According to agroecological principles, pest management involves making agroecosystems less favorable for pests and more attractive for natural enemies via crop diversification. Crop diversification is the norm in urban community gardens, which makes these agroecosystems especially suitable for examining the effects of vegetation complexity on herbivorous insects, their natural enemies, and biological pest control. In this study, I ask: 1) does vegetation complexity (vegetation density, diversity, and connectivity) differentially influence the abundance and richness of herbivores and natural enemies, and 2) how does vegetation complexity affect predation levels provided by natural enemies in garden plots? I sampled 4 garden plots at 8 community gardens in the California central coast over two time points. I surveyed the vegetation (e.g. percent cover of herbaceous plants, number of plant morphospecies, distance between plants) and sampled the insect community using manual sampling, pitfall traps, and sticky traps. I measured predation levels using sentinel plants inoculated with moth eggs, half of which were covered with mesh to exclude natural enemies and half of which were uncovered to allow prey access. I used generalized linear models (GLMs) to test for effects of vegetation complexity variables on herbivores, natural enemies, and moth egg predation.
Results/Conclusions
I sampled 507 herbivores from 8 families and 472 natural enemies representing 18 families. The most abundant herbivores were in the Aphididae (n=304) and Aleyrodidae (n=113) families. The most common enemy families were Formicidae (n=351) and Syrphidae (n=47). I found that herbivores did not show a significant response to vegetation density, diversity, or connectivity. In contrast, garden plots with higher vegetation density had a higher richness of natural enemy families (z = 1.98, p = 0.047). Considering specific natural enemy taxa, plots with high vegetation density were positively associated with spider abundance (z = 2.92, p = 0.003), spider family richness (z = 2.61, p = 0.009), and ant genera richness (z = 1.93, p = 0.054). Moth egg predation was not significantly associated with vegetation density, diversity, or connectivity. These results indicate that high vegetation density in urban community gardens may be especially important for attracting a diversity of natural enemies, which is thought to support biological pest control (although there was no effect on short-term predation measurements in this study). Community gardens include a high baseline level of vegetation diversity which may contribute to the importance of vegetation density for natural enemy richness in this study.
