Reference Evapotranspiration Prediction Using Neural Networks and Optimum Time Lags

The reference evapotranspiration (ET₀) plays a significant role especially in agricultural water management and water resources planning for irrigation. It can be calculated using different empirical equations and forecasted by applying various artificial intelligence techniques. The simulation result of a machine learning technique is a function of its structure and model inputs. The purpose of this study is to investigate the effect of using the optimum set of time lags for model inputs on the prediction accuracy of monthly ET₀ using an artificial neural network (ANN). For this, the weather data time-series i.e. minimum and maximum air temperatures, vapour pressure, sunshine hours, and wind speed were collected from six meteorological stations in Serbia for the period 1980–2010. Three ANN models were applied to monthly ET₀ time-series to study the impacts of using the optimum time lags for input time-series on the performance of ANN model. Achieved results of goodness–of–fit statistics approved the results obtained by scatterplots of testing sets - using more time lags that are selected based on their correlation to the dataset is more efficient for monthly ET₀ prediction. It was realized that all the developed models showed the best performances at Loznica and Vranje stations and the worst performances at Nis station. Simultaneous assessment of the impact of using a different number of time lags and the set of time lags that show a stronger correlation to the dataset for input time-series, on the performance of ANN model in monthly ET₀ prediction in Serbia is the novelty of this study.

     

基于模糊神经网络的参考作物蒸腾量预测研究

为准确估算灌区作物需水量,建立了基于模糊神经网络的参考作物腾发量时间序列预测模型。采用宝鸡地区1954—2004年逐月气象资料,利用主成分分析法提取影响参考作物蒸发蒸腾量的主要影响因子,获得4个综合变量作为输入向量,用彭曼-蒙蒂斯公式计算的参考作物蒸发蒸腾量作为目标向量。运用matlab进行编程,应用模糊神经网络模型预测参考作物腾发量。结果表明:12组测试集样本的平均相对误差绝对值为5%,最大相对误差为11.4%,最小相对误差为0.4%;模糊神经网络模型与用彭曼-蒙蒂斯公式计算值有很高的一致性。     

Developing Equations for Estimating Reference Evapotranspiration in Australia

Quantifying reference evapotranspiration (ET₀) is essential in water resources management. Although, many methods have been developed with different level of accuracy, in this study, two new equations were developed and optimized for estimating ET₀ using Honey-Bee Mating Optimization (HBMO) algorithm. The first eq. estimates ET₀ from extraterrestrial radiation (R_a), relative humidity (RH) and mean daily temperature (T_mean), while the second uses the same parameters except that mean daily temperatures is replaced with maximum daily air temperature (T_max). Both equations were developed using climatic data from eight weather stations in Western Australia and subsequently verified using data from ten sites across Australia. The estimated ET₀ values from both equations versus the FAO56-Penman-Monteith have a coefficient of determination, R², of larger than 0.96. Moreover, the performance of six commonly used methods of estimating ET₀ including Hargreaves-Samani, Thornthwaith, Hamon, Mc Guinness-Bordne, Irmak and Jensen-Haise were assessed and the Hargreaves-Samani method performed better than others. An attempt was made to calibrate the Hargreaves-Samani equation; however, its overall performance did not improved and the two newly proposed equations are suggested to be used in Australia.     

Performance of Recalibrated Equations for the Estimation of Daily Reference Evapotranspiration

An accurate estimation of reference evapotranspiration (ET₀) is of paramount importance for many studies such as hydrologic water balance, irrigation system design and management, crop yield simulation, and water resources planning and management. In the present study, Blaney-Criddle, Jensen-Haise and Hargreaves (temperature based), Priestley-Taylor, Radiation and Makkink (radiation based) and, Pan Evaporation and Christiansen (pan evaporation based) methods have been evaluated and recalibrated with respect to FAO-56 Penman-Monteith method for estimating daily ET₀ in the semi-arid Tirupati, Nellore, Rajahmundry, Anakapalli and Rajendranagar sites of Andhra Pradesh, India. Recalibrated Blaney-Criddle (temperature based), Radiation (Radiation based) and Christiansen (Pan evaporation based) methods showed a satisfactory performance at the sites. Further, recalibrated Blaney- Criddle method showed relatively better performance than Radiation and Christiansen methods in the daily ET₀ estimation. Recalibrated Blaney- Criddle method may therefore be adopted at the sites selected for the present study and also at the sites with similar climatic conditions for satisfactory daily ET₀ estimation.     

Hierarchical Fuzzy Systems Integrated with Particle Swarm Optimization for Daily Reference Evapotranspiration Prediction: a Novel Approach

Water Resources Management - Reference evapotranspiration (ET0) is a crucial element for deriving irrigation scheduling of major crops. Thus, precise projection of ET0 is essential for better...     

Framework for Standardizing Less Data-Intensive Methods of Reference Evapotranspiration Estimation

Evapotranspiration is one of the vital components of water cycle and its accurate estimation is the key to sustainable management of irrigation water. The FAO Penman-Monteith (FAO-PM) method is recommended as the standard method for computing reference evapotranspiration (ET_o) as well as for evaluating other indirect methods. However, due to the lack of weather data such as radiation, relative humidity and wind speed in many regions of the world, especially in developing countries, the FAO-PM method is difficult to use. To address this issue, a fairly robust methodology is proposed in this study to standardize two popular less data-intensive (temperature-based) ET_o methods, viz., Hargreaves-Samani (HS) and Penman-Monteith Temperature (PMT) against the FAO-PM method. To achieve this goal, the daily and monthly biases of these two methods were adjusted using the weather data of 14 locations for the 1979–2003 period. Subsequently, the performance of the standardized (de-biased) less data-intensive methods were verified using salient statistical and graphical indicators for the 2004–2013 period. The results indicated that the HS and PMT methods underestimate ET_o on a monthly time step by 9.62 and 14.77%, respectively. However, the performances of these methods significantly improve after the standardization. The estimates of ET_o by the standardized less data-intensive methods were found to be in close agreement with those by the standard FAO-PM method, thereby suggesting the usefulness and applicability of the proposed framework in data-scarce situations irrespective of agro-climatic conditions.     

Evaluation of Reference Evapotranspiration Equations Under Humid Conditions

Five reference evapotranspiration (ET₀) equations are evaluated using data from seven humid locations. The equations evaluated include Hargreaves, Thornthwaite, Turc, Priestley–Taylor, and Jensen–Haise. The objective of this study is to evaluate ET₀ estimated by these equations against the corresponding values estimated using the standardized FAO-56 Penman–Monteith (PM) equation. For each location, ET₀ estimates by the all equations were statistically compared with FAO-56 PM ET₀ estimates. The Turc equation yielded the smallest root-mean-square-difference (RMSD) values at the all locations except Novi Sad, Serbia. The final ranking of equations was based on the weighted RMSD. The Turc equation has the lowest weighted RMSD and ranking first, and other equations ranked in decreasing order are: Priestley–Taylor, Jensen–Haise, Thornthwaite, and Hargreaves. The Turc equation gives the reliable calculation at all humid locations and it has proven to be the most adjustable to the local climatic conditions. The results obtained from this study, indicate very clearly that the Turc equation is most suitable for estimating reference evapotranspiration at humid locations when weather data are insufficient to apply the FAO-56 PM equation.     

Multiple Linear Correlation Analysis of Daily Reference Evapotranspiration

An accurate estimation of reference evapotranspiration (ET₀) is of paramount importance for many studies such as hydrologic water balance, irrigation system design and management, crop yield simulation, and water resources planning and management. Simple regression techniques, may sometimes, provide adequate estimation of ET₀. Implementation of regression methods considering all the predictor variables may, however, lead to overfit and consequent reduction in the predictive capability. The regression models for ET₀ have been developed in the present study for Tirupati, Nellore, Rajahmundry, Anakapalli and Rajendranagar regions of Andhra Pradesh, India by following step-wise procedure, eliminating superfluous predictor variables based on statistical criteria. The sunshine hours, wind velocity, temperature and relative humidity influenced ET₀ in the study area. The linear regression models developed in terms of predictor variables may conveniently be applied in the regions selected for the present study and, in the regions with similar climatic conditions for satisfactory ET₀ estimation.     

Estimation of Monthly Mean Reference Evapotranspiration in Turkey

Monthly mean reference evapotranspiration (ET ₀ ) is estimated using adaptive network based fuzzy inference system (ANFIS) and artificial neural network (ANN) models. Various combinations of long-term average monthly climatic data of wind speed, air temperature, relative humidity, and solar radiation, recorded at stations in Turkey, are used as inputs to the ANFIS and ANN models so as to calculate ET ₀ given by the FAO-56 PM (Penman-Monteith) equation. First, a comparison is made among the estimates provided by the ANFIS and ANN models and those by the empirical methods of Hargreaves and Ritchie. Next, the empirical models are calibrated using the ET ₀ values given by FAO-56 PM, and the estimates by the ANFIS and ANN techniques are compared with those of the calibrated models. Mean square error, mean absolute error, and determination coefficient statistics are used as comparison criteria for evaluation of performances of all the models considered. Based on these evaluations, it is found that the ANFIS and ANN schemes can be employed successfully in modeling the monthly mean ET ₀ , because both approaches yield better estimates than the classical methods, and yet ANFIS being slightly more successful than ANN.     

石河子参考作物潜在腾发量的多元线性回归分析

利用FAO Penman-Monteith(1992)公式,根据新疆生产建设兵团农八师148团1998年～2008年(除2001年)4月～8月每日的气象资料,建立了参考作物潜在腾发量与其它气象要素的相关关系.利用这些相关关系进行参考作物潜在蒸发蒸腾量的估算,结果表明其精度较高,方法简单.可为今后灌溉管理和产量评估提供较为简便实用的模型.     

Fuzzy Genetic Approach for Estimating Reference Evapotranspiration of Turkey: Mediterranean Region

The applicability of fuzzy genetic (FG) approach in modeling reference evapotranspiration (ET₀) is investigated in this study. Daily solar radiation, air temperature, relative humidity and wind speed data of two stations, Isparta and Antalya, in Mediterranean region of Turkey, are used as inputs to the FG models to estimate ET₀ obtained using the FAO-56 Penman–Monteith equation. The FG estimates are compared with those of the artificial neural networks (ANN). Root mean-squared error, mean absolute error and determination coefficient statistics were used as comparison criteria for the evaluation of the models’ accuracies. It was found that the FG models generally performed better than the ANN models in modeling ET₀ of Mediterranean region of Turkey.     

Evaluation of Reference Crop Evapotranspiration Equations in Various Climates

Evapotranspiration is one of the most important elements for quantifying available water since it generally constitutes the largest component of the terrestrial water cycle. This study evaluated four models (Makkink, Turc, Priestley–Taylor and Hargreaves) commonly used to estimate monthly reference crop evapotranspiration (ET_o) values. The main aim of this study was to determine the model used to estimate ET_o with small data requirements and high accuracy for twelve synoptic stations in four climates of Iran. The results showed that the Turc model was the best suited model in estimating ET_o for cold humid and arid climates. The Hargreaves model turned out to be the most precise model under warm humid and semi-arid climatic conditions. In contrast, the Makkink model presented the poorest estimates in all of the climates exception for cold humid environment. In cold humid climate, the Hargreaves model was the least accurate model in estimating ET_o. In general, the results obtained from this study revealed very clearly that the Makkink and Priestley–Taylor models estimated ET_o values less accurately than Turc and Hargreaves models for the all climates.     

Assessment of Blaney-Criddle Equation for Calculating Reference Evapotranspiration with NOAA/AVHRR Data

Reference evapotranspiration (ET₀) data are desirable for assessing crop water requirements and irrigation needs. A large number of methods have been developed for assessing ET₀ from meteorological data. In several places of the world, the existing network of weather stations is insufficient to capture the spatial heterogeneity of this variable The purpose of this work is to investigate whether it is possible to attain reliable estimation of ET₀ only on the basis of the remote sensing-based surface temperature (T_s) data by Blaney-Criddle (B-C) model under a semi arid environment of Iran. This study has assumed that the daytime surface temperature at the cold pixel obtained from the AVHRR/NOAA sensor can be used instead of air temperature in the Blaney-Criddle (BC) equation for ET₀ estimation in irrigated area. For this purpose, 61 NOAA- AVHRR satellite images acquired between June and September in 2004 and 2005 and weather data measured at two weather located in two irrigation regions with sugar cane located in Khuzestan plain in the southwest of Iran were used to calibrate and test the B-C model. The FAO-56 Penman–Monteith model was used as a reference model for assessing the performance of the calibrated BC model. The results show that calibrated B-C model provided close agreement with the reference values, with an average RMSE of 1.0 mm day^?1and a R² of 0.91.     

参考作物蒸腾量的多重分形特性分析

分析ET(Evapotranspiration,参考作物蒸腾量)的动力学特性,有助于进行中长期需水量的分析与预测。将多重分形特性分析方法应用于1978-2007年间30 a汉江流域3个典型站点(钟祥、天门、武汉)的参考作物ET时间序列。结果表明,逐日参考作物ET序列不仅具有不规则的高频振荡的特征,而且还具有明显的分形行为;在不同时间宽度(日、旬、月)下,逐日参考作物ET序列的多重分形特征最强。进一步的分析结果表明,序列中脉动引起的波动相关性和极端事件引起的胖尾概率等都是引起多重分形特征形成的原因。结合趋势转折分析方法发现:不同时间阶段内的多重分形特征显著;但在不同的时间段内,多重分形谱和局部分维宽度等都受到了极端事件的影响,且影响幅度与所处流域内的位置有关。     

Optimized Neural Network Prediction Model for Potential Evapotranspiration Utilizing Ensemble Procedure

Potential evapotranspiration (ET_o) is an essential hydrologic parameter for having better understanding for hydrologic cycle in certain catchment area. In addition, ET_o is vital for calculating the agricultural demand. In fact, Penman-Monteith (PM) method is considered as reference method for estimating (ET_o), however, this method required a lot of data to be used which is not usually available in many catchment areas. Furthermore, there are several efforts that have been performed as competitor to reach accurate estimation of (ET_o) when there is lack of data to utilize (PM) method, but still required numerous research. Recently, methods based on Artificial Intelligence (AI) have been suggested to provide reliable prediction model for several application in engineering and especially for hydrological process. However, time series prediction based on Artificial Neural Network (ANN) learning algorithms is fundamentally difficult and faces problem. One of the major shortcomings is that the ANN model experiences over-fitting problem during training session and also occurs when a neural network loses its generalization. In this research a modification for the classical Multi Layer Preceptron- Artificial Neural Network (MLP-ANN) modeling namely; Ensemble Neural Network (ENN) is proposed and applied for predicting daily ET_o. The proposed model applied at two different region with two different climatic conditions, Rasht city located north part of Iran and Johor Bahru City, Johor, Malaysia using maximum and minimum daily temperature collected from 1975 to 2005. The result showed that the ENN outperformed the classical MLP-ANN method and successfully predict daily ETo utilizing maximum and minimum temperature only with satisfactory level of accuracy. In addition, the proposed model could achieve accuracy level better than the traditional competitor methods for ET_o.     

Gene-Expression Programming for Short-Term Forecasting of Daily Reference Evapotranspiration Using Public Weather Forecast Information

This study aimed to forecast the daily reference evapotranspiration (ETo) using a gene-expression programming (GEP) algorithm with limited public weather forecast information over Gaoyou station, located in Jiangsu province, China. To calibrate and validate the gene-expression code, important meteorological data and weather forecast information were collected from the local meteorological station and public weather media, respectively. The GEP algebraic formulation was successfully constructed based only on daily minimum and maximum air temperature using the true FAO56 Penman-Monteith (PM) set as reference values. The performance of the models was then assessed using the correlation coefficient (R), root mean squared error (RMSE), root relative squared error (RRSE) and mean absolute error (MAE). The study demonstrated that GEP is able to calibrate ETo (all errors ≤0.990 mm/day, R = 0.832–0.866) and forecast the daily ETo with good accuracy (RMSE = 1.207 mm/day, MAE = 0.902 mm/day, RRSE = 0.629 mm/day, R = 0.777). The model accuracies slightly decreased over a 7-day forecast lead-time. These results suggest that the GEP algorithm can be considered as a deployable tool for ETo forecast to anticipate decision on short-term irrigation schedule in the study zone.     

An Evaluation of Four Reference Evapotranspiration Models in a Subtropical Climate

This paper describes a detailed evaluation of the performance and characteristic behaviour of four equations for estimating reference evapotranspiration (ET₀) at eight meteorological sites in a subtropical climate. The equations assessed were: Makkink (MK), Turc (TC), Priestley?CTaylor (PT) and Hargreaves (HG). The sites were distributed throughout the north of Iran and represent an intermediate humidity regime. The Penman?CMonteith (PM) method was chosen as the standard for comparison and calibration of the above??mentioned four equations. Good correlation was found between the ET₀ values estimated by each of the four empirical equations and the PM method for all the locations; however MK and TC equations produced considerable underestimations. The performance of the PT and HG equations improved slightly after region-specific coefficients were developed for each equation, and the TC and MK equations were improved greatly. The modified PT equation turned out to be the most precise method, demonstrating superiority over the other methods evaluated (0.48?mm?d^?1 of root mean square error (RMSE)). Good performance from the modified HG equation (0.53?mm?d^?1 of RMSE) must be emphasized, given the simplicity of that method, which only requires maximum and minimum air temperature data.     

Utilizing RBF-NN and ANFIS Methods for Multi-Lead ahead Prediction Model of Evaporation from Reservoir

Evaporation as a major meteorological component of the hydrologic cycle plays a key role in water resources studies and climate change. The estimation of evaporation is a complex and unsteady process, so it is difficult to derive an accurate physical-based formula to represent all parameters that effect on estimate evaporation. Artificial intelligence-based methods may provide reliable prediction models for several applications in engineering. In this research have been introduced twelve networks with the RBF-NN and ANFIS methods. These models have applied to prediction daily evaporation at Layang reservoir, located in the southeast part of Malaysia. The used meteorological data set to develop the models for prediction daily evaporation rate from water surface for Layang reservoir includes daily air temperature, solar radiation, pan evaporation, and relative humidity that measured at a case study for fourteen years. The obtained result denote to the superiority of the RBF-NN models on the ANFIS models. A comparison of the model performance between RBF-NN and ANFIS models indicated that RBF-NN method presents the best estimates of daily evaporation rate with the minimum MSE 0.0471 , MAE 0.0032, RE and maximum R² 0.963.     

Prediction as a Basis for Planning and Response

Abstract

Enhanced management of water systems requires knowledge of the amounts of water availability on time scales of hours to decades. Managers of water systems can use the information from combined hydro-meteorological/hydrologic prediction models to take appropriate actions in a realtime basis and also to plan the development of both structural and non-structural measures. Cooperation between the meteorological and hydrological communities is leading to improved models, from flash floods to climate change scenarios but there is a continuing need for dialogue to understand the needs and capabilities of each community. This paper reviews the role of prediction and discusses different kinds of predictions: deterministic, probabilistic, and scenarios based on external forcing. An important issue is the factoring in of the uncertainties of predictions into a risk management approach to water systems management.