A priori prediction of an extreme crash in 2015 for a population network of the alpine butterfly, Parnassius smintheus

Abstract

Prediction provides important validation for scientific hypotheses and models. We update an existing model and present a priori predictions for the growth of a network of 21 populations of the butterfly Parnassius smintheus based on previous population size and climate during the overwintering period. The model predicts that the extremely warm, dry winter of 2015 in the Rocky Mountains will result in a network-wide crash. All populations are expected to show extreme negative growth. Ten of 21 populations are expected to have less than one individual and the 95% confidence intervals of 17 of 21 populations are predicted to overlap zero in 2015. Given the unprecedented nature of climate change, these predictions represent the best estimate based on our understanding of the effects of climate and densitydependence for the population growth for this species. If the predictions prove to be valid, it provides strong support for the predictive ability of the current model and the negative impact of extreme climatic events for the persistence of populations and ultimately species. Copyright: