Improve Flood Forecasting Research Articles

Assimilation of soil moisture and streamflow observations to improve flood forecasting with considering runoff routing lags

Assimilation of either soil moisture or streamflow has been well demonstrated to improve flood forecasting. However, it is difficult to assimilate two different types of observations into a rainfall–runoff model simultaneously because there is a time lag between soil moisture and streamflow owing to the runoff routing process. In this study, we developed an effective data assimilation scheme based on the ensemble Kalman filter and smoother (named as EnKF-S) to exploit the benefits of the two observation types while accounting for the runoff routing lag. To prove the importance of accounting for the time lag, a scheme named Dual-EnKF was used to compare. To demonstrate the schemes, we designed synthetic cases regarding two typical flood patterns, i.e., flash flood and gradual flood. The results show that EnKF-S can effectively improve flood forecasting compared with Dual-EnKF, particularly when the runoff routing has distinct time lags. For the synthetic cases, EnKF-S reduced root–mean–square error (RMSE) by more than 70% relative to the data assimilation scheme without considering runoff routing lags. Therefore, this effective data assimilation scheme holds great potential for short-term flood forecasting by merging observations from ground measurement and remote sensing retrievals.

Assessment and weighting of meteorological ensemble forecast members based on supervised machine learning with application to runoff simulations and flood warning

Numerical weather forecasts, such as meteorological forecasts of precipitation, are inherently uncertain. These uncertainties depend on model physics as well as initial and boundary conditions. Since precipitation forecasts form the input into hydrological models, the uncertainties of the precipitation forecasts result in uncertainties of flood forecasts. In order to consider these uncertainties, ensemble prediction systems are applied. These systems consist of several members simulated by different models or using a single model under varying initial and boundary conditions. However, a too wide uncertainty range obtained as a result of taking into account members with poor prediction skills may lead to underestimation or exaggeration of the risk of hazardous events. Therefore, the uncertainty range of model-based flood forecasts derived from the meteorological ensembles has to be restricted.In this paper, a methodology towards improving flood forecasts by weighting ensemble members according to their skills is presented. The skill of each ensemble member is evaluated by comparing the results of forecasts corresponding to this member with observed values in the past. Since numerous forecasts are required in order to reliably evaluate the skill, the evaluation procedure is time-consuming and tedious. Moreover, the evaluation is highly subjective, because an expert who performs it makes his decision based on his implicit knowledge.Therefore, approaches for the automated evaluation of such forecasts are required. Here, we present a semi-automated approach for the assessment of precipitation forecast ensemble members. The approach is based on supervised machine learning and was tested on ensemble precipitation forecasts for the area of the Mulde river basin in Germany. Based on the evaluation results of the specific ensemble members, weights corresponding to their forecast skill were calculated. These weights were then successfully used to reduce the uncertainties within rainfall-runoff simulations and flood risk predictions.

Improvement of operational flood forecasting through the assimilation of satellite observations and multiple river flow data

Abstract. Data assimilation has the potential to improve flood forecasting. However, it is rarely employed in distributed hydrologic models for operational predictions. In this study, we present variational assimilation of river flow data at multiple locations and of land surface temperature (LST) from satellite in a distributed hydrologic model that is part of the operational forecasting chain for the Arno river, in central Italy. LST is used to estimate initial condition of soil moisture through a coupled surface energy/water balance scheme. We present here several hindcast experiments to assess the performances of the assimilation system. The results show that assimilation can significantly improve flood forecasting, although in the limit of data error and model structure.

Improving flood forecasts for better flood preparedness in the UK (and beyond)

This winter (2013/14) coastal storms and an unprecedented amount of rainfall led to significant and widespread flooding across the southern UK. Despite much criticism and blame surrounding the flood events, the Flood Forecasting Centre, a recent development in national‐level flood forecasting capabilities for the government and emergency response communities, has received considerable praise. Here we consider how scientific developments and organisational change have led to improvements in the forecasting and flood preparedness seen in this winter's flooding. Although such improvements are admirable, there are many technical and communication challenges that remain for probabilistic flood forecasts to achieve their full potential.

Improving Flood Forecasting in a Developing Country: A Comparative Study of Stepwise Multiple Linear Regression and Artificial Neural Network

Due to limited data sources, practical situations in most developing countries favor black-box models in real time operations. In a simple and robust approach, this study examines performances of stepwise multiple linear regression (SMLR) and artificial neural network (ANN) models, as tools for multi-step forecasting Chindwin River floods in northern Myanmar. Future river stages are modeled using past water levels and rainfall at the forecasting station as well as at the hydrologically connected upstream station. The developed models are calibrated with flood season data from 1990 to 2007 and validated with data from 2008 to 2011. Model performances are compared for 1- to 5-day ahead forecasts. With a high accuracy, both candidate models performed well for forecasting the full range of flood levels. The ANN models were superior to the SMLR models, particularly in predicting the extreme floods. Correlation analysis was found to be useful for determining the initial input variables. Contribution of upstream data to both models could improve the forecasting performance with higher R 2 values and lower errors. Considering the commonly available data in the region as primary predictors, the results would be useful for real time flood forecasting, avoiding the complexity of physical processes.

Satellite precipitation in southeastern South America: how do sampling errors impact high flow simulations?

Satellite precipitation estimates are increasingly available at temporal and spatial scales of interest to hydrological applications and with the potential for improving flood forecasts in data-sparse regions. This study evaluates the effect of sampling error on simulated large flood events. Synthetic precipitation fields were generated in Monte Carlo fashion by perturbing observed precipitation fields with sampling errors based on 1, 2 and 6 h intervals. The variable infiltration capacity hydrological model was used to assess the impact of these errors on simulated high flow events in the Iguazu basin, a rain-dominated, subtropical basin in southeastern South America. Results showed that unbiased errors in daily error-corrupted precipitation fields introduced bias in the simulated hydrologic fluxes and states. The overall bias for error-corrupted daily streamflows was positive and its magnitude increased with larger sampling intervals. However, for high flow events, the bias was negative as a result of an increase in simulated infiltration and changes in precipitation variability. Errors in precipitation also affected the magnitude and volume of the peak events but did not change the first two statistical moments of the peaks indicating that non-linearities in the hydrological system preserve the statistical properties of high flows in the basin. Caution is needed when using satellite products for hydrological applications that require the estimation of large peaks and volumes.

Assimilation of Passive Microwave Streamflow Signals for Improving Flood Forecasting: A First Study in Cubango River Basin, Africa

Floods are among the most frequently occurring and disastrous natural hazards in the world. The overarching goal of this study is to investigate the utility of passive microwave AMSR-E signal and TRMM based precipitation estimates in improving flood prediction at the sparsely gauged Cubango River Basin, Africa. This is accomplished by coupling a widely used conceptual rainfall-runoff hydrological model with Ensemble Square Root Filter (EnSRF) to account for uncertainty in both forcing data and model initial conditions. Three experiments were designed to quantify the contributions of the AMSR-E signal to the flood prediction accuracy, in comparison to the benchmark assimilation of in-situ streamflow observations, for both “Open Loop” and “Assimilation” modules. In general, the EnSRF assimilation of both in-situ observations and AMSR-E signal-converted-streamflow effectively improved streamflow modeling performance in terms of three statistical measures. In order to further investigate AMSR-E signals' contribution to extreme events prediction skill, the upper 10th percentile daily streamflow was taken as the threshold. Results show significantly improved skill and detectability of floods as well as reduced false alarm rates. Given the global availability of satellite-based precipitation from current TRMM and future GPM, together with soil moisture information from the current AMSR-E and future SMAP mission at near real-time, this “first attempt” study at a sparsely gauged African basin shows that opportunities exist for an integrated application of a suite of satellite data in improving flood forecasting worldwide by careful fusion of remote sensing and in-situ observations.

An operational flood warning system for poorly gauged basins: demonstration in the Guadalhorce basin (Spain)

This paper deals with the presentation of a flood warning system (GFWS) developed for the specific characteristics of the Guadalhorce basin (3,200 km2, SE of Spain), which is poorly gauged and often affected by flash and plain floods. Its complementarity with the European flood alert system (EFAS) has also been studied. At a lower resolution, EFAS is able to provide a flood forecast several days in advance. The GFWS is adapted to the use of distributed rainfall maps (such as radar rainfall estimates), and discharge forecasts are computed using a distributed rainfall–runoff model. Due to the lack of flow measurements, the model parameters calibrated on a small watershed have been transferred in most of the basin area. The system is oriented to provide distributed warnings and fulfills the requirements of ungauged basins. This work reports on the performance of the system on two recent rainfall events that caused several inundations. These results show how the GFWS performed well and was able to forecast the location and timing of flooding. It demonstrates that despite its limitations, a simple rainfall–runoff model and a relatively simple calibration could be useful for event risk management. Moreover, with low resolution and long anticipation, EFAS appears as a good complement tool to improve flood forecasting and compensate for the short lead times of the GFWS.

Robust multi-objective calibration strategies – possibilities for improving flood forecasting

Abstract. Process-oriented rainfall-runoff models are designed to approximate the complex hydrologic processes within a specific catchment and in particular to simulate the discharge at the catchment outlet. Most of these models exhibit a high degree of complexity and require the determination of various parameters by calibration. Recently, automatic calibration methods became popular in order to identify parameter vectors with high corresponding model performance. The model performance is often assessed by a purpose-oriented objective function. Practical experience suggests that in many situations one single objective function cannot adequately describe the model's ability to represent any aspect of the catchment's behaviour. This is regardless of whether the objective is aggregated of several criteria that measure different (possibly opposite) aspects of the system behaviour. One strategy to circumvent this problem is to define multiple objective functions and to apply a multi-objective optimisation algorithm to identify the set of Pareto optimal or non-dominated solutions. Nonetheless, there is a major disadvantage of automatic calibration procedures that understand the problem of model calibration just as the solution of an optimisation problem: due to the complex-shaped response surface, the estimated solution of the optimisation problem can result in different near-optimum parameter vectors that can lead to a very different performance on the validation data. Bárdossy and Singh (2008) studied this problem for single-objective calibration problems using the example of hydrological models and proposed a geometrical sampling approach called Robust Parameter Estimation (ROPE). This approach applies the concept of data depth in order to overcome the shortcomings of automatic calibration procedures and find a set of robust parameter vectors. Recent studies confirmed the effectivity of this method. However, all ROPE approaches published so far just identify robust model parameter vectors with respect to one single objective. The consideration of multiple objectives is just possible by aggregation. In this paper, we present an approach that combines the principles of multi-objective optimisation and depth-based sampling, entitled Multi-Objective Robust Parameter Estimation (MOROPE). It applies a multi-objective optimisation algorithm in order to identify non-dominated robust model parameter vectors. Subsequently, it samples parameter vectors with high data depth using a further developed sampling algorithm presented in Krauße and Cullmann (2012a). We study the effectivity of the proposed method using synthetical test functions and for the calibration of a distributed hydrologic model with focus on flood events in a small, pre-alpine, and fast responding catchment in Switzerland.

Effects of temporal resolution of input precipitation on the performance of hydrological forecasting

Abstract. Flood prediction systems rely on good quality precipitation input data and forecasts to drive hydrological models. Most precipitation data comes from daily stations with a good spatial coverage. However, some flood events occur on sub-daily time scales and flood prediction systems could benefit from using models calibrated on the same time scale. This study compares precipitation data aggregated from hourly stations (HP) and data disaggregated from daily stations (DP) with 6-hourly forecasts from ECMWF over the time period 1 October 2006–31 December 2009. The HP and DP data sets were then used to calibrate two hydrological models, LISFLOOD-RR and HBV, and the latter was used in a flood case study. The HP scored better than the DP when evaluated against the forecast for lead times up to 4 days. However, this was not translated in the same way to the hydrological modelling, where the models gave similar scores for simulated runoff with the two datasets. The flood forecasting study showed that both datasets gave similar hit rates whereas the HP data set gave much smaller false alarm rates (FAR). This indicates that using sub-daily precipitation in the calibration and initiation of hydrological models can improve flood forecasting.

Improving flood forecasting in Bangladesh using an artificial neural network

A river stage neural network model has been developed to study and predict the water level of Dhaka city. A total of five stations located at the border area of Bangladesh on the Ganges, Brahmaputra and Meghna rivers are selected as input nodes and Dhaka on the Buriganga river is the output node for the neural network. This model is trained with river stage data for a period of 1998 to 2004 and validated with data from 2005 to 2007. The river stage of Dhaka has been predicted for up to ten days with very high accuracy. Values of R2, root mean square and mean absolute error are found ranging from 0.537 to 0.968, 0.607 m to 0.206 m and 0.475 m to 0.154 m, respectively, during training and validation of the model. The results of this study can be useful for real-time flood forecasting by reducing computational time, improving water resources management and reducing the unnecessary cost of field data collection.

Cinrad data quality control and precipitation estimation

Cinrad is the acronym describing China's next-generation weather radar that inherits the technology of the US national weather system Nexrad (WSR-88D) through a joint agreement between the two countries. More than 100 Cinrad systems have currently been deployed in China, but challenges exist for their wider application. The data quality assessment and the development of quantitative precipitation estimation for flood risk management will pose a significant challenge as the national system is both commissioned and further developed. This paper is based on an assessment of the data quality of a Cinrad system deployed in southern China and describes aspects of the initial data quality-control procedure. Precipitation was estimated for the Liuxihe River basin using 11 observed storms. The data quality-control procedure presented in this paper has four components, including volume-scan data, missing-data interpolation, abnormal-reflectivity detection and correction and fine-day reflectivity removal. The results indicate that the radar-estimated precipitation is consistent with that measured by conventional rain-gauge networks; also radar estimated precipitation can provide a better representation of the spatial distribution of the precipitation over the whole basin, and thereby has the potential to improve flood forecasting and warning systems. Given the importance of flooding to China, the continued quantitative development of the radar system is of national importance, particularly in terms of the tropical-cyclone-induced flooding prevalent in south China.

Coupling the Xinanjiang model to a kinematic flow model based on digital drainage networks for flood forecasting

AbstractThe Xinanjiang model, which is a conceptual rainfall‐runoff model and has been successfully and widely applied in humid and semi‐humid regions in China, is coupled by the physically based kinematic wave method based on a digital drainage network. The kinematic wave Xinanjiang model (KWXAJ) uses topography and land use data to simulate runoff and overland flow routing. For the modelling, the catchment is subdivided into numerous hillslopes and consists of a raster grid of flow vectors that define the water flow directions. The Xinanjiang model simulates the runoff yield in each grid cell, and the kinematic wave approach is then applied to a ranked raster network. The grid‐based rainfall‐runoff model was applied to simulate basin‐scale water discharge from an 805‐km2 catchment of the Huaihe River, China. Rainfall and discharge records were available for the years 1984, 1985, 1987, 1998 and 1999. Eight flood events were used to calibrate the model's parameters and three other flood events were used to validate the grid‐based rainfall‐runoff model. A Manning's roughness via a linear flood depth relationship was suggested in this paper for improving flood forecasting. The calibration and validation results show that this model works well. A sensitivity analysis was further performed to evaluate the variation of topography (hillslopes) and land use parameters on catchment discharge. Copyright © 2009 John Wiley & Sons, Ltd.

Role of Passive Microwave Remote Sensing in Improving Flood Forecasts

Accurate information concerning antecedent soil moisture conditions is a key source of hydrologic forecasting skill for regional-scale flooding events occurring over time scales of days to weeks. Remotely sensed surface soil moisture observations are a viable source of such information and can potentially improve flood peak timing and magnitude forecasting in such events. C- and X-band brightness temperature data from the Advanced Microwave Scanning Radiometer (AMSR-E) aboard NASA's Aqua satellite are used here to demonstrate the potential for improving streamflow forecasts by using remotely sensed surface soil moisture during a flooding event in northeastern Australia (Queensland) during January-February 2004. An analysis of AMSR-E brightness temperature imagery reveals a clear anomaly of low AMSR-E brightness temperatures (i.e., high soil moisture conditions) over the affected areas in the four- to five-day period preceding peak streamflow conditions. Land surface conditions are a remotely detectable precursor to subsequent downstream flooding. Use of a simple adaptive model demonstrates that AMSR-E passive microwave observations can add skill to streamflow forecasts during the event.

Relationship between atmospheric conditions at Dhaka, Bangladesh, and rainfall at Cherrapunjee, India

To improve flood forecasting, the understanding of the atmospheric conditions associated with severe rainfall is crucial. We analysed the atmospheric conditions at Dhaka, Bangladesh, using upper-air soundings. We then compared these conditions with daily rainfall variations at Cherrapunjee, India, which is a main source of floodwater to Bangladesh, and a representative sample of exceptionally heavy rainfall events. The analysis focussed on June and July 2004. June and July are the heaviest rainfall months of the year at Cherrapunjee. July 2004 had the fourth-heaviest monthly rainfall of the past 31 years, and severe floods occurred in Bangladesh. Active rainfall periods at Cherrapunjee corresponded to “breaks” in the Indian monsoon. The monsoon trough was located over the Himalayan foothills, and strong westerly winds dominated up to 7 km at Dhaka. Near-surface wind below 1 km had southerly components, and the wind profile had an Ekman spiral structure. The results suggest that rainfall at Cherrapunjee strongly depends on the near-surface wind speed and wind direction at Dhaka. Lifting of the near-surface southerly airflow by the Meghalaya Plateau is considered to be the main contributor to severe rainfall at Cherrapunjee. High convective available potential energy (CAPE) also contributes to intense rainfall.

Influence of uncertain boundary conditions and model structure on flood inundation predictions

In this study, the GLUE methodology is applied to establish the sensitivity of flood inundation predictions to uncertainty of the upstream boundary condition and bridges within the modelled region. An understanding of such uncertainties is essential to improve flood forecasting and floodplain mapping. The model has been evaluated on a large data set. This paper shows uncertainty of the upstream boundary can have significant impact on the model results, exceeding the importance of model parameter uncertainty in some areas. However, this depends on the hydraulic conditions in the reach e.g. internal boundary conditions and, for example, the amount of backwater within the modelled region. The type of bridge implementation can have local effects, which is strongly influenced by the bridge geometry (in this case the area of the culvert). However, the type of bridge will not merely influence the model performance within the region of the structure, but also other evaluation criteria such as the travel time. This also highlights the difficulties in establishing which parameters have to be more closely examined in order to achieve better fits. In this study no parameter set or model implementation that fulfils all evaluation criteria could be established. We propose four different approaches to this problem: closer investigation of anomalies; introduction of local parameters; increasing the size of acceptable error bounds; and resorting to local model evaluation. Moreover, we show that it can be advantageous to decouple the classification into behavioural and non-behavioural model data/parameter sets from the calculation of uncertainty bounds.

Improving flood forecasting in international river basins

In flood‐prone international river basins (IRBs), many riparian nations that are located close to a basin's outlet face a major problem in effectively forecasting flooding because they are unable to assimilate in situ rainfall data in real time across geopolitical boundaries. NASA's propose Global Precipitation Measurement (GPM) mission, which is expected to begin in 2010, will comprise high‐resolution passive microwave (PM) sensors (at resolution ∼3‐6 hours, 10 x 10 square kilometers) that may provide new opportunities to improve flood forecasting in these river basins.Research is now needed to realize the potential of GPM. With adequate research in the coming years, it may be possible to identify the specific IRBs that would benefit cost‐effectively from a preprogrammed satellite‐based forecasting system in anticipation of GPM. Acceleration of such a research initiative is worthwhile because it could reduce the risk of the cancellation of GPM [see Zielinski, 2005].

Designing flood forecasting, warning and response systems from a societal perspective

Improving flood forecasting, warning and response systems (FFWRS) is now a common requirement in Europe and elsewhere. FFWRS need to be designed from a societal as well as a scientific perspective, but social preferences and needs are often not well integrated. This is one of the fundamental reasons for the under-performance of FFWRS, but insufficient flood warning lead time caused by communication time-lags is another. Several conceptual perspectives illuminate how a societal approach can aid FFWRS redesign. Recent experience of redesigning FFWRS in England and Wales using a societal approach is examined. Warning 'customers' require consistent and informative warning messages which they can confirm. New information and communication technologies (ICTs), including automatic voice messaging, dial-and-listen services and the internet, are being harnessed to improve direct warning dissemination and to increase warning lead times. These and other ICTs, including mobile telephones, open up new possibilities for reaching members of typically diverse and heterogeneous floodplain communities. England and Wales is moving towards a Multi-Media Warning Dissemination Service for flood warnings in which 'customers' will be able to select from a range of warning delivery methods those which most suit their different requirements. Increasing the public's awareness and knowledge of flooding, flood warnings and appropriate warning response is now critical to achieving significantly improved warning response and the strategy used in England and Wales is leading to progress.

Flood forecasting reaches new potential

Natural disasters caused by floods challenge scientists to forecast the magnitude and timing of peak flood discharges more accurately. With recent advances in Geographic Information Systems (GIS) and in computer technology, scientists have developed large‐scale surface runoff models to improve flood forecasting. The first version of the model CASC2D, developed at Colorado State University, simulates surface runoff from flash floods caused by intense thunderstorms moving across watershed areas. The hydrologic model has been coupled with data from the CSU‐CHILL polarized dual‐Doppler radar in Greeley, Colorado, which accurately tracks the motion of severe thunderstorms.