Climate-driven species redistribution in changing forests: Butterfly examples from the African and Asian tropics

Timothy C. Bonebrake and Michel Dongmo, School of Biological Sciences, The University of Hong Kong, Hong Kong, Hong Kong, Tsun Fung Au, Department of Geography, Indiana University, Bloomington, IN

Background/Question/Methods
Past and future rates of climate-driven range shifts for species globally have a high dependence on habitat availability and habitat change. However, there is unlikely to be a simple relationship between species redistribution rates and habitat. My lab has investigated this issue in diverse tropical ecosystems across Asia and Africa utilizing a combination of long-term data collection, experimental systems and a range of modelling tools. We investigated how climatic warming and reforestation in Hong Kong is resulting in a reshuffling of the butterfly community, focusing on a newly established Nymphalidae species (Euripus nyctelius). We conducted physiological tests in the lab (i.e. measured critical thermal minimum) and ran simulations with climatic time series. To explore how habitat associations structure evolutionary and plastic responses to climate, my group has also undertaken lab rearing studies of Bicyclus butterfly species in Cameroon with populations from rainforest and ecotone (a forest savannah mosaic landscape). Using lab-reared populations at variable conditions (20°C and 30°C for F2 populations) critical thermal maximum (Ctmax) and minimum (Ctmin) was measured for individuals of the same species originating from rainforest and ecotone habitats.
Results/Conclusions
Since its establishment in Hong Kong in 2007, E. nyctelius has experienced a largely positive thermal safety margin as a consequence of climatic warming since 1990. However, since 1990 Hong Kong has also experienced considerable reforestation which may have played a role in the successful establishment. While we cannot readily distinguish whether forest recovery or warming has caused the establishment, it is likely that both have had impacts. In Cameroon, we studied three species of variable habitat affinity (Bicyclus vulgaris, B. dorothea and B. sandace). Plasticity in Ctmax was always higher for ecotone populations of all three species while plasticity in Ctmin was higher in the ecotone for two species (B. dorothea and B. sandace, but not B. vulgaris). These results were consistent for both adults and larvae. Altogether these results highlight 1) landscape-level climatic and habitat changes can complicate the detection and attribution of species redistribution, and 2) habitat associations can influence evolutionary and plastic climate responses. Tropical forest protection is and should continue to be a core conservation strategy for combatting climate and habitat loss impacts on biodiversity.