Fusion Method Enhances All-Weather Total Precipitable Water Monitoring
Jan 12, 2024
In a big stride towards advancing our understanding of the water cycle and climate change, researchers have unveiled a novel fusion method capable of providing high-resolution Total Precipitable Water (TPW) information even in challenging all-weather conditions. The method, detailed in a recent paper published in the Remote Sensing of Environment addresses the limitations faced by current satellite sensors in obtaining high-resolution TPW and introduces a fusion algorithm leveraging multiple water vapor-related variables from both microwave and near-infrared (NIR) remote sensing.
The proposed method achieves a remarkable spatial resolution of 0.05° × 0.05°, a critical enhancement for studying regional water cycle processes. Moreover, this high-resolution TPW serves as a vital input parameter for atmospheric correction, contributing to the elimination of atmospheric impacts and refining the accuracy of surface parameter retrieval.
Two key advancements distinguish this fusion method. Firstly, a new water vapor-sensitive parameter, the ratio of brightness temperature polarization difference between frequencies 89 and 36.5 GHz (ΔTb89/ΔTb36.5), was introduced into the algorithm. This addition significantly improves the accuracy of the fused TPW, providing enhanced information in both clear and cloudy sky conditions.
Secondly, a two-step fusion approach was proposed to maximize the use of high-quality TPW information from different sources. In the initial step, the method exploits high-quality TPW data from the Advanced Microwave Scanning Radiometer 2 (AMSR2) and ΔTb89/ΔTb36.5 in cloudy conditions. Subsequently, in clear sky conditions, high-quality TPW from the Moderate Resolution Imaging Spectroradiometer (MODIS) is incorporated in the second step.
Validation of the fused TPW was conducted using TPW data from global navigation satellite system (GNSS) stations of the Crustal Movement Observation Network of China (CMONOC). The results revealed an impressive correlation coefficient of 0.96, a root mean square error of 4.23 mm, and a relative root mean square error of 23.67% when compared to GNSS TPW. Importantly, the fused TPW exhibited a substantial improvement of 19.96% and 27.32% in root mean square errors compared to the original TPW from MODIS and AMSR2, respectively.
This fusion method not only overcomes existing limitations in obtaining high-resolution TPW but also represents a promising advancement in enhancing our ability to monitor and understand water cycle dynamics and climate change, even in adverse weather conditions.
Contact: luyq@aircas.ac.cn
News & Events