Premium
Historically calibrated predictions of butterfly species' range shift using global change as a pseudo‐experiment
Author(s) -
Kharouba Heather M.,
Algar Adam C.,
Kerr Jeremy T.
Publication year - 2009
Publication title -
ecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.144
H-Index - 294
eISSN - 1939-9170
pISSN - 0012-9658
DOI - 10.1890/08-1304.1
Subject(s) - species distribution , ecology , range (aeronautics) , climate change , butterfly , ecological niche , niche , environmental niche modelling , environmental science , extinction (optical mineralogy) , biology , habitat , paleontology , materials science , composite material
Global changes have the potential to cause a mass extinction. Predicting how species will respond to anticipated changes is a necessary prerequisite to effectively conserving them and reducing extinction rates. Species niche models are widely used for such predictions, but their reliability over long time periods is known to vary. However, climate and land use changes in northern countries provide a pseudo‐experiment to test model reliability for predicting future conditions, provided historical data on both species distributions and environmental conditions are available. Using maximum entropy, a prominent modeling technique, we constructed historical models of butterfly species' ranges across Canada and then ran the models forward to present‐day to test how well they predicted the current ranges of species. For the majority of species, projections of how we predicted species would respond to known climate changes corresponded with species' observed responses (mean autoregressive R 2 = 0.70). This correspondence declined for northerly and very widely distributed species. Our results demonstrate that at least some species are tracking shifting climatic conditions across very large geographic areas and that these shifts can be predicted accurately using niche models. We also found, however, that models for some species fail when projected through time despite high spatial model accuracies during model training, highlighting the need to base management decisions on species assemblages, not individual species.