Integration of Aphthona spp. flea beetles and herbicides for leafy spurge (Euphorbia esula) control in the habitat of the western prairie fringed orchid (Platanthera praeclara), a threatened species Ann M. Erickson and Rodney G. Lym 1 Summary Leafy spurge is a serious threat to maintaining biodiversity in rangelands and pastures of the northern Great Plains in the United States and Canada. Leafy spurge is threatening the habitat of the western prairie fringed orchid (Platanthera praeclara Sheviak & Bowles), a federally listed threatened species in the US and endangered species in Canada. Aphthona spp. flea beetles, biological control agents for leafy spurge, have established and controlled leafy spurge in the region, but not in habitat of the orchid. Also, current law prohibits use of herbicides in areas where the orchid grows. Previous research had shown that leafy spurge could be controlled with imazapic or quinclorac with minimal or no injury to the orchid. The purpose of this research was to evaluate leafy spurge control and Aphthona spp. establishment in habitat of the orchid, using imazapic or quinclorac in combination with the flea beetles. This combination method in other habitats has resulted in better leafy spurge control than either method used alone and has increased biological control agent establishment. Leafy spurge was treated with Aphthona flea beetles, herbicides, and Aphthona flea beetles plus herbicides. Leafy spurge stem density decreased from about 114 to 4 stems/m 2 the season following treatment with herbicides alone and from an average of 126 to 1 stem/m 2 the following season, when the treatment included both Aphthona flea beetles and herbicides. Stem density of leafy spurge treated with only flea beetles decreased from 150 to 41 stems/m 2. The population of flea beetles the season following release was estimated. Fewer than 1/m 2 Aphthona flea beetles were collected with sweep nets in plots where the flea beetles were not released, 6/m 2 in plots treated only with flea beetles, and 2/m 2 in plots treated with both flea beetles and herbicides. This is the first establishment of a leafy spurge biological control agent in the habitat of this orchid. Keywords: Aphthona, imazapic, IPM, leafy spurge, quinclorac, threatened species. Introduction Platanthera praeclara Sheviak & Bowles, the western prairie fringed orchid, is a native plant of the tallgrass prairie that was placed on the federal threatened species list in 1989 (US Fish and Wildlife Service 1989). Various threats to the survival of P. praeclara exist and include habitat invasion by Euphorbia esula L., leafy 1 Department of Plant Sciences, North Dakota State University, Fargo ND 58105 spurge (Sieg & Bjugstad 1994, US Fish & Wildlife Service 1996, Wolken et al. 2001). E. esula is one of the most widespread and competitive noxious weeds in North America, where it invades mainly non-cultivated areas such as native prairie and rangeland (Hanson & Rudd 1933, Selleck et al. 1962). E. esula decreases forage production for range animals, suppresses native plant species, and decreases biodiversity (Westbrooks 1998). E. esula is very difficult to control with methods other than herbicides, but herbicides cannot be used in areas where the orchid is located due to its status as a 389
Proceedings of the XI International Symposium on Biological Control of Weeds federally listed threatened species. Biological control of E. esula in the habitat of the threatened species would seem to be the least harmful approach. Biological control of E. esula with the use of Aphthona spp. flea beetles in North Dakota began in the mid-1980s (Carlson & Mundal 1990), but establishment of the flea beetles in the habitat of P. praeclara has not yet been successful (Lym 1998; Mundal et al. 2000). An experiment to evaluate herbicides for E. esula control in North Dakota in the early 1990s had to be discontinued two years after establishment due to the appearance of P. praeclara in areas treated with fallapplied herbicides (Kirby et al. 2003). Subsequent research found imazapic and quinclorac provided good E. esula control with little or no injury to the orchid (Kirby et al. 2003, Sterling et al. 2000a, b). Therefore, this study was initiated to evaluate the interaction of imazapic and quinclorac with Aphthona spp. flea beetles for E. esula control in the habitat of P. praeclara. Aphthona spp. flea beetles (Coleoptera: Chrysomelidae) are some of the most promising biological control agents of E. esula. These insects are natural enemies of E. esula in eastern and central Europe. Years of research with Aphthona spp. flea beetles were conducted on several plant species to determine the host range of the insects (Harris et al. 1985). Six species have been released in the US, but the three most effective species of these flea beetles are A. czwalinae Weise., A. lacertosa Ross. and A. nigriscutis Foudr. (Hansen et al. 1997, Mundal et al. 2000). Aphthona spp. flea beetles oviposit eggs at the base of an E. esula plant and the eggs hatch in 14 to 19 days (Mundal et al. 2000). The larvae go through three instars, and with each instar, the larvae feed on progressively larger roots (Gassmann et al. 1996, Hansen et al. 1997, Mundal et al. 2000). Aphthona larvae prefer to feed on previously attacked root sections, so aggregate feeding occurs, which destroys the root sections (Gassmann et al. 1996). A. abdominalis is the only multivoltine Aphthona species released in the US, with four generations per year, and overwintering in the adult stage (Fornasari 1993). All five other Aphthona species released in the US are univoltine (Gassmann et al. 1996). Third instar larvae of univoltine species overwinter and pupate the following spring (Gassmann et al. 1996, Mundal et al. 2000). Adult univoltine flea beetles emerge from late May to early July and live for 2 to 3 months, during which time they feed on the leaves and stems of E. esula, but do not contribute greatly to its control (Gassmann et al. 1996, Hansen et al. 1997). Larval feeding may kill the plant directly by disrupting water and nutrient transport or indirectly by creating pathways for plant pathogens to enter the plant (Hansen et al. 1997). E. esula densities may be reduced when Aphthona spp. flea beetles become established (Kirby et al. 2000), but the control of E. esula with the flea beetles has not occurred in all habitats (Lym 1998). Establishment of Aphthona spp. flea beetles has been variable because they usually do not survive well in habitats that are moist, shady, contain sandy soil, or have high E. esula densities (Lym 1998, Mundal et al. 2000), which are characteristics of the habitat of P. praeclara. The use of Aphthona spp. flea beetles is an ecologically favourable control method for E. esula, but, as noted, the flea beetles generally do not survive well in the sandy, mesic habitat of P. praeclara. In fact, no releases made in the habitat of the orchid have effectively established (Lym 1998). Establishment of Aphthona spp. flea beetles in the habitat of P. praeclara may be improved using herbicides (Lym & Nelson 2002, Nelson 1999). An apparently unsuccessful Aphthona spp. population increased rapidly from an average of 14 flea beetles swept/m 2 to an average of 76 flea beetles swept/m 2 1 year after herbicide application in a study by Lym & Nelson (2002). E. esula stem density decreased from 114 stems/m 2 to 8 stems/m 2 one year after fall application of imazapic, and no E. esula stems remained in the study area two years after herbicide application. E. esula control generally occurred more rapidly and was maintained for longer periods when herbicides were used in conjunction with Aphthona spp. flea beetles than when either method was used alone in a number of experiments by Lym & Nelson (2002). Aphthona flea beetles also are compatible with other methods of E. esula control, such as sheep grazing and prescribed burning, and the integration of the flea beetles with either method controls E. esula better than any of the control methods used alone (Beck & Rittenhouse 2000, Fellows & Newton 1999). Materials and method An experiment to evaluate the interaction of imazapic and quinclorac with Aphthona spp. flea beetles for E. esula control in the habitat of P. praeclara was established in June 2001. The experiment was located near a large population of orchids. The experiment was arranged as a randomized complete block-design with a split-plot arrangement and four replicates. Whole plots were 3.05 m wide and 9.15 m long. Whole plots consisted of herbicides alone, and subplots consisted of flea beetles plus herbicides, flea beetles alone, and an untreated control (neither flea beetles nor herbicides) (Nelson 1999). Measures were compared to an untreated check using an ANOVA, and individual treatment means were separated using Fisher s-protected LSDs calculated at the 95% levels of confidence. The soil at the site was classified as a Hecla Hamar Arveson association, which is sandy, mixed, frigid Oxyaquic Hapludolls; sandy, mixed, frigid Typic Endoaquolls; and coarse-loamy, mixed, superactive, frigid Typic Calciaquolls; respectively (US Soil Conservation Service 1975). The soil was 75:20:5 sand:silt:clay and was analyzed for nutrients (Table 1). 390
Integration of Aphthona and herbicides Table 1. Depth (cm) Western prairie fringed orchid research site soil characteristics at two depths. N (kg/ha) P (ppm) K (ppm) ph EC OM (%) 0 to 15 12 3 140 7.1 0.12 4.3 15 to 30 9 3 75 NA NA NA A mixture of A. czawalinae and A. lacertosa was collected from an established population near Lisbon, North Dakota, approximately 29 km from the experiment s location. Approximately 350 adult A. czawalinae and A. lacertosa were released into insect cages on June 27, 2001, and 100 additional beetles were released on July 17, 2001 to ensure appropriate sex ratios (Olson & Mundal 1999). Cages were 1.8 by 1.8 by 1.8 m with a PVC frame covered by a plastic screen (32 32 Lumite) (Nelson 1999, Lym & Nelson 2002, Nelson & Lym 2003). Imazapic and quinclorac were applied on September 20, 2001 using a CO 2 -pressurized backpack sprayer delivering 80 L/ha at 240 kpa with four flat-fan 8001 nozzles. Cages were removed from the plots for the winter prior to herbicide application (Nelson 1999, Lym & Nelson 2002, Nelson & Lym 2003). The effects of the interaction of imazapic and quinclorac with Aphthona spp. were evaluated by counting the number of larvae that developed into adults from soil cores collected in October 2001 and May 2002, and by counting adults in the field in late June and early July 2002. Four soil cores were collected per subplot using a golf cup-cutter. The golf cup-cutter was placed over an E. esula root crown, and cores 10.8 cm in diameter were cut to a depth of 15 cm. Soil cores collected in the fall were placed in plastic bags, transported, and stored in a refrigeration unit for 75 days at 3 C. Vernalization induced larvae to pupate and emerge as adults. After 75 days, the soil cores were placed into 0.9 L paper cups. Cups were covered with 2 L clear plastic cylinders. The covered cups containing soil cores were maintained in the laboratory at 21 C with a 16h photoperiod under artificial lighting. Soil cores collected in the spring were brought directly into the laboratory. Adults emerging from soil cores from both fall and spring collections were counted and removed from trap chambers daily. Soil cores were discarded 2 weeks after the last adult was collected (approximately 4 weeks) (Nelson 1999, Nelson and Lym 2003). To estimate Aphthona flea beetle density in the field, vegetation in the subplots was swept for adults with a sweep net having a 38 cm diameter hoop. Quarters of the subplot and portions of the subplot border, each totalling 1 m 2, were swept in five sweeps in the spring, and adults captured were counted and returned. E. esula control was monitored by counting E. esula stems in four 0.25 m 2 areas in each subplot both before and after treatment (Lym and Nelson 2002, Nelson 1999). Results Imazapic or quinclorac applied alone or with Aphthona spp. flea beetles reduced E. esula density more than flea beetles alone (Table 2). E. esula stem density was reduced from an average of 150 to 41 stems/m 2 (53% control) with A. czwalinae/lacertosa alone compared to a reduction from an average of 114 to 4 stems/m 2 (96% control) and from an average of 126 to 1 stem/m 2 (99% control) with herbicides alone and herbicides with flea beetles, respectively, one season following treatment. Imazapic and quinclorac provided similar E. esula control one season following treatment regardless of application rate. Table 2. Control of leafy spurge one season following release of Aphthona spp. flea beetles, herbicide application, or both. Leafy spurge density (stems/m 2 ) June 4, 2001 June 5, 2002 Imazapic 140 104 7 Imazapic + Aphthona a 140 115 <1 Imazapic 210 105 1 Imazapic + Aphthona 210 150 0 Quinclorac 840 96 4 Quinclorac + Aphthona 840 132 0 Quinclorac 1120 149 3 Quinclorac + Aphthona 1120 107 2 Control 99 80 Control + Aphthona 150 41 LSD (0.05) 44 11 Imazapic and quinclorac did not affect A. czwalinae/ lacertosa adult emergence from soil cores collected in the fall or spring (Table 3). However, approximately 15 flea beetles emerged per subplot from soil cores collected in the fall while only an average of 7 flea beetles emerged per subplot from soil cores collected in the spring. This 47% decrease in the number of flea beetles that emerged may have been due to low soil temperatures in the winter of 2001 to 2002 caused by little snow cover. A similar study reported that soil temperatures in the winter of 1995 to 1996 that were colder than normal for an extended period of time caused a 60% winter kill of A. nigriscutis adults taken from soil cores collected in the spring compared to soil cores collected in the fall (Nelson 1999, Nelson and Lym 2003). When that study was repeated in 1996, snow accumulated in record amounts in the winter of 1996 to 1997, which insulated the soil and prevented winter kill of larvae. The numbers of A. czwalinae/lacertosa adults collected in the field were higher in subplots without herbicide treatment compared to subplots with herbicide treatment. An average of six Aphthona flea beetles 391
Proceedings of the XI International Symposium on Biological Control of Weeds per m 2 was collected with sweep nets from subplots that received flea beetles alone compared with an average of 2 flea beetles per m 2 collected from subplots that received both imazapic or quinclorac and flea beetles (Table 4). Even though fewer Aphthona adults were collected where herbicides were applied, neither the choice of herbicide nor the application rate of the herbicide affected the numbers of flea beetles collected. Picloram plus 2,4-D did not affect the number of A. nigriscutis adults collected in the field in a similar study (Nelson 1999, Lym & Nelson 2002). However, E. esula density one season following the release of flea beetles and herbicide application was 55 stems/m 2 in the study by Lym & Nelson, whereas it was less than 1 stem/m 2 in this study (Table 2). Fewer Aphthona flea beetles collected from subplots that received imazapic or quinclorac plus flea beetles may have resulted because of the low densities of E. esula for adults to feed on the season following herbicide application. Table 3. Effect of imazapic and quinclorac on Aphthona spp. flea beetle establishment according to emergence from soil cores. Emerged beetles (no./subplot) Fall b Spring c Imazapic + Aphthona a 140 12 5 Imazapic + Aphthona 210 19 8 Quinclorac + Aphthona 840 13 7 Quinclorac + Aphthona 1120 16 4 Control + Aphthona 13 10 LSD (0.05) NS NS b Soil cores were collected on October 18, 2001 and stored for 75 days at 3 C. c Soil cores were collected on May 13, 2002. Table 4. Population of Aphthona spp. flea beetles when estimated per five sweeps the season following flea beetle release and herbicide application. Collected Aphthona flea beetles (no./ m 2 ) June 28, 2002 July 3, 2002 Imazapic 140 <1 0 Imazapic + Aphthona a 140 1 2 Imazapic 210 0 0 Imazapic + Aphthona 210 2 2 Quinclorac 840 <1 0 Quinclorac + Aphthona 840 1 2 Quinclorac 1120 <1 <1 Quinclorac + Aphthona 1120 1 2 Control 1 0 Control + Aphthona 5 6 LSD (0.05) 2 2 Discussion This is the first reported establishment of Aphthona spp. flea beetles in the habitat of P. praeclara. However, an effective establishment may take up to five years. Using imazapic or quinclorac in conjunction with Aphthona spp. flea beetles may be useful to enhance establishment of flea beetles in the habitat of the orchid. Once the flea beetles become established, need for yearly herbicide applications decreases. An integrated approach to leafy spurge control may reduce costs for land managers (Lym & Nelson 2002). References Beck, K.G. & Rittenhouse, L.R. (2000) Leafy spurge management with sheep and flea beetles. Proceedings of the Western Society for Weed Science 53, 48 49. Carlson, R.B. & Mundal, D. (1990) Introduction of insects for the biological control of leafy spurge in North Dakota. North Dakota Farm Research 47, 7 8. Fellows, D.P. & Newton, W.E. (1999). Prescribed fire effects on biological control of leafy spurge. Journal of Range Management 52, 489 493. Fornasari, L. (1993) Life history of the flea beetle, Aphthona abdominalis Duftschmid, on Euphorbia esula L. (leafy spurge) in Italy. Biological Control 3, 161 175. Gassmann, A., Schroeder, D., Maw, E. & Sommer, G. (1996) Biology, ecology, and host specificity of European Aphthona spp. (Coleoptera, Chrysomelidae) used as biocontrol agents for leafy spurge, Euphorbia esula (Euphorbiaceae), in North America. Biological Control 6, 105 113. Hansen, R.W., Richard, R.D., Parker, P.E. & Wendel, L.E. (1997) Distribution of biological control agents of leafy spurge (Euphorbia esula L.) in the United States: 1988 1996. Biological Control 10, 129 142. Hanson, H.C. & Rudd, V.E. (1933) Leafy spurge life history and habits. North Dakota Agricultural College, Agriculture Experimental Station Bulletin No. 266. Harris, P., Dunn, P.H., Schroeder, D. & Vonmoos, R. (1985) Biological control of leafy spurge in North America. Weed Science Society of America Monograph no. 3: Leafy Spurge (ed A.K. Watson), pp 79 92. Champaign, Illinois. Kirby, D.R., Carlson, R.B., Krabbenhoft, K.D., Mundal, D. & Kirby, M.M. (2000) Biological control of leafy spurge with introduced flea beetles (Aphthona spp.). Journal of Range Management 53, 305 308. Kirby, D.R., Lym, R.G., Sterling, J.J. & Sieg, C.H. (2003) Leafy spurge control and soil seedbank composition in western prairie fringed orchid habitat. Journal of Range Management 56, in press. Kuehl, B.D. & Lym, R.G. (1997) Leafy spurge (Euphorbia esula) control with quinclorac. Weed Technology 11, 265 269. Lym, R.G. (1998) The biology and integrated management of leafy spurge (Euphorbia esula) on North Dakota rangeland. Weed Technology 12, 367 373. Lym, R.G. & Nelson, J.A. (2002) Integration of Aphthona spp. flea beetles and herbicides for leafy spurge (Euphorbia esula) control. Weed Science 50, 812 819. Markle, D.M. & Lym, R.G. (2001) Leafy spurge (Euphorbia esula) control and herbage production with imazapic. Weed Technology 15, 474 480. 392
Integration of Aphthona and herbicides Mundal, D.A., Olson, D.L. & Carlson, R.B. (2000) The effect of Aphthona spp. flea beetle (Coleoptera: Chrysomelidae) larval feeding on leafy spurge, Euphorbia esula L., root systems and stem density in North Dakota: 1986 1996. USDA ARS TEAM Leafy Spurge Report. Nelson, J.A. (1999) Integration of the flea beetle, Aphthona nigriscutis Foudras, and herbicides for control of leafy spurge, Euphorbia esula L. Ph. D. Thesis, North Dakota State University, Abstract # AAT 9965430. Nelson, J.A. & Lym, R.G. (2003) Interactive effects of Aphthona nigriscutis and picloram plus 2,4-D in leafy spurge (Euphorbia esula). Weed Science 51, 118 124. Olson, D.L. & Mundal, D.A. (1999) Sex ratio effects on fecundity and fertility of a leafy spurge flea beetle Aphthona lacertosa. Proceedings of the Leafy Spurge Symposium (ed R.J. Lorenz), pp. 9 10. North Dakota State University, Fargo. Selleck, G.W., Coupland, R.T. & Frankton, C. (1962) Leafy spurge in Saskatchewan. Ecological Monographs 32, 1 29. Sieg, C.H. & Bjugstad, A.J. (1994) Five years of following the western prairie fringed orchid (Platanthera praeclara) on the Sheyenne National Grassland, North Dakota. Proceedings of the 13 th North American Prairie Conference (eds R.G. Wickett, P.D. Lewis, A. Woodliffe, & P. Pratt), pp. 141 146. Department of Parks and Recreation, Windsor, Ontario. Sterling, J.J., Kirby, D.R., Lym, R.G. & Sieg, C.H. (2000a) Response of the western prairie fringed orchid to herbicides for leafy spurge. Proceedings of the Western Society for Weed Science, p. 98. Sterling, J.J., Lym, R.G. & Kirby, D.R. (2000b) Response of the prairie fringed orchid to herbicides for leafy spurge control. Research Program Report of the Western Society for Weed Science, pp. 207 209. US Fish and Wildlife Service (1989) Endangered and threatened wildlife and plants; determination of threatened status for Platanthera leucophaea (eastern prairie fringed orchid) and Platanthera praeclara (western prairie fringed orchid). Federal Register 54, 39857 39862. US Fish and Wildlife Service (1996) Platanthera praeclara (western prairie fringed orchid) recovery plan. US Fish and Wildlife Service, Ft. Snelling, Minnesota. US Soil Conservation Service (1975) Soil survey of Richland County, and Sheyenne National Grassland Area of Ransom County, North Dakota. US Government Printing Office, Washington, D.C. Westbrooks, R.G. (1998) Invasive plants, changing the landscape of America: Fact book, pp. 26 28. Federal Interagency Committee for the Management of Noxious and Exotic Weeds, Washington, D.C. Wolken, P.M., Sieg, C.H. & Williams, S.E. (2001) Quantifying suitable habitat of the threatened western prairie fringed orchid. Journal of Range Management 54, 611 616. 393