Land-Use Patterns Surrounding Greater Prairie-Chicken Leks in Northwestern Minnesota Author(S): Michael D
Total Page:16
File Type:pdf, Size:1020Kb
Land-Use Patterns Surrounding Greater Prairie-Chicken Leks in Northwestern Minnesota Author(s): Michael D. Merrill, Kim A. Chapman, Karen A. Poiani, Brian Winter Source: The Journal of Wildlife Management, Vol. 63, No. 1 (Jan., 1999), pp. 189-198 Published by: Allen Press Stable URL: http://www.jstor.org/stable/3802500 . Accessed: 10/08/2011 12:55 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Allen Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Wildlife Management. http://www.jstor.org LAND-USEPATTERNS SURROUNDING GREATER PRAIRIE- CHICKENLEKS IN NORTHWESTERNMINNESOTA MICHAELD. MERRILL,The NatureConservancy, 201 DevonshireStreet, FifthFloor, Boston, MA02110, USA KIMA. CHAPMAN,',2The NatureConservancy, 1313 FifthStreet SE, Suite 320, Minneapolis,MN 55414, USA KARENA. POIANI,The NatureConservancy, Department of NaturalResources, FernowHall, Cornell University, Ithaca, NY 14853, USA BRIANWINTER, The NatureConservancy, Northern Tallgrass Prairie Office, Box 240, RuralRoute 2, Glyndon,MN 56547, USA Abstract: To better managewildlife populations, managers must knowwhich combinationof land uses creates optimal habitat.We used spatial analysisat a landscapescale to describe land-use patternsand patch charac- teristics surroundingleks of greater prairie-chicken(Tympanuchus cupido pinnatus L.) in the Agassiz Beach Ridges (ABR) landscape(2,467 km2)in northwestMinnesota. We hypothesizedthat types and patternsof land use favorableto greater prairie-chickenswould be associated positivelywith lek versus non-lek points, and particularlymore stable (traditional)leks. Using a GeographicInformation System (GIS), we analyzedland- use proportionsand patch characteristicswithin an 810-ha area (1.6-km radius) surroundingtraditional leks, temporaryleks, and randomlylocated non-lek points. We found locationsof greaterprairie-chicken leks were stronglydependent on land use as revealed by a multivariateanalysis of variance(MANOVA; P < 0.001). A discriminantfunction analysisand univariateanalysis of variance(ANOVA) showed that several land-usechar- acteristicswere associatedmost stronglywith leks: smalleramounts of residential-farmstead,smaller amounts and smallerpatches of forest, and greateramounts of ConservationReserve Program (CRP) lands. Comparisons between traditionaland temporaryleks revealed that traditionalleks were surroundedby a lesser proportion of forest and croplandthan were temporaryleks (P < 0.001). UnivariateANOVAs showed that traditionalleks also were associatedwith largerpatches of grassland(P < 0.001), and grassland(P = 0.016) and forest patches (P = 0.017) having more irregularshapes. Our study suggests efforts to manage and conserve greaterprairie- chickenpopulations in the TallgrassPrairie Region should focus on landscape-scaleland-use patterns in addition to local habitat characteristics.Landscape-scale efforts could include enlarginggrasslands around traditional leks by completingprairie restorationsand CRP plantings,while local-scalestrategies should seek to improve the qualityof habitatin existingand new grasslandareas. JOURNALOF WILDLIFEMANAGEMENT 63(1):189-198 Key words: GeographicInformation Systems, GIS, greater prairie-chicken,land-use pattern, leks, patches, Tympanuchus cupido pinnatus. Several researchers have established that per- lands in central North America, the greater manent grassland is vital to the greater prairie- prairie-chicken's range, not surprisingly, has chicken (Hamerstrom et al. 1957, Jones 1963, contracted, and population numbers have de- Robel et al. 1970, Kirsch et al. 1973, Newell et clined (Westemeier 1980). In Minnesota, the al. 1987). Grasslands provide cover for nesting, greater prairie-chicken is listed as a species of brood rearing, roosting, and concealment from special concern (Coffin and Pfannmuller 1988). diverse predators, as well as an abundant and Studies of habitat patterns and preferences of food In supply (Kobriger 1965). general, larger greater prairie-chickens have used radiotelem- are found greater prairie-chicken populations etry or field observations and focused on nest- near (Hamerstrom and larger grassland patches ing, brooding, courtship, roosting, and foraging Hamerstrom 1973). Kirsch (1974) the suggested areas. Landscape-scale studies are needed to minimum area for management greater prairie- understand species' distribution and habitat chickens was 520 ha, with the approximately patterns over large areas (Weins 1989). For smallest unit of grassland equal to 65 ha. Given greater prairie-chickens, leks are an obvious fo- the decrease and fragmentation of native grass- cal point at the landscape scale because they are surveyed relatively easily over large areas. Half I Present address: of ConservationBi- Department of all leks in 1 study (Schroeder and White ology, NaturalResources Science Building,University of Minnesota,St. Paul, MN 55108, USA. 1993) and all leks in another (Svedarsky 1988) 2 E-mail:[email protected] were located within 2 km of the nests of females 189 190 GREATERPRAIRIE-CHICKEN * Merrill et al. J. Wildl. Manage. 63(1):1999 visiting those leks, which suggested that the sur- bors the largest populations of the greater prai- rounding area meets certain habitat require- rie-chicken in Minnesota (Svedarsky et al. ments such as nesting and brood-rearing cover 1997). for females. Not does habitat only high-quality METHODS influence where leks are located, males appear to choose lek locations based on proximity to Land-Use Data females (Schroeder and White 1993). We compiled and analyzed data for the ABR We hypothesized that land-use patterns and landscape in a vector-based GIS (ARCNIEW, some land-use patch characteristics (e.g. patch ARC/INFO). The ABR boundary was digitized size, shape) surrounding leks would be different by the Minnesota State Natural Heritage Pro- than land-use patterns and patch characteristics gram and followed the system of glacial dune surrounding randomly located non-lek points. ridges. The entire landscape was generalized to We also hypothesized that land-use patterns and 14 land-use types interpreted and digitized patch characteristics surrounding more stable from 1:24,000-scale 1990 aerial photos by the leks (i.e. traditional leks) would differ from pat- International Coalition for Land/Water Stew- terns and patch characteristics surrounding ardship in the Red River Basin (cropland, tran- temporary leks, assuming lek permanency was sitional agriculture, forest, wetland, grass- positively related to overall habitat quality. Per- shrub, grassland, urban, farmstead, rural resi- manent leks may be in areas with "better" nest- dence, other rural residential, open water, grav- ing habitat, whereas Schroeder and Braun el pits, exposed soil, unclassified). The mini- (1992) suggested temporary leks consisted of mum mapping unit was 2 ha. Land enrolled in males unable to obtain territories on permanent CRP was not distinguished from other land-use leks. The permanency of leks should help us types via the 1990 aerial photos; thus, we ob- identify preferred greater prairie-chicken habi- tained a separate data coverage of CRP lands tat as well as marginal habitat and suggest which enrolled between 1985 and 1995 from the Min- differences in land-use patterns and patch char- nesota Department of Agriculture. When we acteristics are most important, with implications overlaid CRP with land-use data, approximately for management at the landscape scale. We 92% of CRP coincided with agricultural uses used greater prairie-chicken leks and current (i.e., cropland and transitional agriculture). To land-use data to examine such relations across correct for this data mismatch problem, crop- the ABR landscape of northwest Minnesota. land was decreased by the amount of overlap- CRP. The 8% of CRP was clas- STUDYAREA ping remaining sified as other land-use types such as native The ABR study area covers 2,467 km2 in grasslands, wetlands, and grass-shrub areas. We northwestern Minnesota and spans 6 counties did not change these land-use types to CRP. An within the Red River Valley (Fig. 1). The elon- additional classification error that we did not gated shape of the landscape (about 200 km quantify or correct was that transitional agricul- long and 3-30 km wide) follows the historic ture (generally consisting of CRP and fallow dunes and beach ridges created 10,000 years land) also contained some burned native prairie. ago by former Glacial Lake Agassiz (Krenz and For our analysis, we combined urban, farm- Leitch 1993). The native grassland ecosystem steads, rural residence, and other rural residen- that once dominated the area has been highly tial types into 1 class (i.e., residential-farm- fragmented by agriculture, roads, and urban de- stead). Open water, gravel pits, exposed soil, velopment. Currently, the landscape is >73% and unclassified land-use classes were not ana- cropland. However, since 1985, about 400 km2 lyzed because they accounted for only 0.6% of of cropland have been enrolled in the CRP, the landscape. Thus, we arrived at 8 land-use which has been