A novel method for normalizing the time of day effects on polar orbit satellite land surface temperature has been developed by the Agricultural Remote Sensing Team at the Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences. This method addresses the non-linear variations in land surface temperature during satellite overpasses. The research findings have been published in the international journal "IEEE Transactions on Geoscience and Remote Sensing".
Land surface temperature is a crucial indicator in fields such as agriculture, meteorology, ecology, and hydrology. Currently, inconsistencies in observation times of land surface temperature in the same area acquired by polar orbit satellites due to sensor scanning characteristics severely limit the practical application of land surface temperature data.
To tackle this issue, the agricultural remote sensing team developed differential and ratio models for normalizing polar orbit satellite land surface temperature over time, based on high-temporal-resolution station data and geostationary satellite data. These models enable a more accurate capture of the non-linear temperature trends.
Team members also analyzed the influence of various factors such as land surface characteristics, atmospheric conditions, and solar radiation on temperature. They introduced gradient regression algorithms to optimize parameter estimation for the models, successfully enhancing the accuracy and applicability of the land surface temperature normalization algorithm. This research breakthrough provides important theoretical and methodological support for monitoring and managing agricultural ecological environments at the regional scale.
Dr. Du Wenhui, a postdoctoral fellow at the Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, is the first author of the paper. This research received funding from the National Natural Science Foundation of China's Innovative Research Groups and other projects.
