The estimation of high resolution urban surface temperature using hyperspectral spectral mixture analysis

Abstract

In this paper, a simple method that uses hyperspectral spectral mixture analysis to estimate fine resolution urban LST is discussed. In the first step, multiple endmember spectral mixture analysis is applied to calculate land composition fractions. High resolution urban LST is then estimated through a linear combination of each thermal response component multiplied by their respective abundance in the second step. Evaluation of the proposed method was performed at four spatial scales (4m, 8m, 16m and 32m) over two scenes with different land cover characteristics. Results show a relative modest agreement was found between the estimated LST and the reference LST for both visual interpretation and quantitative accuracy metrics. Overall, this approach enhances the exploitation of hyperspectral data for estimating fine spatial resolution LST, therefore can serve as a promising technical in urban and environmental applications.