Beetles in urban gardens: Local and landscape correlates of abundance

Background/Question/Methods

Urban ecosystems are mosaics of residential, industrial, commercial, and agricultural land that present unique challenges for species survival. The creation of impervious surfaces, frequent human disturbance, pollution, introduced species, and habitat fragmentation all contribute to biodiversity loss within urban areas. However, some habitats in urban areas, such as community gardens, may support biodiversity and promote ecosystem services. For gardens, local characteristics, as well as landscape surroundings, may influence species abundance, richness, and ecosystem processes. We examined the local and landscape factors that best predict beetle abundance within urban gardens and specifically tested the hypothesis that changes in local and landscape characteristics of urban gardens correlate with changes in beetle abundance for all beetles, and for the predatory beetle families Carabidae and Staphylinidae. Between May-September 2013, we sampled 19 community gardens that vary in both local characteristics (vegetation, ground cover) and landscape surroundings (amount of agriculture, developed, and natural habitat). Local variables measured included ground cover (concrete, mulch, soil, vegetation), number of crop, ornamental, and weed species, number of flowers, and number of individuals and species of woody plants. We tested correlates of beetle abundance with regression trees to determine if local or landscape factors were more important predictors.

Results/Conclusions

Overall, we collected a total of 1997 beetles. Approximately 41% of all beetles collected were predatory beetles including Carabidae (17.5% of individuals) and Staphylinidae (23.5%). Beetle abundance varied during different months with on average 28.5 beetles collected per site in May, 16.0 in June, 18.9 in July, 29.3 in August, and 12.9 in September. Total beetle abundance increased in sites with >13.5% mulch cover within 100m² (P=0.001), and was higher in sites with >6 crop species per 1m² (P=0.003). No landscape factors correlated with total beetle abundance. Mulch could provide beetles with camouflage from predators, or buffer the soil from extreme temperature changes. Generally, increases in crop richness increase natural enemy diversity and abundance in rural agroecosystems. Yet, according to preliminary results, no landscape or local factors (including crop richness) correlated with abundance of Carabidae or Staphylinidae beetles in this urban agriculture study. In sum, it appears that local factors more strongly influence beetle abundance than landscape factors, with important results for management of and promotion of ecosystem services and biodiversity conservation within urban community gardens.