Animal pollinators are essential for biodiversity maintenance and crop production, but we have a limited understanding of how natural and anthropogenic disturbances influence pollinator populations in forested landscapes. Native bees are the dominant animal pollinators in many ecosystems, and they often benefit from those disturbance events that increase floral resources and enhance nesting habitat. In this study, we evaluated how wildfire severity and post-fire salvage logging influence bee communities in a mixed-conifer forest ecosystem. We hypothesized that bee abundance and richness would be positively related to wildfire severity because floral resources and nesting habitat availability often increase with fire severity, but that salvage logging would decrease bee abundance and richness due to loss of habitat for wood-nesting bees. To test these hypotheses, we delineated 42 stands within the Douglas Complex, a mixed-severity wildfire that burned in southwestern Oregon in 2013. We sampled bee communities using blue vane traps during four sampling periods from May-September 2016. In addition, we measured floral abundance, floral richness, bare ground, and canopy cover on stands during each collection event and measured coarse woody debris once per stand. We then used generalized linear mixed models to evaluate the link between bee diversity measures and disturbance types.
Results/Conclusions
In 2016, we collected 2,320 bees representing 22 genera and 98 species. Halictus, Lasioglossum, Bombus, Apis, and Xylocopa were the most common genera, representing 89% of the fauna collected. As predicted, stands characterized by low to moderate-low levels of fire severity had significantly fewer bees and species compared to those of moderate-high and high severity. In contrast, bee abundance and richness were similar between logged and unlogged plots. Habitat variables, including flowering plant abundance and canopy openness, were positively related to both fire severity and salvage logging. In addition, the relative abundance of bee functional groups shifted with wildfire severity: moderate high and high severity sites (both logged and unlogged) had a greater proportion of small, solitary, and cavity-nesting bees compared to less disturbed sites. Based on results to date, we conclude that there is a positive relationship between wildfire severity and bee diversity in mixed conifer forest due to increased floral resource availability (by removing canopy shading), and that this relationship is maintained in stands subjected to salvage logging. Additional work in the 2017 field season aims to better understand these patterns and their influence on pollination services within plots.