A fractal approach for the characterization of NOAA-AVHRR NDVI profiles of broad scale ecoregions

Regional variations in climate, vegetation and landform are important in the development of ecosystems. For this reason, it is necessary to recognize regional differences in ecologically relevant factors at the highest level of spatial scale in the ecosystem classification. With the operational program of the National Oceanic and Atmospheric Administration (NOAA), begun in 1990, periodical time series of the Normalized Difference Vegetation Index (NDVI) from the Advanced Very High Resolution Radiometer (AVHRR) are available for large area monitoring of Earth's vegetation phenology and long-term changes resulting from human interactions or global climate alterations. We show here that the fractal dimension (D) of average NDVI profiles associated with broad scale ecoregions can be used to summarize their phenological pattern better than the simple visual examination of NDVI profiles or previous profile-derived indexes do.