Markus Quirmbach

The implementation of a decentralized blue–green infrastructure (BGI) is a key strategy in climate adaptation and stormwater management. However, the integration of urban trees into the multifunctional infrastructure remains insufficiently addressed, particularly regarding rooting space in dense urban environments. Addressing this gap, the BoRSiS project developed the soil-pipe system (SPS), which repurposes the existing underground pipe trenches and roadway space to provide trees with significantly larger root zones without competing for additional urban space. This enhances tree-related ecosystem services, such as cooling, air purification, and runoff reduction. The SPS serves as a stormwater retention system by capturing excess rainwater during heavy precipitation events of up to 180 min, reducing the pressure on drainage systems. System evaluations show that, on average, each SPS module (20 m trench length) can store 1028–1285 L of water, enabling a moisture supply to trees for 3.4 to 25.7 days depending on the species and site conditions. This capacity allows the system to buffer short-term drought periods, which, according to climate data, recur with frequencies of 9 (7-day) and 2 (14-day) events per year. Geotechnical and economic assessments confirm the system stability and cost-efficiency. These findings position the SPS as a scalable, multifunctional solution for urban climate adaptation, tree vitality, and a resilient infrastructure.

Deep learning (DL) approaches to forecast precipitation and inundation areas in the short-term forecast horizon have up until now been treated as independent research problems from the model development perspective. However, for the urban hydrology area, the coupling of these models is necessary in order to forecast the upcoming inundation area maps and is, therefore, of the utmost importance for successful flood risk management. In this paper, three deep-learning-based models are coupled in a systematic modular approach with the aim to analyze the performance of this model chain in an operative setup for urban pluvial flooding nowcast: precipitation nowcasting with an adapted version of the NowcastNet model, the forecast of manhole overflow hydrographs with a Seq2Seq model, and the generation of a spatiotemporal sequence of inundation areas in an urban catchment for the upcoming hour with an encoder–decoder model. It can be concluded that the forecast quality still largely depends on the accuracy of the precipitation nowcasting model. With the increasing development of DL models for both precipitation and flood nowcasting, the presented modular approach for model coupling enables the substitution of individual blocks for better and newer models in the model chain without jeopardizing the operation of the flooding forecast system.

This study employs a temporal fusion transformer (TFT) for predicting overflow from sewer manholes during heavy rainfall events. The TFT utilised is capable of forecasting overflow hydrographs at the manhole level and was tested on a sewer network with 975 manholes. As part of the investigations, the TFT was compared to other deep learning architectures to evaluate its predictive performance. In addition to precipitation measurements and forecasts, the issue of how the additional consideration of measurements in the sewer network as model inputs impacts forecast accuracy was investigated. A varying number of sensors and different measurement signals were compared. The results indicate high performance for the TFT compared to other model architectures like a long short-term memory (LSTM) network or a dual-stage attention-based recurrent neural network (DA-RNN). Additionally, results suggest that considering a single measuring point at the outlet of the sewer network instead of an entire measuring network yields better forecasts. One possible explanation is the high correlation between measurements, which increases model and training complexity without adding much value.

This study presents a deep-learning-based forecast model for spatial and temporal prediction of pluvial flooding. The developed model can produce the flooding situation for the upcoming time steps as a sequence of flooding maps. Thus, a dynamic overview of the forthcoming flooding situation is generated to support the decision of crisis management actors. The influence of different input data, data formats, and model setups on the prediction results was investigated. Data from multiple sources were considered as follows: precipitation information, spatial information, and an overflow forecast. In addition, models with different layers and network architectures such as convolutional layers, graph convolutional layers, or generative adversarial networks (GANs) were considered and evaluated. The data required to train and test the models were generated using a coupled hydrodynamic 1D/2D model. The model setup with the inclusion of all available input variables and an architecture with graph convolutional layers presented, in general, the best results in terms of root mean square error (RMSE) and critical success index (CSI). The prediction results of the final model showed a high agreement with the simulation results of the hydrodynamic model, with drastic reductions in computation time, making this model suitable for integration into an early warning system for pluvial flooding.

Measuring weather data in an urban environment is an important task on the journey towards smart cities. Heavy rain can cause flooding in cities and prevent emergency services from reaching their destination because roads or underpasses are blocked. In order to provide a high-resolution site-specific overview in urban areas during heavy rainfall, a dense measurement network is necessary. To achieve this, a smart low-cost rain gauge is needed. In this paper, the current status of the development of an inductive rain gauge is presented. The sensor is based on the eddy current principle and evaluates the frequency of an electrical resonant circuit. For this purpose, a coil is placed under a metal plate. When raindrops hit the plate, it starts to oscillate, which changes the distance to the coil accordingly and causes changes in the frequency of the resonant circuit. Since the sensor is cost-effective, operates self-sufficiently in terms of energy and transmits data wirelessly via LoRaWAN, it can be used flexibly. This enables dense, area-wide coverage over the urban area of interest. The first experimental investigations show a correlation between the size of the rain droplets and the frequency change. Small droplets cause a shift of about 8 kHz and larger droplets of up to 40 kHz. The results prove that raindrops can be detected and categorized using this measurement principle. These data will be used as a basis for future work on calculating precipitation.

This paper presents an application of radar data (DX-product of the German Weather Service) with a high resolution in space (1° × 1 km) and time (Δt = 5 minutes) in urban hydrology. The radar data and data of rain gauges with different locations in the test catchment were compared concerning their suitability as input into an urban rainfall-runoff model. In order to evaluate the accuracy of model simulation results, five evaluation criteria have been specified which are relevant for an efficient management of sewer systems and wastewater treatment plants. The results demonstrate that radar data should be used in urban hydrology if distances > 4 km between rain gauge and catchment exist and for catchments with a density of rain gauges smaller than 1 rain gauge per 16 km².