Estimating very high resolution urban surface temperature using a spectral unmixing and thermal mixing approach

► Very high resolution (VHR) land surface temperature (LST) is critical information. ► This information is difficult to obtain due to the limitations of sensor technology. ► We propose a very high resolution spectral unmixing and thermal mixing approach. ► The method alleviates confusion among moist soil, shadow, and impervious surfaces. ► Accurate estimates of VHR LST were achieved with relatively low RMSE and MAE.