Testing space‐for‐time transferability of climate effects on occupancy and abundance

Abstract Species distribution models (SDMs) are frequently used to project species' ranges under future climate conditions. Such space‐for‐time substitutions rely on the assumptions that spatial climate‐distribution relationships are causal and that relationships are equivalent over space and time. These assumptions have rarely been tested. Using UK populations of the Orange‐tip butterfly ( Anthocharis cardamines ) as a case study, we demonstrate an approach to test whether these key assumptions are met. First, we identified the seasonal periods over which temperature and precipitation variables best explained the Orange‐tip butterfly's abundance. Then, we compared the effects of climate variables on both occupancy and abundance over space versus time. We found that, over both space and time, temperature in the previous year's flight period has a positive effect on the occupancy (i.e. presence) and abundance of the Orange‐tip butterfly, whereas precipitation appeared not to have a causal effect. We found that temporal effects of temperature on abundance did not differ from spatial effects at colder sites, but at warmer UK sites, the magnitude of these effects significantly differed. Conversely, spatial and temporal effects of temperature on occupancy did not differ at warmer (mid‐range) sites, with effect magnitudes significantly differing at colder sites. Our results demonstrate the importance of identifying causal climate‐distribution relationships before making projections. We also show that the reliability of SDM projections over time can be highly context dependent, even when considering a single species. Therefore, where data are available over space and time, the space versus time modelling approach presented here should be incorporated into the SDM statistical repertoire to improve the reliability of projections. Read the free Plain Language Summary for this article on the Journal blog.