Microwave links in cellular communication networks are employed to transfer data from the antenna of one telephone tower to the antenna of another one. Although these links have not been designed to measure rainfall, they can be used to estimate the rainfall intensity between the telephone towers. The basic principle is as follows. Rainfall attenuates the electromagnetic signals transmitted from one telephone tower to another. By measuring the received power at one end of a microwave link as a function of time, the path-integrated attenuation due to rainfall can be calculated, which can be converted to average rainfall intensities over the length of a link1. The attenuation becomes larger for an increasing number and size of raindrops present along the link. These received powers are often already stored by telecommunication companies to monitor the stability of their connections.
Because these networks cover roughly 20% of the Earth’s land surface, the potential for this new rainfall measurement technique is large. They are particularly interesting for those countries where few surface rainfall observations are available. Rainfall information is highly relevant for e.g. agriculture, hazardous weather warnings, and climate monitoring. New rainfall measurement techniques are especially important because the number of useful rain gauge observations is declining.
The potential for long-term large-scale operational rainfall monitoring has been demonstrated by utilizing a 2.5-year data set from a cellular communication network2. The data set consists of roughly 2,000 links covering the land surface of the Netherlands (35,500 km2). Comparisons of link rainfall maps with maps from radars and gauges confirm their potential for rainfall monitoring. An example of such a comparison is provided by the movie.
The underlying rainfall retrieval and mapping algorithm is freely available as R package RAINLINK3,4. It contains a working example to compute link-based rainfall maps for the entire surface area of The Netherlands. We hope that RAINLINK will promote the application of rainfall monitoring using microwave links in poorly gauged regions around the world.
Evaluation of link and satellite rainfall products for The Netherlands confirms the potential of link rainfall data for ground validation of or merging with satellite rainfall products5. Package RAINLINK has been extended, so that it can also deal with link data from networks with instantaneous sampling strategies. Moreover, we analyzed data from a dedicated research experiment (WURex14), held in Wageningen, in order to increase our understanding of (sources of) errors in link rainfall estimates6. Finally, we intend to apply this technique to data from other countries and climates in order to advance the process toward operational application and upscaling of this new rainfall measurement technique. As a first step this has been performed by applying RAINLINK to data from São Paulo, Brazil7. We also continue to investigate the merging of rainfall data from various sources, i.e. personal weather stations, microwave links and weather radars.
1. Overeem, A., Leijnse, H., Uijlenhoet, R. Country-wide rainfall maps from cellular communication networks, 2013. Proceedings of the National Academy of Sciences of the United States of America, 110, 2741-2745
2. Overeem, A., Leijnse, H., Uijlenhoet, R. Two and a half years of country-wide rainfall maps using radio links from commercial cellular telecommunication networks, 2016b. Water Resources Research, 52.
3. Overeem, A., Leijnse, H., Uijlenhoet, R. Retrieval algorithm for rainfall mapping from microwave links in a cellular communication network, 2016a. Atmospheric Measurement Techniques, 9, 2425-2444.
4.The RAINLINK software
5. Rios Gaona, M.F., Overeem, A., Brasjen, A.M., Meirink, J.F., Leijnse, H., Uijlenhoet, R. Evaluation of rainfall products derived from satellites and microwave links for The Netherlands, 2017. IEEE Transactions on Geoscience and Remote Sensing, 55, 6849-6859.