Southern California's sage scrub (SS) ecosystem is severely threatened by suburban development and invasion by non-native grasses, but how these threats impact the arthropod community is poorly understood. Native ants, which face the additional threat of being displaced by non-native Argentine ants, may be particularly at risk of local and regional extirpation. In this study, we surveyed the ant communities in the SS and non-native grassland habitats at the Robert J Bernard Biological Field Station (BFS) and surrounding suburban habitat, and compared patterns of species richness and composition among habitat types. We also compared ant richness and composition at the BFS to 40 coastal SS fragments previously surveyed in San Diego County to better understand how ant communities in interior and coastal SS fragments differ. Ant composition significantly differed among all three habitat types at and surrounding the BFS, but species richness did not. Comparisons between the BFS and coastal fragments indicate that interior SS fragments harbor unique ant species and more species relative to fragment area. Increased richness and unique ant assemblages are probably associated with the limited ability of invasive Argentine ants to colonize the non-native grassland and SS at the BFS. Because many southern California invertebrates are narrowly endemic to low elevation areas, patterns of habitat specificity seen with ants highlight that maintaining a mosaic of SS and non-native grassland habitat, particularly in interior areas where activity and diversity of nonnative invertebrate species may be restricted, may be critical to preserving biodiversity.

California sage scrub (CSS), a native ecosystem type of low‐elevation areas of Southern California, is increasingly threatened by urban development, altered fire regimes, and vegetation‐type conversion to non‐native grasslands. Using pitfall traps, we examined how suburbanization, type conversion, and fire influence ground‐dwelling spider assemblages in eastern Los Angeles County, CA, by surveying spiders in three habitats (CSS, non‐native grasslands, and suburban areas) before and after a fire that occurred in a small portion of our study site. Spider assemblages in the suburban habitat differed from those in CSS and non‐native grassland habitats, but CSS and grassland assemblages did not significantly differ. This suggests that the urban development, but not vegetation‐type conversion to non‐native grasslands, has significant effects on ground‐dwelling spider assemblages. Fire had no observable effect on assemblages. Because ground‐dwelling spiders were not impacted by fire and type conversion, increased fire frequencies, which often result in the establishment of non‐native grasses, may not deleteriously influence this animal group, a differing pattern from other taxonomic groups. However, the rapid urban development occurring in low‐elevation areas of Southern California means that species requiring non‐suburban sites for their survival (15 species, 24.1%) may be threatened and require conservation assessment.