Population responses to temperature in 12‐year insect time series

Abstract

Abstract Considerable work has examined the distribution of species under projected climate change, but empirical data are needed to support predicting change in abundance. I analysed populations of 17 saproxylic beetle species collected twice nightly from mid‐May to mid‐October from 2004 to 2015 at lights in Shizuoka, Japan, to investigate effects of two 100‐year hot summers (2010 and 2013) and correlated changes with June, July and August maximum temperatures. The first analysis examined the effect of heat stress and the second examined how populations tracked changes in temperature. I hypothesised that the responses would be species‐specific and could be related to interspecific variation in body size, voltinism, primary food source and the strength of density dependence. Heat stress was associated with population declines of five species during the same year and five other species in the following year. Conifer feeders were more likely to decline during the same year as the heat stress. Only one species tracked variation in maximum June temperature. Larger, later emerging species were more strongly detrimentally affected by high June temperatures than smaller, early emerging species. There was no relation between the response to heat stress and tracking June temperatures across species. The influence of summer temperature on saproxylic beetles was detectable from relatively short time series and occurred in a species‐specific manner. The effect of heat stress was more readily detected than correlated changes in abundance and temperature. Short time series may help reveal the response of insect populations to changing temperatures.