18.00 TiTimmee A AEDEDT T

Figure 5.11a: Diurnal response of PRD on stomatal conductance (gs) on 14 February 2003, ± SEM.

-1 15 .s

-2 m SD PRD mol. µ µ

A 10

0 6.00 8.00 10.00 12.00 14.00 16.00 18.00 TiTimmee A AEDEDTT

Figure 5.11b: Diurnal response of PRD on photosynthesis (A) on 14 February 2003, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 124 12

-1 .s

-2 m . l 6 SD mo

µ PRD T 4

0 6.00 8.00 10.00 12.00 14.00 16.00 18.00 Time AEDT Time AEDT Figure 5.11c: Diurnal response of PRD on transpiration (T) on 14 February 2003, ± SEM. 10

-1 ol 6 m m

l. SD 5

mo PRD µ µ 4 A/T

0 6.00 8.00 10.00 12.00 14.00 16.00 18.00

TiTimmee A EAEDDT T Figure 5.11d: Diurnal response of PRD on transpiration efficiency (A/T) on 14 February 2003, ± SEM.

400

350

300

250

-1

mol SD 200 PRD

mol. µ µ i

C 15 0

10 0

0 6.00 8.00 10.00 12.00 14.00 16.00 18.00 TiTimmee A AEDEDT T

Figure 5.11e: Diurnal response of PRD on intercellular CO2 partial pressure (Ci) on 14 February 2003, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 125 TimTime AAEDEDTT

6.00 8.00 10.00 12.00 14.00 16.00 18.00 0.00

-0.20

-0.40

-0.60

-0.80

a -1.00 MP L ⎝ ψ -1.20 SD PRD -1.40

-1.60

-1.80

-2.00

Figure 5.11f: Diurnal response of PRD on leaf water potential (ΨL) on 14 February 2003, ± SEM. TimTime AeE AEDDT T

6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 0.00

-0.20

-0.40 -0.60

-0.80

-1.00 MPa Ψ

s SD

ψ -1.20 PRD -1.40

-1.60

-1.80

-2.00

Figure 5.11g: Diurnal response of PRD on osmotic potential (ΨS) on 14 February 2003, ± SEM 1.2 0

1.0 0

0.80

0.60 MPa

0.40 SD PRD 0.20

0.00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 TiTimmee A AEDEDTT Figure 5.11h: Diurnal response of PRD on turgor on 14 February 2003, ± SEM.

Chapter 5: Physiological response to irrigation and pruning treatments 126

5.6 DISCUSSION

5.6.1 Irrigation Effect on Leaf Gas Exchange

Partial rootzone drying reduced mean (ca. veraison to harvest) gs, A and T compared to standard drip irrigation in seasons 2001-02 and 2002-03. The mean gs, A, T rates between veraison and harvest for both seasons were comparable to previous studies conducted on Vitis vinifera in semi-arid climates (Winkel and Rambal 1993; Naor and Wample 1994; Correia et al. 1995). However, actual values were greater in the second season, particularly for SD. This may be attributed to relatively high soil water content in January 2003 due to a large rainfall event (58 mm) on 31 December 2002. The absolute reduction in gs by PRD was 43 and 34 mmol.m-2.s-1 for seasons 2001-02 and 2002-03, respectively. These reductions are similar to absolute differences reported by Dry (1995) for partial rootzone drying of field-grown Cabernet Sauvignon vines. The reductions in A and T may be attributed directly to stomatal limitation, as the decrease trends in gs by PRD relative to SD in 2001-02 and 2002-03 were similar. There was a curvilinear relationship found between gs and A (R2 = 0.748, p<0.001). Also, the rates of A and T were greater in season 2002-03 compared to the previous season, which corresponded to increases in gs. The percent decrease of A relative to T was less for both seasons for PRD compared to SD. As a result, carbon fixation per unit of water transpired was improved by PRD relative to SD. Improved A/T of grapevines is important for the sustainability of Australia’s wine-grape industry, as water is a limited resource because of water restrictions being imposed on the River Murray.

Ci was not influenced by irrigation practice when seasonal results were pooled. This may suggest that stomatal limitation was not the only factor contributing to the reduction of A by

PRD. Stomatal closure and reduction in mesophyll CO2 availability have been shown to be the main factors responsible for reductions in A as soil water content decreases (Chaumont et al. 1997; Correia et al. 1995). However, other studies indicate that non-stomatal limitations can also reduce A, including photoinhibition (Powles 1984), photoinactivation of photosystem II (PSII) (Flexas et al. 2001) and inhibition of CO2 metabolism by reduction in ATP production and ribulose-1,5-bisphosphate (RuBP) regeneration (Maroco et al. 2002). The summer midday PAR of the Sunraysia region is often > 1800 µmol.m-2.sec-1, therefore photoinhibition may be expected as grapevines are illuminated with more light than can be safely dissipated by photo-chemical and non-radiative mechanisms. Given that the summer

Chapter 5: Physiological response to irrigation and pruning treatments 127 conditions of the Sunraysia region would be similar to those reported by Flexas et al. (1999), photoinactivation of approximately 40-50% of total PSII could be expected during the course of a 15-hour illumination day (Flexas et al. 2001).

5.6.2 Pruning Effect on Leaf Gas Exchange

The influence of pruning treatments on leaf gas exchange was minor compared to the effect of irrigation method on gs, A and T. MIN increased the rate of gs, A and T in season 2002 relative to SPUR and MECH. The increased photosynthetic capacity of MIN vines may be associated with the increased crop load in 2001-02 (refer to section 3.3.4). To sustain the higher yield and produce sufficient photoassimilates to ripen the crop, MIN vines may have increased the rate of A by increasing gs and internal CO2 metabolism. Wünche et al. (2000) showed that a positive linear trend existed between whole canopy gas exchange per unit leaf area and crop load for ‘Braeburn’ apples. However, Poni et al. (1994) found no effect of crop load on leaf gas exchange of field grown Concord grapevines.

Pruning treatments had no effect on leaf gas exchange in the following season (2002-03). These results support previous studies, which have shown no statistical differences in seasonal leaf gas exchange between SPUR and MIN grapevines (Downton and Grant 1992; Sommer and Clingeleffer 1993; Lakso et al. 1996; Poni et al. 2000; Intrieri et al. 2001). Downton and Grant (1992) and Poni et al. (2000) found that leaf gas exchange for MIN vines was higher between budburst and flowering when canopies developed rapidly compared to SPUR vines, but after flowering, vines of both treatments had similar photosynthetic rates. Sommer and Clingeleffer (1993) suggested that the higher yield levels of MIN vines relative to MECH and cane pruned vines were the result of larger canopy size and therefore, higher net photosynthetic capacity rather than increased photosynthetic capacity per unit leaf area.

5.6.3 Irrigation and Pruning Effect on Leaf Gas Exchange

Midday leaf gas exchange was affected by the interaction between irrigation and pruning treatments, in conjunction with seasonal variation. In season 2001-02, an interesting integrated response was observed on gs. PRD irrigation significantly decreased the rate of

Chapter 5: Physiological response to irrigation and pruning treatments 128

gs, A and T on SPUR and MECH vines but had no significant effect on MIN vines. This suggests that SPUR and MECH vines were more sensitive to the water stress induced by the dry rootzone of PRD and thus decreased the rate of gs to minimise water loss by T. The increased physiological sensitivity of severely pruned vines, in response to PRD contrasts to results from previous studies conducted by Lasko et al. (1994), who found no interaction between crop level and late season water stress on vine physiology (gs, A and ΨL). However, they did observe a reduction in gs and A by applying a late season water deficit.

Stomatal conductance was not significantly affected by the interaction between irrigation and pruning in the following season (2002-03). However, a large decrease in photosynthetic capacity was observed on PRD + MECH vines because of the large reduction in gs relative to the other PRD treatments. The stomatal limitation by PRD on transpiration was greater than on photosynthesis, particularly when combined with SPUR vines and as a result, transpiration efficiency was improved. Again, this may have been in response to the greater physiological sensitivity of SPUR vines to the decreased water inputs of PRD. Ci was also reduced by PRD + SPUR, which corresponds to the greater stomatal limitation observed for this treatment.

The greater stomatal sensitivity of SPUR and MECH vines to PRD relative to MIN vines may be associated with canopy morphology (Downton and Grant 1992; Sommer et al. 1993; Intrieri et al. 2001), leaf morphology (Sommer and Clingeleffer 1993; Syvertsen et al. 1995; Palliotti et al. 2000; Poni et al. 2000) and leaf age (Patakas and Noitsakis 2001). Higher whole canopy photosynthesis and respiration of MIN compared to SPUR pruned ‘Concord’ grapevines have been attributed to the early spring development of MIN canopies (Lasko et al. 1996). Also, the ratio of exposed to shaded leaves may be greater on MIN vines due to restricted growth (Sommer and Clingeleffer 1993), which has positive implications for whole canopy physiology because sun-lit leaves have greater photosynthetic capacity (Palliotti et al. 2000). The larger leaves of the SPUR vines may have required a higher plant water status to maintain cell turgor and consequently gs. Whilst, leaves from MIN vines may have been better adapted to water-stress, in terms of partitioning carbohydrates between more sinks, small leaf area, increased leaf thickness and/or reduced susceptibility to photoinhibition due to a higher relative leaf age (Bertamini et al. 2003). Relative leaf age is greater on MIN than SPUR vines, as a result of the earlier spring canopy and leaf

Chapter 5: Physiological response to irrigation and pruning treatments 129 development (Intrieri et al. 2001). Patakas and Noitsakis (2001) found mature leaves of potted Vitis vinifera were less susceptible to water stress than immature leaves, in terms of the photosynthetic rate. When PRD was applied in late November of each growing season, the relative age of leaves on SPUR vines was less than those on MIN vines due to delayed canopy development and therefore possibly more sensitive to reduced water inputs. However, other studies on field grown vines have shown photosynthetic capacity may decrease with increased leaf age (Kriedemann 1968; Schultz 1996; Zuffery et al. 2000), which may suggest leaf morphology may have had a greater influence on stomatal sensitivity than leaf age in this study.

5.6.4 Leaf Gas Exchange at Different Phenological Stages

Grapevine photosynthetic capacity was assessed at midday between veraison and harvest to determine the integrated effects of PRD irrigation with the three pruning treatments, SPUR, MECH and MIN. VPD ranged between 2.5 and 4 kPa. Relative gas exchange results were not affected by variation in VPD but a large rainfall event (58 mm) occurred at veraison 2003 and increased the subsurface volumetric soil water content. This may have accounted for the higher rates of gas exchange at 56, 57 and 59 DAF. Subsurface volumetric soil water content was also increased by the higher irrigation inputs of SD compared to PRD. Differences in the physiological response and plant water status were expected given the difference in soil water content between irrigation strategies.

The photosynthetic capacity of the vines, as determined by gas exchange, was greater at post-veraison than at harvest. The higher rates of gs, A and T may be attributed to leaf ontogeny, vine phenology or subsurface volumetric soil water content. Poni et al. (2001) showed a decline in A associated with increased leaf age. However, the effects of leaf ontogeny on grapevine photosynthesis do not appear to be consistent. Patakas and Noitsakis (2001) reported higher rates of A for mature leaves under both well-watered and water- stressed conditions. The phenological stage of berry development may also affect carbohydrate production by photosynthesis. At veraison and harvest, berries are the predominant sinks for carbohydrates and increased photosynthesis may be associated with rapid rates of sugar accumulation. Alternatively, the increased subsurface soil water content may have increased plant water status and consequently increased gs, A and T. Several

Chapter 5: Physiological response to irrigation and pruning treatments 130

studies have attributed decreased rates of gs, A and T to reduced soil water content and water-stress (Naor and Wample 1994; Poni et al. 1994; McCarthy 1995).

The application of PRD did not influence gs, A or T at post-veraison when subsurface soil water content was relatively high as a result of the rainfall event 11 days prior to measurements. The high soil water content may have increased plant water status of both SD and PRD irrigated vines and increased photosynthetic activity at a leaf level in response to greater gs. Subsequent drying of the subsurface soil profile at around 59 days after flowering caused PRD to have lower gas exchange. The corresponding decrease in A and T might have been associated with partial stomata closure. The stomatal limitation on T was greater than on A. As a result, A/T was improved for PRD irrigated vines. PRD continued to have a large effect on gas exchange until harvest, reducing gs, A and T and increasing

A/T. The decrease in gs, A and T with PRD irrigated vines 95 DAF was associated with an increase in Ci. This suggests that there was partial closure of stomata and a build-up of intercellular CO2. Water-stress may have affected measurements just prior to harvest, as all gas exchange parameters were reduced by PRD, resulting in a reduction in photosynthetic activity.

The effects of pruning practice on leaf gas exchange between post-veraison and harvest were inconsistent. MIN increased the rate of A and subsequently transpiration efficiency on days

56 and 95 after flowering with no associated changes in gs, T or Ci. This suggests that there was an increase in photosynthetic activity (i.e. increased CO2 metabolism). Similarly, SPUR increased A and A/T at 59 DAF, probably as a result of increased photosynthetic activity. These results indicate that pruning has an influence on the carboxylation efficiency of the leaf as opposed to the stomatal effects observed by the imposition of PRD. Carboxylation efficiency may have been increased by greater mesophyll CO2 availability due to different leaf anatomy, increased CO2 metabolism and resistance or tolerance to photoinhibition (Powles 1984; Maroco et al. 2002).

The effects of the interaction between irrigation and pruning on leaf gas exchange were influenced by the stomatal limitations of PRD and carboxylation efficiency of the three pruning treatments from veraison through to harvest. As a result of the variation in pruning treatment effects on measurement days, no integrated trends were apparent. Stomatal

Chapter 5: Physiological response to irrigation and pruning treatments 131 conductance was significantly affected (P<0.05) by the irrigation and pruning interaction on

57, 91, 95 DAF. PRD reduced gs, relative to SD irrigated treatments, and the pruning treatment effects on gs correspond to those explained in the previous paragraph. The reduction of gs by PRD may have been triggered by an ABA response to decreasing soil water content in the dry half of the rootzone (Loveys 1984). The rates of A and T were not directly associated with gs, probably as a consequence of the difference in carboxylation efficiency of the pruning treatments.

5.6.5 Irrigation and Pruning Effects on Carbon Isotope Discrimination

Carbon isotope discrimination (∆) has been used as an indirect measure of plant photosynthetic capacity on a range of crops in response to the ratio between intercellular

CO2 and atmospheric CO2 partial pressures (Farquhar et al. 1982, 1989). PRD reduced ∆ in the three experimental seasons by 2-3%, which suggests a minor decrease in CO2 fixation due to either increased stomatal limitation or decreased carboxylation efficiency of PRD irrigated vines. However reduced ∆ is associated with improved water use efficiency in grapevines, as a result of greater stomatal limitation on T than on A (Gibberd et al. 2002). Therefore, PRD improved the water use efficiency of Shiraz grapevines in the Sunraysia region, as determined by both ∆ and leaf gas exchange.

MIN reduced ∆ in seasons 2001-02 and 2002-03. Hence, this treatment reduced the photosynthetic capacity of the vines at a leaf level. This may be attributed to greater partitioning of photoassimilates between a greater number of leaves, differences in leaf morphology and/or leaf ontogeny. Downton and Grant (1992) observed that MIN vines had greater photosynthetic rates than SPUR vines between budburst and flowering when canopies develop rapidly due to retention of high bud numbers at winter pruning. However, after flowering there were no differences in photosynthetic rates and at harvest, SPUR pruned vines had greater photosynthetic capacity at the leaf level. This reduction in photosynthetic capacity by MIN vines as the season progressed may explain the reduction in ∆ observed in this study.

Chapter 5: Physiological response to irrigation and pruning treatments 132

The reduction in ∆ by PRD dominated the interaction between irrigation and pruning in seasons 2000-01 and 2001-02. However, SPUR and MECH vines were more sensitive than MIN to PRD in 2002-03. As a result, greater reductions in ∆ and leaf gas exchange were found for SPUR and MECH when irrigated by PRD. No relationships were found between yield and ∆ when treatment means were pooled. This suggests that carbon isotope discrimination may have been lowered by increasing photosynthetic capacity whilst maintaining stomatal conductance on treatments with higher crop loads, as has been shown in peanuts (Nageswara Rao and Wright 1994). Alternatively, the reduction of yield associated with PRD would have produced a positive relationship between yield and ∆, whereas the high yields produced in conjunction with MIN may have resulted in a negative relationship between yield and ∆. As a consequence, when integrated treatment yields were pooled and plotted against ∆, the two opposing relationships may have cancelled each other out.

5.6.6 Diurnal response of PRD vines to increasing vapour pressure deficit

VPD is determined by temperature and relative humidity and given the low variation in humidity during summer in the Sunraysia region (see section 2.3); VPD was dominated by diurnal increases in temperature. Stomata opened early in the morning as PAR approached -1 -2 1200 µmol.s .m as shown by a rapid increase in gs and A. The rate of gs and A stabilised mid morning (approximately 11:00 AEDT) as VPD reached its maximum and then slowly declined as VPD decreased in the late afternoon (17:00 AEDT). Similarly, T was reduced in conjunction with VPD, gs and A in the late afternoon. However, the diurnal increase of T to the maximum rate was more gradual than the increase of gs or A. An initial decrease in Ci was associated with the opening of stomatal aperture and loss of CO2 stored in the mesophyll tissue to atmosphere. Levels of Ci then stabilised for SD and PRD irrigation treatments between 200-250 µmol.mol-1. Leaf water potential is accepted as an indicator of plant water status (Kramer 1983) and has been used extensively in viticultural research

(Smart and Coombe 1983). A maximum ΨL was observed at pre-dawn as it approached equilibrium with soil water potential, then decreased rapidly to a minimum after midday followed by a gradual recovery during the late afternoon. The diurnal fluctuation of ΨL in response to VPD agrees with previous diurnal studies conducted on grapevines (Smart and Coombe 1983; van Zyl 1987; Dry 1997). Interestingly, whilst diurnal leaf water potential

Chapter 5: Physiological response to irrigation and pruning treatments 133 decreased to very low levels as VPD increased (-1.8 MPa on the 24/1/02), assimilation rates remained relatively high (>15 µmol.m-2.s-1). Schultz (2003) has shown significant reductions in gas exchange as leaf water potential decreases for cv. Shiraz of mesic origin. However, grapevine physiological experimentation conducted on V. vinifera cv. Cabernet sauvignon also at the Deakin Estate vineyard has produced comparable leaf water potentials and assimilation rates under similar VPD conditions (N. Cooley, pers. comm.). The high photosynthetic capacity of the experimental grapevines, which are of a more xeric origin may be associated with the high levels of PAR and sunlight hours of the experimental site with combined with irrigation. The Shiraz grapevines of this study also have large canopies relative to grapevines of mesic or hydric origins, thus net photosynthetic capacity is expected to be greater.

Diurnally, PRD reduced gs compared to SD irrigation, particularly as VPD increased. As a result, PRD applied a stomatal limitation by partial stomata closure on A and T during the day. The relationship between gs and A was established in section 5.3.1 and agrees with previous work conducted on field-grown grapevines under a range of climatic conditions (Jacobs et al. 1996). These results suggest that the reduction in A and T may be influenced by partial stomatal closure. However, a recent study by de Souza et al. (2003) has also highlighted the importance of decreased carboxylation efficiency (electron transport rate and triose-phosphate utilization) to leaf gas exchange reduction by PRD when compared to full irrigation. In addition to the reduction in gs of PRD irrigated vines in this study, PRD maintained ΨL and ΨS compared to SD irrigated vines. The lack of difference in ΨL between PRD and SD and the absence of osmotic adjustment by PRD irrigated vines may be attributed to irrigation of the wet half of the rootzone.

Given that PRD applied only 50% of the amount of irrigation water compared to SD over the season, yet plant water status was maintained, suggests vines were not displaying a physiological stress response. This differentiates PRD from deficit irrigation practices, such as reduced deficit irrigation (RDI) because the latter typically results in stomatal limitation as well as reduced plant water status (de Souza et al. 2003; McCarthy 1997a). Since plant water status was not altered by PRD, no yield loss should be expected. However, an average 18% decrease in yield, as a result of reduced berry size associated with PRD was observed over the three seasons in the warm region of Sunraysia (section 3.3.3). The reduced

Chapter 5: Physiological response to irrigation and pruning treatments 134 irrigation inputs of PRD compared to SD may have imposed a water stress during the critical stages of berry development, when ΨL was not measured. The effects of post-flowering water stress on berry size is well established (Goodwin and Jerie 1992; Poni et al. 1993; McCarthy 1997a). However, PRD successfully maintained hydraulic water status throughout the remainder of the growing season and no significant differences in soil moisture content between irrigation treatments during berry development were observed. This suggests that an alternative mechanism for berry reduction by PRD may have occurred pre-veraison. PRD may have induced a pre-veraison hormonal and/or chemical response early berry development, possibly ABA. The induced hormonal/chemical synthesis may have reduced pericarp volume and restricted berry expansion by limiting flexibility and enlargement of pericarp cells, as previously shown by Ojeda et al. (1999, 2001).

5.6.7 Proposed mechanisms for Stomatal Closure by PRD

The closure of stomata under water deficit conditions is controlled by hydraulic and/or non- hydraulic (chemical signalling) mechanisms. Hydraulic signalling occurs by changes in xylem sap tension to reduce plant water status and stomatal conductance in response to soil water deficits (Dodd et al. 1996). However, if plant water status is maintained and stomatal conductance is lowered by partial closure of stomata, then it is likely that synthesis of inhibitory chemical signals activates the movement of solutes out of guard cells and the subsequent loss of turgor to close stomata. The partial stomatal closure associated with PRD irrigation coincides with drying of the soil around part of the root system and maintenance of leaf water potential (Dry and Loveys 1999; Lovisolo et al. 2002; Stoll et al. 2000). This suggests drying roots produce a “positive” signal, such as abscisic acid (ABA), which increases in concentration in the xylem sap as the soil dries. Root tips can detect small changes in soil water content and generate ABA, which acts as a growth inhibitor and stress indicator (Zhang and Davis 1991). Accumulation of ABA in the leaves has the capacity to trigger the exodus of potassium ions (K+) and osmotic loss of water from the guard cells. The guard cells than become flaccid and reduce the stomatal aperture, subsequently restricting stomatal conductance and water loss through transpiration. Dry (1997) showed that a reduction in gs in response to partial drying of the root system of grapevines was associated with an increase in the concentration of ABA in xylem sap.

Chapter 5: Physiological response to irrigation and pruning treatments 135

5.6.8 Single Leaf Gas Exchange Measurements

The determination of the photosynthetic capacity of a vine can be conducted on a leaf or whole vine scale. Single-leaf photosynthetic response does not always reflect the response of the whole canopy, as leaf responses vary with age, position and orientation. Intrieri et al. (1997) found that whole canopy net A was one third that of an individual leaf under saturating light intensity. Hence, single-leaf A measurements are not indicative of whole canopy responses and calculations of vine A on a leaf area basis should consider the phenology of the vine because carbon assimilation varies throughout the growing season (Howell 2001). However, assessments of single-leaf A have been used extensively for comparative studies of relative treatment values (Naor and Wample 1994; Poni et al. 1994; Gomez-del-Campo et al. 2002); because the determination of whole-vine canopy responses using canopy chambers can be very difficult.

5.7 CONCLUSIONS

a) PRD irrigation reduced leaf gas exchange (gs, A and T) by partial closure of stomata but

maintained plant water status as determined by ΨL. Transpiration efficiency was improved by PRD because the stomatal limitation on T was greater than on A.

b) PRD reduced ∆ in the three experimental seasons by 2-3%, which suggests a small

decrease in CO2 fixation due to increased stomatal limitation. Reduced ∆ is associated with improved WUE in grapevines (Gibberd et al. 2002). Therefore, PRD improved the WUE of cv. Shiraz grapevines in the Sunraysia region, as determined by both ∆ and leaf gas exchange.

c) The level of pruning had minimal effect on grapevine leaf physiology, which supports previous studies conducted by Sommer and Clingeleffer (1993). Minor gas exchange effects showed that pruning level influenced carboxylation efficiency and not stomatal limitations, as photosynthesis was not directly correlated to stomatal conductance.

Chapter 5: Physiological response to irrigation and pruning treatments 136 d) MIN reduced ∆ in two seasons, which indicates a reduction in the photosynthetic capacity of the vines. This may be attributed to greater partitioning of photoassimilates between greater numbers of leaves.

e) The effects of the interaction between irrigation and pruning on leaf gas exchange from veraison through to harvest were influenced by the stomatal limitations of PRD and carboxylation efficiency of the three pruning treatments.

f) Midday leaf gas exchange was affected by the interaction between irrigation and pruning treatments in season 2001-02. PRD irrigation significantly decreased the rate of

gs, A and T on SPUR and MECH vines but had no significant effect on MIN vines.

g) Diurnally, PRD reduced gs, A and T and maintained hydraulic water status compared to

SD irrigation as VPD increased. The lack of difference in ΨL between PRD and SD and

absence of osmotic adjustment by PRD irrigated vines may be attributed to irrigation of the wet half of the rootzone.

h) The hypothesis that states “the physiological response of cv. Shiraz will be negatively affected by the interaction between PRD and light pruning levels, as it nears the critical ‘source:sink’ relationship” was rejected, since SPUR and MECH vines were more

sensitive to PRD than MIN, in terms of relative reductions in gs, A and T. The sensitivity of severe pruning levels with PRD illustrates the importance of holistic viticultural evaluation.

Chapter 6: Botrytis bunch rot development and bunch architecture 137

6 BOTRYTIS BUNCH ROT DEVELOPMENT AND BUNCH ARCHITECTURE

6.1 INTRODUCTION AND EXPERIMENTAL AIMS

Botrytis cinerea is an important pathogen of winegrapes that may lead to Botrytis bunch rot. B. cinerea can be saprophytic using nutrients from dead or dying plant tissue (Agrios 1997). Latent infection by B. cinerea on grapevines can occur in the receptacle area or cap scar at anthesis up to four months before visible expression of the disease (McClellan and Hewitt 1973; Kellar et al. 200;). Colonisation of loose floral debris within bunches by B. cinerea has also been reported (Bulit and Lafon 1977 cited in Savage and Sall 1984). Spore germination and infection of berries after veraison by B. cinerea can occur during wet periods, ie. rain events or dew. All infection modes are influenced by bunch and canopy microclimate. Botrytis bunch rot can cause substantial yield losses in all wine grape varieties by rotting flower clusters or immature berries or by rotting and desiccation of mature berries (Emmett and Nair 1991). Furthermore, B. cinerea can negatively influence berry composition, vinification of musts, protein stability of wine and increase oxidative breakdown of red wines by laccase activity (Somers 1984).

B. cinerea can be controlled in the vineyard by fungicide application, sanitation and canopy management. Sources of B. cinerea inoculum within the vines can be reduced by sanitation practices, such as removing dead canes and mummified bunches at winter pruning. Mechanisation of harvest and pruning can increase the amount of dead plant material in the vine, which can lead to increased inoculum loads in the vine (Emmett and Nair 1991). Canopy management practices, including pruning level, trellising, bunch thinning and leaf removal may alter canopy microclimate. Increased aeration and reduced humidity in the fruiting zone may reduce the development of Botrytis and other diseases, such as powdery mildew (Carroll and Wilcox 2003). Savage and Sall (1983, 1984) reported that mid-season hedging and wire trellising of Chenin Blanc grapevines resulted in a moderate reduction of Botrytis bunch rot incidence and severity. Light pruning practices, such as minimal pruning have been associated with lower incidence and severity of Botrytis bunch rot in Chardonnay (Emmett et al. 1995). Post-fruit set bunch thinning reduced Botrytis bunch rot on Seyval Blanc grapevines (Smithyman et al. 1991). Gubler et al. (1987) reported Botrytis bunch rot of grapevines can be controlled by removal of leaves from nodes adjacent to the bunch. It

Chapter 6: Botrytis bunch rot development and bunch architecture 138 has been hypothesised that the reduction in Botrytis bunch rot by canopy management is linked to differences in canopy microclimate, characterised by temperature, atmospheric humidity, wind speed and leaf wetness. However, there is only circumstantial evidence in the literature to support this hypothesis (Savage and Sall 1984; English et al. 1989).

Bunch architecture also influences the susceptibility of bunches to Botrytis bunch rot. Cultivars most susceptible to Botrytis bunch rot have compact bunch architecture (Chenin Blanc, Zinfandel, Riesling and Sauvignon Blanc) (Savage and Sall 1984). The relationship between tight, compact bunches and increased Botrytis bunch rot may be attributed to the presence of free water on the berries. Free water is required for the germination of spores of B. cinerea (Jarvis 1980). Compact bunches exposed to rain or heavy dew may take longer to dry than loose bunches because of restricted air movement between berries and/or large, fully hydrated berries in compact bunches may also be more likely to rupture (Sall et al. 1982). The exocarp of a grape berry is its main defence against B. cinerea infection; the cuticle membrane and epicuticular wax provide a physical barrier against invasion by the pathogen. A study by Marois et al. (1985) showed that epicuticular wax played an important role in the resistance of grape berries to infection by B. cinerea. Epicuticular wax influences pesticide/fungicide retention, water retention, adhesive ability of plant pathogens (Baker 1982). Cultivars susceptible to B. cinerea infection have reduced cuticle and epicuticular wax accumulation (Rosenquist and Morrison 1989). Exposed, loose bunches have greater wax production associated with high light intensities and temperatures and low relative humidity (Marois et al. 1986; Percival et al. 1993). Thus, these bunches retain less water on their surfaces and dry faster (Vail and Marois 1991). Increased Botrytis infection in compact bunches may also be due to the alteration of the epicuticular wax, from well- defined platelets to an amorphous structure with shallow depressions, by berry contact within a grape bunch (Marois et al. 1986).

Integrated viticultural practices that manipulate bunch architecture and canopy microclimate have the potential to minimise Botrytis bunch rot development. The objective of this experiment was to determine the optimum integrated viticultural practice for cv. Shiraz that will reduce Botrytis bunch rot development in the field.

Chapter 6: Botrytis bunch rot development and bunch architecture 139

The aims of this study were to:

1) Evaluate the effects of inoculum spore concentration on Botrytis disease development in bunches.

2) Determine the susceptibility of bunches to B. cinerea infection at flowering, fruit-set, veraison and pre-harvest.

3) Assess the influence of integrated pruning and irrigation treatments on Botrytis disease expression at critical stages of bunch development (flowering, fruit-set, veraison and harvest)

4) Examine the effects of pruning and irrigation treatments on bunch architecture and the relationship between bunch architecture and Botrytis disease expression.

The following hypothesis was tested in this study:

The integration of lighter pruning with PRD will reduce the level of bunch infection and disease expression of B. cinerea, as a result of improved canopy microclimate (i.e. reduced shoot vigour and leaf area) and bunch architecture (i.e. small, loose, exposed bunches).

6.2 METHODOLOGY

6.2.1 Isolates of Botrytis cinerea

Isolates of Botrytis cinerea were collected from randomly selected mummified bunches on vines at the experimental site (Deakin Estate, Iraak, Victoria). Single spore isolations were conducted on potato dextrose agar (PDA) and incubated for 24 hours at 25°C. Single germinating spores were transferred onto PDA plates and the cultures were grown in constant temperature (25°C) until sporulation occurred (5-7 days). A collection of isolates was maintained on PDA slopes for sub-culturing for all inoculation experiments. PDA slopes prepared in sterile screw cap glass vials (McCartney bottles) were inoculated with pure B. cinerea isolates and allowed to grow. The colony and PDA slope was completely covered with sterile distilled water to stop the agar from drying out and stored at 4°C.

Chapter 6: Botrytis bunch rot development and bunch architecture 140

Sub-cultures from the isolate collection were prepared on PDA plates at 25°C, 10 days prior to field inoculations to produce spores for inoculum. Spore suspensions were prepared by washing sub-cultures with 20 mL of sterile distilled water and transferring the wash solution to a flask containing 200 mL sterile distilled water. Spores were dispersed in solution by gently shaking for 60 seconds. The concentration of each spore suspension was determined in 0.1 mm3 on 10 repeat sub-samples using an Improved Neubauer haemocytometer (Blaubrand®, Wertheim, Germany) and adjusted to a final concentration of 104 spores.mL-1. The viability of the spore suspensions was assessed by pipetting 50 µL of each spore solution onto five Petri dishes containing PDA. The Petri dishes were sealed for 36 hours at 25°C and the mean percentage of germinating spores determined by light microscopy.

6.2.2 Effects of Spore Concentration

The effect of inoculum spore concentration on Botrytis bunch rot incidence and severity was assessed in seasons 2001-02 and 2002-03. Twelve bunches were randomly selected and tagged on two vines from SD or PRD irrigation treatments combined with three pruning treatments (SPUR, MECH or MIN) (n = 144). Bunches at 80% capfall on the vine were sprayed with spore suspensions of B. cinerea of 3 concentrations: 0, 102 or 104 spores.mL-1. The bunches were enclosed in polyethylene bags containing 2 mL distilled water to create a humid environment for 12 hours to encourage Botrytis disease development. Bunches were left in situ to develop on the vines under ambient conditions.

At harvest (119 days after flowering commenced, DAF) inoculated bunches were picked and surface-sterilized (SS) to evaluate latent infection or non-surface sterilized (NS) to assess both surface and latent infection. Surface-sterilized bunches were immersed in 1% sodium hypochlorite solution for 2 minutes, rinsed once with tap water, rinsed 3 times with distilled water and air-dried on paper towel for 1 hour. All bunches (SS and NS) were stored in clean zip-lock bags and frozen for 12 hours at –20°C to encourage tissue degradation. The bunches were incubated for 7 days at 26°C in zip-lock bags containing damp paper towel to produce a humid environment to promote Botrytis development and sporulation. Botrytis incidence was assessed as the percentage of bunches with sporulating conidiophores of B. cinerea and severity was assessed as the number of sites in bunches where there were diseased tissues with sporulating conidiophores of B. cinerea. As expression of B. cinerea

Chapter 6: Botrytis bunch rot development and bunch architecture 141 was predominantly on berries within a bunch, severity will be referred to as the number of berries infected per bunch throughout the remainder of this chapter.

6.2.3 Botrytis Field Inoculations at Flowering

Early season inoculations were conducted at 80% capfall (10 DAF) in seasons 2001-02 and 2002-03. Thirty-two bunches were randomly selected and tagged on two vines from SD or PRD irrigation treatments combined with three pruning treatments (SPUR, MECH or MIN) (n = 384). Bunches on the vine were wet-inoculated with a B. cinerea spore suspension (104 spores.mL-1) or a control treatment (sterile distilled water). Bunches were enclosed in a humid environment for 12 hours and then left to develop on the vines in situ under ambient conditions.

Bunches were assessed for the incidence and severity of Botrytis disease at four stages of bunch development in both seasons: flowering (17 DAF), fruit-set (28 DAF), veraison (77 DAF) and harvest (119 DAF). At each stage, inoculated bunches were picked and either SS or NS as described in section 6.2.2. All bunches were frozen for 12 hours at –20°C and moist incubated for 7 days at 26°C. Incidence and severity of expressed Botrytis bunch rot were assessed as explained in section 6.2.2.

6.3.4 Late season Botrytis Field Inoculations

Bunches from each combined irrigation and pruning treatment were inoculated with B. cinerea at flowering, veraison and pre-harvest to determine the susceptibility of bunches to Botrytis bunch rot throughout the season. Bunches were randomly selected and inoculated with a B. cinerea spore suspension (104 spores.mL-1) or sprayed with sterile distilled water as a control treatment at flowering (17 DAF), veraison (77 DAF) or pre-harvest (114 DAF) and assessed at harvest (119 DAF) in seasons 2001-02 and 2002-03.

In season 2001-02, two randomly selected bunches were tagged on two vines from each integrated irrigation and pruning treatment and sprayed with the B. cinerea spore suspension or the control treatment (n= 144). All bunches were picked at harvest, frozen for 12 hours at

Chapter 6: Botrytis bunch rot development and bunch architecture 142

–20°C and moist incubated for 7 days at 26°C. Incidence and severity of Botrytis bunch rot was assessed as described in section 6.2.2.

In season 2002-03, the experiment was extended to assess the effects of latent and trash infection, in addition to assessing the susceptibility of bunches to Botrytis bunch rot. Four randomly selected bunches were tagged on 2 vines for each integrated irrigation and pruning treatment and sprayed with the B. cinerea spore suspension or control treatment (n= 288). Bunches were picked at harvest and surface sterilized (SS) or non-surface sterilized (NS) as described in section 6.2.2. All bunches were frozen and moist incubated as in the previous season. Incidence and severity of Botrytis bunch rot were assessed as described in section 6.2.2.

6.2.5 Bunch Architecture

The influence of imposed irrigation and pruning treatments on bunch architecture and subsequent Botrytis bunch rot development was assessed. Attributes of bunch architecture for integrated irrigation and pruning treatments were measured at harvest in seasons 2001-02 and 2002-03. Bunch architecture was characterized by bunch size (weight, volume, length and compactness), by berry parameters (weight and average number.bunch-1) and in relation to yield parameters (bunch number.vine-1 and kg.vine-1). Bunch volume was measured by displacement of water (cm3) and bunch compactness was calculated as bunch weight divided by bunch length (g.cm-1).

6.2.6 Statistical Analysis

Botrytis incidence and severity were analysed by an Analysis of Variance (ANOVA) using GenStat® 6th Edition (Lawes Agricultural Trust, Rothamsted Experimental Station). All interactions were balanced and analysed by general ANOVA; treatment structure of irrigation * pruning * spore load * sterilisation and no blocking structure was implemented. Seasonal variation in bunch architecture was also analysed using ANOVA. The interactions were balanced and analysed by general ANOVA; treatment structure of irrigation * pruning * season and no blocking structure was implemented.

Chapter 6: Botrytis bunch rot development and bunch architecture 143

6.3 RESULTS

6.3.1 Effect of Spore Concentration on Incidence and Severity of Botrytis cinerea

The concentration of the spore suspension used to inoculate bunches at flowering did not influence incidence or severity of Botrytis over all treatments, when bunches were assessed at harvest in either 2002 or 2003. Botrytis incidence did not differ significantly between bunches inoculated with 0, 102 or 104 spores.mL-1 (Table 6.1).

2001-02 2002-03

Spore Suspension Incidence Severity Incidence Severity -1 -1 Concentration (% Infected Bunches) (Infected berries bunch ) (% Infected Bunches) (Infected berries bunch ) 0 Spores.mL-1 70.8 2.77 58.3 1.25 102 Spores.mL-1 87.5 3.59 50.0 1.10 104 Spores.mL-1 71.5 2.94 58.3 1.56 P value ns ns ns ns LSD 17.1 1.17 19.1 0.55

6.3.2 Botrytis Field Inoculations at Flowering

The incidence of Botrytis bunch rot was influenced by the stage of bunch development. Mean incidence was highest on bunches assessed at harvest in seasons 2001-02 and 2002-03 (Figure 6.1). Mean incidence was significantly lower at fruit-set and veraison in 2002 and at flowering and veraison in 2003. Similarly, mean Botrytis severity was higher at harvest in both seasons when bunches were fully mature. Mean Botrytis severity in bunches was lowest at veraison in season 2002 and at flowering in season 2003.

Bunches inoculated with a B. cinerea spore suspension of 104 spores.mL-1 had a general trend towards higher incidence than control bunches at flowering and fruit-set in both seasons. However, the increase in B. cinerea incidence on inoculated bunches was only significant at flowering in 2001-02 (4.8-fold) and fruit-set in 2001-02 (4.5-fold) and fruit-set in 2002-03 (2.0-fold) (Table 6.2 and 6.3). Levels of Botrytis incidence were very low for

Chapter 6: Botrytis bunch rot development and bunch architecture 144 both inoculated and control bunches at veraison in both seasons and no differences were found.

Generally, Botrytis severity was also higher on inoculated bunches compared to control bunches in both experimental seasons at all stages of bunch development. Botrytis severity was significantly higher on inoculated bunches at flowering (11.9-fold) and fruit-set (7.8- fold) in 2001-02 and at fruit-set (2.5-fold) and harvest (1.6-fold) in 2002-03 (Table 6.4 and 6.5).

100 100

Incidence 2002 Incidence 2003 Incidence I nc

s) 80 80 c id en nche LSD = 9.34 LSD = 10.6 c c e

(% In

bu 60 60 (%

I cted n fected

e f f ec t In e d bunc

40 40 b b un nce (%

b c hes) h de es

20 a 20 ) In a a a 0 0

3.5 Severity 2003 3.5

Severity 2002 Severity Seve )

-1 c 3.0 3.0 r nch i LSD = 0.54 LSD = 0.29 ty

(Infect (In

s bu 2.5 2.5 e i f ec rr e te e d B 2.0 2.0 d Berries

d be e r r

c i 1.5 1.5 e

b s

bun (Infect bunch ty ri

1.0 1.0 ch e

v b e -1 ) S

ab ) 0.5 0.5 a a a 0.0 0.0 Flowering Fruit-Set Veraison Harvest Flowering Fruit-Set Veraison Harvest

Phenological Stage Phenological Stage

Chapter 6: Botrytis bunch rot development and bunch architecture 145

The mean incidence and severity of expressed latent infection of B. cinerea was assessed on SS bunches and the mean incidence and severity of expressed latent plus surface infection was determined on NS bunches. Generally, Botrytis incidence and severity of infection was higher on NS bunches than SS bunches at flowering, fruit-set and harvest in both seasons (Tables 6.2 to 6.5). Again, the levels of Botrytis incidence and severity were very low at veraison and there were no differences between latent and surface infection.

The interaction between irrigation (SD or PRD) and pruning (SPUR, MECH or MIN) had no effect on the mean Botrytis incidence (Tables 6.2 and 6.3) or severity (Tables 6.3 and 6.4) at the four stages of bunch development. However, at harvest in season 2001-02 Botrytis incidence was higher on treatments that produced larger bunches (i.e. SD + SPUR, PRD + SPUR and SD + MECH) than on treatments that produced smaller, looser bunches (i.e. PRD + MECH, SD + MIN and PRD + MIN) (refer to Table 6.11 and Figure 6.6). This trend was not observed in the following season and will be discussed in section 6.3.5. A significant interaction between irrigation and pruning treatments was found at veraison 2002-03, since PRD + MECH had a higher Botrytis incidence than the other imposed treatments.

Botrytis Incidence (% Infected Bunches) in 2001-02 Treatments Flowering Fruit-set Veraison Harvest 0 spores.mL-1 8.3a 4.2a 2.1 70.8 104 spores.mL-1 39.6b 18.8b 2.1 71.5 P value *** * ns ns NS 43.8b 20.8b 0.0 81.2b SS 4.2a 2.1a 4.2 61.1a P value *** ** ns * SD + SPUR 18.7 6.2 0.0 93.8 PRD + SPUR 31.2 6.2 0.0 87.5 SD + MECH 18.7 12.5 0.0 81.2 PRD + MECH 31.2 6.2 6.2 56.3 SD + MIN 31.2 25.0 6.3 58.3 PRD + MIN 12.5 12.5 0.0 50.0 P value ns ns ns ns

Chapter 6: Botrytis bunch rot development and bunch architecture 146

Botrytis Incidence (% Infected Bunches) in 2002-03 Treatments Flowering Fruit-set Veraison Harvest 0 spores.mL-1 2.1 18.8a 8.3 52.1 104 spores.mL-1 10.4 38.2b 16.7 70.8 P value ns * ns ns NS 12.5b 34.0 6.3a 56.2 SS 0.0a 22.9 18.8b 66.7 P value * ns * ns SD + SPUR 12.5 37.5 12.5a 56.2 PRD + SPUR 6.2 25.0 12.5a 56.2 SD + MECH 0.0 20.8 6.2a 62.5 PRD + MECH 6.2 18.7 37.5b 81.2 SD + MIN 0.0 25.0 0.0a 50.0 PRD + MIN 12.5 43.7 6.2a 62.5 P value ns ns ** ns

Botrytis Severity (number of infected berries bunch-1) in 2001-02 Treatments Flowering Fruit-set Veraison Harvest 0 spores.mL-1 0.17a 0.04a 0.06 2.77 104 spores.mL-1 2.02b 0.31b 0.04 2.94 P value *** ** ns ns NS 2.15b 0.33b 0.00 3.73 SS 0.04a 0.02a 0.10 1.99 P value *** ** ns ** SD + SPUR 0.62 0.13 0.00 5.06 PRD + SPUR 1.56 0.13 0.00 4.75 SD + MECH 1.00 0.19 0.00 3.12 PRD + MECH 1.56 0.06 0.19 1.37 SD + MIN 1.19 0.38 0.13 1.52 PRD + MIN 0.63 0.19 0.00 1.31 P value ns ns ns ns

Chapter 6: Botrytis bunch rot development and bunch architecture 147

-1 Botrytis Severity (number of infected berries bunch ) in 2002-03 Treatments Flowering Fruit-set Veraison Harvest 0 spores.mL-1 0.02 0.35a 0.19 0.98a 104 spores.mL-1 0.19 0.89b 0.56 1.54b P value ns * ns * NS 0.21b 0.99b 0.17 1.06a SS 0.00a 0.25a 0.58 1.46b P value * ** ns * SD + SPUR 0.19 0.69 0.56 1.25 PRD + SPUR 0.06 0.25 0.31 1.01 SD + MECH 0.00 0.60 0.19 1.19 PRD + MECH 0.19 0.94 1.00 2.06 SD + MIN 0.00 0.63 0.00 0.75 PRD + MIN 0.19 0.63 0.19 1.50 P value ns ns ns ns

The relationship between Botrytis incidence, imposed irrigation and pruning treatments, stage of bunch development and latent infection was evaluated as a factorial experiment in seasons 2001-02 and 2002-03. No significant differences were found between parameters in 2002 (Figure 6.2), as trends at each phenological stage were similar and no interactions occurred. However, incidence was higher on bunches at harvest, particularly on irrigation and pruning treatments that produced large bunches (i.e. SD + SPUR, PRD + SPUR and SD + MECH). Also, incidence was higher on NS bunches at most phenological stages with the exception of veraison. The interaction between irrigation and pruning treatments only influenced Botrytis incidence on NS bunches at harvest. The interaction between incidence, imposed irrigation treatment, phenological stage and latent infection was significant in 2002-03 (Figure 6.3). Incidence of Botrytis was higher in bunches of all irrigation and pruning treatment combinations at harvest. Generally, incidence was higher on SS bunches at harvest, which indicated that latent infection was predominant in season 2002-03, particularly on PRD + MIN treatments. A corresponding high level of latent infection of B. cinerea on PRD + MIN was observed at fruit-set. Incidence was also high on SS bunches from PRD + MECH vines at veraison. This may account for the 100% rate of disease expression in bunches from PRD + MECH vines at harvest.

Chapter 6: Botrytis bunch rot development and bunch architecture 148

100% ns LSD=32.7 90%

80%

70%

60% Incidence %

50%

40%

30%

20%

SD+spur 10% PRD+spur SD+mech Treatment PRD+mech 0% SD+min SS NS PRD+min SS NS NS Harvest NS SS Veraison SS Fruit set Flowering Phenological Stage

100% *** LSD=36.8 90%

80%

70%

60% Incidence %

50%

40%

30%

20%

SD+spur 10% PRD+spur SD+mech Treatment PRD+mech 0% SD+min SS NS PRD+min SS NS SS NS Harvest SS NS Veraison Fruit set Flowering Phenological Stage

Chapter 6: Botrytis bunch rot development and bunch architecture 149

The factorial interaction between severity, imposed irrigation and pruning treatments, phenological stage and latent infection was also assessed in seasons 2001-02 and 2002-03. No significant interactions were observed in 2001-02 (Figure 6.4), as severity was greater on NS bunches on the majority of treatments at each phenological stage, particularly at harvest. Bunches from SPUR-pruned vines were most susceptible to B. cinerea infection at harvest. This may be associated with bunch architecture. The severity of Botrytis was lower in the following season and the interaction was significant (Figure 6.5). The severity of Botrytis was high at fruit-set on NS bunches and at harvest on both SS and NS bunches. Again, PRD + MECH had the highest level of B. cinerea at fruit-set, veraison and harvest, which reflects the high percentage of incidence found.

10.00 ns LSD=1.90 9.00

8.00 Severity (Infected berries bunch 7.00

6.00

5.00

4.00

3.00

-1 ) 2.00

SD+spur 1.00 PRD+spur SD+mech Treatment PRD+mech 0.00 SD+min NS PRD+min NS SS NS SS NS SS Harvest SS Veraison Fruit set Flowering DPheevnolelopmentogical Salta geStag e

Chapter 6: Botrytis bunch rot development and bunch architecture 150

10.00 *** LSD=0.99 9.00

8.00 Severity (Infected berries bunch

7.00

6.00

5.00

4.00

3.00

-1 ) 2.00

SD+spur 1.00 PRD+spur SD+mech Treatment PRD+mech 0.00 SD+min SS NS PRD+min SS NS NS Harvest NS SS SS Veraison Fruit set Flowering Phenological Stage

6.3.3 Late Season Botrytis Field Inoculations

The susceptibility of developing bunches to Botrytis bunch rot was assessed by inoculating bunches with B. cinerea at flowering, veraison and pre-harvest and leaving them in situ until fully mature (harvest). Bunches inoculated at flowering and pre-harvest had a higher level of mean Botrytis incidence than bunches inoculated at veraison in season 2001-02 (Table 6.6). Mean severity of Botrytis was high on bunches inoculated at flowering and veraison but bunches inoculated late in the season (pre-harvest) had the lowest degree of disease expression. No differences in mean incidence or severity were found in bunches inoculated at flowering, veraison or pre-harvest in season 2002-03.

Chapter 6: Botrytis bunch rot development and bunch architecture 151

-1 Botrytis Incidence (% Infected bunches) Botrytis Severity (Infected berries bunch ) Treatments 2001-02 2002-03 2001-02 2002-03 Flowering 81.2b 57.3 3.73b 1.41 Veraison 52.1a 58.3 4.21b 1.04 Pre-harvest 81.2b 61.5 0.94a 1.30 P value *** ns *** ns

The level of natural infection (% incidence of Botrytis in bunches the from control treatment) accounted for >73% of the infection observed at each phenological stage in both seasons, except at harvest 2001-02 when natural infection only accounted for 39% of Botrytis incidence (Table 6.7). At flowering and veraison of season 2002-03, all recorded Botrytis disease was due to natural infection as no difference was found between control and inoculated bunches. Bunches inoculated at flowering and veraison in season 2002-03 had a higher level of latent infection of Botrytis in relation to surface infection. This may be correlated with the high level of natural infection observed on these bunches.

Botrytis Incidence (% Infected bunches) Botrytis Incidence (% Infected bunches)

2001-02 2002-03 Treatments Flowering Veraison Harvest Flowering Veraison Harvest 0 spores.mL-1 79.2 70.8a 29.2a 58.3 58.3 52.1 104 spores.mL-1 83.3 91.7b 75.0b 58.3 58.3 70.8 P value ns * ** ns ns ns NS - - - 41.7a 43.8a 56.2 SS - - - 75.0b 72.9b 66.7 P value *** ** ns

Severity of Botrytis expression was also strongly influenced by natural infection. Botrytis severity on control bunches was not significantly different to the severity of Botrytis on inoculated bunches at flowering and veraison in both experimental seasons (Table 6.8). However, bunches inoculated at pre-harvest were less influenced by natural infection in both

Chapter 6: Botrytis bunch rot development and bunch architecture 152 seasons. The severity of Botrytis was also higher on SS bunches at flowering and veraison. This suggests that disease development was predominantly associated with latent infection.

Botrytis Severity (Infected berries bunch-1) Botrytis Severity (Infected berries bunch-1)

2001-02 2002-03 Treatments Flowering Veraison Harvest Flowering Veraison Harvest 0 spores.mL-1 3.75 3.42 0.54a 1.25 0.98 0.98a 104 spores.mL-1 3.71 5.00 1.33b 1.56 1.10 1.62b P value ns ns * ns ns * NS - - - 0.88a 0.60a 1.06 SS - - - 1.94b 1.48b 1.54 P value - - - *** *** ns

The influence of irrigation and pruning treatments on Botrytis disease development was assessed on bunches inoculated at flowering, veraison and pre-harvest (Table 6.9). Pruning had a large influence on the incidence and severity of Botrytis in season 2001-02. As a result, an interaction between irrigation and pruning was observed. Irrigation and pruning had minor effect on incidence or severity in 2002-03. Bunches produced on vines irrigated by PRD had greater incidence of Botrytis disease in season 2002-03 but a significant difference was only recorded on bunches inoculated at veraison. Pruning had a significant effect on Botrytis incidence at flowering and veraison in season 2001-02. When bunches were inoculated early in the growing season, bunches from SPUR vines had the highest incidence of Botrytis (100%) and bunches from MIN vines had lowest incidence. No pruning effect was found at harvest or at any phenological stage in the subsequent season. When the effects of irrigation strategy combined with pruning level on Botrytis incidence in bunches were evaluated, it was apparent that treatments that produced larger bunches with more berries bunch-1, such as SD + SPUR, PRD + SPUR and SD + MECH, had higher incidence in 2002. Bunches from the aforementioned treatments had 100% incidence at both flowering and veraison. However, the trend between incidence and treatments with large, compact bunches was not replicated in the following season when irrigation and pruning treatment effects were assessed.

Chapter 6: Botrytis bunch rot development and bunch architecture 153

Botrytis Incidence (% Infected bunches) Botrytis Incidence (% Infected bunches)

2001-02 2002-03 Treatments Flowering Veraison Harvest Flowering Veraison Harvest SD 87.5 87.5 45.8 52.1 45.8a 56.2 PRD 75.0 75.0 58.3 64.6 70.8b 66.7 Irrigation ns ns ns ns ** ns SPUR 100.0b 100.0b 56.3 53.1 62.5 56.3 MECH 81.2ab 75.0a 68.8 65.6 56.2 71.9 MIN 62.5a 68.7a 31.3 56.3 56.2 56.3 Pruning * * ns ns ns ns SD + SPUR 100.0 100.0b 50.0 50.0 68.8 56.3 PRD + SPUR 100.0 100.0b 62.5 56.2 56.2 56.2 SD + MECH 100.0 100.0b 75.0 68.7 31.2 62.5 PRD + MECH 62.5 50.0a 62.5 62.5 81.2 81.2 SD + MIN 62.5 62.5a 12.5 37.5 37.5 50.0 PRD + MIN 62.5 75.0ab 50.0 75.0 75.0 62.5 P value ns * ns ns ns ns

The effects of irrigation and pruning on Botrytis severity were also evaluated on bunches inoculated at flowering, veraison and pre-harvest (Table 6.10). PRD irrigation increased Botrytis severity on bunches inoculated at veraison relative to SD in season 2002-03. Significant pruning effects were recorded on bunches inoculated at flowering and veraison in season 2001-02. SPUR vines were more than twice as susceptible to Botrytis than MECH and MIN vines at flowering and veraison. No significant interactions between irrigation and pruning treatments were found in either season. Light pruning levels and PRD reduced botrytis severity at flowering and veraison in the first experimental season but PRD increased severity at each phenological stage (i.e. flowering, veraison and pre-harvest) in the following season under all pruning levels.

Chapter 6: Botrytis bunch rot development and bunch architecture 154

Botrytis Severity (Infected berries bunch-1) Botrytis Severity (Infected berries bunch-1)

2001-02 2002-03 Treatments Flowering Veraison Pre-Harvest Flowering Veraison Pre-Harvest SD 4.46 4.71 0.88 1.17 0.77a 1.06 PRD 3.00 3.71 1.00 1.65 1.31b 1.54 Irrigation ns ns ns ns ** ns SPUR 6.81b 7.06b 1.44 1.38 1.28 1.16 MECH 2.50a 3.00a 0.81 1.44 0.91 1.62 MIN 1.88a 2.56a 0.56 1.41 0.94 1.12 Pruning *** *** ns ns ns ns SD + SPUR 7.25 7.75 1.62 1.25 1.31 1.25 PRD + SPUR 6.38 6.38 1.25 1.50 1.25 1.06 SD + MECH 4.00 4.00 0.88 1.37 0.31 1.19 PRD + MECH 1.00 2.00 0.75 1.50 1.50 2.06 SD + MIN 2.13 2.38 0.13 0.87 0.69 0.75 PRD + MIN 1.62 2.75 1.00 1.94 1.19 1.50 P value ns ns ns ns ns ns

6.3.4 Seasonal Effects on Bunch Architecture Bunch architecture was evaluated in seasons 2001-02 and 2002-03, in terms of bunch number vine-1, bunch weight, berry number bunch-1, berry weight, bunch volume, bunch length, maximum bunch width and bunch compactness (Table 6.11).

Yield Parameters

Bunch number vine-1 was significantly reduced (58%) in 2003 compared to the previous season, which may be attributed to poor bunch initiation in November 2001-02 (discussed further in section 3.3.4). In conjunction with reduced bunch number vine-1 in 2002-03, mean bunch weight, berry number bunch-1 and berry weight increased in relation to the previous season by 1.2-fold, 1.1-fold, 1.1-fold, respectively.

Bunch Architectural Characteristics

Bunch length and maximum width decreased in 2002-03 by 29% and 18%, respectively. (Table 6.11). No significant difference in bunch volume was found between 2001-02 and 2002-03. Since bunch length and width decreased in 2002-03 but bunch weight increased

Chapter 6: Botrytis bunch rot development and bunch architecture 155 and volume was not altered, bunch architecture was altered to produce more compact bunches. Bunch compactness (calculated as bunch weight divided by length) significantly increased in 2003 (1.33-fold) compared to 2002.

Season Treatments 2001-02 2002-03 P value LSD Bunch Number Vine-1 379.0b 159.3a *** 21.69 Mean Bunch Weight (g) 58.7a 71.1b *** 4.25 Berry Number Bunch-1 55.0a 60.6b ** 3.34 Mean Berry Weight (g) 1.01a 1.14b *** 0.03 Bunch Volume (cm3) 61.9 56.6 ns 9.19 Bunch Length (mm) 150.0b 107.0a *** 7.12 Max bunch width (mm) 68.7b 57.1a *** 3.90 Compactness g mm-1 0.51a 0.68b *** 0.04

6.3.5 Integrated irrigation and pruning effects on bunch architecture

The effects of integrated irrigation and pruning on bunch architecture were assessed in seasons 2001-02 and 2002-03 (Figures 6.6 and 6.7). No significant interactions were found between irrigation and pruning for bunch weight, volume, length, maximum width or berry number bunch-1 but a significant interaction was observed for bunch compactness. The lack of significance can be attributed to the additive effect of irrigation and pruning treatments on bunch architectural parameters.

Bunch Weight and Berry Number Bunch-1

A reduction in bunch weight was associated with the imposition of PRD compared to SD and MECH and MIN compared to SPUR (Figure 6.6a). Bunch weight did not differ between seasons when vines were SD irrigated and SPUR pruned but an increase in weight was observed for bunches from both MECH and MIN vines when SD irrigated in 2002-03. Similarly, bunch weight increased on MECH and MIN vines when they were PRD irrigated in 2003 but decreased on SPUR vines irrigated by PRD. A similar trend was observed for berry number bunch-1. MECH and MIN vines produced bunches with more berries in 2002-03 than in 2001-02 when vines were irrigated by SD or PRD (Figure 6.6b).

Chapter 6: Botrytis bunch rot development and bunch architecture 156

Bunch Volume, Width and Length

Bunch volume was predominantly affected by pruning treatments. Bunch volume decreased progressively from SPUR to MECH to MIN vines in both seasons (Figure 6.6c). No seasonal effects on bunch volume were observed on SD irrigated vines but bunch volume decreased in 2002-03 when vines were irrigated by PRD. Bunch width was reduced in 2002-03 for all integrated irrigation and pruning treatments (Figure 6.6d) and decreased from SPUR to MECH to MIN treatments under SD or PRD irrigation treatments. Bunch length was generally unaffected by irrigation and pruning treatments in 2001-02. However, pruning treatments influenced bunch length in 2002-03 (Figure 6.6e). Bunches decreased in length for SPUR relative to MECH to MIN treatments on both SD and PRD irrigated vines.

Bunch Compactness

The interaction between irrigation and pruning treatments over the 2 seasons was significant for bunch compactness. Compactness (bunch weight/length) was greater in 2002-03 than 2001-02 for all integrated irrigation treatments except SD + MECH (Figure 6.6f). The combination of PRD irrigation and light pruning levels reduced bunch compactness in both seasons. Compactness was reduced significantly by PRD relative to SD irrigation in both seasons. Light pruning levels (MIN) reduced bunch compactness compared to SPUR in season 2001-02 but no difference was recorded between pruning levels in 2002-03.

SD + SPUR SD + MECH SD + MIN

PRD + SPUR PRD + MECH PRD + MIN

Figure 6.6: Photograph of representative bunch architecture for each of the combined irrigation and pruning treatments in season 2001-02.

Chapter 6: Botrytis bunch rot development and bunch architecture 157

120 100 a. ns b. ns LSD = 18.4 LSD = 10.3 100 80 B e rr g y

80 Nu 60 ight m

e b

60 e r Bu h W 40

40 nch

Bunc -1 20 20

0 0 100 120 c. d. ns ns LSD = 22.5 LSD = 9.6 100 M 3 80

a x i cm mu

e 80 m W

m 60 lu 60 i d t 40 h

m nch Vo

40 m u B 20 20

0 0 180 e. ns f. *** 1.4 160 LSD = 17.5 LSD = 0.1

1.2 C

140 o mp

mm 1.0 120 a h c t n

100 0.8 e s engt s L 80 0.6 g

mm nch 60 -1 Bu 0.4 40 0.2 20 0 0.0 r r n h n h n h n h i i ur u c c pu p pur Mi e ec Mi e ec Sp S S +M +M M M M M +S SD SD RD+ RD+ SD SD+ RD+ P P SD+ SD+ P PRD+ PRD+ PRD+ 20022001-02 20032002-03

Figure 6.7: Effect of irrigation and pruning treatments at harvest 2001-02 () and 2002-03 () on bunch architectural parameters; a. mean bunch weight, b. berry number vine-1, c. bunch volume, d. maximum width, e. bunch length and f. compactness. Significant differences were calculated by Fisher’s least significant difference (LSD 5% level1) and significance levels are indicated by ***P<0.001, **P<0.01, *P<0.05, ns = not significant.

1 Error bars calculated on LSD of treatments for each season.

Chapter 6: Botrytis bunch rot development and bunch architecture 158

6.3.5 Bunch architecture and Botrytis cinerea incidence and severity at harvest

A significant linear relationship (R2= 0.87, P<0.01) was found between incidence (%) of B. cinerea and average bunch weight of the six combined irrigation and pruning treatments at harvest in season 2002 (Fig 6.8a). Larger, heavier bunches had a greater incidence of Botrytis than smaller bunches. Similarly a significant linear relationship (R2= 0.94, P<0.01) (Fig 6.8b) was observed between severity of Botrytis for each of the imposed irrigation and pruning treatments and average bunch weight for those treatments at harvest in season 2002.

a. Incidence b. Severity

100 6

) )

90 -1 5 s h e

h 80 4 es bunc i r ed bunc

t c e 70 3 ted ber Inf c %

e ( nf e I

c 60 ( 2

rity den i 2 e 2

Inc R = 0.87 R = 0.94

50 Sev 1

40 0 20 40 60 80 100 120 20 40 60 80 100 120 Average bunch weight g Average Bunch Weight g

No linear relationship was found between Botrytis incidence or severity and mean bunch weight of the six combined irrigation and pruning treatments in the following season (Figure 6.9a and b). The lack of relationship is related to a change in bunch architecture and favourable weather conditions prior to harvest in 2002-3. A large rainfall event (37 mm), below average maximum summer temperature (23°C) and 100% relative humidity occurred on the 21 February 2003, which may have induced natural infection of Botrytis, therefore overriding architectural effects. The levels of mean incidence and severity were lower for most integrated treatments in 2002-03 compared to 2001-02.

Chapter 6: Botrytis bunch rot development and bunch architecture 159

a. b. Incidence Se verity 85 2.5

80 2 2 R = 0.009 R =0.014 2.0 75

ion

70 t c 1.5 % e e nc 65 Inf de of i c ts

n 1.0 In 60 Poi 55 0.5 50

45 0.0 20 40 60 80 100 120 20 40 60 80 100 120

Average bunch weight g Average Bunch Weight g

6.4 DISCUSSION

6.4.1 Influence of Spore Concentration on Inoculation Experiments

Spore concentration of inoculum can influence the development of Botrytis in bunches in the field. Warren et al. (2000) reported a difference in the incidence of Botrytis rot on bunches with spore concentrations between 103 and 104 spores.mL-1. However, in the studies reported here, the incidence and severity of expressed Botrytis at harvest on bunches wet inoculated at flowering was not influenced by inoculum spore concentrations between 0 and 104 spores.mL-1 of B. cinerea. These results are not consistent with those of the field inoculation study conducted by Warren et al. (2000).

The lack of difference in Botrytis rot on incubated bunches suggests that either natural infection of B. cinerea occurred without field expression on bunches from inoculum carried over from the previous season, the wet application method encouraged Botrytis development and/or the incubation conditions in the laboratory were highly conducive to Botrytis development. Studies by Nair et al. (1995) indicated that 70% of variation in flower infection on field-grown grapevines was associated with carry-over inoculum from the

Chapter 6: Botrytis bunch rot development and bunch architecture 160 previous season and flower infection accounted for 78% of berry infection. In this experiment, the potential for high levels of carry-over inoculum and berry infection in season 2002-03 was high because the same vines were used for the experiments for two seasons. Also, no fungicides were applied to the Shiraz vineyard to control B. cinerea and dead plant material was present because the surrounding vineyards were mechanically harvested and pruned. As a consequence, levels of natural inoculum were likely to be high. Inoculation by the wet application method may have also encouraged B. cinerea infection. Coertze and Holz (1999) found fresh grape berries, previously considered susceptible to B. cinerea when applied as wet inoculum, were resistant to infection from dry conidia. The authors suggested that spore distribution was restricted to the region of the water droplet from wet inoculation, creating a highly localised zone of disease pressure and the subsequent collapse of host resistance. Inoculation with dry conidia of B. cinerea using settling chambers (Coertze and Holz 1999; Cook et al. 2002) has been used effectively in the laboratory. However, controlled in situ field inoculations were not possible using this method. Finally, removal of bunches and incubation in conditions with optimal temperature and humidity in the laboratory was conducive to Botrytis development and sporulation on infected tissues. The incubation would have promoted the expression of infections arising from natural and artificially applied inoculum.

6.4.2 Botrytis incidence and severity at different bunch development stages

Bunches were less resistant to secondary infection of Botrytis when fully developed and at maximum maturity (TSS >22°Brix). Bunches inoculated at flowering and left in ambient conditions on the vine until harvest had higher incidence and severity of Botrytis compared to bunches assessed earlier in the growing season. This indicates that latent infection by the pathogen occurred during the growing season (flowering to harvest) without expression of the disease. If favourable weather conditions (i.e. cool and wet weather) occurred before harvest, field expression of Botrytis bunch rot would have been expected.

The development of tighter bunches as berry size increased from veraison to harvest and the increase in berry sugar content may have also influenced Botrytis development. Emmett et al. (1995) recorded a lower level of rot caused by B. cinerea on loose bunches. This may be attributed to increased aeration and less epicuticular wax damage by berry-to-berry contact

Chapter 6: Botrytis bunch rot development and bunch architecture 161

(Marois 1986). This may also be extrapolated to developing bunches that are looser when berries are small. Botrytis disease expression on bunches was less apparent earlier in the growing season with low incidence and severity reported at flowering, fruit-set and veraison. The improved microclimate of developing bunches may have been associated with these lower levels of Botrytis. Also, compounds present in pea-size green berries, such as the stilbene stress-metabolite, resveratrol, are believed to inhibit infection by the pathogen. However, recent work by Keller et al. (2003) showed that stilbenes may have a limited role in the inhibition of flower and latent infection.

Generally, the level of naturally occurring B. cinerea in the vineyard was high throughout the two experimental seasons, as no significant differences in Botrytis incidence or severity were observed between control and inoculated bunches at each phenological stage, with the exception of flowering in 2001-02 and fruit-set in 2002-03. This may be attributed to high levels of carry-over inoculum levels from previous seasons, especially in 2002-03, as discussed in section 6.4.1. The source of B. cinerea in bunches at each phenological stage was predominantly a combination of latent infection, probably in the receptacle area of berries (Keller et al. 2003) and surface contamination from spores on the surface of berries or infected trash (i.e. dead plant tissue, such as senescent flower parts) trapped in the bunches.

6.4.3 Susceptibility of Bunches to late season Botrytis Inoculations

Seasonal effects influenced the development of Botrytis in bunches inoculated at flowering, veraison and pre-harvest. In season 2001-02, incidence of Botrytis was higher in inflorescences and mature bunches, whilst severity was higher on inflorescences and bunches at veraison. The stage of bunch development had no effect on susceptibility to B. cinerea in the subsequent season. The seasonal variation in bunch susceptibility may have been associated with more favourable weather conditions for Botrytis development at the time of infection in season 2001-02. Lower mean monthly temperature and higher mean monthly relative humidity were recorded from November 2001 to February 2002 compared to the same growth period in the following season (refer to section 2.3).

Chapter 6: Botrytis bunch rot development and bunch architecture 162

The high susceptibility of inflorescences may have been due to the high levels of decaying flower tissue trapped in the bunch and easier penetration by the pathogen into the receptacle area through the cap scar (Keller et al. 2003). Compact bunch architecture and the resultant bunch microclimate of mature bunches prior to harvest may have produced favourable conditions for Botrytis bunch rot. The high susceptibility of bunches to Botrytis infection at flowering and in the pre-harvest period, highlights the need to monitor for Botrytis development during these growth stages. Also, the application of fungicides for Botrytis control at flowering and pre-harvest is important if weather conditions are conducive to infection.

6.4.4 Influence of Integrated Irrigation and Pruning Treatments on B. cinerea

Light pruning levels produce high numbers of bunches per vine and, consequently, alter bunch architecture. Similarly, PRD irrigation strategies can reduce shoot vigour and reduce berry size, as shown in section 3.3.4. Given the difference in bunch and canopy architecture between severe and light pruning treatments and SD and PRD irrigation strategies, it was hypothesised that combinations of irrigation and pruning practices would influence bunch susceptibility to Botrytis infection throughout the growing season and Botrytis disease expression (cf. p.138).

The combination of irrigation and pruning practices did not influence incidence and severity of Botrytis when assessed at the four stages of bunch development in season 2001-02. However, treatments that produced large, compact bunches within the fruiting zone of the canopy, such as SD + SPUR, PRD + SPUR and SD + MECH increased the incidence and severity of Botrytis at harvest in season 2001-02. In season 2002-03, PRD + MECH had a significantly higher incidence at veraison compared to all other integrated treatments. This result may be attributed to increased maturity of these bunches because of the earlier veraison date associated with PRD irrigation (refer to section 3.2.2). The severity of infection was not influenced by irrigation and pruning treatments in season 2002-03, even though bunches from PRD + MECH vines had high Botrytis incidence at veraison.

The susceptibility of bunches from combined irrigation and pruning treatments to infection by B. cinerea was assessed during bunch growth. Bunches at veraison from PRD irrigated

Chapter 6: Botrytis bunch rot development and bunch architecture 163 vines were more susceptible to B. cinerea infection. This may be associated with the higher sugar levels of berries from PRD compared to SD irrigated vines, refer to section 3.3.5. Inflorescences and small, loose ripening bunches from MIN and MECH vines had a lower incidence and severity of Botrytis than the larger bunches produced on SPUR vines in season 2001-02. However, this result was not reproduced in the subsequent season (discussed further in section 6.4.5). The integrated irrigation and pruning treatments affected Botrytis incidence in bunches at flowering and veraison in season 2001-02. Large, compact bunches produced by severe pruning (SPUR) and high input irrigation (SD) had higher Botrytis incidence than in small, loose bunches produced on light pruning (MIN) and lower input irrigation (PRD). Interestingly, incidence was lowest on ripening bunches from PRD + MECH vines in season 2001-02 and was highest on the same treatment in the following season. This large increase in incidence in the second season may be the consequence of high levels of inoculum carried over from the previous season. Alternatively, it may indicate that vines subjected to this treatment were stressed due to insufficient irrigation input and/or high crop levels and thus more susceptible to infection. Water-stress of grapevines can lead to dehydration and desiccation of berries as a survival mechanism. Consequently, the level of trash (dead and decaying plant tissue) within a bunch increases. It has been reported that bunches with high levels of trash were more susceptible to Botrytis bunch rot (Emmett et al. 1994). Also, dehydrated berries may be more prone to berry splitting at harvest when vines are rapidly re-hydrated by either large irrigation or rainfall events. Sall et al. (1982) showed bunches which contained split berries were highly susceptible to Botrytis infection.

6.4.5 Effects of Irrigation and Pruning on Bunch Architecture and Botrytis infection

It has been established that grapevine cultivars with tight, compact bunches are more susceptible to Botrytis infection than cultivars that produce small, loose bunches (Savage and Sall 1984). Presence of free water (Jarvis 1980), berry splitting (Sall et al. 1982), cultivar variation in cuticle and epicuticular wax (Rosenquist and Morrison 1989) and exocarp damage by berry contact (Marois et al. 1986) are several explanations for the increased susceptibility of large, tight bunches. In view of the aforementioned, the effect of the architecture of bunches on vines with the imposed irrigation and pruning treatments on Botrytis development were evaluated. Strong positive linear relationships between mean bunch weight and Botrytis incidence and severity were found in season 2001-02. Larger and

Chapter 6: Botrytis bunch rot development and bunch architecture 164 more compact bunches, as produced by SD + SPUR and PRD + SPUR, had higher Botrytis incidence and severity. The higher incidence and severity of Botrytis on large, tight bunches was likely to have been associated with a more favourable microclimate in these bunches for Botrytis development, berry damage and differences in berry exocarp, as discussed previously.

However in the following season, there was no relationship between bunch weight and Botrytis development. The lack of relationship may be explained by changes in bunch architecture between the two experimental seasons. Irrigation and pruning treatments produced bunches of distinct bunch architecture in season 2001-02; severe pruning levels and SD irrigation produced large, compact bunches with many berries per bunch, whereas light pruning levels and PRD irrigation produced small, loose bunches with fewer berries per bunch. However in spring 2001, poor bunch initiation on lightly pruned vines (i.e. MIN and MECH) caused large changes in most bunch architectural parameters in season 2002-03. Bunch weight, berry number per bunch and berry weight all increased and bunch length and maximum width decreased for bunches from MIN and MECH pruned vines in 2002-03 compared to the previous season. As a consequence, bunches from MIN and MECH pruned vines became more compact because more berries were concentrated on a shorter rachis. The compactness of bunches from light pruning treatments was further increased when combined with PRD irrigation in season 2002-03. However, bunch architecture was unaffected in the two seasons by SPUR pruning because severe pruning levels controlled crop level and subsequently yield components (i.e. bunch weight, berry weight, berry size and berry number).

This study has shown that light pruning practices combined with PRD irrigation can alter bunch architecture and this, in turn, has the potential to reduce infection by B. cinerea. However, bunch architecture is dependent on crop level (bunch number per vine) and light pruning treatments are more susceptible to seasonal variation in bunch initiation and fruit set. Consequently, reproduction of the types of bunch architecture required to discourage Botrytis infection and development may not be achievable every season because of unfavourable climatic conditions at critical phenological stages (i.e. bunch initiation, flowering and fruit set).

Chapter 6: Botrytis bunch rot development and bunch architecture 165

6.5 CONCLUSIONS

a) Bunches were least resistant to infection by Botrytis when fully developed and at maximum maturity. The development of bunches into tighter clusters as berry size increased from veraison to harvest and the increase in sugar content may have encouraged development of Botrytis.

b) Inflorescences and mature bunches were more susceptible to secondary infection by B. cinerea, possibly because of the high levels of trash trapped in inflorescences at flowering and the compact architecture of mature bunches just prior to harvest.

c) The potential for natural infection of Botrytis in the Shiraz vineyard was high because the incidence and severity of Botrytis at harvest was not influenced by inoculum spore concentrations and incubated bunches had a natural infection rate of more than 73%.

d) The source of expressed Botrytis was generally a combination of surface contamination and latent infection when bunches were inoculated at flowering.

e) This study showed light pruning combined with PRD irrigation produced small, loose bunches in season 2001-02. However, in the following season, bunch architecture was more compact on MIN and MECH vines because vines compensated for reduced crop level by increased fruit-set and average berry size.

f) Strong positive correlations between bunch weight and Botrytis incidence (R2 = 0.87) and Botrytis severity (R2 = 0.94) were found in season 2001-02. However, no significant correlations were shown in the following season.

g) Given the large seasonal variation in results from inoculation experiments in 2001-02 and 2002-03, the hypothesis that light pruning systems combined with PRD would reduce the level bunch infection and disease expression of B. cinerea was not accepted.

Chapter 7: General Discussion 166

7 GENERAL DISCUSSION

7.1 INTRODUCTION TO THE EXPERIMENT

Modern viticultural practices have improved winegrape production and the economic and environmental sustainability of Australia’s wine industry. The widespread practice of mechanised light pruning systems across Australia has reduced production costs and increased tonnages, without negative impact on fruit or wine quality. Similarly, drip irrigation strategies that induce a regulated water deficit or partially dry the rootzone are currently being adopted by industry and can dramatically improve water-use efficiency in the vineyard. The aim of this holistic study was to evaluate the potential of industry- recognised, integrated irrigation and pruning systems to improve production sustainability. The sustainability of winegrape production was determined by important viticultural parameters: yield, fruit and wine composition, vine physiology and disease development. Few other research projects have taken a multi-disciplinary, holistic approach to studying viticultural practices. This study was established to identify optimal integrated management strategies for warm region, irrigated vineyards and to test the hypothesis that:

“Partial rootzone drying integrated with light pruning techniques can improve sustainability of winegrape production in warm, irrigated vineyards.”

The experiment investigated the effects of standard drip (SD) and partial rootzone drying (PRD) irrigation techniques combined with three conventional pruning practices on grape production, vine physiology and pathological susceptibility of field-grown cv. Shiraz grapevines. SD and PRD irrigation application and scheduling was based on conventional industry practice. Hand spur pruning (SPUR), mechanical hedging (MECH) and minimal (MIN) pruning practices were superimposed on the irrigation treatments. A large-scale field trial was conducted in the warm, irrigated region of north-west Victoria at a commercial vineyard at Deakin Estate, Iraak, VIC. The experiment was bi-factorial using a Latin square design and each integrated treatment was replicated six times.

Chapter 7: General Discussion 167

Winegrape production was assessed as tonnes of crop produced, yield components, fruit development, fruit chemical and spectral composition. Wine was produced from each treatment and assessed as an important determinant of winegrape production. Grapevine physiological studies involved the assessment of the effects of the imposed treatments on midday gas exchange. Diurnal gas exchange and leaf water potential measurements were employed to determine if PRD reduced stomatal conductance and maintained leaf water potential, as described by Dry (1997). The pathological component of the project involved field Botrytis inoculation experiments to assess the effects of the imposed treatments and resultant bunch architecture on Botrytis development.

7.2 IRRIGATION EFFECTS ON THE SUSTAINABILITY OF WINEGRAPE PRODUCTION

Partial drying of the grapevine rootzone had a detrimental effect on yield relative to SD irrigation practice, as a result of reduced mean berry weight at harvest. The reduction in berry weight may be attributed to the reduction in irrigation input as PRD applies half the amount of water by halving the number of emitters on each drip line compared to SD irrigation. However, plant water status was maintained by irrigating half of the vine, as determined by leaf water potential. This suggests the reduction in berry weight may have been the result of a pre-veraison hormonal/chemical response (i.e. ABA) induced by partial drying of the rootzone. As a result, pericarp volume may have been reduced and berry expansion restricted by modification of the structural properties of cell components and limitation of subsequent enlargement of pericarp cells.

Given the large decrease in irrigation input (50%) by PRD compared to the relatively small yield reduction (18%), a large improvement in water use efficiency was measured. An improvement in water-use efficiency has important economic and environmental implications. Annual volumetric irrigation licences in the Riverland, Murray Darling, Swan Hill and Riverina regions are subject to continual water restriction. By substantially reducing irrigation amounts, fewer demands are placed on the Murray, Murrumbidgee and Darling Rivers and the potential for vineyard expansion is increased. Reduction in irrigation amounts can also have important environmental benefits, including reduced inputs to ground-water tables, soil waterlogging and soil salinisation.

Chapter 7: General Discussion 168

Irrigation had little influence on basic fruit composition: pH, TA and TSS. However, PRD did improve fruit anthocyanin concentration slightly relative to SD irrigation, possibly due to the effects of co-pigmentation. Bravdo et al. (1985), Hepner et al. (1985), Matthews et al. (1990) and Iland (2000) reported increased berry anthocyanin and phenolic concentration and/or wine colour in association with reduced irrigation. Interestingly, PRD had a strong positive effect on the spectral parameters of wine, including wine density, wine hue, total anthocyanin and phenolic concentration and ionised anthocyanin concentration. The difference between wine spectral and fruit colour and phenolics concentration may be attributed to co-pigmentation between anthocyanin and exocarp tannins (i.e. quercetin) in the wine produced from PRD irrigated vines. The influence of cultural practices, such as PRD irrigation, on co-pigmentation in fruit and wine requires further investigation.

Midday and diurnal leaf gas exchange were affected by partially drying the rootzone. PRD lowered the rates of midday stomatal conductance, photosynthesis and transpiration in both experimental seasons, when compared to SD. The absolute reductions in stomatal conductance were comparable to decreases observed by Dry (1995) during the partial rootzone drying of field-grown Cabernet Sauvignon vines. Partial closure of stomata has been associated with increased synthesis of ABA in the dry roots (Loveys 1984a and b; Loveys and Düring 1984; Stoll et al. 2000) but determination of ABA production was beyond the scope of this project. Stomatal limitation on photosynthesis and transpiration was probable, given the strong positive relationship with stomatal conductance and reduced carbon isotope discrimination by PRD. Internal CO2 concentration was not influenced by irrigation practice. This suggests other factors, such as photoinhibition and inhibition of

CO2 metabolism, may have also been associated with the reduction of photosynthesis. The limitation on the amount of water transpired was greater than on the amount of carbon fixed by photosynthesis. As a result, transpiration efficiency was improved for PRD irrigated vines compared to SD irrigated vines. Leaf water potential and osmotic potential were measured diurnally in conjunction with leaf gas exchange to investigate the response of PRD irrigated vines to increasing VPD. Diurnally, stomatal conductance was reduced by PRD compared to SD. However, leaf water potential was maintained and no osmotic adjustment occurred. This suggests PRD maintained plant water status by irrigation of the wet half of the rootzone but promoted partial closure of the stomata in response to the dry half of the rootzone.

Chapter 7: General Discussion 169

Bunch susceptibility to Botrytis during the growing season was largely unaffected by irrigation treatment. However, bunch architecture was different for SD and PRD irrigated vines, which had implications for Botrytis bunch rot development after artificial infection. PRD irrigation produced looser bunches with smaller berries than SD irrigated vines. A strong positive correlation between bunch size and increased incidence and severity of Botrytis was found in season 2001-02. The small loose bunches of PRD irrigated vines may have altered bunch microclimate and reduced berry contact within the bunch, which consequently reduced infection by B. cinerea.

PRD has many advantages over deficit irrigation strategies that apply a hydraulic water stress to the grapevine, such as RDI. PRD is generally less limited by climatic conditions than RDI, as it is applied for the entire season compared to the short period post-flowering (several weeks). Also, PRD irrigated vines are less susceptible to extreme heat conditions than RDI vines because plant water status is maintained by irrigating one half of the rootzone. However, if PRD is to be implemented in a vineyard, additional irrigation infrastructure is required to alternately water each side of the vine.

7.3 PRUNING EFFECTS ON THE SUSTAINABILITY OF WINEGRAPE PRODUCTION

Yield was strongly influenced by pruning level and the resultant bunch number per vine. The number of nodes retained on the vine after winter pruning and the degree of bunch initiation during the previous season determined the number of bunches per vine in the subsequent growing season. Light pruning levels of MIN vines resulted in 4-fold more bunches per vine than the severe pruning levels of SPUR vines. Mechanical hedged vines had an intermediate number of bunches per vine. The high bunch numbers of MIN and MECH treatments were subject to greater seasonal variation than the highly controlled SPUR pruning treatment. Yield reflected the trend in bunch number per vine. However, MIN and MECH vines compensated for greater carbohydrate partitioning between reproductive sinks by producing smaller bunches with fewer berries per bunch.

Chapter 7: General Discussion 170

Pruning level influenced berry maturation in response to bunch number per vine, since delayed ripening occurred on lightly pruned treatments. Fruit chemical composition was influenced by berry size. The dilution of organic acids in the larger, more hydrated berries produced on SPUR vines affected berry juice pH and titratable acidity. Anthocyanin and phenolic content on a per berry basis was higher for the larger berries produced on SPUR vines. However, anthocyanin and phenolic concentration was higher from MIN vines in season 2001-02. The pruning effects on wine composition reflected fruit composition, as berry size had a strong association with both fruit and wine quality.

Pruning effects on vine physiology were less pronounced. MIN was associated with an increase in photosynthetic capacity in season 2001-02. Higher photosynthesis and transpiration rates may have been linked to greater photo-assimilate demands by higher crop loads in this season. Pruning treatments had no effect on leaf gas exchange in the subsequent season when crop loads were more uniform. These results support previous findings, which have shown no significant differences in seasonal leaf gas exchange between conventionally spur pruned and minimally pruned grapevines (Downton and Grant 1992; Lakso et al. 1996; Poni et al. 2000; Interi et al. 2001). It has been suggested that higher photo-assimilate demands of light pruning treatments are met by their larger canopy size, and therefore, higher net photosynthetic capacity (Sommer and Clingeleffer 1993).

The distinct bunch architecture resulting from the pruning treatments influenced Botrytis bunch rot development. Large, compact bunches from SPUR vines were more susceptible to Botrytis incidence and severity at harvest in season 2001-02. Bunches produced on MIN vines were looser with smaller berries. This bunch architecture allows greater aeration of the bunch and creates a less favourable microclimate for Botrytis disease development. Also, bunches from MIN vines were generally more exposed to sunlight. Therefore, epicuticular wax was expected to be thicker, providing greater protection against pathogen infection. However, low bunch numbers and high fruitset on MIN and MECH vines in season 2002-03 led to a significant change in bunch architecture. Bunches from MIN and MECH vines had reduced rachis length and increased berry numbers per bunch and as a result, became more compact. As a consequence of the increased compactness of bunches in season 2002-03, no linear relationship between Botrytis development and bunch weight was found.

Chapter 7: General Discussion 171

Light pruning techniques increased crop production without any negative effect on fruit or wine composition. In fact, small berries were strongly correlated with improved wine colour, density and hue. The photosynthetic capacity of MIN vines was sufficient to ripen the higher crop level. However, maturation was delayed by up to 13 days. Delayed maturation and harvest may be detrimental in cooler wine regions of Australia but it may be overcome by thinning for crop control. The loose architecture of MIN bunches discouraged the development of Botrytis bunch rot in season 2001-02. However, lighter pruning levels were more susceptible to seasonal variation in bunch initiation and fruitset. Thus, it is difficult to consistently reproduce bunch architecture that discourages Botrytis infection and development.

7.4 INTEGRATED IRRIGATION AND PRUNING EFFECTS ON THE SUSTAINABILITY OF

WINEGRAPE PRODUCTION

The integration of SD or PRD irrigation strategies with different pruning practices had minor effects on grape yield. The additive effect of PRD irrigation treatments combined with light pruning treatments resulted in few statistically significant interactions. Berry weight was the only yield component influenced by the interaction during the three experimental seasons. Berry weight was reduced when PRD was combined with MIN pruning, in response to lower irrigation inputs and higher bunch number. However, yield was not negatively affected by the reduction in berry weight by PRD + MIN because of the increased bunch number associated with lighter pruning. Large yield reductions were measured when PRD was combined with SPUR, as a result of low bunch numbers and reduced berry size. An improvement in WUE was an outcome of the reduced irrigation inputs of PRD combined with the high crop loads of MIN. A 2-fold increase in WUE was observed for PRD + MIN vines relative to SD + SPUR vines. This has important environmental sustainability implications, as discussed previously.

Fruit and wine composition were also largely unaffected by integrated irrigation and pruning treatments, although large berries produced by SD + SPUR had higher anthocyanin and phenolic content on a per berry basis. However, the latter was predominantly influenced by berry weight. Wines produced from PRD + MECH and PRD + SPUR vines were brighter

Chapter 7: General Discussion 172 and redder than their SD irrigated counterparts, as determined by wine hue over the three seasons. The additive effect of PRD combined with light pruning treatments to improve wine spectral properties did not result in any significant interactions between irrigation and pruning treatments. However, wine colour, phenolics and density were generally higher for PRD and lighter pruning. These results are supported by the strong correlation between small berries and improved fruit and wine composition.

Vine physiology, as determined by midday leaf gas exchange was influenced by the interaction between irrigation and pruning systems. An interesting integrated effect on stomatal conductance and carbon isotope discrimination was observed. SPUR and MECH vines combined with PRD were subjected to greater stomatal limitation than MIN vines with PRD. The difference in stomatal sensitivity between SPUR, MECH and MIN vines may be attributed to stress adaptability. MIN vines appear to have been better adapted to reduced water inputs by PRD because of greater partitioning of carbohydrates between more sinks, better suited leaf morphology and/or reduced susceptibility to photoinhibition.

The combination of PRD irrigation and light pruning decreased the incidence and severity of Botrytis bunch rot at harvest in season 2001-02. This was attributed to the looser bunches and improved bunch microclimate. However, the integrated treatment effects on Botrytis disease development were not significant in season 2002-03, when differences in bunch architecture were less pronounced. Those combinations of treatments that produced small, loose bunches on the outside of the canopy had the advantage of reducing Botrytis infection and disease expression, as well as allowing better spray penetration.

In conclusion, the hypothesis that partial drying of the rootzone integrated with light pruning techniques can improve sustainability of winegrape production, in terms of yield, fruit and wine quality, physiology and disease development may be accepted. Seasonal variation influenced results obtained in relation to each aspect that was studied but over the three experimental seasons PRD + MIN was determined to be the preferred strategy to enhance winegrape production in warm, irrigated vineyards.

Chapter 7: General Discussion 173

7.5 RECOMMENDATIONS FOR THE AUSTRALIAN WINE INDUSTRY

The results from this study indicate the combination of PRD irrigation and minimal pruning can maintain yield, improve wine composition and minimise disease development of Botrytis. Additional infrastructure is required when installing PRD irrigation and it may be argued that applying a 50% water deficit to the grapevines by standard drip irrigation practices as opposed to partially drying the rootzone could produce similar results. However, this study has clearly shown that PRD maintained plant water status whilst applying stomatal limitation to transpiration and photosynthesis. It is important to sustain hydraulic water status to avoid negative plant responses associated with water-deficit, such as leaf desiccation, reduced net photosynthetic capacity and yield reduction.

Any yield losses incurred in response to PRD irrigation in this study were compensated for by the increase in bunch number associated with MIN vines. Therefore, yield was maintained at a level equivalent to fully irrigated vines. Vine photosynthetic capacity of PRD + MIN vines was sufficient to fully mature the higher crop load. The combination of PRD and minimal pruning produced small berries that were associated with improved wine quality, improved bunch architecture and lower susceptibility to pathogen infection. The loose bunch architecture of PRD + MIN vines gave rise to a less favourable microclimate for disease development, reduced berry damage within the bunch and may increase fungicide spray penetration.

This study also highlights the importance of fungicide spray applications at flowering and pre-harvest to control Botrytis disease development. This is particularly important when the potential for natural infection is high because of high levels of source inoculum in the field and latent infection. Penetration of spray into vines and within bunches was expected to be higher on open canopies with loose, exposed bunches, as is the case on minimally pruned vines.

References 174

REFERENCES Agrios, G. (1997) Plant Pathology, 4th ed. Academic Press, California.

Allen, M., Lacey, M. and Boyd, S. (1996) Methoxypyrazines of grapes and wines – differences of origin and behaviour. In ‘Proceedings 9th Australian Wine Technical Conference, Adelaide’. Eds. C. Stockley, A. Sas, R. Johnstone and T. Lee, Winetitles, Adelaide. pp. 83-86.

Antolin, C., Baigorri, H., De Luis, I., Aguirrezábal, F., Geny, L., Broquedis, M. and Sánchez-Díaz, M. (2003) ABA during reproductive development in non-irrigated grapevines (Vitis vinifera L. cv. Tempranillo). Aust. J. Grape Wine Res. 9: 169-176.

Baker, E.L. (1982) Chemistry and morphology of plant epicuticular waxes. In ‘The Plant Cuticle’. Eds. D. Cutler, K. Alvin and C. Price, Academic Press, Toronto. pp. 139-161.

Bakker, J., Preston, N. and Timberlake, C. (1986) The determination of anthocyanins in aging red wine: Comparison of HPLC and spectral methods. Am. J. Enol. Vitic. 37: 121-126.

Becker, H. and Kerridge, G. (1972) Methods of small-scale wine making for research purposes in both hot and cool climates. J. Aust. Institute Ag. 38(1): 3-6.

Bertamini, M. and Nedunchezhian, N. (2003) Photoinhibition of photosynthesis in mature and young leaves of grapevine (Vitis vinifera L.). Plant Sci. 164: 635-644.

Botting, D., Dry, P. and Iland, P. (1996) Canopy architecture- implications for Shiraz grown in a hot climate. Aust. Grapegrower Winemaker 390a: 53-57.

Bravdo, B. and Hepner, Y. (1987) Water management and effect on fruit quality in grapevines. In ‘Proceedings of the 6th Australian Wine Industry Technical Conference, Adelaide’. Ed. T. Lee, Australian Wine Research Institute, Urrbrae. pp. 150-158.

Bravdo, B., Hepner, Y., Loinger, C., Cohen, S. and Tabacman, H. (1985a) Effect of crop level and crop load on growth, yield, must and wine composition and quality of Cabernet Sauvignon. Am. J. Enol. Vitic. 36: 125-131.

Bravdo, B., Hepner, Y. Loinger, C., Cohen, S. and Tabacman H. (1985b) Effect of irrigation and crop level on growth, yield and wine quality of Cabernet Sauvignon. Am. J. Enol. Vitic. 36: 132-139.

References 175

Brown, S. and Coombe, B. (1984) Solute accumulation by grape pericarp cells. II. Studies with protoplasts and vacuoles. Biochem. Physiol. Pflanz. 179: 157-171.

Bulit, J. and Lafon, R. (1977) Observations on infection of grapes by Botrytis cinerea Pers. In ‘ Travaux dédiés à Georges Viennot-Bourgin’. Soc. Française Phytopathol. Paris, France.

Bureau of Meteorology (2003) URL: http://bom.gov.au/climate/averages/tables/cw_076031

Buttrose, M., Hale, C. and Kliewer, W. (1971) Effect of temperature on the composition of Cabernet Sauvignon berries. Am. J. Enol. Vitic. 22: 71-75.

Candolfi-Vasconcelos, M., Koblet, W., Howell, G. and Zweifel, W. (1994) Influence of defoliation, rootstock, training system and leaf position on gas exchange of Pinot noir grapevines. Am. J. Enol. Vitic. 45: 173-180.

Carbonneau, A. (1996) General relationships within the whole plant: examples of the influence of vigour status, crop load and canopy exposure on the sink ‘berry maturation’ for the grapevine. Acta Hort. 427: 99-118.

Carroll, J. and Wilcox, W. (2003) Effects of humidity on the development of grapevine powdery mildew. Phytopathol. 93(9): 1137-1144.

Chalmers, D., Mitchell, P. and van Heek, L. (1981) Control of peach tree growth and productivity by regulated water supply, tree density and summer pruning. J. Am. Soc. Hort. Sci. 106(3): 307-312.

Chaumont, M., Morot-Gaudry, J. and Foyer, C. (1994) Seasonal and diurnal changes in photosynthesis and carbon partitioning in Vitis vinifera leaves in vines with and without fruit. J. Exp. Bot. 45: 1235- 1243.

Chaves, M. (1984) Photosynthesis and assimilate partition in fruiting and non-fruiting grapevine shoots. Ad. Photosynthesis Res. 4: 145-148.

Clingeleffer, P. (1983) Minimal pruning- Its role in canopy management and implications of its use for the wine industry. In ‘Advances in Viticulture and Oenology for Economic gain: Proceedings 5th Australian Wine Industry Technical Conference, Perth’. Eds. T. Lee and T. Somers, Australian Wine Research Institute, Urrbrae. pp. 133-140.

References 176

Clingeleffer, P. (1989) Update: minimal pruning of cordon trained vines (MPCT). Aust. Grapegrower Winemaker 304: 78-83.

Clingeleffer, P. (1992) Development of management systems for low cost, high quality wine production and vigour control in cool climate Australian vineyards. In ‘ Proceedings of 3rd International Cool Climate Viticultural Symposium, 8-12 June, Geisenheim, Germany’. Vitic. Enol. Sci. 48:130-134.

Clingeleffer, P. (1993) Vine response to modified pruning practices. In ‘Pruning mechanisation and crop control: Proceedings of the 2nd Nelson J. Shaulis Grape Symposium, 13-14 July 1993, Fredonia, New York’. Eds. R. Pool, New York State Agriculture Experiment Station, New York. pp. 20-30.

Clingeleffer, P. (2001) Mechanization of wine and raisin production in Australian vineyards. In ‘Proceedings of the ASEV 50th Anniversary Meeting, Seattle, Washington, June 19-23 2000’. American Society for Enology and Viticulture, Davis, California. pp. 165-169.

Clingeleffer, P. and Krake, L. (1992) Response of Cabernet Franc grapevines to minimal pruning and virus infection. Am. J. Vitic. Enol. 43: 31-37.

Clingeleffer, P. and Sommer, K. (1995) Vine development and vigour control. In ‘Canopy Management: Proceedings of a seminar organised by the Australian Society of Viticulture and Oenology, Inc. and held on 18 August, 1994 at Mildura, Victoria.’ Eds. P. Hayes, Australian Society of Viticulture and Oenology, Adelaide. pp. 7-17.

Clingeleffer, P., Krstic, M. and Sommer, K. (2000) Production efficiency and relationships among crop load, fruit composition and wine quality. In ‘Proceedings of the ASEV 50th Anniversary meeting, Seattle, Washington, June 19-23, 2000’. American Society for Enology and Viticulture. pp 318-332.

Coertze, S. and Holz, G. (1999) Surface colonization, penetration, and lesion formation on grapes inoculated fresh or after cold storage with single airborne conidia of Botrytis cinerea. Plant Dis. 83(10): 917-924.

Cook, D., Elmer, P. and Hill, R. (2002) Quantitative inoculation with dry conidia of Botrytis cinerea. Aust. Plant Pathol. 31: 217-222.

Coombe, B. (1980) Development of the grape berry. I. Effects of time of flowering and competition. Aust. J. Ag. Res. 31:125-131.

References 177

Coombe, B. (1984) The inception of ripening in the grape berry. Quadernis della Scuola di Specilizzazione in Viticoultra ed Enologia, Universita di Torino, 8:87-99.

Coombe, B., Bovio, M. and Schneider, A. (1987) Solute accumulation by the grape pericarp cells. Relationship to berry size and the effects of defoliation. J. Exp. Bot. 38 (196):1789-1798.

Coombe, B. and Phillips, P. (1982) Development of the grape berry. III. Compositional changes during veraison measured by sequential hypodermic sampling. In ‘Proceedings of International Symposium on Grapes and Wine, June 1980’. Davis, California. Pp 132-136.

Condon, A., Richards, R. and Farquhar, G. (1987) Carbon isotope discrimination is positively correlated with grain yield and dry matter production in field-grown wheat. Crop Sci. 27: 996-1001.

Correia, M., Pereira, J., Chaves, M., Rodrigues, M. and Pacheco, C. (1995) ABA xylem concentrations determine daily leaf conductance of field-grown Vitis vinifera L. plants. Plant, Cell, Enviro. 18: 511- 521.

Crippen, D. and Morrison J. (1986) The effects of sun exposure on the phenolic content of Cabernet Sauvignon berries during development. Am. J. Enol. Vitic. 37: 243-247.

Davies, W. and Zhang, J. (1991) Root signals and the regulation of growth and development of plants in drying soil. Annual Rev. Plant Physiol. Plant Mol. Biol. 42: 55-76.

DaPeng, Z., XueTeng, X., Zilian, Z. and Wensuo, J. (1997) Studies on the veraison mechanism: physiological and biochemical responses of grape berries to exogenous hormone treatments and girdling during the lag phase of berry growth. Acta Hort. Sinica 24: 1-7. de Souza, C., Maroco, J., dos Santos, T., Rodrigues, L., Lopes, C., Pereira, J. and Chaves, M. (2003) Partial rootzone drying: regulation of stomatal aperture and carbon assimilation in field-grown grapevines (Vitis vinifera cv. Moscatel). Funct. Plant Biol. 30:653-662.

Dodd, I., Stikic, R. and Davies, W. (1996) Chemical regulation of gas exchange and growth of plants in drying soil in the field. J. Exper. Bot. 47: 1475-1490.

Dokoozlian, N. and Kliewer, W. (1995) The light environment within grapevine canopies. II. Influence of leaf area density on fruit zone light environment and some canopy assessment parameters. Am. J. Enol. Vitic. 46 (2): 219-226.

References 178

dos Santos, T., Lopes, C., Rodrigues, L., de Souza, C., Maroco, J., Pereira, J., Silva, J. and Chaves, M. (2003) Partial rootzone drying: effects on growth and fruit quality of field grown-grapevines (Vitis vinifera). Funct. Plant Biol. 30:663-671.

Downton, W. and Grant., W. (1992) Photosynthetic physiology of spur-pruned and minimal pruned grapevines. Aust. J. Plant Physiol. 19: 309-316.

Downton, W., Grant, W. and Loveys, B. (1987) Diurnal changes in photosynthesis of field-grown grapevines. New Phytol. 105: 71-80.

Downton, W., Loveys, B. and Grant, W. (1988) Non-uniform stomatal closure induced by water stress causes putative non-uniform stomatal inhibition of photosynthesis. New Phytol. 110: 503-509.

Dry, P. (1997) Response of grapevines to partial drying of the root system. Ph.D. Thesis, The University of Adelaide, South Australia.

Dry, P. and Loveys, B. (1998) Factors influencing grapevine vigour and the potential for control with partial rootzone drying. Aust. J. Grape Wine Res. 4: 140-148.

Dry, P. and Loveys, B. (1999) Grapevine shoot growth and stomatal conductance are reduced when part of the root system is dried. Vitis 38(4): 151-156.

Dry, P., Loveys, B., Botting, D. and Düring, H. (1996) Effects of partial root-zone drying on grapevine vigour, yield, composition of fruit and use of water. In ‘Proceedings 9th Australian Wine Technical Conference, Adelaide’. Eds. C. Stockley, A. Sas, R. Johnstone and T. Lee, Winetitles, Adelaide. pp. 128- 131.

Dry, P., Loveys, B., Iland, P., Botting, D., McCarthy, M. and Stoll, M. (1999) Vine manipulation to meet fruit specifications. In ‘Proceedings 10th Australian Wine Technical Conference, Sydney’. Eds. R. Blair, A. Sas and P. Hayes, Australian Wine Research Institute, Urrbrae. pp. 208-214.

Dry, P. and Smart, R. (1988) The grapegrowing regions of Australia. In ‘Viticulture. Volume 1. Resources in Australia’. Eds. B. Coombe and P. Dry, Australian Industrial Publishers, Adelaide. pp. 37- 60.

Düring, H. (1984) Evidence for osmotic adjustment in grapevines. Acta Hort. 171: 315-322.

References 179

Düring, H. and Loveys, B. (1982) Diurnal changes in water relations and abscisic acid in field-grown Vitis vinifera cvs. 1. Leaf water potential components and leaf conductance under humid temperature and semi arid conditions. Vitis 21: 223-275.

Düring, H., Loveys, B. and Dry, P. (1996) Root signals affect water use efficiency and shoot growth. In ‘Proceedings on Workshop strategies to optimize wine grape quality’. Eds. S. Poni, E. Peterlunger, F. Iacono and C. Intrieri, Acta Hort. 427: 1-13.

Edson, C. and Howell, G. (1993) A comparison of vine architecture systems at different crop loads; leaf photosynthesis, vine yield, and dry matter partition. Die Wein-Wissenscaft 48: 90-95.

Emmett, R., Clingeleffer, P., Wicks, T., Hall, B., Hart, K. and Clarke, K. (1994) Canopy management and disease control. Aust. Grapegrower Winemaker 369: 22-24.

Emmett, R., Clingeleffer, P., Wicks, T., Nair, B., Hall, B., Clarke, K. and Somers, T. (1995) Influence of canopy architecture on disease development. In ‘Canopy Management: Proceedings of a seminar organised by the Australian Society of Viticulture and Oenology, Inc. and held on 18 August, 1994 at Mildura, Victoria.’ Eds. P. Hayes, Australian Society of Viticulture and Oenology, Adelaide. pp. 22-23.

Emmett, R. and Nair, N. (1991) Botrytis rot of grapes in Australia. Aust. Grapegrower Winemaker 333: 19-21.

English, J., Thomas, C., Marois, J. and Gubler, W. (1989) Microclimates of grapevine canopies associated with leaf removal and control of Botrytis bunch rot. Phytopathol.. 79(4): 395-401.

Escalona, J., Flexas, J. and Medrano, H. (1999) Stomatal and non-stomatal limitations of photosynthesis under water stress in field-grown grapevines. Aust. J. Plant Physiol. 26:421-433.

Escalona, J., Flexas, J. and Medrano, H. (1999) Contribution of different levels of plant canopy to total carbon assimilation and intrinsic water use efficiency of Manto Negro and Tempranillo grapevines. In ‘ Proceedings of 1st ISHS Worksop on water relations of grapevines’. Eds. E. Ruhl and J. Schmid, Acta Hort. 493: 141-148.

Esteban, M., Villanueva, M. and Lissarrague, J. (2001) Effects of irrigation on changes in anthocyanin composition of the skin of cv. Tempranillo (Vitis vinifera L.) grape berries during ripening. J. Sci. Food Agric. 81: 409-420.

References 180

Farquhar, G., Ehleringer, J. and Hubick, K. (1989) Carbon isotope discrimination and photosynthesis. Ann. Rev. Plant Physiol., 40:503-537.

Farquhar, G., O’Leary, M. and Berry, J. (1982) On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Aust. J. Plant Physiol. 9: 121-137.

Farquhar, G. and Richards, R. (1984) Isotopic composition of plant carbon correlates with water-use efficiency of wheat genotypes. Aust. J. Plant Physiol. 11: 539-552.

Flexas, J., Escalona, J. and Medrano H. (1999) Water stress induces different levels of photosynthesis and electron transport regulations in grapevines. Plant, Cell, Enviro. 22: 421-433.

Flexas, J., Hendrickson, L. and Soon Chow, W. (2001) Photoinactivation of photosystem II in high light- acclimated grapevines. Aust. J. Plant Physiol. 28: 755-764.

Freeman, B. and Kleiwer, W. (1983) Effect of irrigation, crop level and potassium fertilization on Carignane vines. II Grape and wine quality. Am. J. Enol. Vitic. 34: 197-207.

Freeman, B., Lee, T. and Turkington, C. (1979) Interaction of irrigation and pruning level on growth and yield of Shiraz vines. Am. J. Enol. Vitic. 30(3): 218-223.

Freeman, B., Lee, T. and Turkington, C. (1980) Interaction of irrigation and pruning level on grape and wine quality of Shiraz vines. Am. J. Enol. Vitic. 31(2): 124-135.

Gao, Y. and Cahoon, G. (1994) Cluster shading effect on fruit quality, fruit skin colour and anthocyanin content and composition in Reliance (Vitis hybrid). Am. J. Enol. Vitic. 22: 71-75.

Gibberd., M., Walker, R., Blackmore, D. and Condon, A. (2001) Transpiration efficiency and carbon- isotope discrimination of grapevines grown under well-watered conditions in either glasshouse or vineyard. Aust. J. Grape Wine Res. 7(3): 110-117.

Gladstones, J. (1992) Viticulture and Environment. Winetitles, Adelaide.

References 181

Gomez-del-Campo, M., Ruiz, C. and Lissarrague, J. (2002) Effect of water stress on leaf area development, photosynthesis and productivity in Chardonnay and Airén grapevines. Am. J. Enol. Vitic. 53(2): 138-143.

Goodwin, I. and Jerie, P. (1992) Regulated deficit irrigation: from concept to practice. Aust. N.Z. Wine Industry J. 7: 258-261.

Gowing, D., Davies, W. and Jones, H. (1990) A positive root-sourced signal as an indicator of soil drying in apple, Malus x domestica Borkh. J. Exp. Bot. 41: 1535-1540.

Gray, J., Gibson, R., Coombe, B., Iland, P. and Pattison, S. (1992) Assessment of winegrape value in the vineyard- Survey of cv. Shiraz from South Australian vineyards in 1992. Aust. J. Grape Wine Res. 4:109-116.

Gubler, W., Marois, J., Bledsoe, A. and Bettiga, L. (1987) Control of Botrytis bunch rot of grapes with canopy management. Plant Disease 71: 599-601.

Hardie, W. and Considine, J. (1976) Response of grapes to water-deficit stress in particular stages of development. Am. J. Enol. Vitic. 27(2): 55-61.

Hardie, J. and Martin, S. (1990) A strategy for vine growth regulation by soil water management. In ‘Proceedings of 7th Australian Wine Industry Technical Conference.’ Adelaide, South Australia. Eds. P. William, D. Davidson and T. Lee, Australian Wine Research Institute, Urrbrae. pp. 51-57.

Harley, P. and Baldocchi, D. (1995) Scaling carbon dioxide and water vapour exchange from leaf to canopy in a deciduous forest. I. Leaf model parameterization. Plant Cell Enviro. 18: 1146-1156.

Hepner, Y., Bravdo, B., Loinger, C., Cohen, S. and Tabacman, H. (1985) Effect of drip irrigation schedules on growth, yield, must, composition and wine quality of Cabernet Sauvignon. Am. J. Enol. Vitic. 36: 77-85.

Holzapfel, B., Rogiers, S., DeGaris, K. and Small, G. (1999) Ripening grapes to specification: effect of yield on colour development of Shiraz grapes in the Riverina. Aust. Grapegrower Winemaker 428: 24- 28.

Howell, G. (2001) Sustainable grape productivity and the growth-yield relationship: a review. Am. J. Enol. Vitic. 53(3): 165-174.

References 182

Hubble, G. and Crocker R. (1941) A soil survey of the Red Cliffs Irrigation district, Victoria. Bulletin No. 137, pp 334-35.

Hubick, K. and Farquhar, G. (1989) Carbon isotope discrimination and the ratio of carbon gained to water lost in barley cultivars. Plant Cell Enviro. 12: 795-804.

Iland, P. (2001) A study of glycosides in grapes and wines of Vitis vinifera cv. Shiraz. Ph.D. Thesis, The Adelaide University.

Iland, P., Gawel, R., McCarthy, M., Botting, D., Giddings, J., Coombe, B. and Williams, P. (1996) The glycosyl-glucose assay, its application to assessing fruit composition. In ‘Proceeding of the 9th Australian Wine Industry Technical Conference, 16-19 July 1995’. Eds. C. Stockley, A. Sas, R. Johnstone and T. Lee, Winetitles, Adelaide. pp. 98-100.

Intrieri, C., Poni, S., Rebucci, B. and Magnanni, E. (1997) Effects of canopy manipulation on whole vine photosynthesis: results from pot and filed experiments. Vitis 36:167-173.

Intrieri, C., Poni, S., Lia, G. and Gomez del Campo, M. (2001) Vine performance and leaf physiology of conventionally and minimally pruned Sangiovese grapevines. Vitis 40(3): 123-130.

Iacono, F., Reniero, F. and Bertamini, M. (1993) Relationships among carbon isotope discrimination, fluorescence, yield and water stress in Vitis. In ‘Proceedings of the 4th international symposium on grapevine physiology.’ Eds. Fondazione, Giovanni and Dalmosso, Torino, Italy. pp 283-287.

Jackson, D. (1991) Environmental and hormonal effects on development of early bunch-stem necrosis. Am. J. Enol. Vitic. 42: 290-294.

Jackson, D. and Lombard, P. (1993) Environmental and management practices affecting grape composition and wine quality- a review. Am. J. Enol. Vitic. 44: 409-430.

Jacobs, C., Vandenhurk, B. and Debruin, H. (1996) Stomatal behaviour and photosynthetic rate of unstressed grapevines in semi-arid conditions. Agric. Forest Meteor. 80: 111-134.

Jarvis, W. (1980) Epidemiology. In ‘ The Biology of Botrytis.’ Eds. J. Coley-Smith, K. Verhoeff and W. Jarvis. Academic Press, New York. pp. 219-250.

References 183

Johnstone, R., Clingeleffer, P. and Lee, T. (1996) The composition of Shiraz grape berries: implications for wine. In ‘Proceeding of the 9th Australian Wine Industry Technical Conference, 16-19 July 1995’. Eds. C. Stockley, A. Sas, R. Johnstone and T. Lee, Winetitles, Adelaide. pp. 105-108.

Kang, S., Liang, Z., Hu, W., Zhang, J. (1998) Water use efficiency of controlled alternate irrigation on root-divided maize. Agric. Water Mgmt. 38: 69-76.

Kang, S., Zhang, J., Hu, X., Li, Z. and Jerie, P. (2001) An improved water use efficiency for hot pepper grown under controlled alternate drip irrigation on partial roots. Sci. Hort. 89: 257-267.

Kang, S., Hu, X., Jerie, P. and Zhang, J. (2003) The effects of partial rootzone drying on root, trunk, sap flow and water balance in an irrigated pear (Pyrus communis L.) orchard. J. Hydrology 280: 192-206.

Keller, M., Viret, O. and Cole, F. (2003) Botrytis cinerea infection in grape flowers: defence reaction, latency, and disease expression. Phytopathol. 93(3): 316-322.

Kennedy, J., Matthews, M. and Waterhouse, A. (2002) Effect of maturity and vine water status on grape skin and wine flavonoids. Am. J. Enol. Vitic. 53(4): 268-274.

Kidd, C. (1989) Light pruning experiences. In ‘Introducing vineyard canopy management: Proceedings of Autumn Vineyard Seminar on 22 May 1989, Tocal’. NSW Agriculture & Fisheries, Hunter Valley Vineyard Association. Tocal. pp. 50-51.

Kliewer, W. (1970) Effect of day temperature and light intensity on coloration of Vitis vinifera L. grapes. J. Am. Soc. Hort. Sci. 95: 693-697.

Kliewer, W. and Antcliff, A. (1971) Influence of defoliation, leaf darkening, and cluster shading on the growth and composition of Sultana grapes. Am. J. Enol. Vitic. 21: 26-36.

Kliewer, W., Freeman, B. and Hossum, C. (1983) Effect of irrigation, crop level and potassium fertilization on Carrignane vines. I Degree of water and effect of growth and yield. Am. J. Enol. Vitic. 34: 186-196.

Kriedemann, P. (1968) Photosynthesis in vine leaves as a function of light intensity, temperature and leaf age. Vitis 7, 213-220.

References 184

Kriedemann, P. and Smart, R. (1971) Effects of irradiance, temperature and leaf water potential on photosynthesis of vine leaves. Photosynthetica 5: 6-15.

Lakso, A., Denning, S., Dunst, R., Fendinger, A., and Pool, R. (1996) Comparisons of growth and gas- exchange of conventionally pruned and minimally pruned “Concord” grapevines. In ‘Proceedings of 4th International Symposium on Cool Climate Viticulture and Enology’. 16-20 June Rochester, IV.11-IV.12.

Lavee, S. and Haskal, A. (1982) An integrated high density intensification system for table grapes. In ‘Proceedings of Grape and Wine Centennial Symposium, June 1980, Davis, California’. University of California, California. pp. 390-398.

Levengood, J. (1996) A survey of co-pigmentation in Cabernet Sauvignon wines. M.S. Thesis, University of California, Davis.

Loveys, B. (1984) Diurnal changes in water relations and abscisic acid in field grown Vitis vinifera cultivars. III The influence of xylem-derived ABA on leaf gas exchange. New Phytol. 98: 563-573.

Loveys, B. and Düring, H. (1984) Diurnal changes in water relations and ABA in field-grown Vitis vinifera cultivars. 2. ABA changes under semi arid conditions. New Phytol. 97: 34-47.

Loveys, B., Stoll, M., Dry, P. and McCarthy, M. (1998) Partial rootzone drying stimulates stress responses in grapevines to improve water use efficiency while maintain crop yield and quality. Aust. Grapegrower Winemaker 414a: 108-110,113.

Loveys, B., Dry, P., Stoll, M. and McCarthy, M. (2000) Using plant physiology to improve the water use efficiency of horticultural crops. In ‘Proceedings 3rd International Symposium on Irrigation of Horticultural Crops’. Eds. M. Ferreira and H. Jones, Lisbon, Portugal, Acta Hort. 537: 187-197.

Lovisolo, C., Hartung, W. and Schubert, A. (2002) Whole-plant hydraulic conductance and root-to-shoot flow of abscisic acid are independently affected by water stress in grapevines. Funct. Plant. Biol. 29: 1349-1356.

Lui, L., McDonald, A., Stadenberg, I. and Davis, W. (2001) Stomatal and leaf growth responses to partial drying of root tips in willow. Tree Physiol. 21: 765-770.

Lui, W., Wenkert, W., Allen, L. and E. Lemon (1978) Soil plant water relations in a New York vineyard: Resistances to water movement. J. Am. Soc. Hort. Sci. 103: 226-230.

References 185

Maroco, J., Rodrigues, M., Lopes, C. and Chaves, M. (2002) Limitations to leaf photosynthesis in field- grown grapevine under drought- metabolic and modelling approaches. Funct. Plant Biol. 29: 451-459.

Marois, J., Bledsoe, A. and Gubler, W. (1985) Effect of surfactants on epicuticular wax and infection of grape berries by Botrytis cinerea. Phytopathol. 27: 335-349.

Marois, J., Nelson, J., Morrison, J., Lile, L. and Bledsoe, A. (1986) The influence of berry contact within grape clusters on the development of Botrytis cinerea and epicuticular wax. Am. J. Enol. Vitic. 37(4): 293-296.

Matthews, M. and Anderson, M. (1988) Fruit ripening in Vitis vinifera L.: Responses to seasonal water deficits. Am. J. Enol. Vitic. 39(4): 313-320.

Matthews, M., Ishii, R., Anderson, M., and O’Mahoney, M. (1990) Dependence of wine sensory attributes on vine water status. J. Sci. Food. Agric. 51: 321-335.

May, P. (1965) Reducing inflorescence formation by shading individual Sultana buds. Aust. J. Biol. Sci. 18: 663-73.

May, P., Sauer, M. and Scholefield, P. (1973) Effect of various combinations of trellis, pruning, and rootstock on vigorous Sultana vines. Vitis 12: 192-206.

May, P., Clingeleffer, P., Scholefield, P. and Brien, C. (1976) The response of grape cultivar Crouchen (Australian Syn. Clare Riesling) to various trellis and pruning treatments. Aust. J. Agric. Res. 27: 845- 856.

McCarthy, M. (1997a) The effects of transient water deficit on berry development of cv. Shiraz (Vitis vinifera L.). Aust. J. Grape Wine Res. 3: 102-108.

McCarthy, M. (1997b) Timing of water deficit on fruit development and composition of Vitis vinifera. Ph.D. Thesis, The University of Adelaide.

McCarthy, M. (1998) Irrigation management to improve winegrape quality – nearly 10 years on. Aust. Grapegrower Winemaker 414a: 65-66,68,70-71.

References 186

McCarthy, M. (1999) Weight loss from ripening berries of Shiraz grapevines (Vitis vinifera L. cv. Shiraz). Aust. J. Grape Wine Res. 3: 10-16.

McCarthy, M. and Cirami, R. (1990) Minimal pruning effects on the performance of selections of Vitis vinifera cultivars. Vitis 29: 85-96.

McCarthy, M., Cirami, R. and McCloud, P. (1983) Vine and fruit responses to supplementary irrigation and canopy measurements. S. Afr. J. Enol. Vitic. 4: 67-76.

McClellan, W. and Hewitt, W. (1973) Early Botrytis rot of grapes: Time of infection and latency of Botrytis cinerea Pers. In Vitis viniferea L. Phytopathol. 63: 1151-1157.

Morris, J. and Cawthorn, D. (1982) Effect of irrigation, fruit load and potassium fertilization on yield, quality and petiole analysis of Concord (Vitis labrusca L.) grapes. Am. J. Enol. Vitic. 33:145-148.

Morrison, J. (1989) The effects of shading on Cabernet Sauvignon grape berries. In ‘Proceedings of 2nd International Symposium of Cool Climate Viticulture and Oenology, January 1988, Auckland, New Zealand’. New Zealand Society of Viticulture and Oenology, New Zealand. pp. 144-146.

Nageswara Rao, R. and Wright, G. (1994) Stability of the relationship between specific leaf area and carbon isotope discrimination across environments in peanut. Crop Sci. 34: 98-103.

Nair, N., Guilbaud-Oulton, S., Barchia, I. and Emmett, R. (1995) Significance of carry-over inoculum, flower infection and latency on the incidence of Botrytis cinerea in berries of grapevines at harvest in New South Wales. Aust. J. Exp. Ag. 35:1177-1180.

Naor, A., Bravdo, B. and Gelobter, J. (1994) Gas exchange and water relations in field-grown Sauvignon blanc grapevines. Am. J. Enol. Vitic. 45(4): 423-428.

Naor, A. and Wample, L. (1994) Gas exchange and water relations of field-grown Concord (Vitis labruscana Bailey) Grapevines. Am. J. Enol. Vitic. 45(3): 333-337.

Naor, A., Gal, Y. and Bravdo, B. (2002) Shoot cluster thinning influence vegetative growth, fruit yield, and wine quality of ‘Sauvignon blanc’ grapevines. J. Amer. Hort. Sci. 127(4): 628-634.

Neja, R., Wildman, W., Ayers, R. and Kasimatis, A. (1977) Grapevine response to irrigation and trellis treatment in the Salinas Valley. Am. J. Enol. Vitic. 28: 16-26.

References 187

Ojeda, H., Deloire, A., Carbonneau, A., Ageorges, A. and Romieu, C. (1999) Berry development of grapevines: Relations between the growth of berries and their DNA content indicate cell multiplication and enlargement. Vitis 38: 145-150.

Ojeda, H., Deloire, A. and Carbonneau, A. (2001) Influence of water deficits on grape berry development. Vitis 40(3): 141-145.

Palliotti, A., Cartechini, A. and Ferranti, F. (2000) Morpho-anatomical and physiological characteristics of primary and lateral shoot leaves of Cabernet Franc and Trebbiano Toscano grapevines under two irradiance regimes. Am. J. Enol. Vitic. 51(2): 122-130.

Patakas, A. and Noitsakis, B. (2001) Leaf age effects on solute accumulation in water-stressed grapevines. J. Plant Physiol. 158:63-69.

Pearman, G. (1966) The reflection of visible radiation from leaves of some western Australian species. Aust. J. Biol. Sci. 19: 97-103.

Pearson, R. and Goheen, A. (1988) Compendium of grape disease. American Phytopathological Society, Minnesota. pp 93.

Percival, D., Sullivan, J. and Fischer, K. (1993) Effect of cluster exposure, berry contact and cultivar on cuticle membrane formation and occurrence of Botrytis bunch rot (Botrytis cinerea Pers.) with 3 Vitis viniferea L. cultivars. Vitis 32: 87-97.

Penman, H. (1955) Humidity: Monographs for students. Chapman and Hall, London.

Petrie, P. (2002) The influence of management practices on grapevine growth and gas exchange. Ph.D. Thesis. Lincoln University, N.Z.

Phillip, P., Foott, J., and Righetti, L. (1990) Grape pruning can affect Botrytis bunch rot. California Agric. 44: 9-10.

Poni, S., Lakso, A., Turner, J. and Melious, R. (1994) Interactions of crop level and late season water stress on growth and physiology of field-grown Concord grapevines. Am. J. Enol. Vitic. 45: 252-258.

References 188

Poni, S., Intrieri, C. and Gomez-del-Campo, M. (2001) Vine performance and leaf physiology of conventionally and minimally pruned Sangiovese grapevines. Vitis 40(3): 123-130.

Poni, S., Intrieri, C. and Magnanini, E. (2000) Seasonal growth and gas exchange of conventionally and minimally pruned Chardonnay canopies. Vitis 39(1): 13-18.

Possingham, J., Clingeleffer, P. and Cooper, A. (1990) Vine management techniques for the wine industry. Aust. Grapegrower Winemaker 316: 85-89.

Powles, S. (1984) Photoinhibition of photosynthesis induced by visible light. Ann. Rev. Plant Physiol. Plant Mol. Biol. 35: 15-44.

Rayment, G. and Higginson, F. (1992) Australian laboratory handbook of soil and water chemical methods. Inkata Press, Melbourne.

Reynolds, A., Price, S., Wardle, D. and Watson, B. (1994) Fruit environment and crop level effects on Pinot Noir. I. Vine performance and fruit composition in British Columbia. Am. J. Enol. Vitic. 45 (4): 452-459.

Reynolds, A. and Wardle, D. (2001) Evaluation of minimal pruning upon vine performance and berry composition of Chancellor. Am. J. Enol. Vitic. 52(1): 45-48.

Rodrigues, M., Chaves, M., Wendler, R., David, M., Quick, W., Leegood, R., Stitt, M. and Pereira, J. (1993) Osmotic adjustment in water stressed grapevine leaves in relation to carbon assimilation. Aust. J. Plant Physiol. 20: 309-322.

Rojas-Lara, B. and Morrison, J. (1989) Differential effects of shading fruit or foliage on the development and composition of grape berries. Vitis 28:199-208.

Rosenquist, J. and Morrison, J. (1989) Some factors affecting cuticle and wax accumulation on grape berries. Am. J. Enol. Vitic. 40(4): 241-244.

Sall, M., Teviotdale, B. and Savage, S. (1982) Bunch Rots. In ‘Grape Pest Management.’ University of California, Publication No. 4105. University of California Press, Berkeley. pp 53-56.

Savage, S. and Sall, M. (1983) Botrytis bunch rot of grapes: The influence of selected cultural practices on infection under California conditions. Plant Dis. 67: 771-774.

References 189

Savage, S. and Sall, M. (1984) Botrytis bunch rot of grapes: The influence of trellis type and canopy microclimate. Phytopathol. 74: 65-70.

Schultz (2003) Differences in hydraulic architecture account for near-isohydric and anisohydric behaviour of two field-grown Vitis vinifera L. cultivars during drought. Plant, Cell, Enviro. 26: 1393- 1405.

Schultz, H. (1996) Leaf absorptance of visible radiation in Vitis vinifera L.: estimates of age and shade effects with a simple field method. Sci. Hort. 66: 93-102.

Schultz, H. and Matthews, M. (1993) Growth, osmotic adjustment and cell-wall mechanics of expanding grape leaves during water deficits. Crop Sci. 33: 287-294.

Schultz, H., Kramel, S., Werwitzke, U., Zimmer, T. and Schmid, J. (2001) Adoption and utilization of minimal pruning systems for quality production in cool climates. In ‘Proceedings of the ASEV 50th Anniversary Meeting, Seattle, Washington, June 19-23 2000’. American Society for Enology and Viticulture, Davis, California. pp. 185-190.

Singleton, V. (1972) Effects of red wine quality of removing juice before fermentation to simulate variation in berry size. Am. J. Enol. Vitci. 23(3): 106-113.

Smart, R. (1992) Canopy Management. In ‘Viticulture Volume 2, Practices’. Eds. B. Coombe and P. Dry. Winetitles, Adelaide, South Australia. pp. 85-103.

Smart, R. (1987) The light quality environment of vineyards. In ‘Proceedings of 3rd International Symposium Physiologie de la Vigne, June 1986’. Bordeaux. pp. 378-385.

Smart, R. and Coombe, B. (1983) Water relations of grapevines. In ‘Water deficits and plant growth.’ Eds. T. Kozlowski. Academic Press Inc., New York. pp. 137-196

Smart, R., Dick, J. and Gravett, L. (1989). Shoot devigoration by natural means. In ‘Proceedings of 7th Australian Wine Industry Technical Conference.’ Adelaide, South Australia. Eds. P. William, D. Davidson and T. Lee. Australian Wine Research Institute, Urrbrae. pp. 58-65.

Smart, R., Dick, J., Gravett, I., and Fisher, B. (1990) Canopy management to improve grape yield and wine quality- principles and practices. S. Afr. J. Enol. Vitic. 11: 3-17.

References 190

Smart, R. and Robinson, M. (1991) Sunlight into Wine. A handbook for winegrape canopy management. Winetiltes, Adelaide, South Australia.

Smart, R., Robinson, M., Due, G. and Brien, C. (1985) Canopy microclimate modification for the cultivar Shiraz. I. Definition of canopy microclimate. Vitis 24: 27-31.

Smithyman, R., Howell, G. and Miller, D. (1998) The use of competition for carbohydrates among vegetative and reproductive sinks to reduce fruit set and Botrytis bunch rot in Seyval blanc grapevines. Am. J. Enol. Vitic. 49(2): 163-170.

Soderlund, R., Clarnette, L., Tisdall, J., Adem, H., Dry, P., Smart, R. and Brien, C. (1983) Rejuvenation of low yielding vineyards. In ‘Advances in Viticulture and Oenology for Economic gain: Proceedings 5th Australian Wine Industry Technical Conference, Perth’. Eds. T. Lee and T. Somers, Australian Wine Research Institute, Urrbrae. pp. 67-79.

Somers, T. (1984) Botrytis cinerea- consequences for red wines. Aust. Grapegrower Winemaker. 244: 80-85.

Somers, T. and Evans, M. (1977) Spectral evaluation of young red wines: anthocyanin equilibria, total phenolics, free and molecular SO2, “Chemical Age”. J. Sci. Food Agric. 28: 279-287.

Sommer, K. and Clingeleffer, P. (1993) Comparison of leaf area development, leaf physiology, berry maturation, juice quality, and fruit yield of minimal and cane pruned Cabernet Sauvignon grapevines. In ‘Pruning mechanization and crop control: Proceedings of the 2nd Nelson J. Shaulis Grape Symposium, July 13-14, 1993, Fredonia, New York’. Geneva, New York. pp. 14-19.

Sommer, K. and Lang, A. (1994) Comparative analysis of two indirect methods of measuring leaf area index as applied to minimal and spur pruned grapevines. Aust. J. Plant Physio. 21: 197-206.

Sommer, K., Clingeleffer, P. and Ollat, N. (1993) Effects of minimal pruning on grapevine canopy development, physiology and cropping level in both cool and warm climates. Wein-Wiss 48: 135-139.

Sommer, K., Clingeleffer, P. and Shulman, Y. (1995) Comparative study of vine morphology, growth and canopy development in cane-pruned and minimally pruned Sultana. Aust. J. Exp. Ag. 35: 265-273.

References 191

Stoll, M., Dry, P., Loveys, B., Stewart, D. and McCarthy, M. (2000) Partial rootzone drying- effects on root distribution and commercial application of a new irrigation techniques. Aust. N.Z. Wine Industry J. 15(2): 74-77.

Striegler, K., Lake, C., Morris, J., Threlfall, R., Main, G. and Graves, S. (2002) Minimal input production systems affect yield and juice quality of ‘Sunbelt’ grapes in California’s San Joaquin Valley. HortSci. 37 (6): 867-870.

Syversten, J., Lloyd, J., McConchie, C., Kriedemann, P. and Farquhar, G. (1995) On the relationship between leaf anatomy and CO2 diffusion through the mesophyll of hypostomatous leaves. Plant Cell Enviro. 18: 149-157.

Tardieu, F., Zhang, J. and Davies, W. (1992) What information is conveyed by an ABA signal from maize roots in drying soils? Plant Cell Enviro. 15: 185-191.

Tassie, E. and Freeman, B. (1992) Pruning. In ‘ Viticulture. Volume 2, Practices’. Winetitles, Adelaide. pp. 66-84.

Thomas, C., Marois, J. and English, J. (1988) The effects of wind speed, temperature and relative humidity on development of aerial mycelium and conidia of Botrytis cinerea on grape. Phytopathol. 78: 260-265.

Turner, N. (1988) Measurement of plant water status by pressure chamber technique. Irrig. Sci. 9:289- 308.

Vaadia, M. (1997) Copigmentation in varietal wines from the Napa Valley. M.S. Thesis, University of California, Davis.

Vail, M. and Marois, J. (1991) Grape cluster architecture and susceptibility of berries to Botrytis cinerea. Phytopathol. 81: 188-191. van Zyl, J. (1984) Response of Colombard grapevines to irrigation as regards to quality aspects and growth. S. Afr. Enol. Vitic. 5: 19-28. van Zyl, J. (1987) Diurnal variation in grapevine water stress as a function of changing soil water status and meteorological conditions. S. Afr. J. Enol. Vitic. 8:45-52.

References 192

Virgona, J., Hubick, K., Rawson, H., Farquar, J. and Downes, R. (1990) Genotypic variation in transpiration efficiency, carbon-isotope discrimination and carbon allocation during early growth in sunflower. Aust. J. Plant Physiol. 17: 207-214. von Caemmerer, S. and Farquhar, G. (1981) Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta 153: 376-387.

Warren, M., Laukart, N. and Mebalds, M. (2000) Inoculation experiments and spore monitoring. In ‘Proceedings of Botrytis Researcher’s Workshop- 26-28 May 1999, Knoxfield.’ NRE Victoria. pp. 13- 18.

Williams, L. and Grimes, D (1987) Modelling vine growth- development of a data set for a water balance subroutine. In ‘Proceedings of the Sixth Australian Wine Industry Technical Conference Adelaide, S.A., 14-17 July 1986’. Eds. T. Lee. Australian Industrial Publishers, Adelaide. pp. 167-174.

Williams, L. and Matthews, M. (1990) Grapevine. In ‘Irrigation of Australian crops, Volume 30’. Eds. B. Stewart and D. Nielsen. American Society of Agronomy, Madison. pp. 1019-1055.

Winkel, T. and Rambal, S. (1993) Influence on water stress on grapevines growing in the field: from leaf to whole-plant response. Aust. J. Plant Physiol. 20:143-157.

Winkler, A., Cook, J., Kliewer, W. and Lider, L. (1974) Pruning. In ‘General Viticulture’. University of California Press, Berkley, California. pp. 287-337

Wright, G., Nageswara Rao, R. and Farquhar, G. (1994) Water-use efficiency and carbon-isotope discrimination in peanut under water deficit conditions. Crop Sci. 34: 92-97.

Wünsche, J., Palmer, J. and Greer, D. (2000) Effects of crop load on fruiting and gas-exchange characteristics of ‘Braeburn’/M.26 apple trees at full canopy. J. Am. Soc. Hort. Sci. 125(1): 93-99.

Zhang, J. and Davies, W. (1989a) ABA production in dehydrating roots may enable the plant to measure the water status of the soil. Plant Cell Enviro. 12: 73-81.

Zhang, J. and Davies, W. (1989b) Sequential response of whole plant water relations to prolonged soil drying and the involvement of xylem sap ABA in the regulation of stomatal behaviour of sunflower plants. New Phytol. 113: 277-285.

References 193

Zhang, J. and Davies, W. (1990) Changes in the concentration of ABA in xylem sap as a function of changing water status can account for changes in leaf conductance and growth. Plant Cell Enviro. 13: 277-285.

Zhang, J. and Davies, W. (1991) Antitranspirant activity in xylem sap of maize. J. Exp. Bot. 42: 1125- 1132.

Zuffery, V., Murisier, F. and Schultz, H. (2000) A model analysis of the photosynthetic response of Vitis vinifera L. cvs Riesling and Chasselas leaves in the field: I. Interaction of age, light and temperature. Vitis 39(1):19-26.

APPENDIX 194

APPENDIX A

Schematic map of the research site at Deakin Estate, Iraak, Victoria highlighting in yellow the selected experimental rows and corresponding irrigation treatments.

Location:DEAKIN ESTATE Variety: SHIRAZ N Irrigation Row No. 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 SD 42 43 44 45 46 47 48 PRD 49 50 51 52 53 54 55 56 57 58 59 60 PRD 61 62 SD 63 64 65 66 67 68 69 70 71 72 PRD 73 74 SD 75 76 77 78 79 80 81 82 83 84 85 86 Vine No. 0 102030405060 160

Scale 24 m

APPENDIX 195

APPENDIX B

Schematic representation of irrigation and pruning experimental design at Deakin Estate, Iraak, Victoria.

Experimental Design

Row Irrigation Panel 1 Panel 2 Panel 3 Panel 4 Panel 5 Panel 6 42 SD MECH MIN SPUR SPUR MIN MECH 63 SD SPUR MECH MIN MECH SPUR MIN 75 SD MIN SPUR MECH MIN MECH SPUR

49 PRD SPUR MIN MECH MECH MIN SPUR 61 PRD MECH SPUR MIN SPUR MECH MIN 73 PRD MIN MECH SPUR MIN SPUR MECH

Key SD = Standard dri[p irrigation PRD = Partial rootzone drying irrigation SPUR = Spur pruning MECH = Mechanical hedging MIN = Minimal pruning

Note: each panel = 8 vines

KEY

SD = Standard drip irrigation PRD = Partial rootzone drying SPUR = Spur pruning MECH = Mechanical hedging MIN = Minimal pruning

NOTE: Each panel consists of 8 vines

APPENDIX 196

APPENDIX C

Standardised procedure and formulae for the determination of anthocyanin and phenolic content and concentration of winegrapes (adapted from Iland 2001).

1. Record the weight of a 100 berry sample (0.01g accuracy) (A).

2. Macerate the berries to a homogenous mixture and measure approximately 1g of homogenate into a pre-weighed and record the weight (B).

3. Add 10 mL of 50% (v:v) aqueous ethanol (acid adjusted to pH 2) to the homogenate.

4. After 1 hour, centrifuge the supernatant at 3500 rpm for 10 minutes.

5. Record the volume of the supernatant (C) and add 1.0 mL of the supernatant to 10.0 mL of 1M HCl.

6. Allow the diluted HCl extract to incubate for 3 hours at 25°C.

7. Using a spectrophotometer, read the absorbance of the diluted HCl extract at 520 nm and 280nm, using a 10 mm path-length.

Calculations:

Anthocyanin Content (mg anthocyanins / berry)

=A520 x dilution factor x C x A x 1000 500 100 B 50

Anthocyanin Concentration (mg anthocyanins / g berry weight)

=A520 x dilution factor x C x A x 1000 500 100 B A

Phenolic Content (mg phenolics / berry)

=A280 x dilution factor x C x A x 1 100 B 50

Phenolic Concentration (mg phenolics / g berry weight)

=A280 x dilution factor x C x A x 1 100 B A

APPENDIX 197

APPENDIX D Standardised procedure and formulae for the determination of wine anthocyanin and phenolic content and concentration (adapted from Somers and Evans 2001).

1. Using a spectrophotometer, record the wine spectrum of 400 – 500 nm using a 5 mm path- length cell and note absorbancies at 420 and 520 nm.

2. Add 20 µl sodium metabisulphide (SO2) solution to the above sample, mix thoroughly by inversion for 1 minute and record the absorbance at 520 nm.

3. Add 20 µl acetaldehyde solution to 2 ml wine and incubate for 45 minutes at 25°C. Measure absorbance at 520 nm in a 5 mm cell.

4. Add 100 µl wine to 10 ml 1M HCl and incubate for 3-4 hours. Measure absorbancies at 520 nm and 280 nm in a 10 mm cell and correct for the dilution used.

5. Convert all absorbance readings to absorbance at a 10 mm path-length (A).

Calculations

A values represent absorbancies at 420 nm or 520 nm of the wine, HCl supernatant or SO2 supernatant, corrected to 10 mm path-length.

a. Wine density (AU)= A420 + A520

b. Wine hue (AU) = A420

A520

c. Degree of ionisation of anthocyanins (α) = 100% x A520 - A520(SO2)

A520(HCl) - 5/3 . A520(SO2)

-1 d. Total anthocyanins (mg.l ) = 20(A520(HCl) - 5/3.A520(SO2))

e. Ionised Anthocyanins (mg.l-1) = α x Total anthocyanins

= 20(A520 – A520(SO2))

f. Total phenolics (AU) = A280 – 4

APPENDIX 198

APPENDIX E

ANOVA analysis of irrigation and pruning treatment effects on yield vine-1 for seasons 2001-2003.

Genstat 5 Release 4.1 (PC/Windows NT) 14 March 2003 14:45:10

______

Genstat 5 Fourth Edition - (for Windows)

Genstat 5 Procedure Library Release PL11

______

"Data taken from File: F:/HARVEST 2003/HARVEST 03 GENSTAT.XLS"

DELETE [redefine=yes] Sample,Plot,Row,Vine,Treatment,Irrig,Prune,Yield_vine

VARIATE [nvalues=72] Sample

READ Sample

Identifier Minimum Mean Maximum Values Missing

Sample 1.00 36.50 72.00 72 0

FACTOR [modify=yes;nvalues=72;levels=6] Plot

READ Plot; frepresentation=ordinal

Identifier Values Missing Levels

Plot 72 0 6

FACTOR [modify=yes;nvalues=72;levels=!(42,49,61,63,73,75)] Row

READ Row; frepresentation=ordinal

Identifier Values Missing Levels

Row 72 0 6

VARIATE [nvalues=72] Vine

APPENDIX 199

READ Vine

Identifier Minimum Mean Maximum Values Missing

Vine 3.00 24.61 45.00 72 0

FACTOR [modify=yes;nvalues=72;levels=6] Treatment

READ Treatment; frepresentation=ordinal

Identifier Values Missing Levels

Treatment 72 0 6

FACTOR [modify=yes;nvalues=72;levels=2] Irrig

READ Irrig; frepresentation=ordinal

Identifier Values Missing Levels

Irrig 72 0 2

FACTOR [modify=yes;nvalues=72;levels=3] Prune

READ Prune; frepresentation=ordinal

Identifier Values Missing Levels

Prune 72 0 3

VARIATE [nvalues=72] Yield_vine

READ Yield_vine

Identifier Minimum Mean Maximum Values Missing

Yield_vine 2.83 10.09 23.81 72 0

"General Analysis of Variance."

BLOCK Row*Plot

TREATMENTS Irrig*Prune

COVARIATE "No Covariate"

ANOVA [PRINT=aovtable,information,mean; FACT=32; FPROB=yes; PSE=diff,lsd,means;

LSDLEVEL=5]\Yield_vine

APPENDIX 200

......

***** Analysis of variance *****

Variate: Yield_vine

Source of variation d.f. s.s. m.s. v.r. F pr.

Row stratum

Irrig 1 67.048 67.048 17.49 0.014

Residual 4 15.337 3.834 0.22

Plot stratum 5 134.043 26.809 1.52

Row.Plot stratum

Prune 2 14.848 7.424 0.42 0.663

Irrig.Prune 2 18.972 9.486 0.54 0.593

Residual 21 371.492 17.690 2.98

Row.Plot.*Units* stratum

36 214.000 5.944

Total 71 835.741

………………………………………………..

***** Tables of means *****

Variate: Yield_vine

Grand mean 10.09

Irrig SD PRD

11.06 9.13

Prune SPUR MECH MIN

9.54 10.07 10.65

Irrig*Prune SPUR MECH MIN

SD 10.66 11.58 10.93

PRD 8.42 8.57 10.38

APPENDIX 201

*** Standard errors of means ***

Table Irrig Prune Irrig*Prune rep. 36 24 12 e.s.e. 0.326 0.859 1.044 d.f. 4 21 24.30

*** Standard errors of differences of means ***

Table Irrig Prune Irrig*Prune rep. 36 24 12 s.e.d. 0.462 1.214 1.476 d.f. 4 21 24.30

*** Least significant differences of means (5% level) ***

Table Irrig Prune Irrig*Prune rep. 36 24 12 l.s.d. 1.281 2.525 3.044 d.f. 4 21 24.30