IEIIT, Consiglio Nazionale delle Ricerche, 20133 Milano, Italy.
DICA, Politecnico di Milano, 20133 Milano, Italy.
Sensors (Basel). 2022 Apr 22;22(9):3218. doi: 10.3390/s22093218.
Despite the several sources of inaccuracy, commercial microwave links (CML) have been recently exploited to estimate the average rainfall intensity along the radio path from signal attenuation. Validating these measurements against "ground truth" from conventional rainfall sensors, as rain gauges, is a challenging issue due to the different spatial sampling involved. Here, we assess the performance of a network of CML as opportunistic rainfall sensors in a challenging mountainous environment located in Northern Italy. The benchmark dataset was provided by an operational network of rain gauges and by three disdrometers. Moreover, disdrometer data were used to establish an accurate relationship between path attenuation and rainfall intensity. A new method was developed for assessing CML: time series of rainfall occurrence and rainfall depth, representative of CML radio path, were derived from the nearby rain gauges and disdrometers and compared with the same quantities gathered from the CML. It turns out that, over the very short integration times considered (10 min), CML perform well in detecting rainfall, whereas quantitative rainfall estimates may have large discrepancies.
尽管存在多种不准确因素，但商业微波链路 (CML) 最近已被用于根据信号衰减来估计无线电路径上的平均降雨量强度。由于涉及到不同的空间采样，因此将这些测量值与传统降雨传感器（如雨计）的“地面真实值”进行验证是一个具有挑战性的问题。在这里，我们评估了商业微波链路网络作为在意大利北部具有挑战性的山区环境中的机会性降雨传感器的性能。基准数据集由运营中的雨量计网络和三个雨滴谱仪提供。此外，雨滴谱仪数据用于建立路径衰减和降雨量强度之间的准确关系。开发了一种新的方法来评估 CML：从附近的雨量计和雨滴谱仪中得出代表 CML 无线电路径的降雨发生和降雨深度的时间序列，并将其与从 CML 收集到的相同数量进行比较。结果表明，在考虑的非常短的积分时间（10 分钟）内，CML 在检测降雨方面表现良好，而定量降雨估计可能存在较大差异。
