For effective urban stormwater management information on rainfall at sufficient temporal and spatial resolution is an essential input. The lack of, or insufficient, rainfall data in urban catchments is a global issue that is particularly pronounced in lower-income countries, where the absence of traditional observation systems, combined with rapidly growing urban populations, makes the challenge even more critical. Opportunistic sensing (OS) of precipitation can help in this regard, especially because the two most common OS sensors, i.e. commercial microwave links (CML) and personal weather stations (PWS), are densely distributed in populated areas and are accessible in near-real time. However, there are a number of challenges related to rainfall retrieval using opportunistic sensors. The rainfall data from opportunistic sensors contain high uncertainties and are often noisy, their networks are inhomogeneous, the data can be inconsistent and their interoperability is low. Moreover, the data are owned by private entities and are often not accessible even for scientific purposes.
In response to this situation, the European OpenSense project was launched. It focuses on improving access to OS data, international coordination of OS data standardisation, data processing and follow-up applications in collaboration with a number of European national meteorological services. Our contribution provides an overview of successful community efforts in tackling OS challenges and highlights the evolution of OS techniques from the initial experimental phase to early-stage practical applications. The benefits of OS observations for urban hydrology, along with the enhancement of high-resolution rainfall products are further demonstrated through several case studies.
However, despite significant advances in utilizing OS data for hydrometeorological purposes, a key challenge that remains in its early stage is the upscaling of OS data acquisition and achieving global data availability. Therefore the OpenSense community introduces the concept of a global initiative to allow collection, curation and usage of OS data from CMLs.
How to cite: Bareš, V., Chwala, C., Fencl, M., Messer, H., Ostrometzky, J., Uijlenhoet, R., Overeem, A., van de Beek, R., Olsson, J., Graf, M., Winterrath, T., Thorndahl, S., Seidel, J., Nebuloni, R., and Hanna, N.: One man's noise is another man's signal - the OpenSense project, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19444, https://doi.org/10.5194/egusphere-egu25-19444, 2025.
National-scale precipitation observations in Sweden have traditionally relied on a combination of weather stations and C-band weather radar networks. These observations provide good spatiotemporal coverage and accurate quantification of most stratiform precipitation events across large areas. In the urban context, however, their resolution may be insufficient to capture critical rainfall variations. This limitation is particularly evident for convective rainfall, which is often highly localized (e.g., cloudbursts), and capable of causing severe damage to infrastructure. In light of this, the Swedish Meteorological and Hydrological Institute (SMHI) is exploring complementary ways to monitor rainfall in urban environments.
This study evaluates data from an X-band weather radar (XWR), Commercial Microwave Links (CML), and Private Weather Stations (PWS) to observe a cloudburst event that hit the Bjärehalvön peninsula in in southwestern Sweden in August 2022. The observations are bench-marked with the official monitoring network: SMHI’s weather stations and a C-band radar composite. A maximum volume of 75 mm in 1 hour was reported by a weather station operated by Båstad municipality. This station showed good agreement with long-term observations (2 years) from the nearest SMHI gauge (9 km away) and matched well with the XWR’s measurements of the event. High-resolution (sub-km and 1-minute) XWR data were used to evaluate precipitation variations along a 4.5 km long CML reach, suggesting new potential for correction of CML observations. Additionally, we propose methods for pre-processing of CML and PWS data to ensure consistent precipitation estimates and facilitate cross-referencing with the other sensors.
The results suggest that complementary sensors can add important data on rainfall intensity and volume, enhancing SMHI’s ability to monitor localized heavy rainfall events. However, the opportunistic sensors (CML and PWS) appear to have reached a maximum detectable intensity of rainfall during the event, which likely caused an underestimation of the total volume. Further, the findings highlight the challenge of estimating return periods of convective storms, as the return period varies significantly depending on which sensor that is chosen as the ground truth for the event.
How to cite: Petersson Wårdh, L., Hosseini, H., van de Beek, R., Andersson, J. C. M., Hashemi, H., and Olsson, J.: Have you ever seen the rain? Observing a record summer cloudburst with multiple radars and opportunistic sensors, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3084, https://doi.org/10.5194/egusphere-egu25-3084, 2025.
Rainfall field reconstruction from sparse gauge observations has always been a challenge in hydrometeorology. Traditional geostatistical approaches, such as Ordinary Kriging (OK) and associated geostatistical-based data merging methods, have been widely used (Matheron., 1963; Oliver and Webster, 1990; Sideris et al, 2014). Despite being generally promising, these models often struggle to maintain spatial-temporal consistency while preserving fine-scale features due to the limitation of their underlying statistical assumptions.
Recent research works have aimed to address this limitation with machine learning (Appleby et al., 2020; Harris et al., 2022; Price and Rasp., 2022; Nag et al., 2023; Hsu et al., 2024; Chen et al. 2024). For example, Hsu et al. (2024) introduced a deep-learning approach for downscaling precipitation data, thereby enhancing the representation of fine-scale structures. In contrast, Chen et al. (2024) combined spatial basis function modeling with neural network-driven feature learning to achieve both high accuracy and interpretability in geospatial interpolation.
However, to our knowledge, existing methods have not fully addressed the temporal coherence in precipitation field reconstruction, specifically in maintaining spatial-temporal patterns across consecutive frames. Moverover, many of these methods assume a simple averaging relationship between point measurements and areal precipitation, overlooking the complex scale discrepancy between rain gauge observations and their representative areal means. These deficiencies tend to result in spatially overly smooth fields or low correlation between consecutive frames.
To overcome these limitations, we present a reconstruction method that integrates Convolutional Neural Networks (CNNs) with Generative Adversarial Networks (GANs). Specifically, it incorporates three key innovations: (1) a multi-scale convolution kernel for capturing diverse spatial dependencies, (2) a Fast Fourier Convolution implementation for high-frequency signal preservation, and (3) an adaptive noise injection mechanism that enriches textural details based on local complexity measures.
To evaluate the proposed method, an experiment, using high-resolution radar images over a 64x64 gridded domain, is designed. Within each 10x10 sub-domain, a known point is arbitrarily chosen, and data values at these point locations remain known across the entire event. Our task is to use these known points to predict (or to interpolate) the rest of the image at each time step. The training dataset comprises 1-km Nimrod precipitation fields at 5-min intervals, covering a 64 × 64 km² domain centered on Birmingham city in the UK, spanning from 2016 to 2020. The validation dataset consists of 20 selected storm events between 2021 and 2022. The Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) are used to assess the prediction result. In addition, the Radial Averaged Power Spectral Density (RAPSD) is employed to compare the power spectral density across different frequency ranges, allowing us to assess the reconstruction quality of fine-scale details and overall coarse-scale features in the images.
Preliminary results indicate substantial improvements over traditional Ordinary Kriging methods in both accuracy and computational efficiency; on average, the MAE decreased by 37%, the RMSE reduced by 22%. In addition, the RAPSD results demonstrate an improvement in capturing spatial details. These findings underscore the considerable potential of deep learning techniques for enhancing the spatial-temporal reconstruction of precipitation fields.
How to cite: Wang, B.-Z. and Wang, L.-P.: From points to images: A deep-learning enhanced spatial-temporal reconstruction of precipitation data, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-4537, https://doi.org/10.5194/egusphere-egu25-4537, 2025.
Urban areas significantly influence rainfall patterns, often intensifying rainfall downwind of cities. Depending on various factors, cities can suppress, deflect, split, or intensify incoming storms, with some studies even highlighting storm initiation due to urban heat and dynamics. The location and magnitude of these impacts are largely dependent on the city’s geographic setting, topography, and prevailing wind-flow regime. In India, urban areas have been associated with enhanced monsoon rainfall extremes, with recent research indicating that urban warming is more pronounced over developing tier-II cities than over established metropolitan centers. Limited studies in India have explored the influence of background flow regimes on the preferential locations of rainfall intensification. Against this backdrop, Bhubaneswar, a developing tier-II tropical city in eastern India, serves as an insightful case study.
Bhubaneswar lies approximately 40–50 km inland from the Bay of Bengal, with an average elevation of 45 m above sea level. The city’s terrain rises westward toward the Eastern Ghats and slopes downward eastward toward the ocean, creating a temperature gradient warmer on the western side and cooler on the eastern side due to proximity to the sea. The region receives nearly 80% of its annual rainfall during the monsoon season (JJAS), with most heavy rainfall events driven by low-pressure systems over the Bay of Bengal.
This study utilizes the Weather Research and Forecasting (WRF) model to simulate seasonal monsoon rainfall over Bhubaneswar under varying city-size scenarios. A total of 88 rainfall events were simulated, of which 47 cases exhibited increased rainfall due to urban expansion, while 41 cases showed a reduction. Initial findings indicate that drawing a definitive conclusion on whether urban expansion consistently amplifies or diminishes rainfall is challenging.
During the monsoon season, the region experiences winds from all directions except the north, with southwesterlies being the most dominant. A detailed classification of events based on the prevailing wind regimes provided critical insights. The analysis revealed significant rainfall enhancement over the city and its downwind areas under easterly wind conditions, often associated with cyclonic circulations over the Bay of Bengal. Conversely, westerly winds were found to reduce downwind rainfall. Notably, rainfall enhancement predominantly occurred on the right side of the prevailing wind direction in the case of westerlies.
The study underscores the prominent role of urban location, topography, and prevailing winds in shaping the magnitude and spatial distribution of urbanization-driven rainfall changes during the monsoon. Identifying the preferential location of rainfall enhancement during different wind conditions is crucial for flood mapping and mitigation.
How to cite: Gopinath, N. and Velu, V.: The Interplay between the Location of the City and the Background Winds in Modifying the Rainfall Patterns over an Eastern Indian Tropical City, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-2838, https://doi.org/10.5194/egusphere-egu25-2838, 2025.
Climate change is intensifying short-duration, high-intensity rainfall events in many regions worldwide, highlighting the need to improve rainfall modelling for such storms to support effective stormwater management. A recurring challenge is that many existing rainfall modelling tools fail to account for the evolution of convective cells, potentially leading to under- or over-estimates of rainfall extremes and their hydrological impacts.
To address this challenge, this study presents a spatial-temporal rainfall generator that explicitly incorporates convective cell evolution. In this approach, storm arrivals are modelled by a point process, with storm cells represented by clusters of rainfall objects, of which each is characterised by specific intensity and geometric properties. Whereas most existing generators assume constant cell properties throughout a storm, the properties of our convective cells evolve with time –a more realistic representation of cell lifecycles exhibiting growth and decay. This design also naturally captures the birth of new cells and the dissipation of existing ones during storm events.
The parameters of the generator are derived from an analysis of 167 convective storm events observed in the Birmingham area (UK) between 2005 and 2017. These events were identified and tracked using the enhanced TITAN storm tracking algorithm (Munoz et al., 2018), which extracts convective cell paths and their key properties (e.g., rainfall intensity, spatial extent, storm and cell duration, and movement). The resulting dataset was then used to calibrate a copula-based convective cell lifecycle generator (Tseng et al., 2025), serving as the core mechanism for introducing cell evolution into the rainfall modelling framework.
Preliminary results suggest that our generator not only reproduces the observed standard statistics but also more effectively preserves rainfall extremes than existing generators that assume constant cell properties. In addition, by offering a more realistic representation of cell dynamics and improved spatial-temporal rainfall structures, our generator has the potential to yield more accurate hydrological responses.
How to cite: Tseng, C.-Y. and Wang, L.-P.: Incorporating convective cell evolution into convective storm modelling, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5282, https://doi.org/10.5194/egusphere-egu25-5282, 2025.
Changing precipitation extremes and more attention for water quality (e.g. the revised Urban Wastewater Treatment Directive in the EU) increase the need to understand, model and monitor sewer flow and especially combined sewer overflows (CSO’s), a major source of pollutants in urban areas.
The first step to model water quality correctly, is the correct modelling of water quantity. Additionally, to be able to test different setups and use these models in a complex modelling chain (e.g. in digital twins), there is a need for fast and correct models for the urban hydrological and sewer network.
Here, we present such a fast approach, allowing for a conceptual modelling of the urban sewer network. This is carried out by a combination of linear reservoirs, which resembles distinct zones within the urban area, and a neural network, which is applied to model the dry weather flow. By splitting the rain-driven and dry weather flow, the model can be more easily setup to correctly model sewer overflow, while simultaneously leveraging long-term area-specific relationships between the measured dry weather flow at the waste water treatment plant and the precipitation deficit.
How to cite: Van de Velde, J. and Dewelde, J.: Modelling sewer and combined sewer overflows through a combination of linear reservoirs and neural networks, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16211, https://doi.org/10.5194/egusphere-egu25-16211, 2025.
Commercial microwave links (CMLs) have recently shown great potential in urban drainage modelling due to their ability to provide rainfall-runoff dynamics. Studies investigating potential of CMLs to improve rainfall-runoff modelling typically used mechanistic hydrodynamic models driven by quantitative precipitation estimates (QPEs) derived from CML attenuation data. Naturally, some errors are introduced, primarily related to CML rainfall retrieval model, including uncertainties in wet antenna attention correction, as well as errors originated from path-averaged character of CML QPEs. These processing steps not only generate some new uncertainties but also result in a loss of valuable information contained in raw data. Besides, mechanistic models require high-quality pre-processed input rainfall data, which adds complexity to the application. We address these issues by employing raw CML attenuation data without QPE derivation using a data-driven discharge model.
A Random Forest (RF) model is employed to estimate 2-minute urban runoff using the raw CML data. The study area is a small urban catchment (1.3 km2) with a lag time of approximately 20 minutes. Datasets consist of 1-minute rainfall data from 3 rain gauges, 10-second CML data from 14 CMLs and 2-minute flow data collected during the year 2014 to 2016. A calibrated SWMM hydrological model driven by the 3 local rain gauges is used as a benchmark.
We find that: (1) Compared with rainfall data as inputs, CML attenuation data performs equally well or better in runoff simulation. The RF model with CMLs inputs achieves NSE of 0.90, PCC of 0.95, RMSE of 0.03 m³/s, and MAE of 0.02 m³/s; (2) The RF model produces comparable results to the SWMM model benchmark; (3) In addition, the RF using CML data can be used for runoff prediction exceeding horizon of the lag time. It accurately forecasts runoff up to 40-minute ahead, with NSE greater than 0.77 and PCC exceeding 0.88, whereas the RF using rain-gauge data struggles to forecast runoff more than 30-minute ahead. These results demonstrate that CML raw data can accurately yield runoff dynamics and volumes, and can be used for short-term runoff predictions.
How to cite: Song, Y., Martin, F., and Bareš, V.: The benefits of using raw CML attenuation data to predict urban runoff, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11333, https://doi.org/10.5194/egusphere-egu25-11333, 2025.
Rivers are at the heart of human activity. They provide many ecosystem services: drinking water, agriculture, transport, hydropower, bathing, freshness, etc. They are also hotspots for biodiversity. However, the water quality of these rivers is deteriorated as a result of human activity. The current work focuses on fecal contamination, which is a discriminating criterion for bathing.
In urban watersheds, fecal bacteria contamination comes from point sources related to the operation of the drainage network. During rainy weather, the combined sewer network, mixing both wastewater and stormwater, can become saturated. As a consequence, part of the flow is discharged directly into the river via combined sewer overflows (CSOs). This is the case for the city of Paris. The possible CSO overflow can be modeled by a function linking its discharge to precipitation. This relationship is currently poorly understood, with little related work, and even less for the Seine river.
To build such linking function, we rely on a dataset that includes location and hourly discharged volume of the monitored CSOs in the Seine River within Paris. Urban watersheds have been delineated within the study site. Rainfall height over these watersheds have been obtained from weather radar. We broke down the data timeseries into events. An event begins with the cause, the rain, and ends with the consequence, the overflow. To link rainfall to CSOs a directional graph based on the drainage network map, was created. It represents the wastewater transport from one watershed to another. This highlights which rainfall variables to consider regarding the CSO location. Principal component analysis (PCA) is used to assess for rain characteristics selection. An unsupervised non-linear technique (Isomap) is then used to build linking function structure.
The overflow volume in time can be modeled by a triangular shape. This shape is described by the overflow initial time, its total and maximum volume and the time of the maximum. We expect to retrieve these overflow variables by reducing the number of rainfall event characteristics to single indicators using sequentially PCA and Isomap.
Modeling and forecasting source discharges would enable better management of bathing and water supply risks, and better evaluation of mitigation infrastructures.
How to cite: Cartier, Y., Guillot-Le Goff, A., Carmigniani, R., Métivier, D., Einfalt, T., Vinçon-Leite, B., and Kennouche, P.: Modelling combined sewer overflow based on sewer network graph representation and rain radar data: application to the Seine river , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11965, https://doi.org/10.5194/egusphere-egu25-11965, 2025.
This study aims to analyze the impact of rainfall prediction uncertainty on urban flooding by focusing on the Dorimcheon basin in Seoul during the heavy rainfall event in the metropolitan area on August 8, 2022. Using AWS observed rainfall data provided by the Korea Meteorological Administration as the baseline, the study evaluated the rainfall prediction performance of two predictive rainfall datasets (LDAPS and MAPLE), estimated manhole overflow volumes, and conducted flood simulations based on these overflow volumes. The results show that the predicted rainfall by LDAPS exhibited an NSE of –0.482 and a PBIAS of 87.692, indicating a significant underestimation. Meanwhile, MAPLE demonstrated an NSE of 0.668 and a PBIAS of –4.176, suggesting an overestimation but achieving quantitatively superior performance. The flood simulation results revealed that LDAPS-based predictions matched AWS-based results with a 5.2% hit rate, whereas MAPLE achieved a hit rate of 91.9%, along with an additional 0.856 km² of flooded area. This study highlights that uncertainty in predictive rainfall datasets significantly impacts urban flood prediction accuracy, emphasizing the necessity of calibration for predictive rainfall data to improve flood prediction reliability.
Funding
This research was supported by the Disaster-Safety Platform Technology Development Program of the National Research Foundation of Korea(NRF) funded by the Ministry of Science and ICT. (No. 2022M3D7A1090338)
How to cite: Lee, J. and Lee, G.: Analysis of Urban Flooding Impacts Based on Predicted Precipitation Uncertainty, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15426, https://doi.org/10.5194/egusphere-egu25-15426, 2025.
The THALESruhr project aims to transfer scientific information in the field of sustainability into practice and to transform the Ruhr metropolitan region in Germany into a sustainable and green industrial region. This subproject of THALESruhr tries to promote resilience to climate-related extreme weather events, in particular urban flooding. The intensity and frequency of heavy rain and flash flood events have increased significantly in recent times, and this trend is expected to continue to intensify as climate change progresses (IPCC, 2023). This has highlighted the need to take action to strengthen resilience against such events.
The approach presented here combines concepts of artificial intelligence and innovative sensor technology with the objective of developing an intelligent monitoring system for traffic areas at risk of flooding in Bochum, Germany. The primary component of the system is the development of a sensor network that employs autonomous radar sensors at strategically significant locations, including bridges, tunnels, and topographical low points, to measure water levels in traffic areas in real time. The data obtained is not only employed for immediate monitoring purposes but is also utilised for the validation of heavy rain hazard maps. Based on this data, in conjunction with additional weather forecast data and historical precipitation data, an early warning system is to be established in the long term. This system will utilise artificial intelligence approaches to inform the population of impending urban flooding resulting from heavy rainfall events at an early stage. The incorporation of real-time data into urban monitoring and warning systems enables early flood alerts, allowing the population to minimize risks. Emergency services can be promptly notified of flooded streets, saving time for rescue operations. This also helps bypass dangerous areas more quickly or approach them in a targeted way. Another goal is to adapt urban planning and development to account for extreme events, such as heavy rainfall and flooding.
The project's measures are therefore not only aimed at optimising the monitoring and prevention of flooding in the short term, but also at promoting sustainable and resilient urban development in the Ruhr metropolis in the long term.
IPCC (2023). Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. Geneva, Switzerland: IPCC, pp. 35-115, doi: 10.59327/IPCC/AR6-9789291691647.
How to cite: Hotzel, A. and Mudersbach, C.: THALESruhr: An Intelligent Monitoring System for Urban Flooding in the Ruhr Metropolis, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-1519, https://doi.org/10.5194/egusphere-egu25-1519, 2025.
As climate change has a profound impact on the hydrological cycle, the frequency and intensity of extreme rainfall events have increased significantly, and the distribution of rainfall in time and space has shown significant uneven characteristics. Therefore, rainfall in the catchment area is of vital importance to hydrological research and disaster prevention. Catchment-average rainfall is widely used in hydrological processes, especially in the issue of rainfall-runoff. This study aims to use the maximum rainfall over a catchment area to estimate the average rainfall in the catchment area.
This study first gridded the catchment and then used the ordinary Kriging method and the rainfall of the rain gauges in the catchment to estimate the rainfall in each grid. The rainfall of the grids are used to estimate the average and maximum rainfall over the catchment area. In addition, this study used the concept of probability and the maximum entropy principle to deduce that there is a strong correlation between maximum and average rainfalls and used data from the Xindian River in the north Taiwan to evaluate the feasibility of the model. The results show that the relationship between the maximum rainfall and the average rainfall in the catchment is a linear relationship passing through the origin. That is, the ratio of the average rainfall to the maximum rainfall in the catchment area is a constant that is not affected by time and space. Therefore, the relationship between the maximum rainfall and the average rainfall in the catchment area developed by this study can be used to quickly estimate the real time average rainfall of a catchment.
How to cite: Chen, Y.-C. and Chou, Y.-H.: Estimation of mean rainfall over a catchment area using the relationship of maximum and mean rainfalls, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-4009, https://doi.org/10.5194/egusphere-egu25-4009, 2025.
Planning flood-resilient infrastructures requires an accurate estimation of the flood hazard, which is commonly achieved by modelling the flood responses to synthetic extreme precipitation events known as design storms. Current methods for estimating design storms fail to account for observed joint return period dependencies across different durations within events. The common block-maxima approaches, for example, follow the entire intensity-duration-frequency curve throughout the event. To overcome the limitations of the current design storm approaches, we develop a method based on vine copula and a constrained micro-canonical cascade model to generate design storms that reproduce observed return period dependencies. Taking Zurich (Switzerland) as a case study, we analysed the dependencies between precipitation volumes over duration intervals ranging from 10-min to 6-h and found strong pairwise dependencies between return periods over different duration intervals, with a maximum Kendall’s τ rank correlation coefficient of 0.69. With our new approach, we find high variability in possible duration-frequency profiles, leading to an average reduction in total storm volume compared to common block-maxima approaches. For example, events with a 50-year return period over the 10-min duration interval have a total storm volume that is on average 56% lower than that of design storms generated using the block-maxima approach. Additionally, the variability in possible duration-frequency profiles indicates that multiple design storm events should ideally be used to ensure a robust flood risk assessment.
How to cite: Cache, T., Bevacqua, E., Zscheischler, J., Müller-Thomy, H., and Peleg, N.: Generating realistic storms using a joint return period sampling of intense precipitation, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5942, https://doi.org/10.5194/egusphere-egu25-5942, 2025.
This study investigates the seasonal variations in the relationship between particulate matter (PM) concentrations and rainwater quality in an urban area of South Korea. Rainwater samples (n = 216) were collected during summer (June to August 2020) and winter (December 2020 to February 2021) at Pukyong National University, Busan, and analyzed for pH and electrical conductivity (EC) in the field, and water-soluble ions (Na+, Mg2+, K+, Ca2+, NH4+, Cl-, NO3-, SO42-) were determined using an Ion Chromatography (IC, HIC-ESP, Shimadzu, Japan) at the Integrated Analytical Center for Earth and Environmental Sciences of Pukyong National University. Atmospheric concentrations of PM10 and PM2.5 were obtained from the Automated Weather System (AWS) of the Korea Meteorological Administration (KMA). The results showed significant seasonal differences in PM concentrations and rainwater quality. In summer, daily average concentrations of PM10 and PM2.5 were relatively low (about 18 μg/m³ and 7 μg/m³) with higher rainwater EC (about 26 µS/cm) and moderate levels of cations (about 10 mg/L) and anions (8 mg/L). In contrast, winter showed increased PM10 and PM2.5 concentrations (27 μg/m³ and 18 μg/m³), accompanied by lower EC (15.6 µS/cm) but higher cation (15 mg/L) and anion (12 mg/L) concentrations. Rainfall intensity was markedly higher in summer (3.01 mm/h) than in winter (0.63 mm/h), reflecting seasonal differences in pollutant washout processes. Correlation analysis revealed stronger relationships between PM concentrations and rainwater quality in summer, particularly for pH (r = 0.75), NH4+ (r = 0.67), and K+ (r = 0.43). These findings indicate that rainfall during summer plays a critical role in transporting atmospheric pollutants to the surface, while in winter, meteorological factors such as wind and humidity have a greater influence. This study highlights the importance of considering seasonal and meteorological variations when assessing the environmental impacts of PM.
This research was supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry & Technology Institute(KEITI) funded by the Ministry of Environment(MOE)
How to cite: Lee, H., Park, H., and Yang, M.: Seasonal Variations in the Correlation between Particulate Matter and Rainwater Quality in an Urban Area of Southeast Korea, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-7572, https://doi.org/10.5194/egusphere-egu25-7572, 2025.
Urbanization significantly alters land surface characteristics, thereby influencing precipitation patterns. However, whether urbanization leads to urban wet islands or dry islands remains controversial. Here, we assess the sensitivity of precipitation to urbanization, defined as the slope of the regression between precipitation and the proportion of impervious area at each site or grid within a city, for 290 mega-cities in China, 51 in Europe, and 108 in the United States, using in situ datasets and two satellite-based products (MSWEP and GPM). Our results show that 46–70% of Chinese, 39–78% of European, and 37–71% of US cities exhibit negative sensitivity, depending precipitation product used, which highlights the uncertainties in precipitation products. We further examine how urbanization influences the frequency and intensity of heavy and light rainfall events, and find that it tends to enhance heavy rainfall and reduce light rainfall. Consequently, the reduction in light rainfall predominantly drives the negative sensitivity of annual precipitation to urbanization. Our study reveals the complexity of urban precipitation dynamics, and underscores the need for high-resolution and accurate datasets to better quantify urbanization impacts on hydrological processes.
How to cite: Cai, Y., Wu, Y., Wang, K., and Peng, S.: Impacts of Urbanization on Precipitation of Mega-cities in the Northern Hemisphere, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6380, https://doi.org/10.5194/egusphere-egu25-6380, 2025.
The use of so-called opportunistic rainfall sensors like personal weather stations (PWS) and commercial microwave links has gained much attention over the recent year, as they clearly outnumber professional rain gauges which are operated by national weather services and other. However, the data quality of such sensors is typically low and thus their information cannot be used without thorough quality control. Various quality control algorithms for PWS rainfall data have been developed and published within the EU COST Action CA 20136 "Opportunistic Precipitation Sensing Network" (OPENSENSE) in the past years and are available on OPENSENSE's GitHub (El Hachem et al. 2024).
Some of the new functions for these QC filters include (1) an improved indicator correlation filter which was originally developed by Bárdossy et al. (2019) which now provides a skill score for the accepted PWS to assess quality of the indicator correlation with neighbouring references, (2) an algorithm to correct rainfall peaks in PWS data which may be caused by connection interruptions between the rain gauge and the base station and (3) a Python implementation of the QC algorithms for identifying faulty zeroes, high influxes and station outliers originally developed in R by de Vos et al. (2019).
These new features will subsequently be implemented in the new ‘pypwsqc’ Python package (https://zenodo.org/records/14177798) which is currently under development in the OPENSENSE COST Action. In this poster we present the new features and guidelines for usage.
References:
Bárdossy, A., Seidel, J., and El Hachem, A. (2021), The use of personal weather station observations to improve precipitation estimation and interpolation. Hydrol. Earth Syst. Sci., 25, 583–601.
El Hachem, A., Seidel, J., O'Hara, T., Villalobos Herrera, R., Overeem, A., Uijlenhoet, R., Bárdossy, A., and de Vos, L.W (2024), Technical note: A guide to using three open-source quality control algorithms for rainfall data from personal weather stations, Hydrol. Earth Syst. Sci., 28, 4715–4731.
de Vos, L.W., Leijnse, H.,Overeem, A., and Uijlenhoet, R. (2019), Quality control for crowdsourced personal weather stations to enable operational rainfall monitoring. Geophysical Research Letters, 46, 8820–8829.
How to cite: Seidel, J., Petersson Wårdh, L., Illich, N., and Chwala, C.: Recent developments in the quality control of personal weather stations data, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10988, https://doi.org/10.5194/egusphere-egu25-10988, 2025.
The prediction of pluvial flooding in urban environment shows many uncertainties due to the structure of drainage network and the usual limited amount of available rain gauges. These two aspects, joint with the increasing frequency of intense rain events, highlight the need to a better comprehension of the main impact of temporal and spatial variability of the rain in the modelling process.
Within the PNRR RETURN project, a SWMM model of an urban subnetwork has been used to perform a sensitivity analysis on the influence of the shape and location of a simulated rain event for different return periods and rainfall durations. Spatially heterogeneous rainfall events are simulated as exponential distributions, and the decay constant is used to quantify the degree of spatial heterogeneity of the events. A first explorative phase aims to recognize global indicators to describe multiple response scenario, comparing the effects of rain events with the same rainfall volume and different spatial distributions. Then, the increasing number of simulations should allow to identify the best indicators that will drive to describe the network response through topological techniques.
The results show a non-linear correlation between the number of flooded nodes and the rainfall volume occurred in a specific duration. When the spatial distribution of rainfall is more heterogeneous (i.e. high decay constant) the network faces more severe criticalities. Furthermore, the response of the drainage system is non-linearly correlated to the rainfall volume intercepted by the basin, highlighting the complexity of the response and the central role of the structure of the drainage network.
This study was carried out within the RETURN Extended Partnership and received funding from the European Union Next-GenerationEU (National Recovery and Resilience Plan – NRRP, Mission 4, Component 2, Investment 1.3 – D.D. 1243 2/8/2022, PE0000005)
How to cite: Costamagna, E., Ridolfi, L., and Boano, F.: How the shape of heterogeneous precipitation affects the response of urban drainage networks in SWMM modelling, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15815, https://doi.org/10.5194/egusphere-egu25-15815, 2025.
Understanding how the space-time properties of extreme rainfall shifts due to climate change is essential for assessing risks in water-related hazards. However, future sub-daily rainfall fields, which are the main trigger of pluvial and flash floods, are not readily available for most locations and many climate change scenarios, challenging the assessment of future hazards and risks. An alternative solution to running computationally expensive convection-permitting climate models to obtain future short-duration rainfall fields is morphing recorded rainfall fields considering temperature as a driving factor. Here, we suggest using a Gamma-based spatial quantile mapping (GSQM) method with temperature as a covariate to project an archive of plausible future rainfall fields that can be used to assess future changes in extreme rainfall frequency. Combined with a stochastic storm transposition (SST) method, which can estimate rainfall frequency for arbitrary spatial scales based on gridded rainfall, future changes in regional rainfall extremes can be efficiently projected. Using Beijing as a case study, we employ 22 years of 1 km2 hourly rainfall and hourly air temperature data to demonstrate the validity of the GSQM-SST approach. First, the observed scalings governing changes in rainfall fields with temperature have been explored across various intensities of rainfall. Then, those scalings are used to morph the rainfall fields’ intensities, areas, and spatial coefficients of variation. Finally, future extremes of 2- to 100-year return levels under several warming scenarios are estimated by integrating the GSQM and SST methods.
How to cite: Zou, W., Wright, D. B., and Peleg, N.: Morphing sub-daily rainfall fields based on temperature shifts to project future changes in rainfall extremes, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8580, https://doi.org/10.5194/egusphere-egu25-8580, 2025.
Accurate flood risk assessments are critical for mitigating the impacts of pluvial flooding in densely populated urban areas. However, conventional flood modeling approaches often face limitations due to the lack of measurement information. To address this challenge, we develop, implement, and evaluate a novel framework that integrates crowdsourced data, such as citizen observations, with process-based modeling to enhance the accuracy of urban flood risk assessments. The proposed method utilizes indicator co-kriging techniques to merge citizen-sourced data with auxiliary variables, including inundation maps generated from 1D-2D urban flood models driven by high-resolution radar rainfall estimates. The framework is applied to the Oncheon River catchment in Busan, South Korea, a region highly vulnerable to pluvial flooding due to its urbanization and complex hydrological conditions. To evaluate the method, synthetic citizen observation data were generated based on inundation maps. These synthetic experiments assess the influence of the spatial distribution and quality of citizen observations on urban flood risk predictions. This study examines the integration of citizen observations into urban flood modeling workflows to address uncertainties in models and observations. In particular, we investigate the extent to which distributed citizen observations enhance prediction accuracy and analyze the effects of model bias on the reliability of flood risk assessments. The study quantitatively evaluates the effects of citizen data quality and spatial distribution on the accuracy of urban flood risk mapping. Furthermore, a sensitivity analysis is conducted for co-kriging parameters, focusing on semivariogram model selection and its influence on prediction accuracy.
How to cite: Kim, M., Choi, H., Kim, B., Lee, Y., Lee, H., Kum, J., Lee, M., and Noh, S. J.: Integrating Citizen Data in Urban Flood Risk Modeling: Insights from synthetic experiments, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8244, https://doi.org/10.5194/egusphere-egu25-8244, 2025.
In this work, two different remote sensing technologies were employed to support the assessment of pluvial flooding scenarios: the Smart Rainfall System (SRS) to estimate the rain rate and aerial photos for object detection purposes. The SRS is a recently developed monitoring technique able to estimate the rainfall intensity by processing the attenuation of microwave signals from satellite links measured by low-cost sensors. To accurately identify exposed objects to flood hazard, advanced object detection algorithms based on deep learning techniques are employed. The proposed methodology was applied to a case study located within the metropolitan area of Genoa (Italy), characterized by a flat area of about 1 km2 and recently affected by a pluvial flooding event characterized by rainfall intensities having a return period lower than three years.
The study area is equipped with a traditional tipping-bucket rain gauge station and one SRS. Two further SRSs and two rain gauges are available close to the investigated area. This configuration allows to mimic different rainfall monitoring levels from the ungauged basin to a higher spatial resolution. Pluvial flooding scenarios were modelled using the HEC-RAS 2D software and results show that significant differences in the expected flood volumes and maximum water depth and velocity are obtained using various sources of rainfall data. The obtained differences reveal that the role of opportunistic sensors located within or in the proximity of the study area largely outperforms the contribution of nearby rain gauge data when these are located even only 5 km far from the study area. This is ascribable to the point nature of measurements taken by rain gauge against the more spatially extended rainfall estimation provided by the SRSs.
The object exposed to flood hazard were detected using the You Only Look Once (YOLO) models applied to aerial images at a spatial resolution of 5 cm. The performances of various YOLO models were investigated. YOLO is pretrained for the detection of vehicles while samples of the aerial images selected outside of the study area were used to train the model for the detection of trash-bins. Results for vehicles and trash-bins are characterized by an accuracy of 95% and 69%, respectively. The assessment of the accuracy of the model based on the spatial resolution and the presence of shadows is still ongoing. Results will allow to assess the vulnerability of the urban context and will be combined with the flood hazard maps to obtain flood risk scenarios.
This study was carried out within the RETURN Extended Partnership and received funding from the European Union Next-GenerationEU (National Recovery and Resilience Plan – NRRP, Mission 4, Component 2, Investment 1.3 – D.D. 1243 2/8/2022, PE0000005).
How to cite: Cauteruccio, A., Rajabi, R., Boni, G., and Moser, G.: Pluvial flooding assessment using remote sensors and object detection models, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17803, https://doi.org/10.5194/egusphere-egu25-17803, 2025.
Extreme precipitation events can lead to urban flooding, resulting in significant casualties and economic losses, especially in highly urbanized regions. Precipitation fields exhibit pronounced spatiotemporal heterogeneity, influenced by factors such as atmospheric circulation patterns, topography, and urbanization. This complexity brings great challenges when simulating precipitation fields but recent advancements in remote sensing technology have facilitated the analysis of high-resolution precipitation fields that can be used to parameterize such models. In this study, we analyzed the spatial patterns of precipitation fields using gridded precipitation data from the CMPAS (China Multisource Precipitation Analysis System) product from 2015 to 2020, which offers a spatial resolution of 0.01° × 0.01° and a temporal resolution of one hour. Using Beijing as a case study, we analyze frequency and duration, the temporal autocorrelation, spatial correlation, and variability of the precipitation fields. Building on these analyses, we conducted stochastic simulations using a spatiotemporal Gaussian field process and deep learning methods. Specifically, we employed the AWE-GEN-2d weather generator and deep generative diffusion model to simulate precipitation fields. The results indicate that AWE-GEN-2d effectively reproduces the evolution process of storm events, while the diffusion model excels in capturing detailed spatial patterns. These findings highlight the complementary strengths of the two methods and provide valuable insights for improving precipitation modeling, flood risk management, and climate resilience planning.
How to cite: Li, T., Yin, S., and Peleg, N.: Simulations of Precipitation Fields Using Stochastic Gaussian Process and Deep Learning in Urban Areas, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17196, https://doi.org/10.5194/egusphere-egu25-17196, 2025.
Urban planners and engineers rely on historical climate data to design flood protection infrastructure capable of withstanding extreme flooding events, typically associated with a 1% annual exceedance probability (the 100-year flood). This study examines how hourly precipitation extremes are expected to evolve with rising temperatures and how these changes will influence urban flooding risks. Specifically, we address the often-overlooked impact of short-duration rainfall extremes using a new non-stationary temperature-conditional extreme precipitation scaling method and a novel regional climate convection-permitting model ensemble for +2°C and +3°C global warming scenarios for the whole of Germany. We compare this newly generated non-stationary extreme precipitation dataset with an established dataset, and then assess the implications of the future precipitation changes on flood risks in two pre-alpine communes in Germany using hydrodynamic modeling. Our results reveal that ignoring climate change can lead to significant underestimations of flood risk. Under the +3°C scenario, flood risks increase dramatically, with a 60% rise in the number of buildings affected by high flood levels (water levels of 1 meter or more). These findings suggest that current or recently implemented flood protection infrastructure may be insufficient to address the future challenges posed by climate change, underscoring the need for adaptive planning to mitigate escalating flood risks.
How to cite: Laux, P., Feldmann, D., Marra, F., Feldmann, H., Kunstmann, H., Trachte, K., and Peleg, N.: Future Precipitation Extremes and Urban Flood Risk Under +2°C and +3°C Warming: A Novel Non-Stationary Climate-Hydrodynamic Modeling Chain for Using High-Resolution Radar Data and a Convection-Permitting Climate Model Ensemble, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20737, https://doi.org/10.5194/egusphere-egu25-20737, 2025.
Urbanization is a global phenomenon characterized by the rapid expansion of urban areas, particularly into steeper terrain, affecting the hydrological cycle by increasing paved areas and surface runoff. The risk of occurrences and severity of flash flooding in both urban and surrounding rural regions may therefore also be increased. While river discharge data may reveal the large-scale effect of urbanization, detailed information on the sensitivity of small-scale changes to the hydrological cycle is generally unavailable. Spatial and temporal changes to catchment properties at multiple scales are necessary to better understand flood dynamics and risk. The objective of this study is twofold: (i) develop a method to assess the impact of the interplay between urbanization and topography on surface runoff and (ii) provide numerical case studies of urban surface runoff focussing on the effect of slope shape and steepness.
We define urban structures by the arrangement of road networks, buildings, and green space distribution at the macroscopic scale complemented by microscale features such as sidewalks and the spatial variability of infiltration properties. Urban structures have been generated by representing those spatial features as digital elevation models (DEM) on flat terrain coded in R. This DEM is then merged with landscape DEMS by overlaying both. Different urbanization scenarios are being assessed by modeling surface runoff using the 2D shallow water equations under uniform rainfall events. Because global urban expansion is showing an increasing trend to develop over mild to steep slopes, we focus our analysis on the effect of slope shape and steepness over different spatial scales on spatial dynamics of surface runoff. This approach enables us to provide quantitative insights into the sensitivity of local and global runoff dynamics. The effect of urbanization is being described by gradually increasing the fraction of urban land coverage. The effect of large-scale (urban fraction, slope steepness, and shape) and small-scale changes (urban forms arrangement, presence and absence of sidewalks, spatial variability of infiltration properties) are analyzed by integrated spatial indicators such as the distribution of velocity and water depths, and hot spots maps of high velocity and depth, which are related to large scale indicators such as peak flow and time to peak of the discharge hydrograph.
Findings of this research point to the critical role of spatial scale in urbanization together with topography features and its profound impacts on runoff dynamics and infiltration. The interplay between large-scale and micro-scale factors helps to identify how the adjustment of small-scale features affects peak flow and high-risk hot spots. Slope shape analysis has indicated that concave slopes behave differently from uniform and convex slopes, with maximum velocities occurring on midslope depending on average steepness and curvature. Implications for urbanization are being outlined.
How to cite: Shlewet, M., Kästner, K., Caviedes-Voullième, D., Frechen, N., and Hinz, C.: Assessing the interplay of topography and urbanization on surface runoff: Modelling overland flow in synthetic urban structures as affected by terrain shape and steepness, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17447, https://doi.org/10.5194/egusphere-egu25-17447, 2025.
This study provides a comprehensive analysis of the significance of rainfall simulator (RS) in hydrological research and investigates the effects of spatial variations on rainfall parameters. Rainfall simulators enable detailed examination of environmental variables under controlled laboratory conditions without relying on natural rainfall. However, the assumption of homogeneity in parameters such as rainfall intensity, uniformity, and drop size can lead to the neglect of spatial variations within the study area, thereby limiting the accuracy of the results. This limitation is particularly critical in studies focused on erosion, drainage, and infiltration, where spatial variations play a key role and may lead to misleading conclusions. In this study, performance parameters were evaluated across nine sub-regions along the simulator channel under four different rainfall intensities (40, 70, 80, and 100 mmh⁻¹). The effects of rainfall intensity on spatial uniformity and drop size were thoroughly analyzed. The findings reveal significant spatial variations in rainfall distribution. Notably, higher rainfall intensities were recorded in the middle regions resulting in higher uniformity values in these areas. Although the evaluation of uniformity coefficients for the entire area under 40 mmh⁻¹ and 70 mmh⁻¹ rainfall intensities yielded debatable results, sub-area analyses indicated that this uniformity did not hold true for the majority of the channel. Overall, a predominantly uniform rainfall distribution (>80%) was observed. Regarding drop sizes, spatial differences were identified, with a slight increase in drop size as rainfall intensity increased. These findings emphasize that treating rainfall parameters as a single fixed value for the entire study area may fail to fully capture the dynamic nature of natural rainfall. Considering spatial variations is essential for achieving more reliable and accurate results
How to cite: Kesgin, E., Demircioğlu, A. E., Gezici, K., Şengül, S., and Tan Kesgin, R. İ.: Analyzing the Influence of Spatial Variations on the Performance Metrics of a Rainfall Simulator, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-9374, https://doi.org/10.5194/egusphere-egu25-9374, 2025.
The frequency of urban flooding has been increasing due to the rising occurrence of extreme weather events driven by climate change. According to SSP scenarios, temperature and precipitation levels in the Korean Peninsula are projected to rise, leading to more frequent and intense localized heavy rainfall and hydrological disasters. This underscores the necessity for non-structural measures to mitigate urban flooding. This study empirically analyzed the significance of buildings and dual drainage in urban flood modeling and compared the impact of different modeling approaches on flood forecasting and risk assessment. Using manhole overflow data derived from the SWMM model, a 2D flood analysis model was applied. Additionally, a fully coupled 1D-2D model (H12) incorporating dual drainage concepts via grate inlets was utilized to enhance the accuracy of urban flood prediction and comparative analysis. The findings revealed that the proper incorporation of dual drainage and building structures significantly improved the accuracy of urban flood simulations, emphasizing their importance in enhancing the reliability of urban flood forecasting systems.
Funding
This research was supported by Disaster-Safety Platform Technology Development Program of the National Research Foundation of Korea(NRF) funded by the Ministry of Science and ICT. (No. 2022M3D7A1090338)
How to cite: Kwon, S. and Lee, G.: Analysis of Urban Flood Simulation Considering Dual-Drainage and Buildings, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13944, https://doi.org/10.5194/egusphere-egu25-13944, 2025.
