Winter conditions do not predict subsequent abundance of adults in either habitat, indicating the greater importance of summer mortality in determining abundance, but in different ways in the two habitats. This contrast is an example of the importance of considering specific habitats in the search for effects of changing climate on abundance.