MODELLING NET ECOSYSTEM EXCHANGE AND LUE IN MEDITERRANEAN OAK FOREST BY SATELLITE REMOTE SENSING

摘要

Net Ecosystem Exchange (NEE) is a key factor defining CO_2 fluxes between atmosphere and ecosystems and CO_2 flux measurements at individual eddy covariance flux sites provide valuable information on the seasonal dynamics of NEE. In this work, we developed and validated a satellite-based Light Use Efficiency (LUE) model to estimate NEE for a typical oak forest located in Central Italy. Satellite data were acquired by Moderate resolution spectroradiometer (MODIS) sensor installed on board Terra satellite. Oak forest studied is coppice managed; 2 eddy-covariance towers are located inside two forests parcels having different ages. We proposed to estimate LUE like function of mean brightness temperature, Normalized Difference Water Index (NDWI) and Photochemical Reflectance Index (PRI). Empirical multiple regressions models (MR) and Artificial Neural Network (ANN) were parameterized and validated using subset of data acquired by both the stations. Daily, 8-day and monthly temporal resolutions were investigated and accuracy estimation in space and time was performed.