Flood Forecasting Model Research Articles

Prediction models for urban flood evolution for satellite remote sensing

Accurate and timely flood forecasts are critical for protecting people and infrastructure in a changing climate. Satellite remote sensing provides the necessary wide area coverage and period revisits to measure episodic heavy precipitation and resultant urban floods. We propose two methods to assimilate satellite-observed precipitation into hydrologic models in real time to update flood forecasts, bypassing two previous barriers in this technology: infrequent satellite overpasses and long model run times. Constellations of small satellites overcomes the first barrier by providing frequent flights over an area of interest; however, these constellations require coordination and planning to capture precipitation data where it is most needed to inform flood forecasts. The primary purpose of this paper is to address the second barrier – that is high computational costs that make it infeasible to run flood forecast models on-board these satellites so that they can re-orient to measure where most needed. We develop a simple regression-based approach and a machine learning framework (Long Short-Term Memory (LSTM) models) to provide reliable flood forecasts using satellite-observed precipitation at a fraction of the computational cost of physics-based hydrologic models. We apply these approaches to a test case for the Atlanta metropolitan region using the Weather Research and Forecasting model hydrologic modeling system (WRF-Hydro) to simulate flooding across the model domain for several precipitation events. We compare the accuracy of our proposed approaches to the WRF-Hydro model using different spatial extents and temporal frequencies of precipitation observations to examine different plausible satellite constellation scenarios. The LSTM approach trades performance accuracy and adaptability for computational efficiency, which can be important in a time and resource constrained scenario. The LSTM model reduces total error up to 38% from an initial flood forecast. Additionally, this approach correctly classified flooding to within one flood magnitude category in ∼90% of cases. These new forecasting algorithms can be used onboard constellations of small satellites to observe ongoing flood events, update short term predictions, and schedule observations to maximize useful measurements and thereby improve flood warning systems for protecting residents and properties.

Flood modelling of Air Bengkulu Watershed, Indonesia, using SUH and HEC-HMS

This study aims to develop a flood forecasting model for Air Bengkulu based on Synthetic Unit Hydrograph and HEC HMS. Rainfall data were obtained from the River Region VII Regional Office of Bengkulu Province, from 2000 to 2021. The Synthetic Unit Hydrograph (SUH) method used to estimate the discharge was a synthetic unit hydrograph, namely GAMA-1. Flood simulation was carried out using HEC-HMS software. The peak discharge value produced by HEC-HMS is 690 ms-1 while the peak discharge value from SUH GAMA-1 is 319.23 m s-1. The peak discharge value of the Air Bengkulu flood produced by both SUH GAMA-1 and HEC-HMS has been able to represent the value of the flood discharge of the Air Bengkulu Watershed. This study concludes that SUH GAMA-1 and HEC HMS have been able to model the flood event on April 29, 2019, that occurred in the Air Bengkulu watershed. Validation of the model needs to be studied continuously so that it can produce a model that can represent the actual flood events in Air Bengkulu.

Development of real-time flood forecast model for vamsadhara river through hydrological approach

Development of real-time flood forecast model for vamsadhara river through hydrological approach

Daily time series of river water levels derived from a seasonal linear model using multisource satellite products under uncertainty

Satellite altimetry helps to monitor continental water levels with global coverage and near-real time accessibility. However, a main drawback of satellite altimetry is its low acquisition frequency that hinders the use of altimeters in flood forecasting or reservoir operation processes. This study provides a new method to estimate daily river heights from multisource hydro-climate satellite products without using ground-based data. First, the approach develops a simple statistical model, a seasonal linear model, using satellite altimetry, land surface temperature (LST), satellite precipitation and satellite soil moisture products. Second, a Monte Carlo simulation is used to understand the model accuracy by attributing the model output errors to uncertainties from the multiple satellite products. For the first aim, the study shows that using high-frequency satellite data to improve the temporal resolution of the satellite radar altimetry is better than an approach merging multiple satellite altimetry data. With respect to the second objective, it is found that higher variances are found in seasonal transition periods where there are significant changes in satellite observations. The overall error of the model is attributed to both uncertainties of satellite input data and the interactions among their uncertainties. The results demonstrate the capability of using multisource satellite data to predict daily time series of river heights in data-poor areas. The results also highlight the importance of quantifying uncertainty to increase the reliability of flood forecasting models using satellite products.

Multi-Step Sequence Flood Forecasting Based on MSBP Model

Establishing a model predicting river flow can effectively reduce huge losses caused by floods. This paper proposes a multi-step time series forecasting model based on multiple input and multiple output strategies, and this model is applied to the flood forecasting process of a river basin in Shanxi, which effectively improves the engineering application value of the flood forecasting model based on deep learning. The experimental results show that after considering the seasonal characteristics of the river channel and screening the influencing factors, a simple neural network model can accurately predict the peak value, the peak time and flood trends. On this basis, we proposed the MSBP (Multi-step Back Propagation) model, which can accurately predict the flow trend of the river basin 20 h in advance, and the NSE (Nash Efficiency) is 0.89. The MSBP model can improve the reliability of flood forecasting and increase the internal interpretability of the model, which is of great significance for effectively improving the effect of flood forecasting.

Evaluation of artificial intelligence models for flood and drought forecasting in arid and tropical regions

With the advancement of computer science, Artificial Intelligence (AI) is being incorporated into many fields to increase prediction performance. Disaster management is one of the main fields embracing the techniques of AI. It is essential to forecast the occurrence of disasters in advance to take the necessary mitigation steps and reduce damage to life and property. Therefore, many types of research are conducted to predict such events due to climate change in advance using hydrological, mathematical, and AI-based approaches. This paper presents a comparison of three major accepted AI-based approaches in flood and drought forecasting. In this study, fluvial floods are measured by the runoff change in rivers whereas meteorological droughts are measured using the Standard Precipitation Index (SPI). The performance of the Convolutional Neural Network (CNN), Long-Short Term Memory network (LSTM), and Wavelet decomposition functions combined with the Adaptive Neuro-Fuzzy Inference System (WANFIS) are compared in flood and drought forecasting, with five statistical performance criteria and accepted flood and drought indicators used for comparison, extending to two climatic regions: arid and tropical. The results suggest that the CNN performs best in flood forecasting with WANFIS for meteorological drought forecasting, regardless of the climate of the region under study. Besides, the results demonstrate the increased accuracy of the CNN in applications with multiple features in the input.

Assimilation of probabilistic flood maps from SAR data into a coupled hydrologic–hydraulic forecasting model: a proof of concept

Abstract. Coupled hydrologic and hydraulic models represent powerful tools for simulating streamflow and water levels along the riverbed and in the floodplain. However, input data, model parameters, initial conditions, and model structure represent sources of uncertainty that affect the reliability and accuracy of flood forecasts. Assimilation of satellite-based synthetic aperture radar (SAR) observations into a flood forecasting model is generally used to reduce such uncertainties. In this context, we have evaluated how sequential assimilation of flood extent derived from SAR data can help improve flood forecasts. In particular, we carried out twin experiments based on a synthetically generated dataset with controlled uncertainty. To this end, two assimilation methods are explored and compared: the sequential importance sampling method (standard method) and its enhanced method where a tempering coefficient is used to inflate the posterior probability (adapted method) and reduce degeneracy. The experimental results show that the assimilation of SAR probabilistic flood maps significantly improves the predictions of streamflow and water elevation, thereby confirming the effectiveness of the data assimilation framework. In addition, the assimilation method significantly reduces the spatially averaged root mean square error of water levels with respect to the case without assimilation. The critical success index of predicted flood extent maps is significantly increased by the assimilation. While the standard method proves to be more accurate in estimating the water levels and streamflow at the assimilation time step, the adapted method enables a more persistent improvement of the forecasts. However, although the use of a tempering coefficient reduces the degeneracy problem, the accuracy of model simulation is lower than that of the standard method at the assimilation time step.

Research on flood forecast of dashimen reservoir in xinjiang based on melting snow and runoff yield under excess infiltration

In China, the distribution of water resources is uneven in time and space. The effective allocation of water resources can realize the rational utilization of water resources. Reservoirs are an important part of the basin system, and the rational operation of reservoirs can realize the optimal allocation of water resources. Flood forecast is of great significance for reservoir operation. In order to realize the optimal operation of Dashimen reservoir in Xinjiang, aiming at the flood forecast in the dry area affected by melting snow, this paper expanded and improved the traditional hydrological forecast model, built the flood forecast model in the dry area with the coupling snowmelt based on the northern Shaanxi model and Elman neural network, and using the typical rainfall and runoff data in this area The results show that the relative error of the model is less than 10% and the accuracy is high, which can be used in the actual forecast. This model realizes the flood forecast in the drought area affected by snowmelt, and provides a new idea for the flood forecast of reservoir in the drought area affected by snowmelt, which has a certain guiding significance for the flood forecast in the drought area.

Improving Urban Flood Mapping by Merging Synthetic Aperture Radar-Derived Flood Footprints with Flood Hazard Maps

Remotely sensed flood extents obtained in near real-time can be used for emergency flood incident management and as observations for assimilation into flood forecasting models. High-resolution synthetic aperture radar (SAR) sensors have the potential to detect flood extents in urban areas through clouds during both day- and night-time. This paper considers a method for detecting flooding in urban areas by merging near real-time SAR flood extents with model-derived flood hazard maps. This allows a two-way symbiosis, whereby currently available SAR urban flood extent improves future model flood predictions, while flood hazard maps obtained after the SAR overpasses improve the SAR estimate of urban flood extents. The method estimates urban flooding using SAR backscatter only in rural areas adjacent to urban ones. It was compared to an existing method using SAR returns in both rural and urban areas. The method using SAR solely in rural areas gave an average flood detection accuracy of 94% and a false positive rate of 9% in the urban areas and was more accurate than the existing method.

Flash flood modeling in the data-poor basin: A case study in Matina River Basin

Forecasting flooding hazards is a very effective non-engineering measure for flood control. Presently, the commonly used forecasting method in simulating flash flood events is through a watershed hydrological model. Many Asia-Pacific countries like the Philippines are prone to frequent hydrometeorological hazards such as tropical cyclones, resulting in frequent heavy rainfall events. However, most rivers in the many basins lack water flow observation data, which makes it challenging to use lumped and data-driven models for flash flood forecasting. With the continuous progress of remote sensing (RS) and geographic information system (GIS) techniques, the physically-based distributed hydrological model (PBDHMs) has rapidly advanced. PBDHMs can directly determine the model parameters according to the underlying surface characteristics from remotely-sensed data, which makes it possible for flood forecasting in areas with little to virtually no data. In this study, the Matina River basin in Davao City was selected as a case study in simulating a small data-poor basin in the region. The Liuxihe model was used to formulate a forecasting scheme and simulated the past flash flood events. The results show that there is a good correspondence between the past heavy rainfall events and their corresponding simulated river discharges. The results conform to the hydrological regularities, which can be used for flood forecasting and can serve as a baseline for the development of a flood forecasting system in the rivers of Davao City and elsewhere.

Flood modeling of Sungai Pinang Watershed under the impact of urbanization

Urbanization has been a worldwide development trend, which regulates river courses, impervious surfaces and drainage systems. Urbanization causes hydrological effects, including increased runoff volumes, peak discharges and flow concentrations. This manuscript selects the Malaysian Sungai Pinang watershed as a case study to illustrate these land use, channel and flooding changes of Asian coastal cities. The Landsat satellite remote sensing images were first used to estimate the land use/land cover changes of the Sungai Pinang watershed by using SVM algorithm, and the results shows the urbanization was very rapid in the past decades, with the urbanization rate reached 46.41% in 2018 based on the build area rate. River channel characteristics also changed significantly, from natural river to concrete channel. Some flood resilience measures for coastal cities experiencing urbanization are also proposed for development and flood mitigation. Moreover, a flood forecasting model of the Sungai Pinang watershed is established herein. The simulation results of the Liuxihe model constructed in this study conforms to hydrological regularities and can provide a technical reference for flood control and disaster reduction. However, it is necessary to pay attention to the uncertainty of the forecast results.

Resampling and ensemble techniques for improving ANN-based high-flow forecast accuracy

Abstract. Data-driven flow-forecasting models, such as artificial neural networks (ANNs), are increasingly featured in research for their potential use in operational riverine flood warning systems. However, the distributions of observed flow data are imbalanced, resulting in poor prediction accuracy on high flows in terms of both amplitude and timing error. Resampling and ensemble techniques have been shown to improve model performance on imbalanced datasets. However, the efficacy of these methods (individually or combined) has not been explicitly evaluated for improving high-flow forecasts. In this research, we systematically evaluate and compare three resampling methods, random undersampling (RUS), random oversampling (ROS), and the synthetic minority oversampling technique for regression (SMOTER), and four ensemble techniques, randomised weights and biases, bagging, adaptive boosting (AdaBoost), and least-squares boosting (LSBoost), on their ability to improve high stage prediction accuracy using ANNs. These methods are implemented both independently and in combined hybrid techniques, where the resampling methods are embedded within the ensemble methods. This systematic approach for embedding resampling methods is a novel contribution. This research presents the first analysis of the effects of combining these methods on high stage prediction accuracy. Data from two Canadian watersheds (the Bow River in Alberta and the Don River in Ontario), representing distinct hydrological systems, are used as the basis for the comparison of the methods. The models are evaluated on overall performance and on typical and high stage subsets. The results of this research indicate that resampling produces marginal improvements to high stage prediction accuracy, whereas ensemble methods produce more substantial improvements, with or without resampling. Many of the techniques used produced an asymmetric trade-off between typical and high stage performance; reduction of high stage error resulted in disproportionately larger error on a typical stage. The methods proposed in this study highlight the diversity-in-learning concept and help support future studies on adapting ensemble algorithms for resampling. This research contains many of the first instances of such methods for flow forecasting and, moreover, their efficacy in addressing the imbalance problem and heteroscedasticity, which are commonly observed in high-flow and flood-forecasting models.

Verification of forecasts for extreme rainfall, tropical cyclones, flood and storm surge over Myanmar and the Philippines

Within the humanitarian sector, there is a pressing need to scale up anticipatory action as climate change-related disasters increase. This article evaluates forecasts relating to extreme weather events – extreme rainfall, tropical cyclones, river flooding and storm surge – in Myanmar and the Philippines to assess the feasibility of using such forecasts to develop early warning systems and responses. To make best use of limited extant data, a variety of methods (reliability diagrams, hit rates, false alarm ratios, correlations) are employed. We review the skill of the European Centre for Medium-Range Weather Forecasts (ECMWF) tropical cyclone forecasts and find that whilst errors in cyclone position are relatively small, forecasting intensity is more difficult. When a tropical cyclone has formed, the probabilities provided in the ECMWF track forecast are highly reliable and only slightly overconfident. A tropical cyclone activity product is relatively reliable for forecasts less than a week ahead for North Indian Ocean cyclones affecting Myanmar, but becomes very overconfident beyond this. Looking at flood forecasting models, a comparison of the Global Flood Awareness System (GloFAS, produced by the ECMWF and the European Commission as part of the Copernicus Emergency Management Services) with the Global Flood Forecasting Information System (GLOFFIS, produced by Deltares) demonstrates that both GloFAS and GLOFFIS have difficulty simulating 1 in 2 year return period flows or higher, although GloFAS performance is better than GLOFFIS. GloFAS reforecasts show significantly overconfident probabilities over Myanmar where discharge observations are available, whilst the lack of a GLOFFIS reforecast prevents evaluation of this forecast system directly. Evaluation of the ten-day operational storm surge forecast (the Global Storm Surge Information System, GLOSSIS) produced by Deltares was attempted but lack of any data prevented assessment. These findings present valuable insights into how well forecasts perform, which is crucial information for establishing effective humanitarian action mechanisms.

Improved flood forecasting using geomorphic unit hydrograph based on spatially distributed velocity field

Abstract This paper presents an energy model for determining the overland flow velocity in order to improve the low accuracy problem in flow concentration simulation. It furnishes a novel idea for studying flow concentration in ungauged basins. The model can be widely applied in analysis of spatial velocity field, extraction of instantaneous geomorphic unit hydrograph and development of distributed hydrological model. A distributed flood-forecasting model is constructed for Lianyuan Basin in Hunan Province of China. In the proposed method, gravitational potential energy is transformed into kinetic energy via an analysis of energy distribution of water particles in the basin. Based on the kinetic energy equation, the overland flow velocity simulating the geomorphic unit hydrograph is computed. Rainfall-runoff simulation is then performed by integrating with runoff yield and concentration model. Results indicate that the model based on energy conversion leads to more accurate results. The model has the following advantages: firstly, the spatial distribution of the velocity field is appropriate; secondly, the model has only one parameter, which is easily determined; and finally, flow velocity results can be used for the computation of river network flow concentration.

Development of Short-term Flood Forecast Using ARIMA

The aim of this study is to use the Box-Jenkins method to build a flood forecast model by analysing real-time flood parameters for Pengkalan Rama, Melaka river, hereafter known as Sungai Melaka. The time series was tested for stationarity using the Augmented Dickey-Fuller (ADF) and differencing method to render a non-stationary time series stationary from 1 July 2020 at 12:00am to 30th July 2020. A utocorrelation (ACF) and partial autocorrelation (PACF) functions was measured and observed using visual observation to identify the suitable model for water level time series. The parameter Akaike Information Information Criterion (AIC) and the Bayesian Information Criterion (BIC) were used to find the best ARIMA model (BIC). ARIMA (2, 1, 3) was the best ARIMA model for the Pengkalan Rama, with an AIC of 5653.7004 and a BIC of 5695.209. The ARIMA (2, 1, 3) model was used to produce a lead forecast of up to 7 hours for the time series. The model's accuracy was tested by comparing the original and forecast sequences by using Pearson r and R squared. The ARIMA model appears to be adequate for Sungai Melaka, according to the findings of this study. Finally, the ARIMA model provides an appropriate short-term water level forecast with a lead forecast of up to 7 hours. As a result, the ARIMA model is undeniably ideal for river flooding.

CREST-iMAP v1.0: A fully coupled hydrologic-hydraulic modeling framework dedicated to flood inundation mapping and prediction

Under the impacts of hydrologic extremes, there is a growing need for integrated frameworks for flood inundation mapping. In this study, we introduce a framework: the Coupled Routing and Excess Storage for inundation MApping and Prediction (CREST-iMAP). It utilizes a highly parallelized and fully integrated hydrologic-hydraulic model and aims to improve flood prediction. To highlight the advantages, a synthetic rainfall event and the 500-year Hurricane Harvey event were investigated using the CREST-iMAP, compared with a hydrologic model, a hydraulic model, and a simplified hydrologic-hydraulic model. The results indicate the CREST-iMAP achieves good performances in both hydrologic simulation and floodplain inundation mapping. Moreover, the antecedent soil moisture is the most sensitive parameter to model accuracy, followed by the land surface characteristics; the infiltration process to a flood forecasting model is significant. Overall, the CREST-iMAP delivers accurate and timely flood information, making it one of the latest developments of an integrated hydrologic-hydraulic expert system.

Hydrological and hydraulic model for flood forecasting in Rwanda

This paper aims to forecast the Nyabugogo River flood risk and to propose mitigation measures that can reduce flood impacts using hydrological and hydraulic models. The annual Nyabugogo River floods that have led to human and economic losses are mainly due to the high topography, rainfall, soil texture and urbanization. It was found that the medium to high flood vulnerability is located in cropland, open grassland, open shrubland, settlement, sparse forest and wetlands. The results for 10-, 30- 50- and 100-year return periods show a slight increase in the flood inundation area and in water depth. The simulation shows that the 100-year return period gives a water depth of 3.24 m. Therefore, flood mitigation measures are needed, which can include the construction reservoirs upstream, relocation of infrastructures within the flood plain area, buffer zoning around the river, rainwater harvesting and raising public awareness. This paper provides basic support for decision-makers and can help in land use planning and flood management within the basin.

Modelling hybrid and backpropagation adaptive neuro-fuzzy inference systems for flood forecasting

The ability of the adaptive neuro-fuzzy inference algorithm architecture to simulate floods is explored in this research. The development of models for flood forecasting has been centered on two adaptive neuro-fuzzy inference (ANFIS) algorithms. The Takagi–Sugeno fuzzy inference systems (FIS) generated through subtracted clustering were trained using hybrid and backpropagation training algorithms. Multiple statistical performance evaluators were used to assess the performability of the established models. The validity and predictive power of the models are evaluated by estimating a flood occurrence in the study area. In designing the models, a total of 12 inputs were employed. The best performability was found for the ANFIS model created utilizing a hybrid training algorithm with mean square error (MSE) of 0.00034, co-efficient of correlation (R2) of 97.066%, root mean square error (RMSE) of 0.018, Nash–Sutcliffe model efficiency (NSE) of 0.968, mean absolute error (MAE) of 0.0073 and combined accuracy (CA) of 0.018, indicating the possible usage of exploiting the established model for prediction of floods.

Temporal flood forecasting for trans-boundary Jhelum River of Greater Himalayas

Flood forecasting is a significant tool for river basin management and early flood warning systems. The aim of the present study was to develop annual time series flood forecasting models for River Jhelum in Kashmir Valley by the application of autoregressive integrated moving average (ARIMA) modelling approach. Annual peak flow at three principal flood monitoring sites of Jhelum, namely, Sangam, Ram Munshi Bagh and Asham were predicted using historical peak flow time series data. ARIMA models were developed for forecasting 1-year ahead floods. The prediction accuracy of the developed models was evaluated in terms of root mean square error (RMSE), the mean absolute error (MAE), the mean absolute percentage error (MAPE) and the Nash-Sutcliffe efficiency (NSE). RMSE values for Sangam, Ram Munshi Bagh and Asham stations were 112.31, 61.08 and 126.12 cumecs, respectively, and MAE values were 83.00, 48.50 and 104.25 cumecs at these stations. Also, the MAPE values were 10.70%, 6.19% and 10.78% for Sangam, Ram Munshi Bagh and Asham stations, respectively. NSE values for Sangam, Ram Munshi Bagh and Asham stations were 0.84, 0.87 and 0.85, respectively. Low values of RMSE, MAE and MAPE along with high values of NSE at all sites suggest the suitability of the developed models for flood forecasting in River Jhelum. The lead time of forecast obtained by this method was 1 year, which is much longer as compared to the conventional rainfall-runoff models.

Performance evaluation of artificial intelligence paradigms-artificial neural networks, fuzzy logic, and adaptive neuro-fuzzy inference system for flood prediction.

Flood prediction has gained prominence world over due to the calamitous socio-economic impacts this hazard has and the anticipated increase of its incidence in the near future. Artificial intelligence (AI) models have contributed significantly over the last few decades by providing improved accuracy and economical solutions to simulate physical flood processes. This study explores the potential of the AI computing paradigm to model the stream flow. Artificial neural network (ANN), fuzzy logic, and adaptive neuro-fuzzy inference system (ANFIS) algorithms are used to develop nine different flood prediction models using all the available training algorithms. The performance of the developed models is evaluated using multiple statistical performance evaluators. The predictability and robustness of the models are tested through the simulation of a major flood event in the study area. A total of 12 inputs were used in the development of the models. Five training algorithms were used to develop the ANN models (Bayesian regularization, Levenberg Marquardt, conjugate gradient, scaled conjugate gradient, and resilient backpropagation), two fuzzy inference systems to develop fuzzy models (Mamdani and Sugeno), and two training algorithms to develop the ANFIS models (hybrid and backpropagation). The ANFIS model developed using hybrid training algorithm gave the best performance metrics with Nash-Sutcliffe Model Efficiency (NSE) of 0.968, coefficient of correlation (R2) of 97.066%, mean square error (MSE) of 0.00034, root mean square error (RMSE) of 0.018, mean absolute error (MAE) of 0.0073, and combined accuracy (CA) of 0.018, implying the potential of using the developed models for flood forecasting. The significance of this research lies in the fact that a combination of multiple inputs and AI algorithms has been used to develop the flood models. In summary, this research revealed the potential of AI algorithm-based models in predicting floods and also developed some useful techniques that can be used by the Flood Control Departments of various states/regions/countries for flood prognosis.