Evaluation of Split Window Land Surface Temperature Algorithms for Climate Data Records

摘要

Land surface temperature (LST) is a key proxy of earth surface energy and is used in range of hydrological, meteorological and climatological applications. As needed for most modeling and climate analysis applications, LST products generated from polar orbiting meteorological satellite sensors have spatial resolutions from several hundred meters to several kilometers and have daily temporal resolution. Those sensors include the NOAA AVHRR series, the EOS MODIS series, and the forthcoming Visible and Infrared Imagery Radiometer Suite (VIIIRS), to be flown onboard the National Polar-orbiting Operational Environmental Satellite System (NPOESS; launch ～ 2011) and NPOESS Preparatory Project (NPP; launch 2008) platforms. Generally, split-window algorithms are used with these sensors to produce LST products. In this study, we evaluated leading algorithms, originally developed specifically for AVHRR, MODIS or VIIRS, to determine those most suitable for generating a consistent LST climate data record (CDR) across different satellite sensors and platforms. MODTRAN simulations were used for determining appropriate coefficients for each algorithm and sensor (AVHRR, MODIS and VIIRS) combination. Algorithm accuracy was evaluated over different view zenith angles, atmosphere-surface combinations, and emissivity uncertainties. We conclude that algorithms from Ulivieri et al. (1996) and Prata and Platt (1991), modified with a path length variability term, were the best overall performers.