Invasive Insects Differ from Non-Invasive in Their Thermal Requirements

Abstract

We tested whether two basic thermal requirements for insect development, lower developmental thresholds, i.e. temperatures at which development ceases, and sums of effective temperatures, i.e. numbers of day degrees above the lower developmental thresholds necessary to complete development, differ among insect species that proved to be successful invaders in regions outside their native range and those that did not. Focusing on species traits underlying invasiveness that are related to temperature provides insights into the mechanisms of insect invasions. The screening of thermal requirements thus could improve risk-assessment schemes by incorporating these traits in predictions of potentially invasive insect species. We compared 100 pairs of taxonomically-related species originating from the same continent, one invasive and the other not reported as invasive. Invasive species have higher lower developmental thresholds than those never recorded outside their native ranges. Invasive species also have a lower sum of effective temperatures, though not significantly. However, the differences between invasive and non-invasive species in the two physiological measures were significantly inversely correlated. This result suggests that many species are currently prevented from invading by low temperatures in some parts of the world. Those species that will overcome current climatic constraints in regions outside their native distribution due to climate change could become even more serious future invaders than present-day species, due to their potentially faster development.

Fig 1. Lower developmental thresholds (LDTs) and sums of effective temperatures (SETs) of non-invasive and invasive species.

Average values ± standard deviations of LDTs in °C, (A) and SETs in day degrees [D°] above LDT (B) for pairs of related species of which one is invasive and the other is not. (A) Invasive species have significantly higher LDTs than non-invasive species: t = 4.38, df = 99, P < 0.001 (two-sample t-test not taking into account that paired differences can vary specifically depending on species relatedness); t = 3.841, df = 93, P < 0.001 (linear mixed-effect model on closely related species pairs, analogous to paired t-test); t = 4.35, df = 177.9, P < 0.001 (linear mixed-effect model on nested taxonomic hierarchy). (B) Invasive species have non-significantly lower LDTs than non-invasive species: t = 1.52, df = 87, P = 0.13 (two-sample t-test not taking into account that paired differences can vary specifically depending on species relatedness); t = 1.23, df = 81, P = 0.22 (linear mixed-effect model on closely related species pairs, analogous to paired t-test); t = 1.29, df = 154.9, P = 0.2 (linear mixed-effect model on nested taxonomic hierarchy).

Fig 2. Model of the differences in thermal requirements of invasive (I) and related non-invasive species (N).

General model based on the linear relationship between the developmental rate and temperature. Invasive species have a higher lower developmental threshold (LDT_I) than non-invasive species (LDT_N), i.e. a higher temperature at which the development ceases. However, as shown by the lines describing the increasing development rate of invasive (DR_I) and non-invasive (DR_N) species with increasing temperature, above temperature T_c the invasive species develop faster than non-invasive species. Because the sum of effective temperatures (SET) necessary for a completion of a development is a reciprocal value of the slope of the developmental rate on temperature (S1 File), faster development means lower SET for invasive than non-invasive species.