25th Annual Meeting and Symposium of the
Desert Tortoise Council, April 21-24, 2000
Abstracts

The Effect of Cattle Grazing on Desert Tortoise (Gopherus agassizii) Abundance and Habitat in the Northeastern Mojave Desert

John L. Oldemeyer, Philip A. Medica, and P. Stephen Corn
U.S. Geological Survey , Midcontinent Ecological Science Center, 4512 McMurry Drive
Fort Collins, CO 80525

When the desert tortoise (Gopherus agassizii) was listed as Threatened, livestock grazing was one of the factors identified as threatening long term survival. We did an analysis of some of the factors and basically determined that effects of livestock grazing, particularly that by cattle had not been scientifically evaluated. Thus, in 1992, we initiated research to evaluate the effects of cattle grazing on tortoises and their habitat in southern Nevada and southeastern California. In many evaluations of cattle grazing, exclosures are used to evaluate grazed versus not grazed; however, in the Mojave, there aren't many sizable exclosures that have been in place long enough to really approximate a non-grazed state. Because cattle are dependent on water and graze near water, we sample at 11 water sources in southern Nevada and southeastern California. This approach using a water device as a center of a study has been used in Australia and published as a piosphere. For our design, we used five lines radiating out 4800-6400 m from a water source. We avoided major highways, other water, and other factors that may have influenced cattle or tortoise distribution. We sampled one hectare plots at ten distances from water starting at 200 meters and ending at 6400 meters in 1993 and 4800 meters in 1994 and 1995. We have made the assumption that cattle impacts are less at those distances from water than they are nearer water. We hypothesized that abundance data would fit a logistic curve with distance from water as the horizontal axis. That is, for perennial grass density as an example, density would increase further from water if livestock grazing was indeed affecting perennial grass abundance. We recognize that during cooler seasons, cattle wander further from water but we believed that 50-100 years of grazing would result in the patterns hypothesized.

Within each one-ha plot, we sampled desert tortoise and active burrow density by the removal method. Woody plant abundance was sampled with line intercept and quadrat methods. None of the methodology we used adequately described abundance of perennial grasses, because they occurred so rarely. Thus, using rare plant sampling theory, we tested several quadrat sizes for estimating density of perennial grasses and selected 2x100 m quadrats as providing the best estimate of grass abundance. We identified 22 plots at each of the 11 locations that occurred in a single soil type to sample soils and perennial grasses.

We clustered the soils data into 7 soil types based on minerals, texture, and elevation. Analyses of tortoise, active burrow, Larrea tridentata, Ambrosia dumosa, and perennial grass abundance were performed in the most common soil type to reduce any effects due to soil differences. Sixty-five 1-ha plots occurred in this soil type in Piute, Fenner, and Ivanpah valleys. We tested the logistic hypothesis by trying to fit each of the above variables to a curve with the distance from water as the horizontal axis. We first tested logistic, then polynomial regressions with no success. Finally we ran straight line regressions with the following results. For each variable tested, R² <0.08, the intercept value was significantly (P<0.05) different from 0, and the slope was not significantly (P>0.10) different from 0. This means that the overall mean of the variable was a good estimate of that variable at any distance from water; we could not detect any impact due to grazing.