The potential of different ANN techniques in evapotranspiration modelling

The potential of three different artificial neural network (ANN) techniques, the multi‐layer perceptrons (MLPs), radial basis neural networks (RBNNs) and generalized regression neural networks (GRNNs), in modelling of reference evapotranspiration (ET₀) is investigated in this paper. Various daily climatic data, that is, solar radiation, air temperature, relative humidity and wind speed from two stations, Pomona and Santa Monica, in Los Angeles, USA, are used as inputs to the ANN techniques so as to estimate ET₀ obtained using the FAO‐56 Penman–Monteith (PM) equation. In the first part of the study, a comparison is made between the estimates provided by the MLP, RBNN and GRNN and those of the following empirical models: The California Irrigation Management Information System (CIMIS) Penman (1985), Hargreaves (1985) and Ritchie (1990). In this part of the study, the empirical models are calibrated using the standard FAO‐56 PM ET₀ values. The estimates of the ANN techniques are also compared with those of the calibrated empirical models. Mean square errors, mean absolute errors and determination coefficient statistics are used as comparing criteria for the evaluation of the models' performances. Based on the comparisons, it is found that the MLP and RBNN techniques could be employed successfully in modelling the ET₀ process. In the second part of the study, the potential of ANN techniques and the empirical methods in ET₀ estimation using nearby station data is investigated. Among the models, the calibrated Hargreaves model is found to perform better than the others. Copyright © 2007 John Wiley & Sons, Ltd.     

Modelling evapotranspiration using discrete wavelet transform and neural networks

This study combines wavelet transforms and feed‐forward neural network methods for reference evapotranspiration estimation. The climatic data (air temperature, solar radiation, wind speed, relative humidity) from two stations in the United States was evaluated for estimating models. For wavelet and neural network (WNN) model, the input data was decomposed into wavelet sub‐time series by wavelet transformation. Later, the new series (reconstructed series) are produced by adding the available wavelet components and these reconstructed series are used as the input of the WNN model. This phase is pre‐processing of raw data and the main different of the WNN model. The performance of the WNN model was compared with classical neural networks approach [artificial neural network (ANN)], multi‐linear regression and Hargreaves empirical method. This study shows that the wavelet transforms and neural network methods could be applied successfully for evapotranspiration modelling from climatic data. Copyright © 2009 John Wiley & Sons, Ltd.     

Choice between competitive pairs of frequency models for use in hydrology: a review and some new results

Abstract

The potential is investigated of the generalized regression neural networks (GRNN) technique in modelling of reference evapotranspiration (ET₀) obtained using the FAO Penman-Monteith (PM) equation. Various combinations of daily climatic data, namely solar radiation, air temperature, relative humidity and wind speed, are used as inputs to the ANN so as to evaluate the degree of effect of each of these variables on ET₀. In the first part of the study, a comparison is made between the estimates provided by the GRNN and those obtained by the Penman, Hargreaves and Ritchie methods as implemented by the California Irrigation Management System (CIMIS). The empirical models were calibrated using the standard FAO PM ET₀ values. The GRNN estimates are also compared with those of the calibrated models. Mean square error, mean absolute error and determination coefficient statistics are used as comparison criteria for the evaluation of the model performances. The GRNN technique (GRNN 1) whose inputs are solar radiation, air temperature, relative humidity and wind speed, gave mean square errors of 0.058 and 0.032 mm² day^?2, mean absolute errors of 0.184 and 0.127 mm day^?1, and determination coefficients of 0.985 and 0.986 for the Pomona and Santa Monica stations (Los Angeles, USA), respectively. Based on the comparisons, it was found that the GRNN 1 model could be employed successfully in modelling the ET₀ process. The second part of the study investigates the potential of the GRNN and the empirical methods in ET₀ estimation using the nearby station data. Among the models, the calibrated Hargreaves was found to perform better than the others.     

Evapotranspiration modelling using support vector machines / Modélisation de l'évapotranspiration à l'aide de ‘support vector machines’

Abstract

This study investigates the accuracy of support vector machines (SVM), which are regression procedures, in modelling reference evapotranspiration (ET₀). The daily meteorological data, solar radiation, air temperature, relative humidity and wind speed from three stations, Windsor, Oakville and Santa Rosa, in central California, USA, are used as inputs to the support vector machines to reproduce ET₀ obtained using the FAO-56 Penman-Monteith equation. A comparison is made between the estimates provided by the SVM and those of the following empirical models: the California Irrigation Management System (CIMIS) Penman, Hargreaves, Ritchie and Turc methods. The SVM results were also compared with an artificial neural networks method. Root mean-squared errors, mean-absolute errors, and determination coefficient statistics are used as comparing criteria for the evaluation of the models' performances. The comparison results reveal that the support vector machines could be employed successfully in modelling the ET₀ process.     

Intercomparison of evapotranspiration estimates

Abstract

Five methods of computing reference evapotranspiration from a reference crop (Penman, radiation, Blaney-Criddle, Hargreaves and pan evaporation) have been studied for their applicability under different climatic conditions. The Penman method was taken as the standard and the other four methods were compared against this method. Good correlation was obtained between the values estimated by the four methods and the Penman method although differences in magnitude were found. Regression equations were developed to correct those differences in magnitude. The method suitable for the estimation of reference evapotranspiration for each climatic condition is also suggested.     

Modelling evaporation using an artificial neural network algorithm

This paper investigates the prediction of Class A pan evaporation using the artificial neural network (ANN) technique. The ANN back propagation algorithm has been evaluated for its applicability for predicting evaporation from minimum climatic data. Four combinations of input data were considered and the resulting values of evaporation were analysed and compared with those of existing models. The results from this study suggest that the neural computing technique could be employed successfully in modelling the evaporation process from the available climatic data set. However, an analysis of the residuals from the ANN models developed revealed that the models showed significant error in predictions during the validation, implying loss of generalization properties of ANN models unless trained carefully. The study indicated that evaporation values could be reasonably estimated using temperature data only through the ANN technique. This would be of much use in instances where data availability is limited. Copyright © 2002 John Wiley & Sons, Ltd.     

A wavelet-neural network hybrid modelling approach for estimating and predicting river monthly flows

Abstract

A wavelet-neural network (WNN) hybrid modelling approach for monthly river flow estimation and prediction is developed. This approach integrates discrete wavelet multi-resolution decomposition and a back-propagation (BP) feed-forward multilayer perceptron (FFML) artificial neural network (ANN). The Levenberg-Marquardt (LM) algorithm and the Bayesian regularization (BR) algorithm were employed to perform the network modelling. Monthly flow data from three gauges in the Weihe River in China were used for network training and testing for 48-month-ahead prediction. The comparison of results of the WNN hybrid model with those of the single ANN model show that the former is able to significantly increase the prediction accuracy.

Editor D. Koutsoyiannis; Associate editor H. Aksoy

Citation Wei, S., Yang, H., Song, J.X., Abbaspour, K., and Xu, Z.X., 2013. A wavelet-neural network hybrid modelling approach for estimating and predicting river monthly flows. Hydrological Sciences Journal, 58 (2), 374–389.     

M5 model tree based modelling of reference evapotranspiration

This paper investigates the potential of M5 model tree based regression approach to model daily reference evapotranspiration using climatic data of Davis station maintained by California irrigation Management Information System (CIMIS). Four inputs including solar radiation, average air temperature, average relative humidity, and average wind speed whereas reference evapotranspiration calculated using a relation provided by the CIMIS was used as output. To compare the performance of M5 model tree in predicting the reference evapotranspiration, FAO–56 Penman–Monteith equation and calibrated Hargreaves–Samani relation was used. A comparison of results suggests that M5 model tree approach works well in comparison to both FAO–56 and calibrated Hargreaves–Samani relations. To judge the generalization capability of M5 model tree approach, model created by using the Davis data set was tested with the datasets of four different sites. Results from this part of the study suggest that M5 model tree could successfully be employed in modeling the reference evapotranspiration. Further, sensitivity analysis with M5 model tree approach suggests the suitability of solar radiation, average air temperature, average relative humidity, and average wind speed as input parameters to model the reference evapotranspiration Copyright © 2009 John Wiley & Sons, Ltd.     

Prediction of hourly actual evapotranspiration using neural networks,genetic programming,and statistical models

The complexity of the evapotranspiration process and its variability in time and space have imposed some limitations on previously developed evapotranspiration models. In this study, two data‐driven models: genetic programming (GP) and artificial neural networks (ANNs), and statistical regression models were developed and compared for estimating the hourly eddy covariance (EC)‐measured actual evapotranspiration (AET) using meteorological variables. The utility of the investigated data‐driven models was also compared with that of HYDRUS‐1D model, which makes use of conventional Penman–Monteith (PM) model for the prediction of AET. The latent heat (LE), which is measured using the EC method, is modelled as a function of five climatic variables: net radiation, ground temperature, air temperature, relative humidity, and wind speed in a reconstructed landscape located in Northern Alberta, Canada. Several ANN models were evaluated using two training algorithms of Levenberg–Marquardt and Bayesian regularization. The GP technique was used to generate mathematical equations correlating AET to the five climatic variables. Furthermore, the climatic variables, as well as their two‐factor interactions, were statistically analysed to obtain a regression equation and to indicate the climatic factors having significant effect on the evapotranspiration process. HYDRUS‐1D model as an available physically based model was examined for estimating AET using climatic variables, leaf area index (LAI), and soil moisture information. The results indicated that all three proposed data‐driven models were able to approximate the AET reasonably well; however, GP and regression models had better generalization ability than the ANN model. The results of HYDRUS‐1D model exhibited that a physically based model, such as HYDRUS‐1D, might be comparable or even inferior to the data‐driven models in terms of the overall prediction accuracy. Based on the developed GP and regression models, net radiation and ground temperature had larger contribution to the AET process than other variables. Copyright © 2010 John Wiley & Sons, Ltd.     

Comparison of the Penman‐Monteith method and regional calibration of the Hargreaves equation for actual evapotranspiration using SWAT-simulated results in the Seolma-cheon basin,South Korea

ABSTRACT

The Hargreaves method provides reference evapotranspiration (ET_o) estimates when only air temperature data are available, although it requires previous local calibration for an acceptable performance. This method was evaluated using the data from 71 meteorological stations in the Seolma-cheon basin (8.48 km²), South Korea, comparing daily estimates against those from the Penman‐Monteith (PM) method, which was used as the standard. To estimate reference ET_o more exactly, considering the climatological characteristics in South Korea, parameter regionalization of the Hargreaves equation is carried out. First, the modified Hargreaves equation is presented after an analysis of the relationship between solar radiation and temperature. Second, parameter (K_ET) optimization of the regional calibration of the Hargreaves equation (RCH) is performed using the PM method and the modified equation at 71 meteorological stations. Next, an application was carried out to evaluate the evapotranspiration methods (PM, original Hargreaves and RCH) in the SWAT (Soil and Water Assessment Tool) model by comparing these with the measured actual evapotranspiration (AET) in the basin. The SWAT model was calibrated using 3 years (2007–2009) of daily streamflow at the watershed outlet and 3 years (2007–2009) of daily AET measured at a mixed forest. The model was validated with 3 years (2010‐2012) of streamflow and AET. RCH will contribute to a better understanding of evapotranspiration of an ungauged watershed in areas where meteorological information is scarce.

EDITOR D. Koutsoyiannis ASSOCIATE EDITOR Not assigned     

Model data selection using gamma test for daily solar radiation estimation

Hydrological modelling is a complicated procedure and there are many tough questions facing all modellers: what input data should be used? how much data is required? and what model should be used? In this paper, the gamma test (GT) has been used for the first time in modelling one of the key hydrological components: solar radiation. The study aimed to resolve the questions about the relative importance of input variables and to determine the optimum number of data points required to construct a reliable smooth model. The proposed methodology has been studied through the estimation of daily solar radiation in the Brue Catchment, the UK. The relationship between input and output in the meteorological data sets was achieved through error variance estimation before the modelling using the GT. This work has demonstrated how the GT helps model development in nonlinear modelling techniques such as local linear regression (LLR) and artificial neural networks (ANN). It was found that the GT provided very useful information for input data selection and subsequent model development. The study has wider implications for various hydrological modelling practices and suggests further exploration of this technique for improving informed data and model selection, which has been a difficult field in hydrology in past decades. Copyright © 2008 John Wiley & Sons, Ltd.     

Modelling suspended sediment concentration using artificial neural networks for Gangotri glacier

The dynamics of suspended sediment involves inherent non‐linearity and complexity because of existence of both spatial variability of the basin characteristics and temporal climatic patterns. This complexity, therefore, leads to inaccurate prediction by the conventional sediment rating curve (SRC) and other empirical methods. Over past few decades, artificial neural networks (ANNs) have emerged as one of the advanced modelling techniques capable of addressing inherent non‐linearity in the hydrological processes. In the present study, feed‐forward back propagation (FFBP) algorithm of ANNs is used to model stage–discharge–suspended sediment relationship for ablation season (May–September) for melt runoff released from Gangotri glacier, one of the largest glaciers in Himalaya. The simulations have been carried out on primary data of suspended sediment concentration (SSC) discharge and stage for ablation season of 11‐year period (1999–2009). Combinations of different input vectors (viz. stage, discharge and SSC) for present and previous days are considered for development of the ANN models and examining the effects of input vectors. Further, based on model performance indices for training and testing phase, a suitable modelling approach with appropriate model input structure is suggested. The conventional SRC method is also used for modelling discharge–sediment relationship and performance of developed models is evaluated by statistical indices, namely; root mean square error (RMSE), mean absolute error (MAE) and coefficient of determination (R²). Statistically, the performance of ANN‐based models is found to be superior as compared to SRC method in terms of the selected performance indices in simulating the daily SSC. The study reveals suitability of ANN approach for simulation and estimation of daily SSC in glacier melt runoff and, therefore, opens new avenues of research for application of hybrid soft computing models in glacier hydrology. Copyright © 2015 John Wiley & Sons, Ltd.     

Artificial neural networks and empirical equations to estimate daily evaporation: application to Lake Vegoritis,Greece

ABSTRACT

Evaporation is one of the most important components in the energy and water budgets of lakes and is a primary process of water loss from their surfaces. An artificial neural network (ANN) technique is used in this study to estimate daily evaporation from Lake Vegoritis in northern Greece and is compared with the classical empirical methods of Penman, Priestley-Taylor and the mass transfer method. Estimation of the evaporation over the lake is based on the energy budget method in combination with a mathematical model of water temperature distribution in the lake. Daily datasets of air temperature, relative humidity, wind velocity, sunshine hours and evaporation are used for training and testing of ANN models. Several input combinations and different ANN architectures are tested to detect the most suitable model for predicting lake evaporation. The best structure obtained for the ANN evaporation model is 4-4-1, with root mean square error (RMSE) from 0.69 to 1.35 mm d^?1 and correlation coefficient from 0.79 to 0.92.

EDITOR M.C. Acreman

ASSOCIATE EDITOR not assigned     

Efficacy of neural network and genetic algorithm techniques in simulating spatio‐temporal fluctuations of groundwater

Groundwater modelling has emerged as a powerful tool to develop a sustainable management plan for efficient groundwater utilization and protection of this vital resource. This study deals with the development of five hybrid artificial neural network (ANN) models and their critical assessment for simulating spatio‐temporal fluctuations of groundwater in an alluvial aquifer system. Unlike past studies, in this study, all the relevant input variables having significant influence on groundwater have been considered, and the hybrid ANN technique [ANN‐cum‐Genetic Algorithm (GA)] has been used to simulate groundwater levels at 17 sites over the study area. The parameters of the ANN models were optimized using a GA optimization technique. The predictive ability of the five hybrid ANN models developed for each of the 17 sites was evaluated using six goodness‐of‐fit criteria and graphical indicators, together with adequate uncertainty analyses. The analysis of the results of this study revealed that the multilayer perceptron Levenberg–Marquardt model is the most efficient in predicting monthly groundwater levels at almost all of the 17 sites, while the radial basis function model is the least efficient. The GA technique was found to be superior to the commonly used trial‐and‐error method for determining optimal ANN architecture and internal parameters. Of the goodness‐of‐fit statistics used in this study, only root‐mean‐squared error, r² and Nash–Sutcliffe efficiency were found to be more powerful and useful in assessing the performance of the ANN models. It can be concluded that the hybrid ANN modelling approach can be effectively used for predicting spatio‐temporal fluctuations of groundwater at basin or subbasin scales. Copyright © 2014 John Wiley & Sons, Ltd.     

Comparison of the Hargreaves and Samani equation and the Thornthwaite equation for estimating dekadal evapotranspiration in the Free State Province,South Africa

Reference evapotranspiration (ET) is an important parameter that needs to be estimated accurately to enhance its utility in numerous applications. Although the widely recommended procedure for calculating this index involves using the FAO Penman–Monteith equation (ETo), the latter’s effectiveness is constrained by its considerable data requirements. To overcome this constraint, alternative methods using the limited data available have to be explored. In this study the ability of the Hargreaves and Samani (ET_HS) and Thornthwaite (ET_T) equations to estimate ET was investigated using multi-year data (1999–2008) from eight weather stations in the semi-arid Free State Province of South Africa. Results for non-calibrated equations are closely correlated, with ET_HS tending to underestimate ET for the July to December period while ET_T underestimates ET for all months of the calendar year. Although estimates from calibrated equations are also closely correlated, they have smaller deviations compared to the original equations with the calibrated Hargreaves and Samani equation (ET_CHS) estimating reference evapotranspiration better than its calibrated Thornthwaite (ET_CT) counterpart. The former’s better performance suggests that in data-scarce areas, the Hargreaves and Samani model is capable of giving results within acceptable ranges of accuracy.     

A Modified Neural Network for Improving River Flow Prediction/Un Réseau de Neurones Modifié pour Améliorer la Prévision de L'Écoulement Fluvial

Abstract

Artificial neural network (ANN) models provide huge potential for simulating nonlinear behaviour of hydrological systems. However, the potential of ANN is yet to be fully exploited due to the problems associated with improving the model generalization performance. Generalization refers to the ability of a neural network to correctly process input data that have not been used for calibrating the neural network model. In the hydrological context, better generalization performance implies higher precision of forecasting. The primary objectives of this study are to explore new measures for improving the generalization performance of an ANN-based rainfall–runoff model, and to evaluate the applicability of the new measures. A modified neural network model (entitled goal programming (GP) neural network) for modelling the rainfall–runoff process has been developed, in which three enhancements are made as compared to the widely-used backpropagation (BP) network. The three enhancements are (a) explicit integration of hydrological prior knowledge into the neural network learning; (b) incorporation of a modified training objective function; and (c) reduction of network sensitivity to input errors. Seven watersheds across a range of climatic conditions and watershed areas in China were selected for examining the alternative networks. The results demonstrate that the GP consistently outperformed the BP both in the calibration and verification periods and three proposed measures yielded improvement of performance.     

A simple temperature method for the estimation of evapotranspiration

Accurate estimation of evapotranspiration (ET) is essential in water resources management and hydrological practices. Estimation of ET in areas, where adequate meteorological data are not available, is one of the challenges faced by water resource managers. Hence, a simplified approach, which is less data intensive, is crucial. The FAO‐56 Penman–Monteith (FAO‐56 PM) is a sole global standard method, but it requires numerous weather data for the estimation of reference ET. A new simple temperature method is developed, which uses only maximum temperature data to estimate ET. Ten class I weather stations data were collected from the National Meteorological Agency of Ethiopia. This method was compared with the global standard PM method, the observed Piche evaporimeter data, and the well‐known Hargreaves (HAR) temperature method. The coefficient of determination (R²) of the new method was as high as 0.74, 0.75, and 0.91, when compared with that of PM reference evapotranspiration (ET_o), Piche evaporimeter data, and HAR methods, respectively. The annual average R² over the ten stations when compared with PM, Piche, and HAR methods were 0.65, 0.67, and 0.84, respectively. The Nash–Sutcliff efficiency of the new method compared with that of PM was as high as 0.67. The method was able to estimate daily ET with an average root mean square error and an average absolute mean error of 0.59 and 0.47 mm, respectively, from the PM ET_o method. The method was also tested in dry and wet seasons and found to perform well in both seasons. The average R² of the new method with the HAR method was 0.82 and 0.84 in dry and wet seasons, respectively. During validation, the average R² and Nash–Sutcliff values when compared with Piche evaporation were 0.67 and 0.51, respectively. The method could be used for the estimation of daily ET_o where there are insufficient data. Copyright © 2013 John Wiley & Sons, Ltd.     

Checks on the measurement of potential evapotranspiration using water balance data and independent measures of groundwater recharge

A twelve-year record of daily evaporation and evapotranspiration measurements at the Coleraine campus of the University of Ulster in Northern Ireland is analysed. Potential evapotranspiration (PE) is independently derived from: (i) Penman PT estimates; (ii) irrigated grass lysimeters PE(L); (iii) measurements of tank evaporation, PE(T). Both PE(T) and PE(L) are higher in winter than PT and have more prolonged summer peaks. Examination of soil moisture deficits during the period shows that actual evapotranspiration (AE) rarely falls below the potential rate and that PE and AE are therefore equal for most of the year. The availability of rainfall, stream discharge and groundwater data from an instrumented river catchment on the University campus enables water balances to be constructed for the period of study. Separate water balances using each of the PE estimates show that Penman PT most satisfactorily reflects catchment storage changes monitored independently. Penman PT is therefore confirmed as the most appropriate estimate of PE for the climatic, soil and vegetation conditions of the region. The use of Penman PT in water balance determinations, however, does not secure perfect agreement between estimated recharge and depletion of catchment storage on the one hand, and observed changes in water-table level on the other. The combined effects of error in surface water balance determinations are estimated at about 13%.     

Spatial estimation of water-table depth by artificial neural networks in light of ancillary data

Proper estimation of the spatial distribution of water-table depth is highly important in most groundwater studies. Groundwater depth is measured at specific and limited points and it is estimated for other parts using spatial estimation methods. In this study, two multivariate methods, artificial neural network (ANN) and multiple linear regression (MLR), are examined to estimate water-table depth in an unconfined aquifer located in Shibkooh, Iran. The different ancillary data, including spatial coordinates, digital elevation model (DEM), aquifer bed elevation, specific resistivity and aquifer thickness were used to improve estimates based on these methods. It was proved that performance of the ANN surpasses that of the MLR method. Using the spatial coordinates, the aquifer bed elevation and aquifer thickness resulted in the optimum spatial estimation of the water-table depth. These parameters, directly or indirectly, affect the water-table depth estimation through techniques such as ANN capable of modelling of nonlinear relationships.     

Modelling hourly and daily open‐water evaporation rates in areas with an equatorial climate

Evaporation rate estimation is important for water resource studies. Previous studies have shown that the radiation‐based models, mass transfer models, temperature‐based models and artificial neural network (ANN) models generally perform well for areas with a temperate climate. This study evaluates the applicability of these models in estimating hourly and daily evaporation rates for an area with an equatorial climate. Unlike in temperate regions, solar radiation was found to correlate best with pan evaporation on both the hourly and daily time‐scales. Relative humidity becomes a significant factor on a daily time‐scale. Among the simplified models, only the radiation‐based models were found to be applicable for modelling the hourly and daily evaporations. ANN models are generally more accurate than the simplified models if an appropriate network architecture is selected and a sufficient number of data points are used for training the network. ANN modelling becomes more relevant when both the energy‐ and aerodynamics‐driven mechanisms dominate, as the radiation and the mass transfer models are incapable of producing reliable evaporation estimates under this circumstance. Copyright © 2006 John Wiley & Sons, Ltd.