Butterflies are a reliable biological indicator species, with changes in habitat ranges correlating with regional climate trends. In this study we explored if butterfly communities at ten sites in northern California changed over the last thirty years, and whether these changes depended on the life history characteristics (weedy or non-weedy) of the species. We also wanted to find if the exhibited change correlated with average daily temperature trends. Our hypotheses were that the communities were changing and that weedy species were changing less because of their greater tolerance to changes in host plant availability. We used non-metric multidimensional scaling (NMDS) to first characterize each year using relative butterfly species abundances. We analyzed the change in communities over time by first partitioning the years into early, mid-, and late periods. We then analyzed the differences among the categories using a multivariate extension of ANOVA called ANOSIM. In addition we performed NMDS on the years based on average daily temperature at the sites. The resulting resemblance matrix among the years was compared to the resemblance matrix of the butterfly abundance data to determine if there was a relationship between the changes in temperature and the changes in butterflies.
Results/Conclusions
We found that all of the butterfly communities exhibited change over time based on the early, mid- and late time categories. (P-values for all sites were < 0.05), but there were no significant differences in the amount of change between the weedy and non-weedy butterflies, suggesting that the differences over time were not dependent on life histories. We also found a significant association between the changes in the butterfly community and mean temperature changes, but only for non-weedy species. The higher elevation sites showed the largest difference in association of weather variables and butterflies for both weedy and non-weedy species. The significance of these findings is that the butterfly communities’ change seems to be independent of their life history characteristics, but strongly associated with the changes in temperature.