Canada Goose Crop Damage Abatement in South Dakota

Human–Wildlife Interactions 5(2):315–320, Fall 2011 Canada goose crop damage abatement in South Dakota

TROY M. RADTKE,1 Department of Biology and Microbiology, South Dakota State University, Brookings, SD 57007, USA [email protected] CHARLES D. DIETER, Department of Biology and Microbiology, South Dakota State University, Brookings, SD 57007, USA Abstract: Canada geese (Branta canadensis) can cause considerable damage to crops during summer when geese are ﬂ ightless. We evaluated the effectiveness of a program designed to alleviate crop damage on soybeans by Canada geese in South Dakota, USA. The applications of electric fences, feeding stations, and propane cannons reduced the area of crop damage by 90% in 2006 and 80% in 2007, but the timing was important. Fields where abatement practices were applied early in the growing season had less damage than ﬁ elds where they were applied later. Abatement practices that were properly applied as soon as damage started and that were maintained throughout the growing season were effective at reducing damage to soybeans by Canada geese.

Key words: abatement, Branta canadensis, Canada goose, crop damage, electric fence, human–wildlife confl icts, South Dakota, molt, soybean Many Canada goose (Branta canadensis) by Canada geese is especially intense during populations have increased over the last 4 this brood-rearing and molting period when decades, and Canada geese are considered a goslings have high energy demands for growth, nuisance in many areas (Conover and Chasko and breeding females need to replenish lost 1985), including agricultural areas where they reserves aft er egg-laying and incubation can cause considerable crop damage (Schaible (Raveling 1979, Alisauskas and Ankney 1992). et al. 2005). While damage to crops by geese has In addition, molting geese have increased been diffi cult to quantify, surveys of agricultural energy demands because of feather production. producers indicate that it may be severe in some Adult geese exhibit a synchronous molt, which areas (Heinrich and Craven 1987, Conover renders them fl ightless for a period of about and Decker 1991), resulting in substantial 25 days (Bellrose 1980). The fl ightless period economic losses. Various abatement techniques is oft en the most problematic for agricultural have been used to reduce crop damage by producers because it coincides with early crop migrating Canada geese. Chemicals sprayed growth and the limited mobility of geese results on vegetation have been tested (Conover 1985, in greater persistence at foraging in crop fi elds. Cummings et al. 1991, 1992, 1995), but have South Dakota Game, Fish and Parks (SDGFP) seen litt le agricultural use because of high instituted a program in 1996 to reduce crop cost or ineff ectiveness. Hazing geese from damage caused by Canada geese. This program fi elds using sonic deterrents, such as propane costs roughly $250,000 per year and is funded exploders or shell crackers, or the use of visual by a $5 surcharge on all hunting licenses sold in deterrents, such as mylar fl ags and refl ective South Dakota. Any agricultural producer that tape, or human effi gies all reduced goose fi led a complaint of goose damage was eligible numbers in some fi elds (Heinrich and Craven to receive abatement assistance by SDGFP. 1990). However, most of this research has been Complaints of goose damage in South Dakota conducted on geese that had the ability to fl y, have been reported on corn, wheat, oats, alfalfa, with litt le research having been conducted on and sorghum (Gigliott i 2007), but in South abatement of crop damage by fl ightless geese Dakota >90% of the crop damage occurs in (Nelson and Oett ing 1998). soybeans fi elds (Schaible et al. 2005). Soybean Geese are fl ightless in early summer when damage continues well into the growing season adults are molting and young geese have not (Schaible et al. 2005) and can result in fi nancial yet att ained the ability to fl y. Crop damage losses to farmers. Here, we evaluate the

1 Present address: Department of Biology, University of Regina, 3737 Wascana Parkway, Regina, SK S4S 0A2, Canada 316 Human–Wildlife Interactions 5(2) eff ectiveness of the SDGFP program in reducing used in conjunction with feeding stations; the area of soybean fi elds that received damage the latt er were intended to lure geese away by Canada geese. from crops and provide an alternative source of food. Electric fences consisted of 1 piece Study area of 1.27-cm-wide white polytape installed 10 The study was conducted in McCook, cm from the ground on polyethylene posts Minnehaha, Miner, Lake, Moody, Brookings between the water and the fi eld. Fences were and Kingsbury counties in eastern South energized by solar units or deep cycle batt eries. Dakota, USA (Figure 1). These counties are Fences were removed in late August aft er geese located within the Prairie Pothole Region, had att ained the ability to fl y. Feeding sites which is characterized by numerous small consisted of an area where shelled corn was depressions left by retreating glaciers and is dumped on a mowed site adjacent to the same mostly devoid of drainage networks. Current wetland where geese were causing damage. land use is dominated by agriculture. The soil Corn was replenished as needed. Single-bang is Chernozem and is productive for row crops, propane canons were used when geese had predominantly soybeans, corn, and wheat. The the ability to fl y. Particularly problematic geese area’s climate is classifi ed as humid continental were sometimes shot; however, this occurred and the average annual precipitation ranges infrequently. from 53 to 61 cm (Hogan 1991). South Dakota Game, Fish and Parks began receiving reports of goose damage when crops Methods began growing, typically in late April, and We compared the area of damage (m2) in reports continued through July. Abatement fi elds where SDGFP abatement practices were practices were always applied within 1 week applied to reference fi elds (no abatement) of the complaint being fi led, and nearly all to determine the eff ectiveness of the SDGFP complaint fi elds were used for the study. Only damage-reduction program. For all abatement fi elds with resident geese were included in fi elds, SDGFP personnel decided on the types the study. The few fi elds that were not used of abatement practices that were applied. In had complaints fi led early in the growing some abatement fi elds, electric fences were season and involved geese that were migrating

0 25 50 100 150 200 Kilometers

Figure 1. Map of South Dakota counties. Bolded counties indicate location of study area, 2006–2007. Goose crop damage • Radtke and Dieter 317 through the area. We considered geese to area of damage variable was square root be residents if they were nesting in the area. transformed (χ0.5) for all analyses to improve No data was collected during spring or fall normality. All data analyses were conducted migration. Reference fi elds were those in which in JMP 7.0.2 (JMP, version 7.02, SAS Institute, landowners fi led complaints but agreed to forgo Cary, N.C., 2007). abatement in return for monetary compensation for damage. Compensation for damage in place Results of abatement was off ered to every landowner Mean area of crop damage in abatement who fi led a complaint. If landowners declined fi elds was smaller than in reference fi elds in this off er, their fi elds became abatement fi elds. 2006 (F1,54 = 38.89, P < 0.001) and 2007 (F1,61 = If they accepted, their fi elds became reference 28.19, P < 0.001). Area of damage in 2006 was fi elds. We assumed that any diff erences be- 0.10 + 0.03 ha (mean + SE) for abatement fi elds, tween landowners accepting compensation 1.02 + 0.24 ha for reference fi elds, and in 2007 and landowners accepting abatement did not area of damage was 0.30 + 0.09 ha for abatement have a substantial eff ect on the results. We also fi elds and 1.55 + 0.46 ha for reference fi elds. We assumed that reference fi elds and abatement saw evidence of goose foraging aft er abatement fi elds are similar. practices were installed in 14 of 47 abatement We measured areas of crop damage by fi elds in 2007. Geese continued to forage in 15 walking the perimeter of the damaged area of 16 reference fi elds aft er complaints had been with a global positioning system unit (Garmin fi led. We measured crop damage at diff erent GPS 12), which recorded the area of damage times during the growing season each year; (m2). Measuring area this way results in errors thus, we could not make a comparison of of <5% for most areas >500 m2 (Webster and damage between years. Date of application of Cardina 1997). Most areas we measured were abatement practices was positively related to >500 m2. We included all areas for which plants area of damage in 2006 (R2 = 0.16, P = 0.004) and were foraged by geese, regardless of degree of 2007 (R2 = 0.34, P < 0.001; Figure 2), indicating damage. Geese caused nearly all crop damage. greater eff ectiveness at reducing crop damage The ground near all damaged areas was litt ered with earlier application of abatement practices. with goose feces, and geese could commonly be seen foraging in the damaged areas. During Discussion 2006, the damage measurements were taken The SDGFP program was eff ective at reducing in early August. However, in many damaged the area of crop damage caused by Canada fi elds, soybeans recovered enough that it was geese, especially if abatement practices were diffi cult to determine the area of initial damage. applied as soon as damage started. Fields with a Therefore, in 2007, we att empted to take substantial amount of damage at the end of the measurements as soon as damage had stopped growing season typically had a large amount and the extent of damage was clear. Damage of damage when abatement practices were was initially measured when the complaint initially applied. Although some fi elds received was fi led. Fields were then checked throughout additional damage aft er practices were applied, the growing season and remeasured if damage the amount of additional damage was typically had continued. When multiple measurements small. Fields that received additional damage of the same fi eld were taken, the largest area oft en did so because of an inoperable fence of damage measured was used for analysis or because geese had gained the ability to fl y because it represented the total area of damage. immediately aft er the molt and were fl ying over We measured 43 abatement fi elds and 13 fences. In both cases, additional damage was reference fi elds in 2006 and 47 abatement fi elds minor because inoperable fences were usually and 16 reference fi elds in 2007. repaired within a week, and goose foraging Area of damage was compared among aft er molt quickly shift ed from soybeans to abatement and reference fi elds using ANOVA. other food sources. Further, damage done aft er Regression was used to predict area of damage the application of abatement practices oft en in abatement fi elds by Julian date of application was on the same areas that were previously of abatement practices in 2006 and 2007. The damaged and did not result in an increase in the area of damage. 318 Human–Wildlife Interactions 5(2)

Similar research in 2003 1.4 in South Dakota found no diff erence in the amount of 1.2 damage between abatement and non-abatement fi elds 1 (Schaible et al. 2005). In the 2003 study, crop damage was 0.8 measured in all crop fi elds, not just in soybean fi elds as 0.6 in the current study. Type of crop should have litt le eff ect on the eff ectiveness of 0.4 abatement practices because 0.2 similar abatement practices are applied in all fi elds. Reasons for the diff erent 0 results between the 2 studies remain unknown. 4

Other studies have evaluated techniques for 3.5 the abatement of goose 3 crop damage. Applying chemicals to crops can deter 2.5 geese but has seen limited use, possibly because of 2 high cost (Conover 1985, Cummings et al. 1995). 1.5 Many techniques are intended to frighten geese 1 (Heinrich and Craven 1990, 0.5 Gilsdorf et al. 2002). While frightening devices can 0 reduce crop damage, geese 135 145 155 165 175 185 195 oft en habituate to them (Heinrich and Craven 1990), making them less eff ective over time (Summers and Figure 2. Relationship between Julian date of application of abatement practices and area of damage by Canada geese in South Dakota, Hillman 1990). Habituation 2006–2007. Square-root regression lines are shown. is particularly problematic with fl ightless geese because limited mobility of the fences. However, some geese still fed on prevents geese from traveling to other food soybeans, even when alternative food from the sources. Most crop damage by geese in South feeding site was available. Because geese show Dakota occurs during the fl ightless period, site fi delity (Anderson et al. 1992), feeding sites and electric fences were eff ective at deterring may actually increase the number of geese that fl ightless geese. Geese do not become habituated return to that area each year, perpetuating the to electric fences because electric fences are a need for management of goose problems in physical barrier that provide a negative stimulus that area. Propane canons were used prior to (McKillop and Sibly 1988). Feeding sites were or aft er the molt, oft en for >1 week, which was sometimes used in conjunction with fences when suffi cient to deter geese. there was not enough natural food in the areas Willingness to accept compensation by where geese had damaged soybeans in past landowners may have been partly determined years. Feeding sites increased the eff ectiveness by the time of year in which the complaint Goose crop damage • Radtke and Dieter 319 was fi led. When complaints were fi led late in Acknowledgments the growing season, there may not have been Funding for this research was provided by enough time for soybeans to recover once SDGFP; South Dakota State University; and the damage stopped. These landowners may Federal Aid to Wildlife Restoration Fund (Project have been more likely to accept compensation W-75-R Amendment #161) administered by the than landowners with damage earlier in the SDGFP. We thank SDGFP personnel S. Bork growing season. We do not believe that this and L. Vanderstroet for asistance throughout had a substantial aff ect on the results. Because the project and N. Dieter for fi eld assistance. no abatement was applied, the date of the Editing by B. Pinno and anonymous reviewers complaint did not aff ect goose foraging. improved the manuscript. T. White provided It is possible that abatement practices resulted valuable GIS assistance. in a shift in the location of goose damage in some areas. When abatement practices were Literature cited applied to a fi eld, geese may have moved to Alisauskas, R. T., and C. D. Ankney. 1992. The diff erent fi elds where no abatement practices cost of egg laying and its relationship to nutri- had been installed (Radtke 2008). Because most ent reserves in waterfowl. Pages 30–61 in B. damage went unreported, abatement practices D. J. Bart, A. D. Afton, M. G. Anderson, C. D. may not have reduced damage substantially, Ankney, D. H. Johnson, J. A. Kadlec, and G. but simply caused a shift in the location of L. Krapu, editors. Ecology and management damage. Neighboring landowners sometimes of breeding waterfowl. University of Minnesota fi led complaints, but it is unknown whether Press, Minneapolis, Minnesota, USA. these new complaints were the result of recent Anderson, M. G., J. M. Rhymer, and F. C. Rohwer. abatement activities in neighboring fi elds or if 1992. Philopatry, dispersal, and the genetic damage was occurring previous to abatement structure of waterfowl populations. Pages 365– in neighboring fi elds. The availability of 395 in B. D. J. Bart, A. D. Afton, M. G. Anderson, additional food near an abatement fi eld likely C. D. Ankney, D. H. Johnson, J. A. Kadlec, and contributed to the eff ectiveness of the SDGFP G. L. Krapu, editors. Ecology and management program. If additional food is readily available, of breeding waterfowl. University of Minnesota abatement practices likely are more eff ective. Press, Minneapolis, Minnesota, USA. However, geese may have been more persistent Bellrose, F. C. 1980. Ducks, geese and swans of at accessing an abatement fi eld if few other food North America. Stackpole. Harrisburg, Penn- sources were available (Summers and Hillman sylvania, USA. 1990). Conover, M. R. 1985. Alleviating nuisance Canada Despite the eff ectiveness of the SDGFP goose problems through methiocarb-induced program, most damaged fi elds went unreport- aversive conditioning. Journal of Wildlife Man- ed. In talking with producers, we discovered agement 49:631–636. several reasons why damaged areas may have Conover, M. R., and G. G. Chasko. 1985. Nuisance gone unreported. Some agricultural producers Canada goose problems in the eastern United were not aware of the goose damage that was States. Wildlife Society Bulletin 13:228–233. occurring. Other producers were unaware of the Conover, M. R., and D. J. Decker. 1991. Wildlife SDGFP program. Still other producers indicated damage to crops: perceptions of agricultural that damage earlier in the year was minimal, and wildlife professionals in 1957 and 1987. and they decided not to fi le a complaint until Wildlife Society Bulletin 19:46–52. damage became more substantial. Cummings, J. L., J. R. Mason, D. L. Otis, and J. F. To keep crop damage to a minimum, Heisterberg. 1991. Evaluation of dimethyl and agricultural producers and managers should methyl anthranilate as a Canada goose repel- remain vigilant of crops that are subject to lent on grass. Wildlife Society Bulletin 19:184– damage. Agricultural producers should act 190. immediately when damage starts and regardless Cummings, J. L., D. L. Otis, and J. E. Davis Jr. of the techniques used, managers must revisit 1992. Dimethyl and methyl anthranilate and sites for basic maintenance and to ensure that methiocarb deter feeding in captive Canada geese have not continued to damage crops. geese and mallards. Journal of Wildlife Man- agement 52:349–355. 320 Human–Wildlife Interactions 5(2)

Cummings, J. L., P. A. Pochop, J. E. Davis Jr, and H. W. Krupa. 1995. Evaluation of ReJeX-it AG- 36 as a Canada goose grazing repellent. Jour- nal of Wildlife Management 59:47–50. Gigliotti, L. M. 2007. Wildlife damage management program: 2006 landowner survey/evaluation report. South Dakota Department of Game, Fish and Parks. HD-5-07. Gilsdorf, J. M., S. E. Hygnstrom, and K. C. Ver- Cauteren. 2002. Use of frightening devices in TROY M. RADTKE is a Ph.D. degree can- wildlife damage management. Integrated Pest didate at the University of Regina in Regina, SK, Management Reviews 7:29–45. Canada.His current research focuses on the effects of small consumers on grassland community struc- Heinrich, J., and S. Craven. 1987. Distribution ture. He earned his B.S. degree in wildlife from the and impact of Canada goose crop damage University of Wisconsin–Stevens Point and his M.S. degree from South Dakota State University. in east-central Wisconsin. Proceedings of the Eastern Wildlife Damage Control Conference. 3:18–19. Heinrich, J. W., and S. R. Craven. 1990. Evalu- ation of three damage abatement techniques for Canada geese. Wildlife Society Bulletin 18:405–410. Hogan, E. P. 1991. The geography of South Da- kota. Pine Hill Press, Freeman, South Dakota, USA. McKillop, I. G., and R. M. Sibly. 1988. Animal be- haviour at electric fences and the implications for management. Mammal Review 18:91–103. Nelson, H. K., and R. B. Oetting. 1998. Giant Can- CHARLES D. DIETER is a professor in the ada goose fl ocks in the United States. Pages Department of Natural Resources at South Dakota State University (SDSU). He received an M.S. 483–495 in Proceedings of the international degree in wildlife and fi sheries science and a Ph.D Canada goose symposium, Milwaukee, Wis- degree in biological sciences from SDSU. His consin, USA. research includes work on waterfowl, upland birds, waterbirds, songbirds, beavers, otters, jackrabbits, Radtke, T. M. 2008. Crop damage by resident fl ying squirrels, and herps. He is a certifi ed wildlife Canada geese in eastern South Dakota. The- biologist and teaches mammalogy, vertebrate zool- ogy, animal behavior, animal diversity, and environ- sis, South Dakota State University, Brookings, mental toxicology at SDSU. South Dakota, USA. Raveling, D. G. 1979. The annual cycle of body composition of Canada Geese with special reference to control of reproduction. Auk 96:234– 252. Schaible, D., C. D. Dieter, R. Losco, and P. Mam- menga. 2005. Quantifying crop damage by giant Canada geese in Day County, South Da- kota, 2003. Proceedings of the South Dakota Academy of Science 84:259–264. Summers, R. W., and G. Hillman. 1990. Scaring brent geese Branta bernicla from fi elds of winter wheat with tape. Crop Protection 9:459–462. Webster, T. M., and J. Cardina. 1997. Accuracy of a global positioning system (GPS) for weed mapping. Weed Technology 11:782–786.