Simulering av avloppsflöden med regndata från mobiltelefonnät i Stockholm
Municipal water utilities often only have access to a sparse network of rain gauges (ca. 10 in Stockholm). With sparse networks there is a risk to under- or overestimate sewer flow calculations when showers pass between gauges, or when a spatially concentrated shower hits a gauge. Moreover, significant resources are required to keep the gauge network operational and reliable. Microwave links in telecommunication networks can measure rainfall. The link network is much denser than the gauge network (here we study 500 links in Stockholm), and is continuously maintained. Such links can hence be a valuable complement to gauges. This study initially compares link-based and gauge-based rainfall data. Generally, we observed lower maximum intensities for the links compared with the gauges: the shorter the duration the larger the difference. Subsequently, we investigate how increasing network density and different types of rain data affects runoff, pipe flow and overflow in the sewer network using a hydraulic model. The results show large differences in calculated flow between varying network densities (sometimes +/-80 %), and that doubling the density often was sufficient during the seven studied rain events. Increasing the density using microwave links also gave better correspondence with measured inflow to the Henriksdal sewage treatment plant. In summary, operational telecommunication networks have a large potential to complement rain gauges for water utilities.