Spatio-temporal analysis using a multiscale hierarchical ecoregionalization

We address the need for spatio-temporally explicit analysis techniques linking the scales of ecosystem, observation, and analysis, using a hierarchical ecoregionalization to examine,remotely sensed data at spatial scales of ecological and management significance. Long- and short-term changes in vegetation functioning are a key indicator of ecological processes. We predict net primary production (NPP) at monthly temporal resolution for 16 years (1981-1996) at an 8-km spatial resolution for the approximately 10(6) km(2) area of Ontario, Canada. We calculate landscape-level light use efficiency values that are tuned to monthly and long-term ecoclimates, and the Normalized Difference Vegetation Index from the NOAA-AVHRR sensor. Applying our spatio-temporal analysis tools, we show evidence for increasing NPP across most of the province. This increase varies seasonally and annually across Ontario, and its magnitude and distribution varies with the spatial scales of analysis. Bridging the gap between local and global studies, this research supports monitoring and analysis of ecosystem functions.