Current and future CML-rainfall estimation in Germany: Improved data processing, real-time rainfall maps and fusion with weather radar data

Many cell phone base stations are connected by a network of commercial microwave links (CMLs). At the typically used frequencies between 15 GHz and 40 GHz, precipitation along the path of a CML leads to significant attenuation of the signal. The path-averaged rain rate along a CML can therefore be derived from measurements of the attenuation.

In cooperation with Ericsson, we record attenuation data of 4000 CMLs across Germany with our own open source data acquisition software. The data is acquired every minute and is available to us in real time. The dataset is continuously growing and now spans more than two and a half years. 

Here we present and discuss results from our current processing chain for hourly country-wide CML-derived rainfall fields. We show the effect of improved rain event detection in the raw attenuation time series and the necessity to correct for wet antenna attenuation (Graf et al., 2019). Validation is done via the gauge-adjusted radar product RADOLAN-RW of the German meteorological service. For summer months the pearson correlation between CML and radar data reaches up to 0.85, but is substantially worse during the winter months. The presented processing chain is fast enough to be applied in real-time, which will be illustrated in a live-demo. Furthermore, since Germany has both, a large network of CMLs and a modern weather radar network, we also work on the combination of these data sources. We will present first results of an approach where CMLs are used as an additional source for weather radar rain rate adjustment similarly to the existing gauge-adjustment done in RADOLAN.

Graf, M., Chwala, C., Polz, J., and Kunstmann, H.: Rainfall estimation from a German-wide commercial microwave link network: Optimized processing and validation for one year of data, Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2019-423, 2019