Effects of Prairie Dogs on Sustainability of Cattle Grazing in Mixed-Grass Prairie

The project was carried out in south-central South Dakota on mixed-grass prairie rangelands. The climate of the region is continental and semiarid with hot summers and cold winters. Annual precipitation averages approximately 19 inches near Mission, SD (NOAA, 2000) and over half the yearly precipitation falls during the growing season. The soils in the study area are predominantly silts and loams (USDA-SCS, 1974). Vegetation is typical of the mixed-grass prairie with a variety of warm- and cool-season species. Common grass species in the area include western wheatgrass (Pascopyrum smithii [Rydb.] A. Love), little bluestem (Schizachyrium scoparium (Michx.) Nash), needleandthread (Stipa Comata Trin. & Rupr.), prairie sandreed (Calamovilfa longifolia [Hook.] Scribn.), blue grama (Bouteloua gracilis (H.B.K.) Lag. Ex Griffiths), and buffalograss (Buchloe dactyloides (Nutt.) Engelm.) (USDA-SCS, 1974).

The project involved three mixed-grass prairie pastures grazed by cattle with varying degrees of prairie dog town coverage (percent of pasture area). Data collection took place on the three pastures during the summers of 2002 and 2003. Vegetation samples were collected during two sampling periods in each year: late spring (June) and mid-summer (July) in an effort to adequately sample both the cool- and warm-season species common to the region. Two sites were selected on each study pasture. One site was the prairie dog town (on-town site). The second site (off-town site) was near the on-town site, but had no prairie dog activity. These off-town sites were chosen based on similarity of soils with their paired on-town site. Soils on all on-town and off-town sites were evaluated by a Natural Resources Conservation Service soil scientist.

It was important to this study that the soils were matched within each pair of on- and off-town sites, because sites with the same soils have the same potential for production of plant communities and biomass. Differences in species and/or production can then be more closely tied to differences in use. On-town sites in this study were dominated by annual grasses and forbs as well as by the shorter perennial grasses including blue grama, buffalograss, sedge species (grouped as shortgrasses), and red threeawn. Off-town sites were dominated by perennial forbs and grasses including the shortgrasses, needleandthread, and prairie sandreed.

Three types of plots were used in this study: 1) uncaged plots available for grazing by both prairie dogs and cattle (NONE EXCL), 2) plots fenced/caged so as to exclude cattle but allow use by prairie dogs (CATTLE EXCL), and 3) plots covered by cages which excluded both prairie dogs and cattle (ALL EXCL). All plots were 0.25 m2.

Forty NONE EXCL plots were established on each prairie dog town and 20 on each off-town site. These plots were “permanent”, that is, the same plots were used for all sampling periods. To avoid marking these plots with stakes or flags which would attract herbivores to them, we obtained a GPS location for each, marked the corners of each plot with metal stakes driven into the ground, and then used those coordinates and stakes to relocate each plot at each sample date.

Ten CATTLE EXCL plots were established on each prairie dog town. These were fenced by creating an exclosure with barbed wire and fence posts to ensure availability to prairie dogs (we were concerned prairie dogs might not go into exclusion cages, even though the wire mesh would have permitted their entry). Another 10 CATTLE EXCL plots were established on each off-town site. These were covered with an exclusion cage constructed with a large mesh wire over a metal frame.

Ten ALL EXCL plots were established on each prairie dog town; none were established on the off-town sites since these sites were not utilized by prairie dogs. The ALL EXCL plots were covered with an exclusion cage constructed of chicken wire mesh over a metal frame to prevent access by prairie dogs and cattle.

At the end of each 3-week sampling period (one in June of each year and one in July of each year), cages were removed and all plots were estimated for biomass by species. Estimates for aboveground biomass by species were obtained using two methods. For all plots in 2002 and the NONE EXCL plots in 2003, double sampling methods were used appropriate for each species, including the use of cover, reference units (Andrew et al., 1979, 1981), and plant volume (Johnson et al., 1988) as estimators. In 2003, plots underneath the movable cages were clipped and sorted by species. Biomass estimations were made for individual species with the exception of two groups which were lumped together: blue grama, buffalograss, and sedge species were grouped together as shortgrasses and downy brome and Japanese brome were classified as annual brome.

Differences between ALL EXCL plots and the CATTLE EXCL plots on towns gave an estimate of forage disappearance (destruction + consumption) due to prairie dogs. Differences between ALL EXCL plots and NONE EXCL plots gave an estimate of total disappearance by all herbivores. By subtracting disappearance due to prairie dogs from total disappearance, disappearance due to large herbivores was estimated for on-town sites. For off-town sites, differences between CATTLE EXCL plots and NONE EXCL plots provided an estimate of forage disappearance due to cattle in areas not grazed by prairie dogs.

The experimental design for this study was a complete randomized block design (pastures are blocks) with three factors: year, season, and grazing treatment (prairie dogs on-town, cattle on-town, and cattle off-town). The dependent variable was forage disappearance. Analysis of variance was used to test the effects of grazing treatment, year, and year x grazing treatment on forage disappearance. Year and year x grazing treatment effects were analyzed with repeated measures using PROC MIXED with AR(1) covariance model structure. Treatment mean differences were separated using the LSMEANS statement with the PDIFF option in PROC MIXED.