Rainfall-runoff Prediction Research Articles

Performance of ANN for rainfall-runoff prediction

Performance of ANN for rainfall-runoff prediction

Water resources assessment and prediction in China

Abstract. Water resources assessment in China, can be classified into three groups: (i) comprehensive water resources assessment, (ii) annual water resources assessment, and (iii) industrial project water resources assessment. Comprehensive water resources assessment is the conventional assessment where the frequency distribution of water resources in basins or provincial regions are analyzed. For the annual water resources assessment, water resources of the last year in basins or provincial regions are usually assessed. For the industrial project water resources assessment, the water resources situation before the construction of industrial project has to be assessed. To address the climate and environmental changes, hydrological and statistical models are widely applied for studies on assessing water resources changes. For the water resources prediction in China usually the monthly runoff prediction is used. In most low flow seasons, the flow recession curve is commonly used as prediction method. In the humid regions, the rainfall-runoff ensemble prediction (ESP) has been widely applied for the monthly runoff prediction. The conditional probability method for the monthly runoff prediction was also applied to assess next month runoff probability under a fixed initial condition.

Integrating Artificial Neural Networks into the VIC Model for Rainfall-Runoff Modeling

A hybrid rainfall-runoff model was developed in this study by integrating the variable infiltration capacity (VIC) model with artificial neural networks (ANNs). In the proposed model, the prediction interval of the ANN replaces separate, individual simulation (i.e., single simulation). The spatial heterogeneity of horizontal resolution, subgrid-scale features and their influence on the streamflow can be assessed according to the VIC model. In the routing module, instead of a simple linear superposition of the streamflow generated from each subbasin, ANNs facilitate nonlinear mappings of the streamflow produced from each subbasin into the total streamflow at the basin outlet. A total of three subbasins were delineated and calibrated independently via the VIC model; daily runoff errors were simulated for each subbasin, then corrected by an ANN bias-correction model. The initial streamflow and corrected runoff from the simulation for individual subbasins serve as inputs to the ANN routing model. The feasibility of this proposed method was confirmed according to the performance of its application to a case study on rainfall-runoff prediction in the Jinshajiang River Basin, the headwater area of the Yangtze River.

Rainfall-Runoff Prediction Using Dynamic Typhoon Information and Surface Weather Characteristic Considering Monsoon Effects

In meteorology and engineering, the prediction of quantitative precipitation and streamflow during typhoon events is a vital research topic. In Southern Taiwan, typhoons often occur in the summer. The interaction between the typhoon circulation and southwesterly monsoon flow frequently transports abundant moisture into Southern Taiwan leading to the substantial pouring rains. This study proposes a rainfall-runoff prediction methodology for addressing the complicated inflow forecasts of southwest monsoon rainfall during typhoons in the upper Tsengwen River in Southern Taiwan. This paper is novel in that it incorporates various data types (reservoir inflows, watershed rainfalls, typhoon information, and ground-weather characteristics) that were applied as model inputs. The most frequently used support vector regressions were employed to construct the rainfall-runoff models on the basis of three designed data combination scenarios. Typhoons Kalmaegi (2008), Fung-wong (2008), Jangmi (2008), and Morakot (2009) were used as validation typhoons. The model cases, involving lead times of 1 h to 6 h, were evaluated. Six performance criteria were used in the three scenarios to highlight the scenario capable of identifying the optimal performance level. In addition, this study compared the error rates between accumulation observations and accumulation predictions. The results showed that Scenario 3, which considered typhoon information and ground-weather characteristics simultaneously, had superior watershed rainfall and runoff predictions to those of the other scenarios. Thus, this study demonstrated the feasibility of using the proposed methodology to increase the accuracy of rainfall-runoff predictions.

Adaptive hydrological flow field modeling based on water body extraction and surface information

Suranaree University of Technology, School of Computer Engineering,111 University Avenue, Muang, Nakhon Ratchasima 30000, ThailandAbstract. Hydrological flow characteristic is one of the prime indicators for assessing flood.It plays a major part in determining drainage capability of the affected basin and also in thesubsequent simulation and rainfall-runoff prediction. Thus far, flow directions were typicallyderived from terrain data which for flat landscapes are obscured by other man-made structures,hence undermining the practical potential. In the absence (or diminutive) of terrain slopes, waterpassageshaveamorepronouncedeffectonflowdirectionsthanelevations.Thispaper,therefore,presents detailed analyses and implementation of hydrological flow modeling from satellite andtopographicimages.Herein,gradualassignmentbasedonsupportvectormachinewasappliedtomodified normalized difference water index and a digital surface model, in order to ensure reli-able water labeling while suppressing modality-inherited artifacts and noise. Gradient vectorflow was subsequently employed to reconstruct the flow field. Experiments comparing theproposed scheme with conventional water boundary delineation and flow reconstruction werepresented. Respective assessments revealed its advantage over the generic stream burning.Specifically, it could extract water body from studied areas with 98.70% precision, 99.83%recall, 98.76% accuracy, and 99.26% F-measure. The correlations between resultant flowsand those obtained from the stream burning were as high as 0.80 0.04 (p ≤ 0.01 in all res-olutions).

위성토양수분과 지점강우량을 이용한 지역 선행습윤조건 분석

Soil moisture is one of the most important interests in hydrological response and the interaction between the land surface and atmosphere. Estimation of Antecedent Wetness Conditions (AWC) which is soil moisture condition prior to a rainfall in the basin should be considered for rainfall-runoff prediction. In this study, Soil Wetness Index (SWI), Antecedent Precipitation Index (API5), remotely sensed Soil Moisture (SMrs), and 5 days ground Soil Moisture (SMg5) were selected to estimate the AWC at four study area in the Korean Peninsula. The remotely sensed soil moisture data were taken from the AMSR-E soil moisture archive. The maximum potential retention (Sobs) was obtained from direct runoff and rainfall using Soil Conservation Service-Curve Number (SCS-CN) method by rainfall data of 2011 for each study area. Results showed the great correlations between the maximum potential retention and SWI with a mean correlation coefficient which is equal to –0.73. The results of time length representing the time scale of soil moisture showed a gap from region to region. It was due to the differences of soil types and the characteristics of study area. Since the remotely sensed soil moisture has been proved as reasonable hydrological variables to predict a wetness in the basin, it should be continuously monitored.

Development of Integrated Drought Evaluation and Monitoring System: Case Study of Aharchay River Basin

AbstractIn this paper, an integrated drought management scheme (IDMS) is developed for the Aharchay River Basin, located in northwestern Iran. This region is located in a semiarid region and has experienced considerable drought damage in recent years. Therefore, application of the proposed IDMS could greatly help to mitigate drought damage. The developed IDMS includes different tools and models needed for drought management. The utilized tools are classified into four groups, as follows: (1) data collection and validation, (2) rainfall-runoff prediction, (3) reservoir operation, and (4) drought analysis and monitoring. Future rainfall and runoff are projected using two computer models. Optimal reservoir operation is determined using a Bayesian stochastic genetic algorithm and Bayesian stochastic dynamic programming. These models are capable of incorporating new data of reservoir inflow and expert opinion in decision-making. For reservoir operation in drought conditions, hedging rules are also developed ba...

Performance of artificial neural network and regression techniques for rainfall-runoff prediction

Different types of methods have been used in runoff prediction involving conceptualand empirical models. Nevertheless, none of these methods can be considered as a single superior model. Owing to the complexity of the hydrological process, the accurate runoff is difficult to be predicted using the linear recurrence relations or physically based watershed. The linear recurrence relation model does not attempt to take into account the nonlinear dynamic of the hydrological process. The Artificial Neural Network (ANN) is a new technique with a flexible mathematical structure that is capable of identifying complex non-linear relationships between input and output data when compared to other classical modelling techniques. Therefore, the present study aims to utilize an Artificial Neural Network (ANN) to predict the rainfall-runoff relationship in a catchment area located in a Tanakami region of Japan. The study illustrates the applications of the feed forward back propagation with hyperbolic tangent neurons in the hidden layer and linear neuron in the output layer is used for rainfall prediction. To evaluate the performance of the proposed model, three statistical indexes were used, namely; Correlation coefficient (R), mean square error (MSE) and correlation of determination (R2). The results showed that the feed forward back propagation Neural Network (ANN) can describe the behaviour of rainfall-runoff relation more accurately than the classical regression model. Key words: Rainfall-runoff, artificial neural network, linear regression.

A WebGIS-based system for rainfall-runoff prediction and real-time water resources assessment for Beijing

A WebGIS-based system designed to predict rainfall-runoff and assess real-time water resources for Beijing was developed to provide support for scientific decision making regarding solving water shortages while effectively reducing urban flood threats in the city. The system adopts a Browse Server (B/S) structure and combines the distributed hydrologic modeling and WebGIS techniques. For this system, a distributed hydrologic model of Beijing that adopts a grid cell-size of 1 km by 1 km and covers the city's entire area of 16,400 km 2 was developed and validated. This model employs a simple, yet practical rainfall-runoff correlation curve method to predict runoff, as well as prediction approaches for rainfall, evaporation, subsurface runoff and recharge to groundwater. In addition, a framework for the assessment of real-time water resources assessment based on hydrologic monitoring stations and the distributed model was established. Finally, a WebGIS-based system for rainfall-runoff prediction and real-time water resources assessment for Beijing was developed by integrating a data platform, the professional models and the WebGIS techniques. This system was successfully integrated into the hydrologic prediction practices of the General Station of Hydrology, Bureau of Beijing Water Affairs in 2005, and the demonstration version of the system can be seen on the Web at http://123.127.143.23/enewRF/login/login.aspx?ReturnUrl=%2fenewRF%2ftemp.aspx.

ENSEMBLE RAINFALL-RUNOFF PREDICTION WITH RADAR IMAGE EXTRAPOLATION AND ITS ERROR STRUCTURE

New attempt of ensemble rainfall-runoff prediction is presented with radar rainfall prediction and spatial random error field simulation. A radar image extrapolation model gives deterministic rainfall predictions, and its prediction error structure is analyzed by comparing with the observed rainfall fields. With the analyzed error characteristics, spatial random error fields are simulated using covariance matrix decomposition method. The simulated random error fields successfully keep the analyzed error structure and improve the accuracy of the deterministic rainfall predictions; then the random error fields with the deterministic fields are given to a distributed hydrologic model to achieve an ensemble runoff prediction.

Application of a Genetic Algorithm in the Development and Optimisation of a Non-linear Dynamic Runoff Model

Daily rainfall and runoff data for the Kashinagar watershed of the Vamsadhara river basin, Orissa, India collected for the years 1984–1995 have been used for the development and optimisation of a runoff prediction model. The prediction model has been developed by considering the process as non-linear and dynamic in nature. Model orders are determined by multiple correlation analysis and then based on ‘ t-test’ values. All positive values of the t-test, obtained through multiple correlation, are considered for model formulation. Genetic algorithm has been employed for the estimation of model parameters and function optimisation. The predicted values by the non-linear rainfall–runoff prediction model were 1–2 days ahead of the measured values. This time problem was corrected by applying backward shift operator technique which improved the correlation coefficient by about 15%.

A METHOD FOR PREDICTING SEDIMENT RUNOFF CAUSED BY EROSION OF STREAM CHANNEL BED

The present study describes a method for predicting sediment runoff caused by erosion of sediment stored in river channels. The method is developed by combining a newly developed channel model, a rain-fall runoff model and a storage-type sediment runoff model. The channel model, which is based on a unit channel having two inflow points and one outflow point, can easily reproduce stream channels distributed in a drainage basin. A kinematic wave runoff method is employed in rainfall runoff prediction, and the sediment runoff model is developed by integrating partial differential equations for water and sediment with respect to the reach of unit channel. The present method can predict temporal and spatial change of water discharge, sediment discharge, sediment storage and sediment size when rainfall data are given as in-put. It is applied to the upper basin of Kumano river, and its applicability is verified by comparing the predicted results and reservoir sedimentation data.

Study on Rainfall Runoff Prediction System for Radar Raingauge

Radar rainfall gauge gives a better information of rainfall distribution than the conventional Thiesen method. In the present study, a system for the prediction of rainfall-runoff is constructed. This system includes packages to obtain a channel network from the digital map, to combine the radar rainfall information with the channel network, and to calculate the runoff using the kinematic wave model.This paper described the each package of the system. And parameters of the kinematic wave model were obtained for 20 runoff events in 1990 and 91. Application of this rainfall runoff prediction system to dam reservoiurs is possible with several minor improvements of the system.

Discussion of “Barnes on Rainfall-Runoff Prediction”

Discussion of “Barnes on Rainfall-Runoff Prediction”

Closure to “Linsley and Ackermann on Rainfall-Runoff Prediction”

Closure to “Linsley and Ackermann on Rainfall-Runoff Prediction”