Commercial Microwave Links (CMLs) have emerged as one of the most widely utilized opportunistic sensors for rainfall monitoring. However, rainfall retrieval using microwave links continues to face significant challenges in terms of accuracy, particularly for shorter path lengths. In recent years, a statistical approach has been demonstrated to effectively enhance retrieval accuracy. Concurrently, studies have shown that the selection of localized parameters can further optimize CML retrieval results. In this study, we evaluate and calibrate the probabilistic–statistical retrieval method proposed in a previous study for the Chinese region. Following their framework, we replace the global parameters with a Gamma rainfall distribution derived from local rain gauge observations, making the method more suitable for local climatic conditions. To validate the effectiveness of the improved method, we deployed three experimental microwave links with path lengths ranging from 420 m to 3.50 km and simultaneously recorded path attenuation along with rainfall data from surrounding rain gauges. The results show that the coefficient of determination and correlation coefficient between the proposed method and rain gauge observations reach 0.85 and 0.86, respectively, indicating a significant improvement over traditional models. The calibrated method performs particularly well during high-intensity rainfall events, demonstrating the importance of parameter localization for improving retrieval accuracy.
Shen et al. (Wed,) studied this question.