Swedish new radar system for enhanced urban stormwater management
1. Climate challenges, mitigation and adaptationHossein Hashemi1, Rolf Larsson1, Jonas Olsson2, Henrik Aspegren3, Ronny Berndtsson1
1 Lund University, Division of Water Resources Engineering, P.O. Box 118, SE-22100 Lund, Sweden.
2 Swedish Meteorological and Hydrological Institute, SE-601 76 Norrköping, Sweden.
3 Sweden Water Research, the Joint R&D company of NSVA, Sydvatten and VA SYD, SE-223 70 Lund, Sweden.
Abstract text
More frequent torrential downpours and pluvial floods with climate change threaten the liveability of cities globally. Urbanization and densification of cities exacerbate the situation by more contribution and exposure to runoff. It is not only excess runoff but also more polluted water, e.g., from combined sewer overflows, that needs to be dealt with to avoid larger socio-economic and environmental losses than the present. However, spatiotemporal precipitation, especially for extreme events, can be highly variable and challenging to predict. Thus, high-resolution rainfall observations across time and area are becoming essential to managing stormwater flooding. The need is further emphasized for required less uncertainty in the prediction and control of peak flows in heterogenous urban catchments with dominant impervious surfaces. Both historical data and accurate nowcasting of local rainstorms are crucial for the design and real-time operation of many flood protection facilities, from pumping stations to nature-based solutions, or, when, e.g., certain parking lots and sports fields must be evacuated for collecting flood water as a secondary function. Other benefits can also be mentioned such as rainwater harvesting, design and maintenance of green roofs, and creation of innovative blue-green environments such as those in Gothenburg’s plan of becoming the world’s best “rainy city”.
In view of the above, sub-km and minute rainfall observations were initiated through two operational ground-based scanning X-band weather radars (WRs) in south Sweden (Figure 1, and Hosseini et al., 2023). Although these resolutions are unique (vs. traditional rain gauges that are tedious and expensive to maintain), the remotely sensed data require extensive verifications and corrections. This study introduces the recent efforts in Sweden in this direction, which, for example, includes merging data from multiple-level scans of two X-band WRs using artificial neural networks. Also, a Micro Rain Radar (MRR) is analyzed for real-time correction of the X-band WRs. MRR more continuously (every 10 sec) monitors the vertical profile of rainfall to detect, among other things, hydrometeors’ melting layer, which can introduce considerable errors in the WRs scanning at certain elevation angles of the antenna. Although the technologies and research are mainly focused on Skåne, the developed methods can be generalized to other areas. The methods building on Artificial Intelligence (AI) solutions, e.g., can be expanded to other regions of Sweden or internationally to advocate consistent strategies towards more reliable flood protection of cities via new radar intelligence.
Selected references
Hosseini S.H., Hashemi H., Larsson R., Berndtsson R., Merging dual-polarization X-band radar network intelligence for improved microscale observation of summer rainfall in south Sweden, Journal of Hydrology (2023) 129090. DOI: 10.1016/j.jhydrol.2023.129090
