Inundation Depth Research Articles

V-FloodNet: A video segmentation system for urban flood detection and quantification

Effective monitoring and forecasting of urban flooding are crucial for climate change adaptation and resilience around the world. We proposed a novel and automatic system for urban flood detection and quantification. Our software takes image/video data of flooding as inputs because the such data source is easy to obtain and widely available compared with conventional water level sensors or flood gauges. First, the kernel of our system is a robust water region segmentation module that detects flooded regions together with surrounding reference objects from the scene. We combine image and video segmentation technologies to make the system reliable under varying weather and illumination conditions. Second, our system uses the detected situated objects to determine the inundation depth. Field experiments demonstrate that our segmentation results are accurate and reliable; and our system can detect flooding and estimate inundation depths from images and time-lapse videos. Our code is available at https://github.com/xmlyqing00/V-FloodNet.

Inundation Depth Shape Phenotypic Variability of Phragmites australis in Liaohe Estuary Wetland, Northeast China

Inundation shifts plant growth, species diversity and ecosystem stability, but it remains unclear how inundation depth shapes the phenotypic variability of clonal plants in an estuary wetland. To reveal the response of clonal plant populations to inundation depth, we calculated the variation of Phragmites australis using shoot height, leaf length, leaf width, leaf biomass and panicle length in the Liaohe estuary wetland. Reproductive allocation was defined by the ratio of panicle length to shoot height. Linear regression showed that shoot height, leaf length, leaf width, leaf biomass and panicle length were negatively correlated with inundation depth, while the ratio of panicle length to shoot height was negatively correlated with inundation depth (p < 0.0001). Based on data regarding the statistical differences of plant phenotypic traits among P. australis, we recognized populations had generated variation differentiation. Compared with other functional traits, the coefficient of variation of leaf-related traits were at a high level. Therefore, leaf parameters would be the most suitable, and they increased the area and weight to support the action of plants during floods. Multivariate statistical analysis suggested that P. australis populations in the Liaohe estuary wetland were divided into two phenotypic clusters, consistent with geographical distance and morphological similarity. Our results provide a novel perspective on the ecological strategy of cloned plants under inundation change and offer theories for the conservation and restoration of estuarine wetland ecosystems.

Projections of coastal flooding under different RCP scenarios over the 21st century: A case study of China's coastal zone

Among sea level rise (SLR)-induced threats to coastal regions, flooding inundation is considered the most serious, especially in China's coastal zone (CCZ). Previous attempts to simulate the inundation area caused by SLR at the macro spatial scale usually employed the bathtub model based on a large number of historical tide station data. However, these efforts mainly focused on simplifying the calculation of the inundation process at a large scale, but failed to improve the accuracy mainly due to the failure to exclude isolated depressions. Here, based on regional sea level data of the Integrated Climate Data Center (ICDC), the University of Hamburg, a physical ocean model (Finite Volume Coast and Ocean Model, FVCOM) was employed to simulate the maximum possible regional SLR in offshore China from 2021 to 2100 under the representative concentration pathways (RCPs) 2.6, 4.5, and 8.5 scenarios, respectively, and the hydraulic connectivity model was applied based on multi-source data to improve the simulation accuracy of the inundation area. Through these improved methods, the maximum possible inundation characters of CCZ over the 21st century were observed as follows: (1) Under RCP2.6/4.5/8.5 scenarios, the macro spatial patterns of inundation areas are similar, with the inundation area mainly distributing in the coastal areas around the Bo Hai, the coastal areas from Jiangsu to the estuary of the Yangtze River, the Pearl River Delta, and the western coastal areas of Taiwan, and the inundation area gradually expands and spreads over time. (2) Under the three scenarios, the maximum possible inundation area shows a wavelike rising trend, and by 2100, the inundation area is expected to exceed 48.89 × 103 km2 (RCP 2.6) and reach a maximum of 53.50 × 103 km2 (RCP 8.5). (3) There are significant differences in the inundation depth and flood frequency under the three scenarios, with the inundation characters of the high-emission scenario being the most extreme. Overall, our work can provide meaningful information for proper coastal urban planning, disaster prevention, and mitigation measures selection in China. The methodology can also be used for predicting the inundation areas in other coastal countries facing the threat of SLR.

Relationship of Rainfall and Flood Return Periods through Hydrologic and Hydraulic Modeling

In order to examine the relationship between rainfall return periods and flood return periods, the design storm approach is compared to the rainfall–runoff continuous simulation and flood frequency analysis approach. The former was based on rainfall frequency analysis and event-based hydrological simulations, while the latter was based on continuous hydrological simulations and flood frequency analysis. All hydrological simulations were undertaken employing the HEC-HMS software. For the rainfall frequency analysis, the Generalized Extreme Value (GEV) probability distribution was used. For the flood frequency analysis, both the Extreme Value Type I (Gumbel) and GEV theoretical distributions were used and compared to each other. Flood hazard (inundation depth, flow velocities and flood extent) was estimated based on hydrodynamic simulations employing the HEC-RAS software. The study area was the Pineios catchment, upstream of Larissa city, Greece. The results revealed that the assumption of equivalent return periods of rainfall and discharge is not valid for the study area. For instance, a 50-year return period flood corresponds to a rainfall return period of about 110 years. Even if flow measurements are not available, continuous simulation based on re-analysis datasets and flood frequency analysis may be alternatively used.

Assessing Coastal Flood Risk in a Changing Climate for Dublin, Ireland

With increasing urban expansion and population growth, coastal urban areas will be increasingly affected by climate change impacts such as extreme storm events, sea level rise and coastal flooding. To address coastal inundation risk for impact studies, integrated approaches accounting for flood hazard modelling, exposure and vulnerability of human and environmental systems are crucial. In this study, we model the impacts of sea level rise on coastal inundation depth for County Dublin, the most extensively urbanized area in Ireland, for the current period and for 2100 under two Representative Concentration Pathways RCP 4.5 and 8.5. A risk-centred approach has been considered by linking the information on coastal flood-prone areas to the exposure of the urban environment, in terms of potential future land cover changes, and to the socioeconomic vulnerability of the population. The results suggest significant challenges for Dublin city and the surrounding coastal areas, with an increase of around 26% and 67% in the number of administrative units considered at very high risk by the end of the century under a RCP 4.5 and 8.5, respectively. This study aims to contribute to existing coastal inundation research undertaken for Ireland by (i) providing a first-level screening of flooding hazards in the study area, (ii) demonstrating how land cover changes and socioeconomic vulnerability can contribute to the level of experienced risk and (iii) informing local authorities and at-risk communities so as to support them in the development of plans for adaptation and resilience.

Hydrodynamic Modelling and Flood Risk Analysis of Urban Catchments under Multiple Scenarios: A Case Study of Dongfeng Canal District, Zhengzhou.

In recent years, urban flooding has become an increasingly serious problem, posing a serious threat to socio-economic development and personal safety. In this paper, we consider the Dongfeng Canal area in Zhengzhou City as an example and build a 1D/2D coupled urban flood model using the InfoWorks ICM. This study area uses six scenarios with rainfall return periods of 5 a, 20 a, and 50 a, corresponding to rainfall ephemeris of 1 h and 2 h to assess the flood risk. The results of the study show that (1) The flood depth, inundation duration, and extent of inundation in the study area vary with the return period and rainfall history. Generally, most of the water accumulation is concentrated in the low-lying areas adjacent to the river and near the roadbed. (2) As the rainfall recurrence period and rainfall duration increase, the proportion of overflow at the nodes becomes more pronounced and the overload from the pipe network flows mainly to the overload. (3) The high-risk areas under the different scenarios are mainly distributed on both sides of the river, and most of the low-risk areas transform into medium- and high-risk areas as the rainfall recurrence period and rainfall duration increase. This study analyses the flood risk situation under different scenarios, as well as the elements and areas that should be monitored in case of flooding, with the aim of providing a reference for flood prevention and control in the study area and formulating corresponding countermeasures. It also serves as a reference for flood risk analysis in other areas with similar situations.

A coupled multi-model framework for waterlogging projection: Towards achieving sustainable development goal 11.5

Modeling potential changes in flooding from urbanization enhances urban risk perception and supports decision-makers in preventative actions of high-risk areas. This study proposes a coupled multi-model framework for projecting the distribution and severity of urban floods and tests for Shenzhen City in China under multi-scenarios. The results show that the coupled multi-model framework has a high accuracy level at 78%. In Shenzhen, the flood-prone areas are more frequently found in the west and less frequently in the east, while the valley regions in the southeast are vulnerable to deep waterlogging. The growth rate of the inundated region slows down as the inundation depth rises after rainfall intensity exceeds that of the 20-year event. The projected results show that the flooding area and waterlogging disaster intensity (WDI) in each watershed in 2035 would be larger those in 2015. In the basins of Chi'ao Reservoir, Ejing Reservoir, and Guangming Farm in the Maozhou River midstream watersheds, the risk of the flood area expanding would be greater than the risk of the flood deepening. The findings provide a spatial reference for flood control and strategic urban drainage design and facilitate the achievement of sustainable development goal 11.5 (SDG11.5) in Shenzhen and other regions.

Impact of Storm Surge on the Yellow River Delta: Simulation and Analysis

Storm surges can lead to serious natural hazards and pose great threats to coastal areas, especially developed deltas. Assessing the risk of storm surges on coastal infrastructures is crucial for regional economic development and disaster mitigation. Combining in situ observations, remote sensing retrievals, and numerical simulation, storm surge floods in the Yellow River Delta (YRD) were calculated in different scenarios. The results showed that NE wind can cause the largest flooding area of 630 km2, although the overall storm surge risk in the delta is at lower levels under various conditions. The coastal oilfields are principally at an increasing storm surge risk level. E and NE winds would result in storm surges of 0.9–1.4 m, increasing the risk of flooding in the coastal oilfields. Nearshore seabed erosion in storm events resulted in a decrease in inundation depths and inundation areas. To prevent and control storm surge disasters, we should adapt to local conditions. Different measures should be taken to prevent the disaster of storm surges on different seashores, such as planting saltmarsh vegetation to protect seawalls, while the key point is to construct and maintain seawalls on high-risk shorelines.

Floodplain Mapping Using Hydraulic Simulation and Geographic Information System

Objectives- A flood is a strangely high stage in a river generally, the level at which the river overflows its banks inundates the adjacent area. Flood occurs due to intense rainfall, dense population, industrialization, illegal settlement along riverbanks, bank erosion, high tide, and urbanization, and it causes harmful damage to human beings and properties. It can be taken care of by adopting flood risk measures. One of the measures is the development of flood inundation maps. Hence in the current study, flood inundation maps are developed with the integration of GIS and HEC-RAS along the Bhīma River reach in Maharashtra, India. Methods: In this study, using the series of annual peak discharges, flood flows were estimated for different recurrence intervals. Inundated areas marked along the Bhīma River represent peak flows of 25 years and 100 years return periods using HEC-RAS, ArcGIS handling of spatial data, and HEC-GEORAS for the interfacing between HEC-RAS and ArcGIS. Simulation for 2, 5, 10, 25, 50, and 100-year return periods is done, and inundation maps have been developed for the 100-years reoccurrence interval. Findings:The analysis on the left and right banks has been done using water surface elevations generated from the HEC-RAS model. The significant findings revealed that the left bank representing the east side of the river is more vulnerable to water spills than the right bank representing the west side of the river. The flooded areas along the Bhīma River are 32.4 km2 and 43.82 km2 for 25-year and 100-year reoccurrence intervals, respectively. Generally, this study revealed that urban and agricultural areas downstream of river reach are more susceptible to flooding. Novelty: The developed inundation map will help the local authority in effective decision-making and disaster management in flooding situations, especially on the left bank side of Bhīma River. Obtained water surface elevations and flood inundation depths can help to decide the necessity of structural measures that can reduce the deteriorating effects of flooding Keywords: Inundation map; flood frequency; DEM; Sensitivity analysis; Return Period; HEC-RAS

Long-Term Flooding Maps Forecasting System Using Series Machine Learning and Numerical Weather Prediction System

Accurate real-time forecasts of inundation depth and area during typhoon flooding is crucial to disaster emergency response. The development of an inundation forecasting model has been recognized as essential to manage disaster risk. In the past, most researchers used multiple single-point forecasts to obtain surface flooding depth forecasts with spatial interpolation. In this study, a forecasting model (QPF-RIF) integrating a hydrodynamic model (SOBEK), support vector machine–multi-step forecast (SVM-MSF), and a self-organizing map (SOM) were proposed. The task of this model was divided into four parts: hydrodynamic simulation, point forecasting, inundation database clustering, and spatial expansion. First, the SOBEK model was used in simulating inundation hydrodynamics to construct the flooding maps database. Second, the SVM-MSF yields water level (inundation volume) forecasted with a 1 to 72 h lead time. Third, the SOM clustered the previous flooding maps database into several groups representing different flooding characteristics. Finally, a spatial expansion module produced inundation maps based on forecasting information from forecasting flood volume and flood causative factors. To demonstrate the effectiveness of the proposed forecasting model, we presented an application to the Yilan River basin in Taiwan. Our forecasting results indicated that the proposed model yields accurate flood inundation maps (less than 1 cm error) for a 1 h lead time. For long-term forecasting (46 h to 72 h ahead), the model controlled the error of the forecast results within 7 cm. In the testing events, the model forecasted an average of 83% of the flooding area in the long term. This flood inundation forecasting model is expected to be useful in providing early flood warning information for disaster emergency response.

Analysis of the influencing factors on industrial resilience to flood disasters using a semi-markov recovery model: A case study of the Heavy Rain Event of July 2018 in Japan

Industrial sectors are progressively threatened by the risk associated with flood disasters. They are also increasingly aware that building resilience is critical for their continuity, competitiveness, and survival. However, empirical evidence of industrial resilience to flood disasters is rarely provided, especially for the impact of infrastructure disruptions, resourcefulness, and other socioeconomic factors on industry resilience. Hence, this study presents a parametric semi-Markov recovery model to quantitatively estimate the business recovery rate, which is conditional on initial damage and can evaluate the recovery of industrial sectors to flood disasters. Additionally, the susceptibility of business resilience to inundation depth, deposited sediment, lifeline services (i.e., electricity, water, gas), and transportation were investigated by incorporating these covariates into the recovery function. The proposed model was applied to a case study of the Heavy Rain Event of July 2018 in Japan. The recovery data from 535 individual businesses were analyzed, revealing an estimated average production capacity loss rate of 7.16% and 6.13% in manufacturing and non-manufacturing sectors, respectively. By comparing the PCLR under different damage scenarios, results indicate that inundation depth and lifeline supply status significantly affect total losses. Analyzing the dynamic resilience of firms could help calculate the production capacity losses caused by flood disasters. It can also provide empirical evidence for decision-makers and business managers to allocate reconstruction resources in the aftermath of disasters and to establish business continuity plans to avoid potential losses in the future.

Probable Flood Inundation Depth and Extent in the Chao Phraya River Basin for Different Return Periods

Understanding the potential flood inundation depth and extent over river basins can provide a reference for understanding and mitigating the flood risk. However, the development of design hyetographs for the evaluation of flood inundation under extreme rainfall is challenging. We evaluated the flood inundation depth and extent in the Chao Phraya River Basin, one of the largest river basins in Asia, which is often vulnerable to water-related disasters. Rainfall data from 119 stations within the basin were collected for a frequency analysis. After processing the 36-year daily rainfall data, a frequency analysis of the maximum monthly rainfall was performed at each station using the Gumbel distribution. The maximum monthly rainfall for various return periods varied substantially among stations. For an inundation analysis over the entire river basin, we produced design hyetographs by integrating extreme rainfall values for each month according to the return period. These design hyetographs were included in a rainfall-runoff-inundation model to simulate the maximum inundation depth profile over the basin for different return periods. The Maximum inundation depths were 8.3, 9.0, 9.7, and 10.5 m for return periods of 50, 100, 200, and 500 years, respectively, over the Chao Phraya River Basin. Similarly, approximately 16.3%, 17.1%, 17.8%, and 18.6% of the basin area was inundated (depth > 0.5 m) over the return periods, respectively. The results of this study provide a good reference for risk analyses and evaluations of the Chao Phraya River Basin.

An Approach to Flood Hazard Mapping for the Chao Phraya River Basin Using Rainfall-Runoff-Inundation Model

Flooding is a major natural hazard that can cause significant damage to socioeconomic and ecological systems. This study presents an approach to producing the maximum flood inundation and flood duration maps over the Chao Phraya River Basin (CPRB), Thailand. An integrated numerical model and spatial analysis tool were utilized in this study. The Rainfall-Runoff-Inundation (RRI) model was first used to simulate both river discharge and inundation depth. Then, the maximum flood inundation and flood duration maps with different return periods were estimated using a Geographical Information System (GIS) tool. The results illustrate that the flood inundation areas were spread out, starting from Nakhon Sawan Province, which is located in the central part of the basin. The maximum flood inundation depth could reach up to approximately 7.71, 8.28, and 8.78 m for the flood return periods of 50, 100, and 200 years, respectively. The results also indicate that the inundation areas over the CPRB could cover approximately 21,837, 23,392, and 24,533 km2 for flood return periods of 50, 100, and 200 years, respectively. The longest flood durations for return periods of 50, 100, and 200 years were approximately 159, 177, and 198 days, respectively. The longest flood duration occurred in the vicinity of the Nakhon Sawan. This study suggests that flood inundation areas and duration mapping could provide supporting information regarding the impacts caused by varying degrees of flood hazards and can be used to enhance comprehensive disaster risk management planning.

Simulation-based assessment of inundation risk potential considering the nonstationarity of extreme flood events under climate change

The projected increase in the frequency of extreme rainfall events with climate change has serious implications for flood risk management. This study developed a novel method to evaluate future flood inundation risk in the Kakehashi River basin, Japan. First, a pseudo global warming (PGW) method and ensemble simulation techniques were combined to develop rainfall projections with different return periods, based on the Weather Research and Forecasting (WRF) model. Second, runoff analyses were performed for the target river basin, and inundation simulations, including levee breaches, were conducted at high spatial resolution (40 × 40 m). Finally, the potential inundation risk for the river basin was estimated. Rainfall simulation results indicated intensification of short-duration heavy rainfall events. The PGW ensemble simulations indicated that extreme rainfall scenarios with long return periods (>1000 years, based on current conditions) are likely to occur more frequently in the future, highlighting the possibility of flood intensification. Regardless of the length of return period, areas with infrequent flooding but large inundation depth and those with more frequent flooding but small inundation depth were classified as high risk. The proposed method provides a baseline framework for future flood risk assessments and the results highlight the importance of incorporating (currently) rare heavy rainfall events in future flood management evaluation and planning.

Gaussian process emulation of spatio-temporal outputs of a 2D inland flood model

The computational limitations of complex numerical models have led to adoption of statistical emulators across a variety of problems in science and engineering disciplines to circumvent the high computational costs associated with numerical simulations. In flood modelling, many hydraulic and hydrodynamic numerical models, especially when operating at high spatiotemporal resolutions, have prohibitively high computational costs for tasks requiring the instantaneous generation of very large numbers of simulation results. This study examines the appropriateness and robustness of Gaussian Process (GP) models to emulate the results from a hydraulic inundation model. The developed GPs produce real-time predictions based on the simulation output from LISFLOOD-FP numerical model. An efficient dimensionality reduction scheme is developed to tackle the high dimensionality of the output space and is combined with the GPs to investigate the predictive performance of the proposed emulator for estimation of the inundation depth. The developed GP-based framework is capable of robust and straightforward quantification of the uncertainty associated with the predictions, without requiring additional model evaluations and simulations. Further, this study explores the computational advantages of using a GP-based emulator over alternative methodologies such as neural networks, by undertaking a comparative analysis. For the case study data presented in this paper, the GP model was found to accurately reproduce water depths and inundation extent by classification and produce computational speedups of approximately 10,000 times compared with the original simulator, and 80 times for a neural network-based emulator.

Impact of rising sea levels on future Nankai–Tonankai earthquake tsunamis: a case study of Osaka, Japan

ABSTRACT The present study aims to clarify the impact of future sea level rise (SLR) on the anticipated Nankai–Tonankai earthquake tsunami, using Osaka, Japan as the study area. A total of 110 simulations were performed, incorporating 11 tsunami source models, five sea levels and the presence or absence of coastal defences. The simulated results indicated that while maximum nearshore tsunami heights would increase by almost the same magnitude as the rising sea level, changes in the maximum inundation depth, depth-velocity product and momentum flux over land did not reveal simple relationships to sea levels. In addition, if Osaka’s seawalls immediately failed due to the ground shaking transmitted by the earthquake, it was estimated that around 38,000 buildings and 58,000 people would be affected by the tsunami at present (for SLR +0 m), rising to around 120,000 buildings and 150,000 people in the future. In contrast, if they survived the ground shaking, the number of affected buildings and people were shown to be significantly lower, with 7,000 buildings and 18,000 people for SLR +1.0 m. It is thus essential for Osaka to maximise the resilience of the current coastal defence system and property maintain them into the near future.

An artificial neural network‐hydrodynamic coupled modeling approach to assess the impacts of floods under changing climate in the East Rapti Watershed, Nepal

AbstractRecurring floods have devastating consequences on the East Rapti Watershed (ERW), but effective mitigation/adaptation measures are lacking. This article aims at establishing a rainfall‐runoff (RR) relationship; estimating depth and extent of inundation under climate change scenarios; assessing impacts on the socio‐economy; and identifying and evaluating adaptation strategies in the ERW. Artificial Neural Network (ANN) was used to generate peak flows which were then entered into a hydraulic model to simulate inundation. Results were validated with field survey. The calibrated and validated RR and hydraulic models were fed with projected future climate (2021–2050) derived from multiple regional‐climate‐models to assess the changes in inundation. Results showed the peak discharge likely exceeds 10,500 m3/s at the ERW outlet in the extreme future flood scenario with corresponding inundation of 80 km2 and up to a depth of 11 m sweeping away over 1000 houses and 19 km2 of agricultural land in the critical areas. Constructing a 17 km long embankment in the critical areas along the right bank of the East Rapti River could reduce the flood spread by 35%, safeguarding 78% of the houses and saving 51% agricultural land compared with the scenarios without the embankment.

Flood inundation in the Lancang-Mekong River Basin: Assessing the role of summer monsoon

• Monsoons have positive impacts on inundation frequency, area and depth. • Inundation timing is either advanced or delayed when monsoons strengthen. • A change of monsoon can cause on average up to 14.2% change in inundation. • Different monsoons dominate different inundation characteristics. Although it is well known that precipitation and flood pulses in the Langcang-Mekong River Basin (LMRB) are largely impacted by monsoons, it is unclear to what extent flood inundation characteristics (i.e., inundation frequency, depth, area, and timing) in the basin respond to different monsoon types and monsoon combined effect, i.e., the Indian summer monsoon (ISM), the Western North Pacific Monsoon (WNPM), and their combined effect (ISWN). In this study, flood inundation in the LMRB during 1967–2015 was simulated by a hydrological-hydrodynamic model, from which the inundation characteristics were extracted and calculated. The monsoon impact on inundation characteristics was then quantified using the slope from linear regression model. The results show the monsoons and the ISWN overall have a positive impact on inundation frequency, depth, and area, while the inundation timing is usually advanced when the WNPM or the ISWN strengthens but delayed when the ISM strengthens. On average, a unit change in different monsoons can cause, 7.7%–14.2% change in inundation frequency, 5.3%–8.1% change in inundation depth and 4.3 days–5.8 days change in inundation timing for depth, which can also lead to 1.0%–4.3% change in inundation area and 2.8 days–3.8 days change in inundation timing for area. Also, the relative contributions of different monsoons and spatial distributions of the dominant monsoon were discussed. The results indicate different monsoons regulate different inundation characteristics, and suggest the coexistence of monsoon impacts. If the impact of the ISWN is ignored, the WNPM will play a more important role than the ISM in affecting the inundation.

Spatial-temporal flood inundation nowcasts by fusing machine learning methods and principal component analysis

• Build a novel forecast model (PCA-SOM-NARX) by statistical & machine learning tools. • 2-dimensional SOBEK simulates the full-scale flooding processes of rainfall patterns. • PCA extracts principal components to present spatial distributions of inundation. • SOM clusters inundation data to form a spatial–temporal topological feature map. • PCA-SOM-NARX makes reliable and accurate multi-step-ahead regional flood nowcasts. The frequency and severity of floods have noticeably increased worldwide in the last decades due to climate change and urbanization. This study aims to build an urban flood warning system for reducing the impact of flood disasters. A great number of storm-induced rainfall data were collected in Taipei, Taiwan, and the corresponding 2-D inundation maps were simulated for illustrating urban rainfall-flood inundation processes. We proposed a novel urban flood forecast methodology framed by machine learning and statistical techniques to mine the spatial–temporal features between rainfall patterns and inundation maps for making multi-step-ahead regional flood inundation forecasts. The proposed methodology (PCA-SOM-NARX) integrated the advantages of Principal Component Analysis (PCA), Self-Organizing Map (SOM), and Nonlinear Autoregressive with Exogenous Inputs (NARX). PCA was used to extract principal components representing the different spatial distributions of urban inundation. SOM was used to cluster high dimensional inundation datasets to form a two-dimensional topological feature map. NARX was used to establish multi-step-ahead flood forecast models for the next hour at a 10-minute scale. The results show that the PCA-SOM-NARX approach not only produced more stable and accurate multi-step-ahead forecasts on flood inundation depth but was also more indicative of the spatial distribution of inundation caused by torrential rain events, compared to the SOM-NARX approach (the benchmark). The results demonstrate the proposed methodology can adequately grasp the inundation status associated with different rainfall distributions to reliably and accurately forecast regional flood inundation depths, which can help decision makers respond to flooding earlier and mitigate flood disasters.

Flood Risk Assessment Using TELEMAC-2D Models Integrated with Multi-Index Analysis in Shenzhen River Basin, China

An urban flood simulation model based on TELEMAC-2D was constructed, and the inundation data of two measured rainstorms (7 June 2018 and 16 September 2018) were selected to validate the model. Flooding processes were simulated under 12 designed rainfall scenarios with rainfall return periods of 20, 50 and 100 years and rainfall peak coefficients of 0.2, 0.4, 0.6 and 0.8, respectively. The hazard-vulnerability (H-V) method was used for urban flood risk assessment. The selected hazard factors included inundation depth, flood velocity, elevation and slope. The vulnerability factors included land use type, population density and property distribution. The analytic hierarchy process (AHP) method was used to calculate the weight values of each indicator factor, and ArcGIS software was used for overlay calculation. The results of the analysis show that as the rainfall peak coefficient factor increases, the area of each risk zone increases to varying degrees. The larger the rainfall peak coefficient factor, the more serious the flooding. As the rainfall return period increases, the effect of the rainfall peak coefficient factor of the change in the area of the highest risk zone diminishes. The highest risk zone is the largest within Luohu District (LHD), accounting for 46.38%, 60.92% and 45.54% of the total highest risk area, respectively. As the return period increases, the area of the highest risk zone within Futian District (FTD) increases, but its proportion has a decreasing trend, and the proportion of the highest risk area within Longgang District (LGD) has an increasing trend. The risk zoning map can better reflect the risk distribution of the basin and provide a scientific basis for early warning of flood prevention and drainage in the Shenzhen River basin.