Maximum Flood Depth Research Articles

Application of hydrodynamic models for designing structural measures for river flood mitigation: the case of Kulfo River in southern Ethiopia

Flood is one of life-threatening events in different parts of Ethiopia. The causes of flood might vary from place to place, for instance, rates of deforestation, agricultural expansion, urbanization, wetland drainage, climate change, siltation of river bed, and several other types of land use change might be the cause for flood in different rivers banks. This study aims to assess, investigate, and design suitable river training works on lower Kulfo river reach. The study reach has been seriously affected by extreme floods due to the above reasons. For minimizing the loss due to flood, and to use the advantage of flood for different developmental activities various flood control measures should be adopted. The flood control measures which should more correctly be termed as “flood management” can be planed either through structural engineering measures or non-structural measures. Structural measures comprise retarding structures which store flood water, channel improvements which increase flood-carrying capacity of the river, embankments, and levees which keep the waterway from flood-prone area, detention basin which retards and absorbs most of the floodwater. Within the identified 6 km reach, field investigation including secondary data collection has been done to predict the flood extent using 1D hydrodynamic model, HECRAS and HEC Geo RAS. Estimated flood depth and extent helped us in fixing the dimension of different river training structures selected. The modelling result indicates a maximum channel bed flood depth of 4.3 m and flood plain flood depth 2.3 m obtained using a 100-year return flow. Analysis of soil samples indicates that the lower reach of Kulfo river is gravel and sand dominated meandering river, with estimated scour depth up to 3.41 m along the river course. Levees have been designed in conjunction with Groynes to protect the Upstream farm located at the prison and upper part of the Limat households. Frequent floods happening near both bridges shall be reduced by using Guide banks without influencing the bridges and diversion structures. Consideration is given for the ecosystem to stay in equilibrium, by providing suitable outer slopes so that plantation is possible on the top and side slope of the levees and guide banks.

Combined 1D Modelling with HEC-RAS for Delineation Floodplain Area: A Case Study of Hennops River in the Centurion Area

Flooding is among the most extreme weather events, endangering lives and causing significant property damage each year. Flood events in the Centurion area occur every year during the rainy season. This causes considerable damage to road infrastructure in both residential and commercial areas. The objective of this study is to incorporate hydraulic/hydrological (i.e., HEC-RAS/HEC-GeoRAS) models with geo-spatial techniques to predict flood extent and depth along Hennops River in Centurion area, Tshwane Metropolitan Municipality, Gauteng Province. In this study, floodplain inundation areas with different return periods were predicted in a 3.1 km distance of the Hennops River that passes through the Centurion area. Flood hazard analysis indicated that areas at close proximity to the Hennops River were submerged by a minimum and maximum flood depth of 0.4 m to more than 1.1 m for both 50 and 100 year flood recurrence interval. The study’s findings show that integrating GIS with HEC-RAS/HEC-GeoRAS techniques is a useful tool for floodplain mapping and analysis. Hence, the findings of this study are expected to be used as a foundation for the identification of causative factors of flash floods and the prediction of flash floods within the study area in future. The floodplain delineation maps developed in this study will be useful to policy-makers and the relevant authorities, as well as to local residents, in finding suitable measures for residential development along the floodplain while reducing flood risk in the study area.

Analysis of Major Rainfall Factors Affecting Inundation Based on Observed Rainfall and Random Forest

It is important to analyze urban floods according to uncertain rainfall patterns. This study was carried out to determine key statistical characteristics of rainfall affecting the occurrences of flooding by using observed rainfall, one- and two-dimensional hydraulic analysis models, and random forests. The target area selected was the Centum City area of Busan, where flooding occurs frequently with heavy rainfall. For analysis, the rainfall data from the Automatic Weather System (AWS) and radar observatory in the entire city of Busan were collected. Using the results from the one- and two-dimensional numerical models, the occurrence of flooding was determined according to the observed rainfall events. The random forest was used to classify the presence or absence of flooding according to the statistical characteristics of rainfall, and the importance evaluation function was applied to determine the impact of each statistical factor. The three main statistical factors of rainfall with the greatest impact on the occurrence of flooding were analyzed as the 1-hr maximum rainfall, the average rainfall, and the skewness. On the other hand, the three main factors affecting the maximum flood depth were the 3-hrs maximum rainfall, the 1-hr maximum rainfall, and the total rainfall. It is judged that the results of this study will be useful in suggesting quantitative urban flood forecasting standards when rainfall is examined in real-time by using various observation equipment.

Resilience

Resilience

Flood hazard delineation in an ungauged catchment by coupling hydrologic and hydraulic models with geospatial techniques-a case study of Koraiyar basin, Tiruchirappalli City, Tamil Nadu, India.

Flooding in urban basins is a major natural catastrophe that leads to many causalities of life and property. The semi-urbanized Koraiyar River basin in Tamil Nadu has important cities like Tiruchirappalli and many towns located in it. The basin unfailingly experiences a flood event in almost every decade. It is anticipated that the basin will undergo rapid unplanned urbanization in the years to come. Such fast and erratic urban developments will only increase the risk of urban floods ultimately resulting in loss of human lives and extensive damages to property and infrastructure. The effects of urbanization can be quantified in the form of land use land cover (LULC) changes. The LULC change and its impacts on urban runoff are studied for the continuous 30-year present time period of (1986-2016) to reliably predict the anticipated impact in the future time period of (2026-2036). The analysis of land cover patterns over the years shows that urbanization is more prevalent in the northern part of the basin of the chosen study area when compared with the other regions. The extreme rainfall events that occurred in the past, and the probable future LULC changes, as well as their influence on urban runoff, are studied together in the current study. In order to minimize flood damages due to these changing land use conditions, certain preventive and protective measures have to be adopted at the earliest. There are some inevitable limitations while applying traditional measures in flood modeling studies. This investigative work considers a case study on the ungauged Koraiyar floodplains. The spatial scale risk assessment is assessed by coupling geographic information systems, remote sensing, hydrologic, and hydraulic modeling, to estimate the flood hazard probabilities in the Koraiyar basin. The maximum flood flow is generated from the Hydrologic Engineering Centre-Hydrologic Modeling System (HEC-HMS), the hydrologic model adopted in the present study. The maximum flood flow is given as input to the Hydrologic Engineering Centre-River Analysis System (HEC-RAS), an effective hydraulic model that generates water depth and flood spread area in the basin. The flood depth and hazard maps are derived for 2, 5, 10, 50, and 100-year return periods. From the analysis, it is observed that the minimum flood depth is less than 1.2m to a maximum of 4.7m for the 100-year return period of past to predicted future years. The simulated results show that the maximum flood depth of 4.7m with flood hazard area of 4.32% is identified as high hazard zones from the years 1986-2036, located in the center of the basin in Tiruchirappalli city. The very high hazard flood-affected zone in the Koraiyar basin during this period is about 198.85km2. It is noticed that the very low hazard zone occupies more area in the basin for the present and future simulations of flood hazard maps. The results show that the increase in peak runoff and runoff volume is marginally varied.

A deep convolutional neural network model for rapid prediction of fluvial flood inundation

Most of the two-dimensional (2D) hydraulic/hydrodynamic models are still computationally too demanding for real-time applications. In this paper, an innovative modelling approach based on a deep convolutional neural network (CNN) method is presented for rapid prediction of fluvial flood inundation. The CNN model is trained using outputs from a 2D hydraulic model (i.e. LISFLOOD-FP) to predict water depths. The pre-trained model is then applied to simulate the January 2005 and December 2015 floods in Carlisle, UK. The CNN predictions are compared favourably with the outputs produced by LISFLOOD-FP. The performance of the CNN model is further confirmed by benchmarking against a support vector regression (SVR) method. The results show that the CNN model outperforms SVR by a large margin. The CNN model is highly accurate in capturing flooded cells as indicated by several quantitative assessment matrices. The estimated error for reproducing maximum flood depth is 0–0.2 m for the 2005 event and 0–0.5 m for the 2015 event at over 99% of the cells covering the computational domain. The proposed CNN method offers great potential for real-time flood modelling/forecasting considering its simplicity, superior performance and computational efficiency.

An Overland Flood Model for Geographical Information Systems

A variety of flood models and commercial flood simulation software are provided in the literature, with different accuracies and precisions changing from coarse to fine, depending on model structure and detailed descriptions of basin and hydrologic properties. These models generally focus on river processes, taking overland processes as inputs of 1D or 2D hydrodynamic or hydrologic river flow models. Due to the discrete structure of overland flow and unknown-dynamic boundary conditions, such classical approaches are not cable of fast and reliable spatio–temporal estimations for overland flows, and require detailed and well-organized spatial data that cannot be immediately obtained during an emergency. A spatially-distributed Geographical Information Systems (GIS) based flood model is developed in this study to simulate overland floods, using cellular automata principles. GIS raster cells are considered hydrologic homogeneous areas throughout which hydrologic properties remain constant. Hydrodynamic flow principles, conservations of mass, momentum and energy are applied at pixel level to simulate floodwaters. The proposed GIS model is capable of directly manipulating spatio–temporal pixel level data (e.g., topography, precipitation, infiltration, surface roughness etc.) for modeling of rainfall-induced overland floods; therefore, it can provide fast, temporal and spatial flood depth estimations as well as maximum flood depths and times of concentration for all pixels throughout a study area. The model is quite simple and easy to apply via easily creatable GIS input layers, and is thus very convenient for preliminary engineering applications that need quick and fast response. Its main advantage is that it does not need a predefined flood boundary and boundary conditions. This advantage is especially valuable for coastal plains where delineation of a basin is generally too difficult. Floodwaters of Cyclone Nargis/Myanmar were simulated to test the model. Sensitivity analyses were applied to evaluate the effects of the model parameters (i.e., surface roughness and infiltration rates) on simulation results. The study shows that the proposed GIS model can be readily applied for the fast and inexpensive modeling of rainfall caused floods in areas where flood boundaries and boundary conditions cannot be clearly identified.

Flood Hazard Analysis on Dulecha Woreda Located in Middle Awash River Basin

Flood is a natural disaster and causes loss of life and property damage. Awash River is causing damage of farm land and other properties at Dulecha Woreda as it is located at flood plain area. The objective of the study is to analyze flood hazard of Dulecha Woreda located in Middle Awash River Basin. The hydrological modeling in combination with hydraulic modeling is used for this study. The HEC-HMS model simulated runoff for the storm event of 14 th July 1996 and 25 th August 2008 were respectively found to be 448.9 m 3 /s and 461.8 m 3 /s during calibration and validation period. HEC-HMS generated results during calibration and validation were used as input of unsteady flow analysis in HEC-RAS model. The hydraulic model calibration is performed with observed stage data after estimating representative Manning’s “n” for the study area. Manning’s roughness coefficient ‘n’ of 0.045 and 0.055 produced the best fit against the observed data for the river main channel and flood plain respectively and used for validation of HEC-RAS model. The flood inundation depth and area exported from HEC-RAS simulation of RAS Mapper has been processed by ArcGIS for further analysis of flood inundation map. The generated inundation map is the model prediction for the peak flow of year 2008 which is obtained from rainfall depth of 130.5 mm. The flood hazard extent which is obtained for 25-Aug-08 (0.001–3.51m) gave the maximum flood depth 3.51 m. Both structural and non-structural mitigation measures are recommended for the study area. Keywords: flood inundation mapping, flood depth, HEC-HMS; HEC-RAS, ArcGIS Cite this Article: Abeya Lencha, Fitsum Tesfaye. Flood Hazard Analysis on Dulecha Woreda Located in Middle Awash River Basin. International Journal of Water Resources Engineering. 2020; 6(1): 47–66p.

Flood Inundation Mapping in an Ungauged Basin

An increase in severe precipitation events of higher intensity are expected to occur in the southeastern Mediterranean due to intensification of the hydrological cycle caused by climate change. Results of the climate change model’s precipitation data for the period 1970–2100 show a decreasing trend of daily precipitation but of higher intensity. Post-flood field investigation from a severe rainfall event in a small ungauged basin located in northwest Crete produced a validated flow hydrograph, and in combination with two high-resolution digital elevation models (DEMs), were used in the 1D/2D HEC-RAS (Hydrologic Engineering Center’s River Analysis System model), in order to determine the flooded area extent. Lateral structures were designed along the stream’s overbanks, hydraulically connecting the 1D streamflow with the 2D flow areas behind levees. Manning’s roughness coefficient and the weir coefficient were the most crucial parameters in the estimation of floodplain extent. The combined 1D/2D hydraulic model provides more detailed results than the 1D model with regards to the floodplain extent at the peak outflow, maximum flood depths, and wave velocities. Furthermore, modeling with a DEM at 2 m spatial resolution showed more precise water depth output and inundated floodplains. Scenarios of increasing peak precipitation for the same event precipitation depth were used to identify the flood extent due to an increase in daily rainfall recorded by adjacent meteorological stations. These simulation results can be useful in flood risk mapping and informing civil protective measures in flood basin management, for an effective adaptation to increased flood risk caused by a changing climate.

Hydrodynamic simulation of a dam breach of Cipanas Dam using HEC-RAS 5.0.5

Dams have many benefits for living creatures. On the other hand, dam breaks may also become disasters that cause potential losses of life and property. The Cipanas Dam has been planned for construction in Sumedang, West Java, and will have a storage capacity of up to 210 million m3. The area downstream of the Cipanas Dam contains many major infrastructures such as the Cikopo-Palimanan Toll Road and Kertajati International Airport that can be impacted by dam breach floods. This study analysed Cipanas Dam breach flow using HEC-RAS 5.0.5. The simulation assumed that a dam breach occurred through piping with 5 scenarios of breach time formation. The simulation resulted in flood hydrographs and flood inundation maps of the area downstream of the dam. The findings revealed that a breach formation time of 1 hour had the greatest flood effect. The maximum flood discharge was 38, 823.3 m3s-1 with a released water volume from the reservoir of 178, 814, 590 m3. The arrival time of the flood at the Cikopo-Palimanan Toll Road Bridge was 3 hours 45 minutes after the breach occurred, and the maximum flood depth was 15.78 m. Kertajati International Airport was not affected by the flood discharge.

A unified framework for the assessment of multiple source urban flash flood hazard: the case study of Monza, Italy

ABSTRACTThis paper aims to analyze the flood hazard in the city of Monza, which periodically suffers from overflows from both the river Lambro and the urban drainage system during wet weather conditions. A unified Urban Flash Flood Model of Monza is therefore developed to capture both sources of hazard. The model is first validated by comparing the results of simulations with historical data related to the area of Monza, then is exploited to obtain maximum flood depth and velocity maps for three different return periods. Overall, the results show the different nature of the flood-related hazard and the sewer-related hazard: the former is concentrated across the rivers whereas the latter spreads patchy over the territory. Furthermore, the former features much larger water levels and flow velocities, which result in more serious hazard to citizens.

An overview of the impact of Hurricane Harvey on chemical and process facilities in Texas

Category 4 Hurricane Harvey with maximum sustained winds of approximately 212 km/h and more than 1000 mm of rainfall reached the densely populated Houston area, Texas, resulting in unprecedented flooding and damage to the onshore industrial facilities. In order to analyze the impact of flood, rainfall, and wind on onshore chemical and process facilities during Harvey, an investigation of the damage suffered by onshore chemical and process facilities is conducted using government incident databases. Based on the locations of the damaged equipment, the corresponding maximum flood depth, accumulated rainfall, and wind speed were identified through online GIS interactive maps and the software ArcGIS. In line with the previous studies in the field of natural-technological accidents (Natechs), atmospheric storage tanks were identified as the most vulnerable type of process equipment. Flooding and unprecedented rainfall were identified as the main causes of Natechs during hurricane Harvey. Further, compared to the common failure modes observed during previous hurricanes, the roof sinking of tanks with floating roofs due to excessive accumulated rainfall was identified as a serious failure mode during hurricane Harvey. In order to reduce the risk of roof sinking in future hurricanes, the design standards of floating-roof storage tanks need to be updated to enable withstanding excessive rainfalls in the future.

Prediction of a multi-hazard chain by an integrated numerical simulation approach: the Baige landslide, Jinsha River, China

Successive major landslides during October and November 2018 in Baige village, eastern Tibet, dammed the Jinsha River on two occasions, and the subsequent dam breaches instigated a multi-hazard chain that flooded many towns downstream. Analysis of high-resolution aerial images and field investigations unveiled three potentially unstable rock mass clusters in the source area of the landslides, suggesting possible future failures with potential for river-damming and flooding. In order to evaluate and understand the disaster chain effect linked to the potentially unstable rock mass, we systematically studied the multi-hazard scenarios through an integrated numerical modelling approach. Our model begins with an evaluation of the probability of landslide failure, including runout and river damming, and then addresses the dam breach and resultant flood—hence simulating and visualising an entire disaster chain. The model parameters were calibrated using empirical data from the two Baige landslides. Then, we predict the future cascading hazards via seven scenarios according to all possible combinations of potential rock mass failure. For each scenario, the landslide runouts, dam-breaching, and flooding are numerically simulated with full consideration of uncertainties among the model input parameters. The maximum dam breach flood extent, depth, velocity, and peak arrival time are predicted at sequential sites downstream. As a first attempt to simulate the full spectrum of a landslide-induced multi-hazard chain, our study provides insights and substantiates the value provided by multi-hazard modelling. The integrated approach described here can be applied to similar landslide-induced chains of hazards in other regions.

Developing spatial prioritization criteria for integrated urban flood management based on a source-to-impact flood analysis

Integrated flood management is essential in urban planning to align flood protection and mitigation with the complex social and physical infrastructure in cities, and involves managing surface water by retaining and transferring it along its pathway across multiple infrastructure systems. However, despite many potential flood management solutions (natural and engineered), spatial prioritization to implement these solutions from a catchment perspective remains difficult. A transferable, source-to-impact flood analysis is developed to identify locations with high surface flooding (impact) and locations contributing to this flooding (source), to define spatial prioritization criteria for flood management intervention. The analysis was applied to Newcastle-upon-Tyne (UK) and combined a spatial rainfall cell dependency analysis with the hydrodynamic flood model CityCAT. Locations within the study area were then classified based on four priority criteria: contribution to (i) total flood extent; (ii) maximum flood depths; (iii) flooded green spaces and roads; and (iv) likelihood of flood exposure. The results illustrate the importance of considering the catchment holistically and identify hydrological linkages between flood source and impact areas, and the corresponding impact on (and interaction with) different infrastructure systems. Different criteria lead to different spatial prioritization information, which stresses the importance of combining criteria that address the specific needs and targets of the desired flood management strategy (e.g. Blue-Green infrastructure). The concept offers a basis for developing a systematic, high-level approach to inform catchment-scale prioritization for flood management intervention, which can be applied prior to developing flood alleviation schemes. In doing so, the approach will help identify opportunities to combine multiple water management solutions and allocate resources more efficiently.

Flood Exposure Affects Long-Term Tree Survival in Compensatory Mitigation Wetlands

Survival of planted trees is commonly used as a performance metric for compensatory mitigation wetlands. However, establishing floodplain forest through planting is difficult due to flood-induced tree mortality. We used multiyear tree census and hydrologic data from 17 compensatory mitigation projects in Illinois, USA, to relate planted and volunteer tree establishment to flood frequency, depth, and duration. Annual survival of planted trees decreased with greater annual maximum flood depth and duration. By the end of official compliance monitoring, sites with greater flood exposure had greater planted tree mortality. We resurveyed 10 sites that were at least 10 years old, and found that long-term tree survival was significantly lower in sites with greater flood exposure. Naturally colonizing trees differed in species composition from planted trees; specifically, wind dispersed species were well-represented among volunteer trees, whereas hard mast species were absent. There was no clear relationship between volunteer tree recruitment and measured flood variables. Across all sites, compliance with tree survival standards was poor, but influenced by hydrologic conditions. Current performance standards for tree survival may be unrealistic in restored wetlands that are exposed to long-duration floods. Regulators and practitioners should seek alternative methods to establish desired floodplain forest structure and function.

Relationships between seed bank composition and an invasive plant in a floodplain wetland of the Murray–Darling Basin

Seed banks are an important characteristic of wetland plant assemblages, enabling the storage of dormant propagules through wet and dry periods until the next favourable period for growth and reproduction. In this study, we use a seed bank emergence experiment to investigate whether the seed bank of a grassy floodplain wetland located in Australia’s Murray–Darling Basin has been impacted by the invasion of Juncus ingens N.A. Wakef. River regulation and altered flood regimes have encouraged the spread of this species throughout grassy wetland areas, particularly at Barmah Forest, a Ramsar Convention-listed wetland of international significance in Victoria. We particularly focus on changes in the seed bank of an ecologically important, but declining, floodplain grass, Pseudoraphis spinescens (R.Br.) Vickery, and implications for restoration. We found that sites invaded by J. ingens had a higher density of emerged J. ingens plants, a lower density of P. spinescens and a lower overall native species richness. J. ingens-dominated sites were also characterised by a significantly deeper maximum flood depth than P. spinescens-dominated sites. The overall density of P. spinescens plants emerging from the soil was very low in comparison to most other species, and largely restricted to shallow sites where the species was already present. This suggests that restoration efforts may need to focus on encouraging vegetative regrowth from existing grassy swards, rather than expecting recovery from a viable seed bank, and highlights the importance of conserving the remaining P. spinescens patches at Barmah Forest.

Analysis of Flood Inundation in Ungauged Mountainous River Basins: A Case Study of an Extreme Rain Event on 5–6 July 2017 in Northern Kyushu, Japan

The heavy rainfall event that occurred on 5–6 July 2017 in Northern Kyushu, Japan, caused extensive flooding across several mountainous river basins and resulted in fatalities and extensive damage to infrastructure along those rivers. For the periods before and during the extreme event, there are no hydrological observations for many of the flooded river basins, most of which are small and located in mountainous regions. We used the Gridded Surface Subsurface Hydrologic Analysis (GSSHA) model, a physically based model, to acquire more detailed information about the hydrological processes in the flood-affected ungauged mountain basins. We calibrated the GSSHA model using data from an adjacent gauged river basin, and then applied it to several small ungauged basins without changing the parameters of the model. We simulated the gridded flow and generated a map of the possible maximum flood depth across the basins. By comparing the extent of flood-affected areas from the model with data of the Japanese Geospatial Information Authority (GSI), we found that the maximum flood inundation areas of the river networks estimated by the GSSHA model are sometimes less than those estimated by the GSI, as the influence of landslides and erosion was not considered in the modeling. The model accuracy could be improved by taking these factors into account, although this task would be challenging. The results indicated that simulations of flood inundation in ungauged mountain river basins could contribute to disaster management during extreme rain events.

Population and building vulnerability assessment by possible worst-case tsunami scenarios in Salinas, Ecuador

Ecuador has been prone to experience earthquakes and tsunamis linked to events in the subduction zone between the Nazca and the South American plates. The main objective of this investigation was to assess the population and buildings vulnerability by a possible worst-case Tsunami scenario in the highly touristic city of Salinas—Ecuador. The vulnerability of buildings was investigated by Fragility Functions (FFs) and Vulnerability Index, while the population vulnerability was assessed by FFs. The population of permanent residents (42,860 inhabitants) and tourists (highly variable, but reaching up to 40,163 tourists/day and 4790 tourists/night) were separately studied during nine public holydays/long weekends (i.e., when the population density reaches critical levels), and during daytime/nighttime. In the selected scenario (i.e., hypocenter: 100 km southwest of Salinas, ocean depth: 2 km, and 8.0 moment magnitude), the flood area covered 43% of Salinas county and 43–85% of urban parishes. The most populated areas were exposed to inundation. According to FFs analysis, between 16,380 and 45,410 people would be affected by a tsunamigenic wave during the day and between 7386 and 10,037 during the night of Christmas and Declaration of Independence holydays, respectively. Elderly, handicapped, underage, and tourist were the most vulnerable groups. A total of 2227 structures would be affected by tsunamigenic wave (FFs), representing 40% of exposed structures to the flood area (i.e., 2.03–6.63 m maximum flood depths). A total of 3160 buildings showed Vulnerability Indexes ranging from medium to high. Results from this study would assist in the identification of hazard areas, safe zones, shelter buildings, evacuation routes/times in this densely populated touristic city.

Development and Comparison of Two Fast Surrogate Models for Urban Pluvial Flood Simulations

Detailed full hydrodynamic 1D-2D dual drainage models are a well-established approach to simulate urban pluvial floods. However, despite modelling advances and increasing computational power, this approach remains unsuitable for many real time applications. We propose and test two computationally efficient surrogate models. The first approach links a detailed 1D sewer model to a GIS-based overland flood network. For the second approach, we developed a conceptual sewer and flood model using data-driven and physically based structures, and coupled the model to pre-simulated flood maps. The city of Ghent (Belgium) is used as a test case. Both surrogate models can provide comparable results to the original model in terms of peak surface flood volumes and maximum flood extent and depth maps, with a significant reduction in computing time.

Urban flash flood forecast using support vector machine and numerical simulation

Abstract In order to provide urban flood early warning effectively, two support vector machine (SVM) models, using a numerical model as data producer, were developed to forecast the flood alert and the maximum flood depth, respectively. An application in the urban area of Jinlong River Basin, Hangzhou, China, showed the superiority of the proposed models. Statistical results based on the comparison between the results from SVM models and numerical model, proved that the SVM models could provide accurate forecasts for estimating the urban flood. For all the rainfall events tested with an identical desktop, the SVM models only took 2.1 milliseconds while the numerical model took 25 hours. Therefore, the SVM model demonstrates its potential as a valuable tool to improve emergency responses to alleviate the loss of lives and property due to urban flood.