Biological Control 23, 191–212 (2002) doi:10.1006/bcon.2001.0994, available online at http://www.idealibrary.com on COMMENTARY Potential for Biological Control of Phragmites australis in North America Lisa Tewksbury,* Richard Casagrande,* Bernd Blossey,† Patrick Ha¨fliger,‡ and Mark Schwarzla¨nder‡,1 *Department of Plant Sciences, University of Rhode Island, Kingston, Rhode Island 02881; †Department of Natural Resources, Fernow Hall, Cornell University, Ithaca, New York 14853; and ‡CABI Bioscience Centre Switzerland, CH-2800 Dele´mont, Switzerland Received September 15, 2000; accepted September 5, 2001; published online December 12, 2001 lis will require considerable dialogue. This decision Phragmites australis is a cosmopolitan plant that is needs to weigh the current negative ecological and undergoing a population explosion in freshwater and economic impacts of P. australis and the benefits and tidal wetlands on the east coast of North America. The risks of a biological control program. © 2001 Elsevier Science rapid spread of P. australis in recent years and the Key Words: biological control; Phragmites australis; virtual absence of native herbivores feeding on P. aus- common reed; Platycephala planifrons; Archanara tralis have led wetland ecologists to believe that ei- geminipuncta; Arenostola phragmitidis; Rhizedra lu- ther the species or more aggressive genotypes were tosa; Phragmataecia castaneae; Chilo phragmitella; introduced. The historical record of the occurrence of Schoenobius gigantella; Giraudiella inclusa; acciden- P. australis in North America and the scarcity of in- tal introductions; insect herbivores; invasions. digenous herbivores provide conflicting evidence for the status of the species as native or introduced. A comparison of P. australis populations from North America and other continents using advanced genetic INTRODUCTION techniques is underway to help determine the status of current and historic North American genotypes. Lit- Common reed, Phragmites australis (Cav.) Trin. ex erature and field surveys reveal that of the 26 herbi- Steudel, is a cosmopolitan angiosperm believed by vores currently known to feed on P. australis in North many to be the most widely distributed reed species in America (many accidentally introduced during the the world, ranging all over Europe, Asia, Africa, Amer- last decade), only 5 are native. In Europe, over 170 ica, and Australia (Holm et al., 1977). A native of the herbivore species have been reported feeding on P. Old World, P. australis is able to grow in a wide range australis, some causing significant damage. Of these of habitats and displays high phenotypic and genotypic herbivores, rhizome-feeding species with consider- plasticity (Haslam, 1972a; van der Putten, 1997). Typ- able negative impact on P. australis performance in- ically, P. australis grows in open wet areas and clude the lepidopterans Rhizedra lutosa (already marshes, along riverbanks and roadsides, and in present in North America), Phragmataecia castaneae, ditches and other watercourses. Low nitrogen or phos- Chilo phragmitella, and Schoenobius gigantella. Stem- phorous availability, high salinity, extensive tidal boring moths in the genera Archanara and Arenostola flooding, and anaerobic soils may limit the growth of and the chloropid fly Platycephala planifrons can this clonal species (Chambers, 1997). P. australis is have large detrimental impacts on P. australis in Eu- wind pollinated but self-incompatible, and its seeds are rope and should be evaluated for their potential as dispersed by wind and water (Haslam, 1972a). Recruit- biological control agents. In addition, the interaction ment from seed is thought to be low but may be quite of potential control agents with accidentally intro- variable and important in the spread to new sites duced P. australis herbivores needs to be evaluated in (Haslam, 1972a; Fournier et al., 1995; McKee and North America. Regardless of the results of the genetic Richards, 1996; Meyerson et al., 2000). Vegetative analyses, any decision to introduce additional host- specific herbivores in an attempt to control P. austra- propagation through dispersal of rhizome fragments by water currents, animals, and construction equipment is another important means of colonization of new ar- 1 Present address: Department of Plant, Soil and Entomological eas. Once established, expansion of a stand occurs pri- Sciences, University of Idaho, Moscow, ID 83844-2339. marily through vegetative growth of the extensive be- 191 1049-9644/01 $35.00 © 2001 Elsevier Science All rights reserved. 192 TEWKSBURY ET AL. lowground rhizome system. Approximately two thirds to eradicate, can obstruct views, and block canals, of the biomass is allocated to the rhizome, which can streams, and drainage ditches (Holm et al., 1977). reach a depth of 2 m (Szczepansky, 1969; Haslam, The abundance of herbivores in European reed beds 1972b). This growth pattern produces homogenous (Tscharntke, 1990, 1992a,b,c) forms the base of a mul- clones with up to 200 culms/m2 that can reach4min tilayered food web. P. australis stands are important as height. staging and feeding grounds for many bird species Throughout its large distribution, P. australis plays during their annual migrations between Africa and quite different roles within the ecosystem. Extensive Eurasia (Ormerod, 1990; Berthold et al., 1993). A group reed beds are a highly valued (even considered endan- of European warblers (reed warblers) has evolved a gered) ecosystem in Europe (Tscharntke, 1992c) and close association with P. australis as an exclusive are protected because of their important ecological breeding habitat (Berthold et al., 1993). Management functions. In contrast, the rapid expansion of P. aust- of reed beds is important to maintain the quality of the ralis in North America during the past decades is con- habitat for the conservation of 13 bird species that nest sidered a threat to biodiversity in natural areas and in common reed (Tscharntke, 1992c). Recent declines of has resulted in aggressive control attempts (Marks et P. australis in Europe have caused great concern and al., 1994). The purpose of this paper is to summarize prompted the formation of EUREED, a European re- European literature focusing on P. australis protec- search program on reed dieback (van der Putten, 1997; tion, herbivores, and food-web communities and con- Brix, 1999). Ironically, habitat destruction and manip- trast it with research in North America focusing on ulation of hydrolic regimes, eutrophication, pollution, control and negative ecological impacts of P. australis. and increased disturbance, often believed responsible We present summary information on life history and for the population explosion of P. australis in North distributions of herbivores associated with P. australis America, are considered key contributing factors for in Europe and North America and discuss the status reed declines in Europe (Ostendorp, 1989; van der and evidence of P. australis as a native or introduced Putten, 1997). species in North America. Finally, we use this infor- Over the past several decades, P. australis popula- mation to evaluate the benefits and potential risks of tions in North America have dramatically increased in developing a biological control program for P. australis both freshwater and brackish wetlands, particularly in North America. along the Atlantic Coast (Marks et al., 1994). Although 3500-year-old fossil Phragmites rhizomes were found ECOLOGICAL AND ECONOMIC IMPACTS in some North American peat cores (Niering et al., OF P. AUSTRALIS 1977; Clark, 1986; Orson et al., 1987), many wetland Many cultures, past and present, have found P. aus- ecologists, seeing the rapid spread of P. australis in tralis useful for various purposes. In the former Soviet recent years, believe that it is an introduced species Union, P. australis is used for fodder and cellulose; in (Mikkola and Lafontaine, 1994). The replacement of Romania, where reeds have become an important part diverse wetland vegetation by P. australis monocul- of the economy, they are turned into pulp for paper, tures has caused declines in water birds and other cardboard, cellophane, synthetic fibers, alcohol, insula- wetland wildlife (Thompson and Shay, 1989; Jamison, tion materials, wood substitute for heating purposes, 1994; Meyerson et al., 2000) and decreases in plant and fertilizer (Rodewald-Rodescu, 1974; Holm et al., diversity and alterations in nutrient cycling and hydro- 1977; Graneli, 1984). In Egypt and throughout Europe, logic regimes (Marks et al., 1994; Chambers, 1997). A common reed is used for matting (Holm et al., 1977) wide variety of control measures are used to slow the and thatching of roofs (Haslam, 1972b). In the United invasion of P. australis (Marks et al., 1994). The U.S. Kingdom, wetlands have been constructed to treat Fish and Wildlife Service recommendations include the point-source pollution, the majority using pure reed use of herbicides, mowing, disking, dredging, flooding, bed treatment systems (Hawk and Jose, 1996). In the draining, burning, and grazing (Cross and Fleming, Netherlands, P. australis plays an important role in 1989). According to Howard et al. (1978), the most land reclamation, where it serves as a soil binder, effective control methods are cutting, draining, saltwa- preventing erosion and washouts. Polders (arable land ter flushing, herbicides, and various combinations of reclaimed from the sea), which have been constructed these methods. Summer burning of P.