A retrieval method for land surface temperatures based on UAV broadband thermal infrared images via the three-dimensional look-up table

Xue Zhong,Lihua Zhao,Jie Wang,Xiang Zhang,Zichuan Nie,Yingtan Li,Peng Ren

DOI: https://doi.org/10.1016/j.buildenv.2022.109793

IF: 7.093

2022-11-11

Building and Environment

Abstract:Land surface temperatures (LSTs) represent a crucial parameter to characterize the urban thermal environment. Although numerous methods have been proposed to retrieve LSTs based on data from satellites or unmanned aerial vehicles (UAVs), challenges remain for acquiring precise pixel-level LSTs of highly heterogeneous urban surfaces on the micro scale. This study proposed a method of temperature retrieval for UAV-derived broadband thermal infrared images via a three-dimensional look-up table (TRUB-3DLUT) to retrieve LSTs of complicated urban surfaces. Based on radiative transfer between surfaces and the atmosphere, the method comprehensively quantified the impacts of radiation emitted by surfaces, wavelength, land surface emissivity (LSE) and atmospheric parameters on the retrieved LSTs. Then, the 3DLUT was directly used to generate the urban low-altitude atmospheric products and retrieve the LSTs via the UAV-derived images with a high-spatial resolution. Validations demonstrated that the retrieved LSTs were well-aligned with the in-situ LSTs and that the absolute differences were less than 1.5 K for sunlit surfaces with higher emissivity. The TRUB-3DLUT offered practical utility in reducing the uncertainty of the inputs in the radiative transfer equation and thus improving the accuracy of the retrieved LSTs. Therefore, such a method is envisioned as an effective tool for quantitatively analyzing the spatial variations of urban thermal environments on the micro scale.

engineering, environmental,construction & building technology, civil