Butterfly declines in protected areas of Illinois: Assessing the influence of two decades of climate and landscape change

Abstract

Despite increasing concern regarding broad-scale declines in insects, there are few published long-term, systematic butterfly surveys in North America, and fewer still that have incorporated the influence of changing climate and landscape variables. In this study, we analyzed 20 years of citizen science data at seven consistently monitored protected areas in Illinois, U.S.A. We used mixed models and PERMANOVA to evaluate trends in butterfly abundance, richness, and composition while also evaluating the effects of temperature and land use. Overall butterfly richness, but not abundance, increased in warmer years. Surprisingly, richness also was positively related to percent impervious surface (at the 2 km radius scale), highlighting the conservation value of protected areas in urban landscapes (or alternately, the potential negative aspects of agriculture). Precipitation had a significant and variable influence through time on overall butterfly abundance and abundance of resident species, larval host plant specialists, and univoltine species. Importantly, models incorporating the influence of changing temperature, precipitation, and impervious surface indicated a significant overall decline in both butterfly abundance and species richness, with an estimated abundance decrease of 3.8%/year and richness decrease of 1.6%/year (52.5% and 27.1% cumulatively from 1999 to 2018). Abundance and richness declines were also noted across all investigated functional groups except non-resident (migratory) species. Butterfly community composition changed through time, but we did not find evidence of systematic biotic homogenization, perhaps because declines were occurring in nearly all functional groups. Finally, at the site-level, declines in either richness or abundance occurred at five of seven locations, with only the two largest locations (>300 Ha) not exhibiting declines. Our results mirror those of other long-term butterfly studies predominantly in Europe and North America that have found associations of butterflies with climate variables and general declines in butterfly richness and abundance.