Optimal spatial assessment of short-time step precipitation for hydrologicalmodelling is still an important research question considering the poorobservation networks for high time resolution data. The main objective ofthis paper is to present a new approach for rainfall observation. The ideais to consider motorcars as moving rain gauges with windscreen wipers assensors to detect precipitation. This idea is easily technically feasible ifthe cars are provided with GPS and a small memory chip for recording thecoordinates, car speed and wiper frequency. This study explorestheoretically the benefits of such an approach. For that a validrelationship between wiper speed and rainfall rate considering uncertaintywas assumed here. A simple traffic model is applied to generate motorcars onroads in a river basin. Radar data are used as reference rainfall fields.Rainfall from these fields is sampled with a conventional rain gauge networkand with several dynamic networks consisting of moving motorcars, usingdifferent assumptions such as accuracy levels for measurements and sensorequipment rates for the car networks. Those observed point rainfall datafrom the different networks are then used to calculate areal rainfall fordifferent scales. Ordinary kriging and indicator kriging are applied forinterpolation of the point data with the latter considering uncertainrainfall observation by cars e.g. according to a discrete number ofwindscreen wiper operation classes. The results are compared with the valuesfrom the radar observations. The study is carried out for the 3300 km2Bode river basin located in the Harz Mountains in Northern Germany. Theresults show, that the idea is theoretically feasible and motivate practicalexperiments. Only a small portion of the cars needed to be equipped withsensors for sufficient areal rainfall estimation. Regarding the requiredsensitivity of the potential rain sensors in cars it could be shown, thatoften a few classes for rainfall observation are enough for satisfactoryareal rainfall estimation. The findings of the study suggest also arevisiting of the rain gauge network optimisation problem.
展开▼
机译：考虑到高分辨率数据的观测网络较差，对水文模拟进行短时降水的最佳空间评估仍然是一个重要的研究问题。本文的主要目的是提出一种新的降雨观测方法。想法是将汽车视为带有雨刷的移动雨量计，以作为检测降水的传感器。如果汽车配备有GPS和一个用于记录坐标，汽车速度和雨刮器频率的小型存储芯片，则该想法在技术上很容易实现。这项研究从理论上探讨了这种方法的好处。为此，这里假设考虑不确定性的雨刮器速度与降雨率之间存在有效关系。一个简单的交通模型被应用于在流域的公路上生成汽车。雷达数据用作参考降雨场，这些场的降雨是通过常规假设的雨量计网络以及由移动汽车组成的几个动态网络进行采样的，其中使用了不同的假设，例如测量的精确度和汽车网络的传感器设备率。然后，将来自不同网络的那些观测点降雨数据用于计算不同规模的区域降雨。点数据的插值采用普通克里金法和指标克里金法，而后者考虑了不确定的降雨观测，例如汽车根据不连续的挡风玻璃刮水器操作类别。将结果与雷达观测值进行比较。这项研究是针对位于德国北部哈尔茨山脉的3300 km 2 Bode流域进行的。结果表明，该思想在理论上是可行的，可以激发实践实验。仅一小部分汽车需要配备传感器，以进行足够的区域降雨估算。关于汽车中潜在的雨水传感器的灵敏度要求，可以看出，通常只有几类降雨观测值足以满足令人满意的区域降雨估计。该研究的发现也表明正在考虑雨量计网络优化问题。
展开▼
