A Continuous NDVI Record
The long term AVHRR-NDVI record provides a critical historical perspective on vegetation dynamics necessary for global change research. The remote sensing community is still struggling to create sensor-independent datasets that allow the simultaneous use of spectral vegetation indices from several sensors in the same time series analysis. To overcome the non-linear and non-stationary aspects of NDVI, we use a Neural Network to map the NDVI indices from AVHRR to those from MODIS using atmospheric, surface type and sensor-specific inputs to account for the differences between the sensors. The NDVI dynamics and range of MODIS NDVI data at one degree is matched and extended through the AVHRR record. Three years of overlap between the two sensors is used to train a neural network to remove atmospheric and sensor specific effects on the AVHRR NDVI.
This site presents the resulting corrected dataset, its relationship to MODIS data, and a validation of the product.