Get reliable rangeland science

In California's Central Valley 98% of native grasslands have been destroyed or degraded due to invasion, farming, development, and fragmentation. Grassland restoration is generally assumed to provide improved wildlife habitat, increasing wildlife abundance and diversity relative to unrestored, invaded annual grasslands. To investigate that assumption, we compared relative wildlife utilization at paired restored and unrestored (control) grasslands at four locations in Yolo and Sacramento counties using live traps, camera traps, snake boards, and observational surveys in four seasons from 2014-2015. Restored sites were planted with native perennial grasses 10-20 years ago but are now invaded by Mediterranean annual grasses and forbs. Control sites contained similar non-native plant species assemblages but did not have any native grass cover. In general, mouse, vole, and snake utilization was higher at control relative to restored sites across all four seasons. Raptor surveys in three seasons revealed greater species diversity, foraging time, and attack rates at control sites as well, likely in response to greater rodent abundance. Within sites, species-specific responses were related to vegetative cover and percent bare ground. For example, Peromyscus maniculatus (deer mouse) was associated with high bare ground and low vertical cover, regardless of site type (restored/control). Substantial changes in rodent community composition could also be achieved over short periods of time (< 3 months) by manipulating vegetative cover. These results reveal that native grassland restoration may not automatically confer increases in habitat utilization or species diversity, and rapid changes in vegetative structure could strongly impact species composition, suggesting a more nuanced approach could be required for the restoration of desired biodiversity.