Conservation biology has traditionally focused on natural and rural areas. Recently, there has been an upswing of interest in urban landscaping with pollinator-friendly plants. It is clear that many pollinator species use these gardens. However, it is not clear whether urban gardens could support viable populations in the absence of immigration from natural areas. In this talk, we test whether current or realistic future levels of pollinator gardening are likely to be sufficient to maintain populations of two butterfly species, the Baltimore checkerspot (Euphydryas phaeton) and monarch (Danaus plexippus) in Somerville, Massachusetts USA. Populations of both species previously occurred in the metropolitan area. Monarchs are still common in the city. We are not aware of extant Baltimore checkerspot populations in the city, although populations are still found in surrounding natural areas. We approach this question in a two-step process. First, we use empirically paramaterized, spatially implicit integrodifference equations to calculate the minimum proportion of the landscape that would need to be high-quality habitat to support populations of each species. Then, we used images of homeowners' front yards from Google Street View to calculate the cover of pollinator-friendly landscaping.
Results/Conclusions
Baltimore checkerspots have higher minimum area requirements (0.4-1.9% cover of high-quality habitat, depending on the surrounding matrix) than monarch butterflies (0.07-0.9% high-quality habitat, depending on the surrounding matrix). In Somerville, 2% of yards have at least 25% cover of pollinator-friendly species. If front yards represent are representative of the total landscape, then the cover of pollinator gardens is 0.5%, which is near the low end of habitat needed to maintain Baltimore checkerspots, but noticeably higher than the minimum area requirements for monarch butterflies. Furthermore, the monarch's larval hostplant, milkweed (Asclepias spp.) is common in gardens and as a weed, but the checkerspot's primary larval hostplant, white turtlehead (Chelone glabra), is much less common in gardens (although it is commercially available and not infrequently used). We explore the levels of planting of each species that would need to be used to maintain each butterfly population. Pollinator gardens were also spatially clustered in Somerville. This pattern is consistent with the hypothesis that gardening behavior is influenced by local neighbor's landscaping choices. Overall, our data support the notion that pollinator gardens could support butterfly populations, if a critical mass of people adopt this landscaping behavior.