Imaging spectrometry provides new opportunities for quantitative assessment of ecosystem states and processes at larger scales than can be easily measured on the ground, and provides a useful experimental tool for developing new remote sensing methods. Imaging spectrometry can be used to assess vegetation productivity (photosynthesis, gross or net primary productivity), typically using some form of the light-use efficiency (LUE) model. At the same time, imaging spectrometry can be applied to assess biodiversity, for example, by mapping critical habitat or dominant species composition. Biodiversity can also be assessed through “optical diversity,” the variation in information content present in reflectance spectra over space or time. In our study, we used the coefficient of variation (CV) of spectral reflectance in space as a metric of optical diversity.
Building on previous studies that demonstrate separate capabilities for productivity and diversity studies, we applied airborne imaging spectrometry for an integrated assessment of productivity and biodiversity in a prairie ecosystem (Mattheis Ranch, Alberta Canada), using the LUE model and biomass as metrics of productivity, and optical diversity (CV) as a proxy for vegetation richness and evenness. This combined assessment allowed us to explore the relationship between biodiversity and productivity, which has been controversial for grassland ecosystems. Eddy covariance, vegetation maps, and field sampling of biomass and species composition were used as calibration and validation.
Results/Conclusions
Results indicate that optical diversity (expressed as the coefficient of variation of spectral reflectance in space) is significantly correlated with independent field-based metrics of vegetation biodiversity. This metric provides an objective, statistically-based measure of variation in optical properties that reflects functional differences in plant resource use, leaf traits, and canopy display. Similarly, biodiversity, expressed as optical diversity, species richness, or Shannon Index, is significantly correlated with productivity, expressed as biomass or photosynthetic rates. These results support the hypothesis of a positive diversity-productivity relationship for this prairie ecosystem, and demonstrate the utility of imaging spectrometry for landscape-level ecological studies. Unanswered topics include the scale dependence of the diversity-productivity relationship in space, time, and spectral space. The further development of these approaches could lead to routine, operational methods for assessing biodiversity, productivity, and other aspects of ecosystem function using imaging spectrometry.