Background/Question/Methods In the face of urban expansion, areas like urban parks have immense potential as providers of resources for birds. Parks have been likened to islands in cities, as they differ in size and proximity to natural areas, and provide refuge to wildlife in many cities throughout the world. The Greater Los Angeles metropolitan region (LA) is known for its diverse assemblage of birds. However, LA also suffers from one of the lowest park scores out of any major city; numerous locations are without parks or green infrastructure. As such, parks may be of particular value for birds in low-income neighborhoods – historically located in the most heavily urbanized and sparsely vegetated parts of cities. Here, we studied the role of LA’s urban parks in providing a refuge for birds. We surveyed birds and habitat features in 55 parks situated in low- (n=18), medium- (n=19), and high-income (n=18) areas. We used area search methods to survey birds within parks and derived habitat features using an image classification analysis with satellite imagery. We then used a series of generalized linear modeling analyses to understand (a) island biogeographic patterns of LA’s urban parks and the influence of (b) park habitat and (c) adjacent landscape habitat on park bird communities. Results/Conclusions We found that parks in low-income neighborhoods were located further away from natural areas than parks in medium- (p = 0.013) and high-income (p = 0.002) areas. We also found landscape composition varied most between parks in low- and high-income areas – low-income parks were surrounded by higher proportions of impervious surfaces (p = 0.014) while high-income parks were surrounded by greater proportions of tree cover (p < 0.001). These landscape differences also affected bird populations in parks. The relative importance of island biogeography (park size and distance to natural area), park and landscape composition differed for each of the bird guilds analyzed (synanthropic, open woodland, shrubland, forest breeding birds). Park size was the best predictor of forest bird abundance in parks (p = 0.009). The best predictor for shrubland and synanthropic birds was at the park-scale. Shrubland bird densities within parks decreased with greater impervious surface cover (p = 0.003), while synanthropic bird densities increased in parks with more bare ground (p < 0.001). These results highlight the effects of socioeconomic inequities in urban development on park biodiversity and can be used to create targeted park management guidelines for LA and beyond.
