95th ESA Annual Meeting (August 1 -- 6, 2010)

OOS 56-9 - Mapping ash trees (Fraxinus spp) using Landsat-scale phenology for improved management of the emerald ash borer (Agrilus planipennis Fairmaire)

Friday, August 6, 2010: 10:50 AM
317-318, David L Lawrence Convention Center
Bernard N. Isaacson1, Shawn P. Serbin2, Aditya Singh1 and Philip Townsend1, (1)Department of Forest and Wildlife Ecology, University of Wisconsin - Madison, Madison, WI, (2)Brookhaven National Laboratory, Upton, NY
Background/Question/Methods

Ash trees across North America are threatened by the invasive emerald ash borer (EAB), a buprestid beetle that continues to kill tens of millions of ash trees (Fraxinus spp) in 15 states and provinces with no signs of abatement.  Invasion by EAB presents a challenge to resource managers who must mitigate its effects on host trees despite poor knowledge of the hosts’ distribution.  We set out to create a method to map ash trees across a broad spatial extent using remote sensing.   We developed a technique that allows discrimination of forest species phenology at 30m resolution using Landsat and MODIS satellite imagery. This allowed us to utilize the unique phenology of ash in an effort to map host trees for EAB.  Using field data collected in the summers of 2008 and 2009, we mapped ash trees in state forests in southeastern and northern Wisconsin using satellite-derived phenology.   

Results/Conclusions

Although our study areas span a climatic gradient and hence display different phenology, our method accounted for this difference to describe the same species in both areas in consistent phenological terms so that we had a basis on which to extend our predictive model to a broader area. We found that relative basal area of the summed proportion of all ash species was predicted with a R2=0.72 using only our phenological data.  This relationship was extended to a wider area to map ash trees in surrounding forests and across Wisconsin, with validation results showing good agreement between our map of ash and ancillary data at levels of ash abundance important to managers. This provides decision-makers in Wisconsin with an accurate, high-resolution statewide map of ash trees.  In addition, our method represents an opportunity to examine other systems with species of interest that possess unique phenology.