Prototyping of MODIS LAI and FPAR algorithm with LASUR and LANDSAT data

This paper describes results from prototyping of the moderate resolution im aging spectroradiometer (MODIS) radiative transfer-based synergistic algori thm for the estimation of global leaf area index (LAI) and fraction of phot osynthetically active radiation (FPAR) absorbed by vegetation using land su rface reflectances (LASUR) and Landsat data, The algorithm uses multispectr al surface reflectances and a land cover classification map as input data t o retrieve global LAI and FPAR fields. Our objectives are to evaluate its p erformance as a function of spatial resolution and uncertainties in surface reflectances and the land cover map. We analyzed reasons the algorithm can or cannot retrieve a value of LAI/FPAR from the reflectance data and justi fied the use of more complex algorithms, instead of NDVI-based methods, The algorithm was tested to investigate the effects of vegetation misclassific ation on LAI/FPAR retrievals. Misclassification between distinct biomes can fatally impact the quality of the retrieval, while the impact of misclassi fication between spectrally similar biomes is negligible. Comparisons of re sults from the coarse and fine resolution retrievals show that the algorith m is dependent on the spatial resolution of the data. By evaluating the dat a density distribution function, we can adjust the algorithm for data resol ution and utilize the algorithm with data from other sensors.