OOS 9-5 - Using phenology in predicting the northward expansion of the Africanized honey bee

Tuesday, August 9, 2011: 9:20 AM
17B, Austin Convention Center
Catherine Jarnevich1, Wayne Eaias2, Peter Ma2, Jeffrey T. Morisette3, Jaime Nickeson4, Thomas J. Stohlgren5 and Tracy R. Holcombe1, (1)Fort Collins Science Center, U.S. Geological Survey, Fort Collins, CO, (2)Sigma Space, NASA Goddard Space Flight Center, Greenbelt, MD, (3)North Central Climate Science Center, U.S. Geological Survey, Fort Collins, CO, (4)Sigma Space Corp at Goddard Space Flight Center, Greenbelt, MD, (5)Natural Resource Ecology Laboratory, Fort Collins
Background/Question/Methods

The Africanized honey bee (Apis mellifera scutellata) can have significant effects on existing pollinator populations, agricultural production, and human health. Furthermore, agriculturally important migratory beekeeping of European honey bees would be compromised by the addition of the Africanized bees. Knowing their limits of expansion can assist management of European honey bees used for pollination, ease public safety concerns, provide focus areas for education of the general public and first responders, and help evaluate potential shifts in the Africanized bees’ habitat due to climate change.  Previous maps of the northern extent of the Africanized bees are relatively coarse and show inconsistencies with observed data.  We used ensemble modeling,  combining four commonly used binary techniques, in conjunction with new remote sensing derived phenology layers, to predict the northern extent of their expansion.  We created a model for the US and two regional models for the Southeast and Southwest.

Results/Conclusions

Our results show a more nuanced and spatially detailed forecast of the spread of the Africanized bees.  Models for the Southwest performed best according to our evaluation metrics, while the Southeast results were least favorable.  This result may stem from the different variables driving distributions in the two regions, where we hypothesize phenology may be more important in the Southeast.  Phenology metric variables were selected as predictors in almost all models.  Thus, the models indicate that a combination of both climate and vegetation factors are influencing the range of the bees.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.