Flash Flood Events Research Articles

Interpretable machine learning for predicting urban flash flood hotspots using intertwined land and built-environment features

Pluvial flash floods are fast-moving hazards and causes significant disruptions in urban areas. With the increase in heavy precipitations, the ability to proactively identify flash floods hotspots in cities is critical for flood nowcasting and predictive monitoring of risks. While rainfall runoff models and hydrologic models are useful models for flash flood prediction, these models are computationally expensive and effort intensive to be used for flood nowcasting. To address this challenge, this study presents interpretable machine learning models for predicting urban flash flood hotspots based on intertwined land and built environment features. The task of predicting flash flood hotspots is formulated as a binary classification problem, and three recent flash flood events in U.S. cities are selected for data collection and model validation. Various features related to land and built environment characteristics are constructed using diverse datasets, and the occurrences of flash floods are captured using crowdsource data from the events. Using these features and datasets, the flash flood hotspots of cities are predicted with two ensemble models based on decision trees. The results demonstrate that the models can achieve good accuracy (0.8) in identifying flooded/non-flooded locations. Especially, the models can achieve high true positive rate (0.83–0.89) and low missing rate, demonstrating the methods' practicability for accurately predicting flooded hotspots. The model interpretation results indicate that land features related to hydrological and topological features have greater impacts on flash flood risk, than built environment features. Further analysis reveals that the feature importance, model performance, and model transferability performance vary among cities and localized specifications of the models are needed for accurate prediction of flash flood for a particular city. The data-driven machine learning models presented in this study provide a useful tool for predicting flash flood hotspots based on the intertwined features of land and the built environment in cities to enable nowcasting and proactive monitoring of flash flood hotspots for emergency response and also inform integrated urban design and development towards flash flood risk reduction.

Improving Flash Flood Hydrodynamic Simulations by Integrating Leaf Litter and Interception Processes in Steep-Sloped Natural Watersheds

More frequent high-intensity, short-duration rainfall events increase the risk of flash floods on steeply sloped watersheds. Where measured data are unavailable, numerical models emerge as valuable tools for predicting flash floods. Recent applications of various hydrological and hydrodynamic models to predict overland flow have highlighted the need for improved representations of the complex flow processes that are inherent in flash floods. This study aimed to identify an optimal modeling approach for characterizing leaf litter losses during flash floods. At a gauged watershed in the Hidegvíz Valley in Hungary, a physical-based model was calibrated using two distinct rainfall–runoff events. Two modeling methodologies were implemented, integrating canopy interception and leaf litter storage, to understand their contributions during flash flood events. The results from the model’s calibration demonstrated this approach’s effectiveness in determining the impact of leaf litter on steep-sloped watersheds. Soil parameters can estimate the behavior of leaf litter during flash flood events. In this study, hydraulic conductivity and initial water content emerged as critical factors for effective parametrization. The findings underscore the potential of a hydrodynamic model to explore the relationship between leaf litter and flash flood events, providing a framework for future studies in watershed management and risk-mitigation strategies.

Validation of WRF Rainfall Model during the November 4th, 2021 Flash Flood Event in Batu City – Indonesia

Flash floods in Batu City, Indonesia on November 4th, 2021, caused damage to property, agricultural land, settlements, death of livestock, and loss of human life. One of the important factors triggering this flash flood is heavy rainfall. This study aims to analyse the spatial and temporal dynamics of rainfall as a trigger for the event. Rainfall modelling was conducted using the WRF model with 2 microphysical schemes and 3 cumulus schemes. The data used is GFS data on October 30th, 2021, with a resolution of 0.25 x0.25 degrees which is used to predict rain events until November 4, 2021. Contingency table verification, RMSE value verification, and verification using tolerance limits are used to verify the results of modelling data based on BMKG observation data. The results of this study show that this model can predict rain and non-rain events very well. However, this model is not good enough to predict the thickness of rain until November 4, 2021. The best scheme in modelling rainfall is scheme 5 (Lin and Betts Miller Janjic). Scheme 5 has the smallest RMSE value of 77, the data fit is appropriate and the medium error rate in scheme 5 is 84%. Scheme 5 with GFS data input started from 02 November 2021 can record rain events on 04 November 2021 with extreme rain categories. This extreme rain is influenced by convective clouds. The temperature of the cloud tops changes significantly from 13.00 WIB to 14.00 local time, namely from −4 degrees, and decreases to −76 degrees Celsius.

Determining sediment deposition dynamics influenced by check dams in a semi‐arid mountainous watershed

AbstractSemi‐arid environments are characterized by infrequent large magnitude rainfalls that produce flash flood events with high sediment concentration. Control structures such as check dams are widely used in this environment for mitigation. However, their impact on the overall sediment balance of watersheds, particularly those severely affected by anthropogenic activity, is sparsely documented. This study used topographic measurements, sediment analysis, and fallout isotope techniques to assess the effectiveness and service life of 18 rock and masonry check dams that were constructed in the 1930s for controlling sediment fluxes in an 11 ha mountainous watershed in southern Arizona. All of the dams are currently filled with sediment resulting in reduction of local channel gradients by 35 to 71% to between 0.04 and 0.28 depending on location within the reach. Sedimentation occurred over multiple decades at a relatively slow average rate of 0.59 t ha−1 y−1 indicating low instantaneous retention efficiency. The smaller headwater dams were filled soon after construction; however, their share of the overall storage capacity was minor. Although evaluation of 137Cs was an effective method for dating sediment, the 210Pb dating method was not satisfactory because of sediment sorting effects and complex deposition patterns. The abundance of similar control structures in the region points to the opportunity to better understand the process impacts of check dams over multiple decades to inform planning and design of their use in future mitigation projects.

Assessing streamflow forecast accuracy for flash flood events in Puerto Rico

Study regionPuerto Rico. Study focusA comprehensive understanding of the hydrological cycle is crucial, where models play a key role in monitoring and predicting flood events, enabling better analysis of event behavior. This study evaluated the National Water Model version 2.1, currently operated and monitored by the National Oceanic and Atmospheric Administration. The study focused on assessing the model's ability to produce reliable streamflow forecasts in Puerto Rico, which was recently added to the model in 2021. The evaluation was conducted with five significant rainfall events, including four flash floods and one hurricane in 2022. The forecasted streamflow was determined using 83 river gauge measured data across Puerto Rico during each event. New hydrological insights for the regionBased on the assessment conducted for the NWM in Puerto Rico, the model has the ability to produce streamflow forecasts and obtain correct trends, only in some cases. The model's accuracy and precision are limited to the forcing supplied during its operation. The model performs optimally in conditions where rainfall forcing is similar to observed, and deviations lead to an unreliable forecast. This study established the primary concern for making a reliable forecast is the forcing supplied. The currently operating model produces overestimated or underestimated forecasts with minimal close estimates, but its outputs can still always provide a general idea of streamflow activity across the island.

Development of a hazard risk map for assessing pedestrian risk in urban flash floods: A case study in Cúcuta, Colombia

AbstractThe rapid growth of impervious areas in urban basins worldwide has increased the number of impermeable surfaces in cities, leading to severe flooding and significant economic losses for civilians. This trend highlights the urgent need for methodologies that assess flood hazards and specifically address the direct impact on pedestrians, which is often overlooked in traditional flood hazard analyses. This study aims to evaluate a methodology for assessing the risk to pedestrians from hydrodynamic forces during urban floods, with a specific focus on Cúcuta, Colombia. The methodology couples research outcomes from other studies on the impact of floodwaters on individuals of different ages and sizes with 1D/2D hydrological modeling. Advanced computational algorithms for image recognition were used to measure water levels at 5‐s intervals on November 6, 2020, using drones for digital elevation model data collection. In Cúcuta, where flood risk is high and drainage infrastructure is limited, the PCSWMM (Computer‐based Urban Stormwater Management Model) was calibrated and validated to simulate extreme flood events. The model incorporated urban infrastructure details and geomorphological parameters of Cúcuta's urban basin. Four return periods (5, 10, 50, 100), with extreme rainfall of 3 h, were used to estimate the variability of the risk map. The output of the model was analyzed, and an integrated and time‐varying comparison of the results was done. Results show that the regions of high‐water depth and high velocity could vary significantly along the duration of the different extreme events. Also, from 5 to 100 years return period, the percentage of area at risk increased from 9.6% to 16.6%. The pedestrian sensitivity appears much higher than the increase in velocities or water depth individually. This study identified medium to high‐risk locations, which are dynamic in time. We can conclude dynamics are spatiotemporal, and the added information layer of pedestrians brings vulnerability information that is also dynamic. Areas of immediate concern in Cúcuta can enhance pedestrian safety during flash flood events. The spatiotemporal variation of patterns requires further studies to map trajectories and sequences that machine learning models could capture.

Morfometría de una cuenca de los Andes Centrales, San Juan, Argentina: evaluación de los factores condicionantes del terreno a la ocurrencia de aluviones

The present work aims to determine flash flood occurrence prone areas, in the Jáchal river basin, San Juan Province, Argentina. The Jáchal river basin covers ~29,000 km2 and has a perimeter of ~2,400 km, while the eponymous river extends for nearly 570 km. It is a 7th order basin according to Strahler’s classification, subdivided into four 6th order sub-basins. Various morphometric basin characteristics are analysed and some basic and derived parameters are calculated. These calculations allow predicting the basin behaviour due to heavy rainfall, which translates into large volumes of water with peak flows capable of generating significant flash floods. The basin has a torrential character, so in case of intense rainfall the terrain conditions would contribute to increasing the flow of the main rivers. An inventory of alluvial deposits was carried out and their relationship with the conditioning variables of the terrain was evaluated. At the confluence zones, the river tributaries show more intense overflows than the main collector. This promotes the generation of flash floods. The areas most likely to be affected by flash flood events correspond to low-humidity, east-facing slightly rough terrains, with slopes of up to 15° as well as low-lying flood plain sectors, and at the confluence of tributaries with the main rivers. Populated centres and provincial and national routes are prone to be affected by alluvial events in the study area.

Rain-on-snow responses to warmer Pyrenees: a sensitivity analysis using a physically based snow hydrological model

Abstract. Climate warming is changing the magnitude, timing, and spatial patterns of mountain snowpacks. A warmer atmosphere may also induce precipitation phase shifts, resulting in a decreased snowfall fraction (Sf). The combination of Sf and snowpack directly influences the frequency and intensity of rain-on-snow (ROS) events, a common cause of flash-flood events in snow-dominated regions. In this work, we investigate ROS patterns and their sensitivity to temperature and precipitation changes in the Pyrenees by modeling ROS through a physically based snow model. This model is forced with reanalysis climate data for elevations of 1500, 1800, and 2400 m perturbed using a range of temperature and precipitation values consistent with 21st century climate projections. ROS patterns are characterized by their frequency, rainfall quantity, and snow ablation. The highest ROS frequency for the historical climate period (1980–2019) is found in the 2400 m zones of the southwest Pyrenees (17 d yr−1). The maximum ROS rainfall amount is detected in 1800 m areas of the southeast (45 mm d−1, autumn), whereas the highest ROS ablation is found in the 2400 m zones of the northwest (−10 cm d−1, summer). When air temperature increases from 1 to 4 ∘C compared to the historical climate period, ROS rainfall amount and frequency increase at a constant rate during winter and early spring for all elevation zones. For the rest of the seasons, non-linear responses of ROS frequency and ablation to warming are found. Overall, ROS frequency decreases in the shoulders of the season across eastern low-elevation zones due to snow cover depletion. However, ROS increases in cold, high-elevation zones where long-lasting snow cover exists until late spring. Similarly, warming induces greater ROS ablation (+10 % ∘C−1) during the coldest months of the season, 2400 m elevations, and northern sectors, where the deepest snow depths are found. In contrast, small differences in ROS ablation are found for warm and marginal snowpacks. These results highlight the different ROS responses to warming across the mountain range, suggest similar ROS sensitivities in near-mid-latitude zones, and will help anticipate future ROS impacts in hydrological, environmental, and socioeconomic mountain systems.

Comparative gene co-expression networks show enrichment of brassinosteroid and vitamin B processes in a seagrass under simulated ocean warming and extreme climatic events.

Ocean warming combined with extreme climatic events, such as marine heatwaves and flash flooding events, threaten seagrasses globally. How seagrasses cope with these challenges is uncertain, particularly for range-edge populations of species such as Posidonia australis in Shark Bay, Western Australia. Analyzing gene expression while manipulating multiple stressors provides insight into the genetic response and resilience of seagrasses to climate change. We conducted a gene expression study on a polyploid clone of P. australis during an 18-week mesocosm experiment to assess the responses to single and combined future climate change-associated stressors. Plants were exposed to (1) future ocean warming temperature (baseline +1.5°C) followed by a simulated marine heat wave (baseline +5.5°C), (2) light deprivation simulating observed marine heatwave driven turbidity (95% shade) at baseline temperatures, or (3) both stressors simultaneously. Basal leaf meristems were sampled for gene expression analysis using RNA-seq at four time points during the experiment. Weighted gene co-expression network analysis, GO term enrichment, and KEGG pathway enrichment analyses were used to identify stress responses. Shaded plants showed specific gene enrichment for shade avoidance (programmed cell death) after three weeks of stress, and before any heated tanks showed a specific heat response. Shaded plants were positively correlated with programmed cell death and stress-related processes at the end of the experiment. Once ocean warming temperatures (+1.5°C) were in effect, gene enrichment for heat stress (e.g., ROS scavenging and polyamine metabolism) was present. Vitamin B processes, RNA polymerase II processes. and light-related meristematic phase changes were expressed with the addition of simulated MHW. Heated plants showed meristematic growth signatures as well as trehalose and salicylic acid metabolism. Brassinosteroid-related processes were significantly enriched in all stressor treatments at all time points, except for the isolated heat-stressed plants three weeks after stressor initiation. Gene expression responses to the interaction between heat waves and turbidity-induced light reduction support the observed geographical scale mortality in seagrasses observed for P. australis in Shark Bay, suggesting that even this giant polyploid clone will be negatively impacted by more extreme climate change projections.

County-Level Flash Flood Warning Framework Coupled with Disaster-Causing Mechanism

Climate change has intensified the risk of extreme precipitation, while mountainous areas are constrained by complex disaster mechanisms and difficulties in data acquisition, making it challenging for existing critical rainfall threshold accuracy to meet practical needs. Therefore, this study focuses on Yunnan Province as the research area. Based on historical flash flood events, and combining remote sensing data and measured data, 12 causative factors are selected from four aspects: terrain and landforms, land use, meteorology and hydrology, and population and economy. A combined qualitative and quantitative method is employed to analyze the relationship between flash floods and triggering factors, and to calibrate the parameters of the RTI (Rainfall Threshold Index) model. Meanwhile, machine learning is introduced to quantify the contribution of different causative factors and identify key causative factors of flash floods. Based on this, a parameter η coupling the causative mechanism is proposed to optimize the RTI method, and develop a framework for calculating county-level critical rainfall thresholds. The results show that: (1) Extreme rainfall, elevation, slope, and other factors are direct triggers of flash floods, and the high-risk areas for flash floods are mainly concentrated in the northeast and southeast of Yunnan Province. (2) The intraday rainfall has the highest correlation with the accumulated rainfall of the previous ten days; the critical cumulative rainfall ranges from 50 mm to 400 mm. (3) The county-level critical rainfall threshold for Yunnan Province is relatively accurate. These findings will provide theoretical references for improving flash flood early warning methods.

A Tale of Two Hazards: Studying Broadcast Meteorologist Communication of Simultaneous Tornado and Flash Flood (TORFF) Events

Broadcast meteorologists are the primary source of weather information for the public, and thus are key to messaging the multiple weather hazards that can occur during simultaneous tornado and flash flood, or TORFF, events. Due in part to the challenge and cost needed to study broadcast coverage, there has been limited study into how broadcasters present these hazards to their viewers during TORFF events. To begin to address this knowledge gap, we developed the Coding Algorithm for Storm coverage Transcripts, or CAST. Bot, a simple algorithm that can efficiently and inexpensively compare the mentions of tornado and flash flood hazards made by meteorologists during on-air coverage. For this study, we used CAST.Bot to quickly analyze 39 segments of coverage from eight TORFF events. Findings suggest that broadcasters generally favor mentions of tornadoes more than flash flooding during TORFF events with many tornado warnings, with more balanced coverage identified during events with similar numbers of tornado and flash flood warnings. Additional study of two cases, 1) the El Reno/Oklahoma City, Oklahoma, tornado and flash flood on 31 May 2013, and 2) Hurricane Harvey in Houston, Texas, on 26 August 2017, suggests that TORFF event coverage on television is subject to differences across stations and the way that the tornado and flash flood hazards in a TORFF unfold. Future work should seek to better understand how changes in the focus of messaging for TORFF events can impact viewers decisions and identify how context can influence TORFF message content. Options for use of the CAST.Bot algorithm to aid broadcasters during multi-hazard event coverage are also discussed.

Flow prediction in Kabul River: An artificial intelligence based technique

This study employs random forest regressor to forecast future flow in Kabul River. Utilizing CMIP6 projected climate data for the SSP585 scenario from the IPSL-CM6A-LR climate model. The random forest regressor demonstrate efficacy in predicting flow, achieving an R2 of 0.77. The study highlights the importance of modern artificial intelligence-based techniques for precise flow and flood predictions and suggests an increase in flash flood events in Kabul River in response to a warming climate.

Representation of a Post-Fire Flash-Flood Event Combining Meteorological Simulations, Remote Sensing, and Hydraulic Modeling

Wildfires are an escalating global threat, jeopardizing ecosystems and human activities. Among the repercussions in the ecosystem services of burnt areas, there are altered hydrological processes, which increase the risks of flash floods. There is limited research addressing this issue in a comprehensive way, considering pre- and post-fire conditions to accurately represent flood events. To address this gap, we present a novel approach combining multiple methods and tools for an accurate representation of post-fire floods. The 2019 post-fire flood in Kineta, Central Greece is used as a study example to present our framework. We simulated the meteorological conditions that caused this flood using the atmospheric model WRF-ARW. The burn extent and severity and the flood extent were assessed through remote sensing techniques. The 2D HEC-RAS hydraulic–hydrodynamic model was then applied to represent the flood event, using the rain-on-grid technique. The findings underscore the influence of wildfires on flooding dynamics, highlighting the need for proactive measures to address the increasing risks. The integrated multidisciplinary approach used offers an improved understanding on post-fire flood responses, and also establishes a robust framework, transferable to other similar cases, contributing thus to enhanced flood protection actions in the face of escalating fire-related disasters.

A Laboratory Study of the Role of Nature-Based Solutions in Improving Flash Flooding Resilience in Hilly Terrains

Nature-based solutions (NBSs) always provide optimal opportunities for researchers and policymakers to develop sustainable and long-term solutions for mitigating the impacts of flooding. Computing the hydrological process in hilly areas is complex compared to plain areas. This study used a laboratory-scaled hillslope model to study rainfall-runoff responses considering the natural hillslope conditions prevailing in hill torrents creating flash floods. The objective of this study was to estimate the impact of nature-based solutions on time-to-peak for flash flooding events on hilly terrains under different scenarios. Many factors decide the peak of runoff generation due to rainfall, like land use conditions, e.g., soil porosity, vegetation cover, rainfall intensity, and terrain slope. To reduce these complexities, the model was designed with thermopore sheets made of impermeable material. A hillslope model using NBS was designed to evaluate flood hydrograph attenuation to minimize the peak discharge (Qp) and increase time-to-peak (Tp) under varying rainfall, land cover, and drainage channel slope conditions. A rainfall simulator was used to analyze the formation of hydrographs for different conditions, e.g., from barren to vegetation under three different slopes (S0, S1, S2) and three rainfall intensities (P1, P2, P3). Vegetation conditions used were no vegetation, rigid vegetation, flexible vegetation, and the combination of both rigid and flexible vegetation. The purpose of using all these conditions was to determine their mitigation effects on flash flooding. This experimental analysis shows that the most suitable case to attenuate a flood hydrograph was the mixed vegetation condition, which can reduce the peak discharge by 27% to 39% under different channel slopes. The mixed vegetation condition showed an increase of 49% in time-to-peak (Tp) compared to the no vegetation condition. Additionally, under P1 rainfall and a bed slope of 0°, it reduced the peak discharge by up to 35% in the simulated flood and effectively minimized its potentially destructive impacts.

Introducing Flashiness‐Intensity‐Duration‐Frequency (F‐IDF): A New Metric to Quantify Flash Flood Intensity

AbstractFlash flooding is a damaging weather event, yet it remains challenging to quantify its severity. We propose a development—the Flashiness‐Intensity‐Duration‐Frequency (F‐IDF) curve—to quantify flash flood intensity based on the frequency and duration of the event. As a proof‐of‐concept, we mapped Contiguous US (CONUS)‐wide F‐IDF values at 3,722 stream gage locations and explored their relations with basin attributes. It is found that (a) The return periods of flash flood events are highly associated with the return peroids of rainfall events; (b) Climatological precipitation amounts exhibit the most positive correlation with flashiness while a basin's drainage area is the most negatively correlated; (c) Correlation of flashiness with basin attributes decreases with increasing F‐IDF return periods and shorter event durations. Both aspects are attributable to the rainfall signal overwhelming the underlying basin attributes as the intensities become more extreme. This new metric has implications for hydrology and emergency responders.

FF-BERT: A BERT-based ensemble for automated classification of web-based text on flash flood events

The web is a rich information repository that can be mined to uncover additional data about past flash flood (FF) events, currently missing from existing structured databases. However, this information originates from multiple sources (news articles, government records, and weather records among others) and may cover several topics. Furthermore, these topics may be disproportionately covered on the web. The large size and heterogenous nature of web information render manual review difficult. To address this challenge, we have developed a multi-label text classification model, FF-BERT. FF-BERT is designed to classify FF-related web paragraphs into one or more of seven categories: (1) Damage and Economic Impact (DI), (2) Fatalities, Injuries, and Rescue (FIR), (3) Hydrometeorology (HM), (4) Warning and Emergency (WE), (5) Response and Recovery (RR), (6) Public Health (PH), and (7) Mitigation (MG). To develop FF-BERT, we labeled 21,180 paragraphs from FF-related webpages and performed experiments with multiple model architectures based on the widely used language model Bidirectional Encoder Representation from Transformers (BERT). Our final model outperforms the baseline by 11.83%, as measured by the micro-F1 score. In addition, FF-BERT significantly improves the prediction of minority labels (RR-32.1%, PH-260.4%, and MG-138.6%). We demonstrate using real world examples that FF-BERT can be used to uncover new information about flash flood events. This information can be used to enhance existing databases, such as NOAA’s Storm Events Database.

Understanding Flash Floods in Hilly Tropical Watersheds: A Trigger Factor Analysis

Flash floods are catastrophic phenomena known for their rapid and unpredictable occurrence. They frequently display precursory indicators, which, when comprehensively studied, serve as vital clues for effectively anticipating and preparing for these events, thus reducing their adverse effects. This study aimed to investigate the key trigger factors contributing to a flash flood event in the upper Ciliwung Watershed in January 2021. Data collection involved various parameters, including morphology, precipitation patterns, land use, hydrological characteristics, and soil conditions collected from the affected area. These data were subsequently analyzed using geospatial methods and enriched with information from local community members. The analysis found multiple contributing factors to flash floods, including high soil moisture content due to prolonged precipitation, geological characteristics, and steep topography. The local community recognized landslides, creating a natural dam about a month before the flash flood. Unfortunately, inadequate measures were taken to address the flood due to limited understanding and resources. The heavy rainfall before the flash flood put immense pressure on the natural dam, causing it to fail and carry debris downstream. The steep banks of the surrounding river worsened the situation. The study’s findings can provide valuable insights for local authorities, leading to better resource allocation.

Spatial and temporal variability of saturated areas during rainfall-runoff events

Abstract Spatially distributed hydrological model Mike SHE was used as a diagnostic tool to provide information on possible overland flow source areas in the mountain catchment of Jalovecký Creek (area 22.2 km2, elevation range 820–2178 m a.s.l.) during different rainfall-runoff events. Selected events represented a sequence of several smaller, consecutive events, a flash flood event and two large events caused by frontal precipitation. Simulation of hourly runoff was better for runoff events caused by heavy rainfalls of longer duration than for the flash flood or consecutive smaller runoff events. Higher soil moisture was simulated near the streamflow network and larger possibly saturated areas were located mainly in the upper parts of mountain valleys. The most pronounced increase in the areal extent of possibly saturated areas (from 6.5% to 68.6% of the catchment area) was simulated for the event with high peak discharge divided by a short rainfall interruption. Rainfall depth exceeding 100 mm caused a large increase in the potentially saturated areas that covered subsequently half of the catchment area or more. A maximum integral connectivity scale representing the average distance over which individual pixels were connected varied for the selected events between 45 and 6327 m.

Soil Erosion and Landslide Susceptibility Mapping in Western Attica, Greece: A Rock Engineering System Approach

Two of the eight main soil degradation processes with which soils worldwide are confronted are soil erosion and landslides. Specifically, landslides are a major threat in particular areas across Europe, often leading to serious impacts on population, property, and infrastructure. Regarding the abovementioned processes, the case study of the fatal Mandra flash flood (November 2017) in the Attica Region (Greece), which caused 24 deaths, and much infrastructure and building damage, is presented with the intention of assessing the relationship between soil erosion and landslide incidents. Investigations were executed from 2018 to 2022, and their outcomes were taken into consideration by the Technical Authority of the Attica Region. Soil erosion lines were delineated in a GIS and were validated using a previously generated regional Web-GIS landslide susceptibility map. The study presents soil erosion types from the Mandra fatal flash flood event and correlates them with already existing landslide susceptibility analyses for the Attica Region. The produced susceptibility map is a cartographic product on a regional scale (1:100,000) generated via a semiquantitative heuristic methodology named the Rock Engineering System (RES). The way in which both soil erodibility and landslide susceptibility maps were generated and validated could be the basis for proposing modeling approaches that can respond to new developments in European landslide policies.

Assessing the ability of a new seamless short-range ensemble rainfall product to anticipate flash floods in the French Mediterranean area

Abstract. Flash floods have dramatic economic and social consequences, and efficient adaptation policies are required to reduce their impacts, especially in the context of global change. Developing more efficient flash flood forecasting systems can largely contribute to these adaptation requirements. The aim of this study was to assess the ability of a new seamless short-range ensemble quantitative precipitation forecast (QPF) product, called PIAF-EPS (Prévision Immédiate Agrégée Fusionnée ensemble prediction system) and recently developed by Météo-France, to predict flash floods when used as input to an operational hydrological forecasting chain. For this purpose, eight flash flood events that occurred in the French Mediterranean region between 2019 and 2021 were reanalysed, using a hydrological-modelling chain similar to the one implemented in the French Vigicrues Flash operational flash flood monitoring system. The hydrological forecasts obtained from PIAF-EPS were compared to the forecasts obtained with different deterministic QPFs from which PIAF-EPS is directly derived. The verification method applied in this work uses scores calculated on contingency tables and combines the forecasts issued on each 1 km2 pixel of the territory. This offers a detailed view of the forecast performances, covering the whole river network and including the small ungauged rivers. The results confirm the added value of the ensemble PIAF-EPS approach for flash flood forecasting, in comparison to the different deterministic scenarios considered.