B41N-03:
Remote Sensing of Mycorrhizae? Detection of Mycorrhizal Association from Canopy Spectral Properties.
Thursday, 18 December 2014: 8:30 AM
Joshua B Fisher1, Sean Sweeney2, Edward R Brzostek3, Tom P Evans4, Norman A Bourg5 and Richard Phillips3, (1)NASA Jet Propulsion Laboratory, Pasadena, CA, United States, (2)Indiana University Bloomington, Center for the Study of Institutions, Populations, and Environmental Change (CIPEC), Bloomington, IN, United States, (3)Indiana University Bloomington, Department of Biology, Bloomington, IN, United States, (4)Indiana University Bloomington, Center for the Study of Institutions, Populations, and Environmental Change (CIPEC) and Department of Geography, Bloomington, IN, United States, (5)Smithsonian Conservation Biology Institute, Conservation Ecology Center, Front Royal, VA, United States
Abstract:
Nearly all tree species form symbiotic relationships with one of two types of mycorrzhae—arbuscular mycorrhizae (AM) and ectomycorrhizal (ECM) fungi. AM- and ECM-dominated forests often have distinct nutrient economies, and there is strong interest in mapping or detecting mycorrhizae over large areas. We explored remotely sensed tree canopy spectral properties to “detect” underlying mycorrhizal association across a gradient of AM- and ECM-dominated forest plots. We used a combination of principal components analysis and statistical mining of reflectance and band differencing across moderate/high-resolution Landsat data in conjunction with phenological signals from stitched seasonal time series and topographic features. This approach was trained and validated against measurements of tree species and mycorrhizal association across more than 60,000 trees throughout the central and eastern US. Using this approach, we were able to predict 81% of the variation in mycorrhizal association (p<0.001). Differences in phenological characteristics between AM- and ECM-associated trees drove the relatively high prediction skill.
