PBG datasets are associated with a long-term, large-scale study that is addressing the effects of fire-grazing interactions in the context of a Patch-Burn Grazing management system designed to promote grassland heterogeneity. Effects of patch-burn grazing management on plant and animal diversity and the nature and variety of wildlife habitat are being assessed in two replicate management units, each consisting of three pastures (watersheds) designated C03A/C03B/C03C and C3SA/C3SB/C3SC. In each patch-burn grazing unit, one watershed is burned and two that are left unburned in a given year. The burning treatments are rotated annually so that each pasture is burned every third year. Each patch-burn grazing unit is paired with an annually-burned pasture for comparison with traditional grazing systems (C01A and C1SB). All grazing units are stocked with cow/calf pairs from approximately 1 May until 1 Oct at a stocking density equal to 3.2 ha per cow/calf. To examine the impact of patch burning and grazing in all 8 units, we monitor changes in plant species composition, residual biomass, grassland bird populations, insect populations, small mammal populations, soil nutrients, and stream water quality1 (1C3SA/C3SB/C3SC unit only). The KSU Department of Animal Science monitors cattle performance, including weight gain and body condition to assess the economic feasibility of using patch-burn management on a widespread basis.
This data set focuses on detecting population and community level responses of small mammals to patch-burn grazing. Grids are trapped for 3 consecutive nights approximately monthly. Watersheds include C1A, C3A, C3B, C3C, & KIB (formerly K4A).
Examine the effects of patch-burn grazing on small mammal biodiversity and population dynamics.
Field Methods: We established two trap grids for sampling small mammal community and population dynamics in each of the three PBG patches (C3A-C) and two controls (C1A, K4A), for a total of ten grids. Grid locations were selected at random, but subject to two constraints. To maintain independence among trap grids, grids were separated by at least 200 m, which corresponds to twice the length of the longest published home range axis for deer mice (Peromyscus maniculatus), the most abundant species of small mammal encountered in native prairie. Trap grids were also located at least 100 m from unit boundaries to avoid potential boundary effects, and greater than 50 m from permanent or regularly flowing water to avoid flooding of traps during runoff from thunderstorms (Konza LTER datasets: GIS210 and GIS211).
Each trap grid was a five-by-five square design with 25 stations and 20 m spacing between adjacent trap stations for a total area of 0.64 ha. Two extra-large Sherman live traps were set at each trap station for a total of 50 traps per grid (Model LNG 12, H.B. Sherman Trap Company, Tallahassee, FL, USA). Traps were baited with a mixture of peanut butter and rolled oats, and each trap was provisioned with polyester fiberfill to keep animals warm during October to May. To reduce heat stress to diurnal mammals, wooden A-frame structures (hereafter, trap shelters) were placed over traps for shading. Trap shelters were left in place all year for weathering and to minimize potential neophobic responses of small mammals to trap stations.
During our 3.5-year study from June 2011 to December 2014, small mammals were trapped for three consecutive nights each month at ten trapping grids. We marked small mammals with passive integrated transponders (PIT tags hereafter; Model AB10320, FDX-B 7 x 1.35 mm, Loligo Systems, Tjele, Denmark; or 'Skinny' FDX-B 8 x 1.4 mm, Oregon RFID, Portland, Oregon USA), and read tags with a handheld reader (Model APR 350 FDX/HDX Reader, Agrident, Manassas, VA, USA; or DataTracer FDX/HDX Reader, Oregon RFID, Portland, Oregon, USA). PIT tags were injected subcutaneously under loose skin at the nape, and massaged away from the insertion site to ensure tag retention. To obtain an estimate of PIT tag retention, 28% of the rodents were tagged with numbered monel ear tags (model 1005-1, National Band and Tag Company, Newport, Kentucky, USA). PIT tag losses were rare (less than 1%). All procedures were approved by the Kansas State University Institutional Animal Care and Use Committee (protocols 3034 and 3443), and conducted under state wildlife permits from Kansas Department of Wildlife, Parks, and Tourism.