Mostafa Farrag
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mostafa Farrag has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Bruno Merz
Section Hydrology, GFZ Helmholtz Centre for Geosciences, Potsdam, 14473, Germany
5 shared publications - Heidi Kreibich
Section Hydrology, GFZ Helmholtz Centre for Geosciences
5 shared publications - Kai Schröter
Technische Universität Braunschweig
5 shared publications - Nivedita Sairam
Section Hydrology, GFZ Helmholtz Centre for Geosciences, Potsdam, 14473, Germany
5 shared publications - Sergiy Vorogushyn
GFZ Helmholtz Centre for Geosciences, Section Hydrology, Potsdam, Germany
5 shared publications Viet Dung Nguyen
GFZ Helmholtz Centre for Geosciences, Section Hydrology
3 shared publications
Latest Publications
- Open Access
- 17 August 2026
- Published by Frontiers Media SA in Frontiers in Water
- Vol. 8, 1873312
Exploring heavy-tailed behavior of flood frequency curves: an analysis of dike overtopping effects
Introduction: Flood Frequency Analysis (FFA) plays a crucial role in the design of hydraulic structures and flood risk management. Conventional estimates of flood exceedance quantiles are highly dependent on the underlying flood frequency distribution, especially its upper tail, which is particularly difficult to estimate from observed data. Methods: We use a simulation-based approach to investigate the transformation in upper tail behavior of flood peak distributions when dikes are overtopped in the large and complex river network of the River Rhine. We account for the effects of dike overtopping, floodplain storage, upstream-downstream interactions and interplay of tributaries. A coupled hydrologic-hydraulic modeling system, driven by long synthetic time series of meteorological fields from many realizations of a stochastic weather generator, is used to simulate dike overtopping and inundation. Results: Overtopping results in a consistent reduction of the shape parameters of the Generalized Extreme Value distribution at mainstream and tributary gauges. At the same time, the variability of shape parameters between model realizations increases in most cases Discussion: On average, reaches with large overtopping volumes and storage capacities show stronger reduction of shape parameters, whereas, in individual model realizations, the timing of overtopping and the rate at which overtopping volumes are generated seem to control shape parameter reduction.
- 17 August 2026
- Published by Frontiers Media SA in Frontiers in Water
- Vol. 8, 1873312
Open AccessExploring heavy-tailed behavior of flood frequency curves: an analysis of dike overtopping effects
Introduction: Flood Frequency Analysis (FFA) plays a crucial role in the design of hydraulic structures and flood risk management. Conventional estimates of flood exceedance quantiles are highly dependent on the underlying flood frequency distribution, especially its upper tail, which is particularly difficult to estimate from observed data. Methods: We use a simulation-based approach to investigate the transformation in upper tail behavior of flood peak distributions when dikes are overtopped in the large and complex river network of the River Rhine. We account for the effects of dike overtopping, floodplain storage, upstream-downstream interactions and interplay of tributaries. A coupled hydrologic-hydraulic modeling system, driven by long synthetic time series of meteorological fields from many realizations of a stochastic weather generator, is used to simulate dike overtopping and inundation. Results: Overtopping results in a consistent reduction of the shape parameters of the Generalized Extreme Value distribution at mainstream and tributary gauges. At the same time, the variability of shape parameters between model realizations increases in most cases Discussion: On average, reaches with large overtopping volumes and storage capacities show stronger reduction of shape parameters, whereas, in individual model realizations, the timing of overtopping and the rate at which overtopping volumes are generated seem to control shape parameter reduction.
- Open Access
- 15 May 2023
- Published by Copernicus GmbH
Spatially consistent flood risk assessment for Germany
Flood risk assessments are an important basis for risk management. For larger regions, these assessments are often based on small-scale modelling, which is subsequently compiled into a large-scale picture. However, this approach neglects spatial interactions, such as decreasing risk due to upstream dike breaches, and does not provide realistic risk statements for larger regions. This paper presents the ‘derived flood risk analysis’ as an alternative approach and its implementation for Germany. A model chain consisting of hydrological, hydraulic, and damage models simulates the occurrence of extreme runoff, inundation, and direct economic damages. This model chain is driven by a weather generator that provides spatially consistent fields of climate variables. The generation of very long (several thousand years) time series with daily resolution allows the estimation of extreme runoff and corresponding damages. The consideration of the spatial relations in all model components, from the weather generator to the damage model, is able to provide consistent large-scale risk statements. This avoids the significant overestimates typical of many large-scale flood risk assessments.
- 15 May 2023
- Published by Copernicus GmbH
Open AccessSpatially consistent flood risk assessment for Germany
Flood risk assessments are an important basis for risk management. For larger regions, these assessments are often based on small-scale modelling, which is subsequently compiled into a large-scale picture. However, this approach neglects spatial interactions, such as decreasing risk due to upstream dike breaches, and does not provide realistic risk statements for larger regions. This paper presents the ‘derived flood risk analysis’ as an alternative approach and its implementation for Germany. A model chain consisting of hydrological, hydraulic, and damage models simulates the occurrence of extreme runoff, inundation, and direct economic damages. This model chain is driven by a weather generator that provides spatially consistent fields of climate variables. The generation of very long (several thousand years) time series with daily resolution allows the estimation of extreme runoff and corresponding damages. The consideration of the spatial relations in all model components, from the weather generator to the damage model, is able to provide consistent large-scale risk statements. This avoids the significant overestimates typical of many large-scale flood risk assessments.
- Open Access
- 4 March 2022
- Published by Taylor & Francis in Hydrological Sciences Journal
- Vol. 67 (4) , 508-534
On the role of floodplain storage and hydrodynamic interactions in flood risk estimation
Hydrodynamic interactions, i.e. the floodplain storage effects caused by inundations upstream on flood wave propagation, inundation areas, and flood damage downstream, are important but often ignored in large-scale flood risk assessments. Although new methods considering these effects sometimes emerge, they are often limited to a small or meso scale. In this study, we investigate the role of hydrodynamic interactions and floodplain storage on flood hazard and risk in the German part of the Rhine basin. To do so, we compare a new continuous 1D routing scheme within a flood risk model chain to the piece-wise routing scheme, which largely neglects floodplain storage. The results show that floodplain storage is significant, lowers water levels and discharges, and reduces risks by over 50%. Therefore, for accurate risk assessments, a system approach must be adopted, and floodplain storage and hydrodynamic interactions must carefully be considered.
- 4 March 2022
- Published by Taylor & Francis in Hydrological Sciences Journal
- Vol. 67 (4) , 508-534
Open AccessOn the role of floodplain storage and hydrodynamic interactions in flood risk estimation
Hydrodynamic interactions, i.e. the floodplain storage effects caused by inundations upstream on flood wave propagation, inundation areas, and flood damage downstream, are important but often ignored in large-scale flood risk assessments. Although new methods considering these effects sometimes emerge, they are often limited to a small or meso scale. In this study, we investigate the role of hydrodynamic interactions and floodplain storage on flood hazard and risk in the German part of the Rhine basin. To do so, we compare a new continuous 1D routing scheme within a flood risk model chain to the piece-wise routing scheme, which largely neglects floodplain storage. The results show that floodplain storage is significant, lowers water levels and discharges, and reduces risks by over 50%. Therefore, for accurate risk assessments, a system approach must be adopted, and floodplain storage and hydrodynamic interactions must carefully be considered.
- 1 December 2021
- Published by Elsevier in Water Security
- Vol. 14
Large-scale flood risk assessment and management: Prospects of a systems approach
- 1 December 2021
- Published by Elsevier in Water Security
- Vol. 14
Large-scale flood risk assessment and management: Prospects of a systems approach
- Open Access
- 5 October 2021
- Published by American Geophysical Union (AGU) in Earth's Future
- Vol. 9 (10)
Process‐Based Flood Risk Assessment for Germany
Large‐scale flood risk assessments are crucial for decision making, especially with respect to new flood defense schemes, adaptation planning and estimating insurance premiums. We apply the process‐based Regional Flood Model (RFM) to simulate a 5000‐year flood event catalog for all major catchments in Germany and derive risk curves based on the losses per economic sector. The RFM uses a continuous process simulation including a multisite, multivariate weather generator, a hydrological model considering heterogeneous catchment processes, a coupled 1D–2D hydrodynamic model considering dike overtopping and hinterland storage, spatially explicit sector‐wise exposure data and empirical multi‐variable loss models calibrated for Germany. For all components, uncertainties in the data and models are estimated. We estimate the median Expected Annual Damage (EAD) and Value at Risk at 99.5% confidence for Germany to be €0.529 bn and €8.865 bn, respectively. The commercial sector dominates by making about 60% of the total risk, followed by the residential sector. The agriculture sector gets affected by small return period floods and only contributes to less than 3% to the total risk. The overall EAD is comparable to other large‐scale estimates. However, the estimation of losses for specific return periods is substantially improved. The spatial consistency of the risk estimates avoids the large overestimation of losses for rare events that is common in other large‐scale assessments with homogeneous return periods. Thus, the process‐based, spatially consistent flood risk assessment by RFM is an important step forward and will serve as a benchmark for future German‐wide flood risk assessments.
- 5 October 2021
- Published by American Geophysical Union (AGU) in Earth's Future
- Vol. 9 (10)
Open AccessProcess‐Based Flood Risk Assessment for Germany
Large‐scale flood risk assessments are crucial for decision making, especially with respect to new flood defense schemes, adaptation planning and estimating insurance premiums. We apply the process‐based Regional Flood Model (RFM) to simulate a 5000‐year flood event catalog for all major catchments in Germany and derive risk curves based on the losses per economic sector. The RFM uses a continuous process simulation including a multisite, multivariate weather generator, a hydrological model considering heterogeneous catchment processes, a coupled 1D–2D hydrodynamic model considering dike overtopping and hinterland storage, spatially explicit sector‐wise exposure data and empirical multi‐variable loss models calibrated for Germany. For all components, uncertainties in the data and models are estimated. We estimate the median Expected Annual Damage (EAD) and Value at Risk at 99.5% confidence for Germany to be €0.529 bn and €8.865 bn, respectively. The commercial sector dominates by making about 60% of the total risk, followed by the residential sector. The agriculture sector gets affected by small return period floods and only contributes to less than 3% to the total risk. The overall EAD is comparable to other large‐scale estimates. However, the estimation of losses for specific return periods is substantially improved. The spatial consistency of the risk estimates avoids the large overestimation of losses for rare events that is common in other large‐scale assessments with homogeneous return periods. Thus, the process‐based, spatially consistent flood risk assessment by RFM is an important step forward and will serve as a benchmark for future German‐wide flood risk assessments.
- Open Access
- 26 April 2021
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 9 (5) , 467
Spatio-Temporal Hydrological Model Structure and Parametrization Analysis
Many grid-based spatial hydrological models suffer from the complexity of setting up a coherent spatial structure to calibrate such a complex, highly parameterized system. There are essential aspects of model-building to be taken into account: spatial resolution, the routing equation limitations, and calibration of spatial parameters, and their influence on modeling results, all are decisions that are often made without adequate analysis. In this research, an experimental analysis of grid discretization level, an analysis of processes integration, and the routing concepts are analyzed. The HBV-96 model is set up for each cell, and later on, cells are integrated into an interlinked modeling system (Hapi). The Jiboa River Basin in El Salvador is used as a case study. The first concept tested is the model structure temporal responses, which are highly linked to the runoff dynamics. By changing the runoff generation model description, we explore the responses to events. Two routing models are considered: Muskingum, which routes the runoff from each cell following the river network, and Maxbas, which routes the runoff directly to the outlet. The second concept is the spatial representation, where the model is built and tested for different spatial resolutions (500 m, 1 km, 2 km, and 4 km). The results show that the spatial sensitivity of the resolution is highly linked to the routing method, and it was found that routing sensitivity influenced the model performance more than the spatial discretization, and allowing for coarser discretization makes the model simpler and computationally faster. Slight performance improvement is gained by using different parameters’ values for each cell. It was found that the 2 km cell size corresponds to the least model error values. The proposed hydrological modeling codes have been published as open-source.
- 26 April 2021
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 9 (5) , 467
Open AccessSpatio-Temporal Hydrological Model Structure and Parametrization Analysis
Many grid-based spatial hydrological models suffer from the complexity of setting up a coherent spatial structure to calibrate such a complex, highly parameterized system. There are essential aspects of model-building to be taken into account: spatial resolution, the routing equation limitations, and calibration of spatial parameters, and their influence on modeling results, all are decisions that are often made without adequate analysis. In this research, an experimental analysis of grid discretization level, an analysis of processes integration, and the routing concepts are analyzed. The HBV-96 model is set up for each cell, and later on, cells are integrated into an interlinked modeling system (Hapi). The Jiboa River Basin in El Salvador is used as a case study. The first concept tested is the model structure temporal responses, which are highly linked to the runoff dynamics. By changing the runoff generation model description, we explore the responses to events. Two routing models are considered: Muskingum, which routes the runoff from each cell following the river network, and Maxbas, which routes the runoff directly to the outlet. The second concept is the spatial representation, where the model is built and tested for different spatial resolutions (500 m, 1 km, 2 km, and 4 km). The results show that the spatial sensitivity of the resolution is highly linked to the routing method, and it was found that routing sensitivity influenced the model performance more than the spatial discretization, and allowing for coarser discretization makes the model simpler and computationally faster. Slight performance improvement is gained by using different parameters’ values for each cell. It was found that the 2 km cell size corresponds to the least model error values. The proposed hydrological modeling codes have been published as open-source.
- Open Access
- 4 March 2021
- Published by Copernicus GmbH
Process-based flood risk assessment for Germany
Floods affect people worldwide and account for more than USD 100 billion losses on average every year. Hazard, Exposure and Vulnerability are the three components that influence flood risk. Flood Risk Management (FRM) decisions especially, with respect to new flood defense schemes and resilience initiatives are generally taken based on the assessment of impacts for hazard scenarios. Current large-scale studies are comprehensive in terms of sectors covered in impact assessment. However, these studies often deploy generalized data and methods on the model components resulting in coarse risk estimates with low spatial resolution.In this study, we use process-based models with 100m resolution on the national scale within a systems approach to develop and simulate a 5000 year flood event catalogue for Germany. The events are then analyzed per economic sector, including residential, commercial and agriculture sectors. The risk chain includes continuous simulation of high-resolution hazard maps, obtained from coupled hydrology and hydraulic models; NUTS3-level exposure asset values further disaggregated to ATKIS land-use data and calibrated object-level vulnerability models that provide high-resolution quantification of economic damage. Spatial dependence of flood events is addressed by the continuous simulation approach. For each model component in the risk assessment (hazard, exposure and vulnerability), uncertainty in data and methods are integrated into the risk predictions. Based on these simulations, we present a sector-wise flood risk assessment for Germany along with the reliability of the risk estimates. This process-based, systemic flood risk assessment is valuable for policy making, adaptation planning and estimating insurance premiums.
- 4 March 2021
- Published by Copernicus GmbH
Open AccessProcess-based flood risk assessment for Germany
Floods affect people worldwide and account for more than USD 100 billion losses on average every year. Hazard, Exposure and Vulnerability are the three components that influence flood risk. Flood Risk Management (FRM) decisions especially, with respect to new flood defense schemes and resilience initiatives are generally taken based on the assessment of impacts for hazard scenarios. Current large-scale studies are comprehensive in terms of sectors covered in impact assessment. However, these studies often deploy generalized data and methods on the model components resulting in coarse risk estimates with low spatial resolution.In this study, we use process-based models with 100m resolution on the national scale within a systems approach to develop and simulate a 5000 year flood event catalogue for Germany. The events are then analyzed per economic sector, including residential, commercial and agriculture sectors. The risk chain includes continuous simulation of high-resolution hazard maps, obtained from coupled hydrology and hydraulic models; NUTS3-level exposure asset values further disaggregated to ATKIS land-use data and calibrated object-level vulnerability models that provide high-resolution quantification of economic damage. Spatial dependence of flood events is addressed by the continuous simulation approach. For each model component in the risk assessment (hazard, exposure and vulnerability), uncertainty in data and methods are integrated into the risk predictions. Based on these simulations, we present a sector-wise flood risk assessment for Germany along with the reliability of the risk estimates. This process-based, systemic flood risk assessment is valuable for policy making, adaptation planning and estimating insurance premiums.
- Open Access
- 3 March 2021
- Published by Copernicus GmbH
RIMurban – A generalized GPU-based model for urban pluvial flood risk modelling and forecasting
Urban flash floods caused by heavy convective precipitation pose an increasing threat to communes world-wide due to the increasing intensity and frequency of convective precipitation caused by a warming atmosphere. Thus, flood risk management plans adapted to the current flood risk but also capable of managing future risks are of high importance. These plans necessarily need model based pluvial flood risk simulations. In an urban environment these simulations have to have a high spatial and temporal resolution in order to site-specific management solutions. Moreover, the effect of the sewer systems needs to be included to achieve realistic inundation simulations, but also to assess the effectiveness of the sewer system and its fitness to future changes in the pluvial hazard. The setup of these models, however, typically requires a large amount of input data, a high degree of modelling expertise, a long time for setting up the model setup and to finally run the simulations. Therefor most communes cannot perform this task. In order to provide model-based pluvial urban flood hazard and finally risk assessments for a large number of communes, the model system RIMurban was developed. The core of the system consists of a simplified raster-based 2D hydraulic model simulating the urban surface inundation in high spatial resolution. The model is implemented on GPUs for massive parallelization. The specific urban hydrology is considered by a capacity-based simulation of the sewer system and infiltration on non-sealed surfaces, and flow routing around buildings. The model thus considers the specific urban hydrological features, but with simplified approaches. Due to these simplifications the model setup can be performed with comparatively low data requirements, which can be covered with open data in most cases. The core data required are a high-resolution DEM, a layer of showing the buildings, and a land use map.The spatially distributed rainfall input can be derived local precipitation records, or from an analysis of weather radar records of heavy precipitation events. A catalogue of heavy rain storms all over Germany is derived based on radar observations of the past 19 years. This catalogue serves as input for pluvial risk simulations for individual communes in Germany, as well as a catalogue of possible extreme events for the current climate. Future changes in these extreme events will be estimated based on regional climate simulations of a ΔT (1.5°C, 2°C) warmer world.RIMurban simulates the urban inundation caused by these events, as well as the stress on the sewer system. Based on the inundation maps the damage to residential buildings will be estimated and further developed to a pluvial urban flood risk assessment. Because of the comparatively simple model structure and low data demand, the model setup can be easily automatized and transferred to most small to medium sized communes in Europe and even beyond, if the damage estimation is modified. RIMurban is thus seen as a generally appölicable screening tool for urban pluvial flood risk and a starting point for adapted risk management plans.
- 3 March 2021
- Published by Copernicus GmbH
Open AccessRIMurban – A generalized GPU-based model for urban pluvial flood risk modelling and forecasting
Urban flash floods caused by heavy convective precipitation pose an increasing threat to communes world-wide due to the increasing intensity and frequency of convective precipitation caused by a warming atmosphere. Thus, flood risk management plans adapted to the current flood risk but also capable of managing future risks are of high importance. These plans necessarily need model based pluvial flood risk simulations. In an urban environment these simulations have to have a high spatial and temporal resolution in order to site-specific management solutions. Moreover, the effect of the sewer systems needs to be included to achieve realistic inundation simulations, but also to assess the effectiveness of the sewer system and its fitness to future changes in the pluvial hazard. The setup of these models, however, typically requires a large amount of input data, a high degree of modelling expertise, a long time for setting up the model setup and to finally run the simulations. Therefor most communes cannot perform this task. In order to provide model-based pluvial urban flood hazard and finally risk assessments for a large number of communes, the model system RIMurban was developed. The core of the system consists of a simplified raster-based 2D hydraulic model simulating the urban surface inundation in high spatial resolution. The model is implemented on GPUs for massive parallelization. The specific urban hydrology is considered by a capacity-based simulation of the sewer system and infiltration on non-sealed surfaces, and flow routing around buildings. The model thus considers the specific urban hydrological features, but with simplified approaches. Due to these simplifications the model setup can be performed with comparatively low data requirements, which can be covered with open data in most cases. The core data required are a high-resolution DEM, a layer of showing the buildings, and a land use map.The spatially distributed rainfall input can be derived local precipitation records, or from an analysis of weather radar records of heavy precipitation events. A catalogue of heavy rain storms all over Germany is derived based on radar observations of the past 19 years. This catalogue serves as input for pluvial risk simulations for individual communes in Germany, as well as a catalogue of possible extreme events for the current climate. Future changes in these extreme events will be estimated based on regional climate simulations of a ΔT (1.5°C, 2°C) warmer world.RIMurban simulates the urban inundation caused by these events, as well as the stress on the sewer system. Based on the inundation maps the damage to residential buildings will be estimated and further developed to a pluvial urban flood risk assessment. Because of the comparatively simple model structure and low data demand, the model setup can be easily automatized and transferred to most small to medium sized communes in Europe and even beyond, if the damage estimation is modified. RIMurban is thus seen as a generally appölicable screening tool for urban pluvial flood risk and a starting point for adapted risk management plans.