Simple physics link the distribution and colour of dragonflies across North America and Europe

Submitted by editor on 13 September 2016. Get the paper!
Assemblages of dragonflies are darker coloured in colder regions and lighter coloured in warmer regions! – A general pattern for insects? Photo: Stefan Pinkert.

By Stefan Pinkert

In this paper, we investigated the predictions of the thermal melanism hypotheses (TMH) that in regions with low temperatures, darker coloured ectotherms have an advantage over lighter coloured ectotherms because they absorb more solar irradiance (Gates 1980), which leads to higher heating rates and equilibrium temperatures (Kalmus 1941, Watt 1968). By contrast, species inhabiting hot climates are expected to profit from enhanced reflectance facilitated by light colours that reduce heat load and avoid overheating (Gates 1980). Specifically, we tested the TMH for nearly 10,000 dragonfly assemblages across North America and Europe.

Whereas endotherms ('warm-blooded'), such as mammals and birds, retain relatively constant body temperature across broad environmental gradients, ectotherms ('cold-blooded'), such as insects, need to receive energy from their environment to be active (see a sun-basking male of Aeshna mixta in the above figure). Ectotherms are therefore more sensitive to the climate regime in which they live and their geographical distribution is directly linked to those traits that affect how ectotherms are influenced by and interact with the environment.

So far, a number of studies have documented support for the TMH within and between ectothermic species (summarised in Clusella-Trullas et al. 2007). The association of colour lightness of body surfaces with temperature might, however, depend also on other functions of colouration, such as camouflage, pathogen resistance, UV protection and sexual selection (Roulin 2014) as well as other mechanisms of thermoregulation (e.g. wing whirring, obelisk posture; May 1976). Even if the underlying ecological processes are universal, these other functions might thus offset the importance of thermal melanism for single species or localities. We conducted a broad-scale assemblage-level analysis of the colour lightness of dragonfly assemblages across North America and Europe to test the generality an association of colour lightness and the distribution of dragonfly species.

Generating colour lightness data for over 200 species and 400 individuals is extremely difficult, time consuming and typically involves months, if not years, of fieldwork. But we used a trick to do that! There are field guides available that include brilliant drawings of dragonfly species, so why not use them? For this we alienated digital image analysis – a tool that was originally developed for counting bacteria with different colours on an agar plate (PNG-package) – to estimate the colour lightness (the average of the red, green and blue colour channel). Of course, we still needed days to prepare the drawings for this analysis but once the body was clipped from the wings, antennae and legs we could automatically generate a set of standardized colour lightness data within minutes.

I know what you think: Are drawings a reliable basis? Well, we provide a lot supplementary material, which shows that the colour lightness of males and females as well as drawings from two different artists are strongest correlated (Appendix ECOG-02578 at <>). In the following we aggregated the colour lightness values of species as well as environmental variables for each assemblage… and there it was – a strikingly detailed spatial pattern in colour lightness across the two continents (see for instance how explicit the Great Basin desert in North America and the Alps in Europe accented from their surrounding; Fig 1, below).

In accordance with the THM, we found that dragonfly assemblages in both North America and Europe were composed of more light-coloured species in the south and more dark-coloured species in the north, with temperature variables being by far the strongest predictors of this pattern. Colour lightness carried a strong phylogenetic signal and by decomposing the colour-lightness of species into a phylogenetic (ancestral predicted) and specific part we found a hint that thermal melanism might also be of importance for the evolution of dragonflies. More importantly, however, although idiosyncrasies of the continents seem to modify the overall pattern, our results clearly highlight the generality of colour lightness for the distribution of dragonfly species. As similar pattern are also known from physiological studies and local-scale studies on butterflies and beetles, our broad-scale analysis of the spatial variation in colour-lightness provides a necessary next step in recognizing thermal melanism as a potential ecogeographical pattern in insects (see also Zeuss et al. 2014, Schweiger and Beierkuhnlein 2015 in the September issue of Ecography).

Map of the spatial variation in average colour lightness of North American and European dragonfly assemblages. Colour lightness ranges from 0 (absolute black) to 255 (pure white). Colour scale intervals follow an equal-frequency classification, ranging from blue (darkest) to red (lightest). Because of different data sources for North America and Europe, only the classes but not the values can be directly compared in this figure (an additional step was needed to make them comparable). The datasets comprise 8,127 half-degree grid cells in North America and 1,839 approximately half-degree grid cells in Europe (EPSG: 4326 and EPSG: 3537; rectangular latitude and longitude grid).



Clusella-Trullas, S. et al. 2007. Thermal melanism in ectotherms. - J. Therm. Biol. 32: 235–245.

Gates, D. M. 1980. Biophysical Ecology. - Springer.

Gibert, P. et al. 1998. Light body pigmentation in Indian Drosophila melanogaster: A likely adaptation to a hot and arid climate. - J. Genet. 77: 13–20.

Kalmus, H. 1941. Physiology and ecology of cuticle colour in insects. - Nature 148: 428–431.

May, M. L. 1976. Thermoregulation and adaptation to temperature in dragonflies (Odonata: Anisoptera). - Ecol. Monogr. 46: 1–32.

Roulin, A. 2014. Melanin-based colour polymorphism responding to climate change. - Glob. Change Biol. 20: 3344–3350.

Schweiger, A. H. and Beierkuhnlein, C. 2015. Size dependency in colour patterns of Western Palearctic carabids. - Ecography: 39: 846–857.

Zeuss, D. et al. 2014. Global warming favours light-coloured insects in Europe. - Nat. Comm. 5: 4847.

Watt, W. B. 1968. Adaptive significance of pigment polymorphisms in Colias butterflies. I. Variation of melanin pigment in relation to thermoregulation. - Evolution 22: 437–458.