Relationships between species composition, fractional cover and land surface temperature in a mediterranean ecosystem

摘要

HyspIRI is a proposed satellite mission combining two 60 m spatial resolution sensors: a Visible-Shortwave Infrared (VSWIR) imaging spectrometer and a multispectral thermal infrared (TIR) scanner. HyspIRI offers the potential to combine the ability of a VSWIR sensor to discriminate plant species and estimate accurate surface fractions of biotic and abiotic materials using multispectral emissivity with improved Land Surface Temperature (LST) retrieval from the TIR sensor. We evaluate potential synergies between AVIRIS maps of species/cover and fractions and MASTER LST utilizing multiple flight lines acquired in July 2011 in the Santa Barbara area. Cover was mapped using Multiple Endmember Spectral Mixture Analysis using a spectral library derived from the 7.5 m imagery and endmembers selected using Iterative Endmember Selection. Temperature Emissivity Separation (TES) was accomplished using the MASTER TES algorithm. Pixel-based accuracy was 55% for 21 classes, but 84.7% based on pixel majority in reference polygons. An inverse relationship was observed between green vegetation (GV) fractions and LST, but varied by plant species, generating unique LST-GV clusters.