Knowledge Discovery from Remotely Sensed Vegetation Indices

The objective of this research is to develop KDD (knowledge discovery in databases) techniques for spatio-temporal geo-data, and use these techniques to examine seasonal and inter-annual vegetation health signals. The underlying hypothesis of the research is that the signatures of inter-annual variability of climate on vegetation dynamics as represented by the statistical descriptors of vegetation index variations depend upon a variety of attributes related to the climate, physiography, topography, and hydrology. Several scientific questions related to the identification and characterization of the inter-annual variability ensue as a consequence of this hypothesis. Various vegetation indices will be enlisted to represent vegetation health, such as NDVI (normalized difference vegetation index), EVI (enhanced vegetation index), LAI (leaf area index), FPAR (fraction of photosynthetically active radiation), PSN (photosynthesis), and NPP (net primary product). Relationships between these indices and topography and its derivatives (slope, aspect, etc.), nearness to water bodies, precipitation, temperature, etc. will be analyzed. Preliminary investigations were performed using 13 years of 1 km resolution NDVI data from the AVHRR instrument on NOAA's POES (polar-orbiting operational environmental satellite). Deviations from the 13-year average were utilized in order to identify anomalous behavior. The pilot KDD technique includes distance-based clustering algorithms and Apriori association rule algorithms adapted for spatial-temporal data. Future work will incorporate more complex algorithms such as density-based clustering and constraint-based association mining algorithms.