Effects of grazing and invasive grasses on desert vertebrates in California.
Much of California's San Joaquin Valley is a desert and, like portions of other North American deserts, is experiencing an ecological shift from being dominated by ephemeral native forbs, with widely spaced shrubs, to fire-prone non-native annual grasses. Small terrestrial vertebrates, many of which are adapted to open desert habitats, are declining. One hypothesis is that the invasive plants contribute to the decline by altering vegetative structure. Although cattle may have originally been a factor in the establishment of these non-native plants, their grazing may benefit the terrestrial vertebrates by maintaining an open structure, especially during average or wet winters when the exotic grasses grow especially dense. We experimentally tested the effect of cattle grazing on invasive plants and a community of small vertebrates at a site in the southwestern San Joaquin Desert. We established and monitored 4 treatment (grazed) and 4 control (ungrazed) plots from 1997 to 2006, and assessed the abundances of blunt-nosed leopard lizards (Gambelia sila), giant kangaroo rats (Dipodomys ingens), short-nosed kangaroo rats (Dipodomys nitratoides nitratoides), and San Joaquin antelope squirrels (Ammospermophilus nelsoni), all of which are listed as threatened or endangered by state or federal agencies. We also recorded abundances of the non-protected western whiptail lizards (Aspidoscelis tigris), side-blotched lizards (Uta stansburiana), San Joaquin pocket mice (Perognathus inornatus inornatus), and Heermann's kangaroo rats (Dipdomys heermanni). Based on repeated measures analysis of variance (ANOVA) and a 0.05 alpha level, only Heermann's kangaroo rats showed a treatment effect; they were more abundant on the control plots. However, this effect could be accounted for by the natural re-establishment of saltbush (Atriplex spp.) on part of the study site. Saltbush return also favored western whiptail lizards and San Joaquin antelope squirrels. A regression analysis indicated that populations of blunt-nosed leopard lizard and giant kangaroo rat increased significantly faster in grazed plots than the ungrazed controls, and abundances of 6 of 8 species were negatively correlated with increased residual dry matter. With relaxed alpha values to decrease Type II error, populations of blunt-nosed leopard lizards (500% greater), San Joaquin antelope squirrels (85% greater), and short-nosed kangaroo rats (73% greater) increased significantly on grazed plots over the course of the study compared to ungrazed plots. We did not find grazing to negatively affect abundance of any species we studied. When herbaceous cover is low during years of below average rainfall in deserts and other arid areas, grazing may not be necessary to maintain populations of small vertebrates. However, if cattle grazing is closely monitored in space and time to minimize adverse effects on the habitat, it could be an effective tool to control dense stands of non-native grasses and benefit native wildlife.