Long-Term Data Records of Biophysical Parameters from Multiple Satellite Systems

Abstract:

Long-term satellite-derived vegetation biophysical products like the Leaf Area Index (LAI) and the Fraction of Photosynthetically Active Radiation (FPAR) are widely popular in the scientific research community for monitoring ecosystem health and vegetation dynamics as well as for a number of climate-ecosystem models that use LAI/FPAR as an input. The availability of continuous time series data from heritage sensors like the MODIS, AVHRR, SPOT, Landsat and MISR have allowed a number of research teams to generate biophysical products from these sensors using a wide variety of physical and emperical modeling techniques. However, the consistency and the continuity in the generated products from these various sensors have always been a concern for application scientists. Deriving consistent products require sophisticated algorithm formulation that takes into account the differences in sensor characteristics like spectral response function, field of view, spatial resolution and view/azimuth angle, to name a few. There have been recent advances in algorithms that take into account these differences in order to create a consistent long-term data. Another important advancement has been the generation of hybrid products that use information from two different sensors to drive a physical model. This paper presents a treatise on the existing long-term biophysical products and their usability in terms of global change research and applications that may range from a watershed level to climate model simulations. In addition, the paper will present (a) state-of-the-art LAI/FPAR products that are being generated from the Landsat sensor at a continental scale and its application to the NASA Carbon Monitoring System (CMS) and, (b) some of the recent efforts on extending the physical algorithm to next generation satellite sensor systems like the Landsat 8 and Sentinel-2.