For more information on temporal profiles research at UMN, please contact Marvin Bauer.
Project Overview
AVHRR and MODIS data have great potential to monitor and compare annual vegetation cycles at regional to global scales. In this project, several new approaches are being developed to examine and improve this potential from the three-state RESAC region down to the county level. The methods we have devised for AVHRR Normalized Difference Vegetation Index (NDVI) data can also be applied to MODIS data which is better suited to regional scale study.
Multi-year archives of AVHRR/NDVI data make it possible to analyze the vegetation cover profiles from 1990 to the present. We have applied time synchronization to the widely used bi-weekly Maximum Value Composite (MVC) processed NDVI data to eliminate the inter-annual variation caused by asynchronous observations, and derived numerical metrics to describe the resulting greenness profiles more accurately.
A database containing multi-tier NDVI profile information has been built to enable data mining analysis on vegetation spatial and temporal dynamics. It also facilitates the exploration of connections among vegetation cover changes and the regional weather factors (temperature and precipitation, etc.). It also makes it possible to investigate modeling crop yield at the county level based on NDVI profile or anomaly information.
