Reference evapotranspiration forecasting based on local meteorological and global climate information screened by partial mutual information

Abstract In this study, reference evapotranspiration (ET0) forecasting models are developed for the least economically developed regions subject to meteorological data scarcity. Firstly, the partial mutual information (PMI) capable of capturing the linear and nonlinear dependence is investigated regarding its utility to identify relevant predictors and exclude those that are redundant through the comparison with partial linear correlation. An efficient input selection technique is crucial for decreasing model data requirements. Then, the interconnection between global climate indices and regional ET0 is identified. Relevant climatic indices are introduced as additional predictors to comprise information regarding ET0, which ought to be provided by meteorological data unavailable. The case study in the Jing River and Beiluo River basins, China, reveals that PMI outperforms the partial linear correlation in excluding the redundant information, favouring the yield of smaller predictor sets. The teleconnection analysis identifies the correlation between Nino 1 + 2 and regional ET0, indicating influences of ENSO events on the evapotranspiration process in the study area. Furthermore, introducing Nino 1 + 2 as predictors helps to yield more accurate ET0 forecasts. A model performance comparison also shows that non-linear stochastic models (SVR or RF with input selection through PMI) do not always outperform linear models (MLR with inputs screen by linear correlation). However, the former can offer quite comparable performance depending on smaller predictor sets. Therefore, efforts such as screening model inputs through PMI and incorporating global climatic indices interconnected with ET0 can benefit the development of ET0 forecasting models suitable for data-scarce regions.

[1]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[2]  Hung Soo Kim,et al.  Neural networks and genetic algorithm approach for nonlinear evaporation and evapotranspiration modeling , 2008 .

[3]  A. Sullivan,et al.  Asymmetry in ENSO Teleconnection with Regional Rainfall, Its Multidecadal Variability, and Impact , 2010 .

[4]  Guohe Huang,et al.  Discrete principal‐monotonicity inference for hydro‐system analysis under irregular nonlinearities, data uncertainties, and multivariate dependencies. Part I: methodology development , 2016 .

[5]  K. P. Sudheer,et al.  Models for estimating evapotranspiration using artificial neural networks, and their physical interpretation , 2008 .

[6]  Holger R. Maier,et al.  Review of Input Variable Selection Methods for Artificial Neural Networks , 2011 .

[7]  J. Arnold,et al.  HYDROLOGICAL MODELING OF THE IROQUOIS RIVER WATERSHED USING HSPF AND SWAT 1 , 2005 .

[8]  Mohammad Karamouz,et al.  Input data selection for solar radiation estimation , 2009 .

[9]  Sarel van Vuuren,et al.  Relevance of time-frequency features for phonetic and speaker-channel classification , 2000, Speech Commun..

[10]  Amin Elshorbagy,et al.  Modelling the dynamics of the evapotranspiration process using genetic programming , 2007 .

[11]  S. Chatterjee,et al.  Influential Observations, High Leverage Points, and Outliers in Linear Regression , 1986 .

[12]  Yashar Falamarzi,et al.  Estimating evapotranspiration from temperature and wind speed data using artificial and wavelet neural networks (WNNs) , 2014 .

[13]  Narendra Singh Raghuwanshi,et al.  Estimating Evapotranspiration using Artificial Neural Network , 2002 .

[14]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[15]  Ozgur Kisi,et al.  Evapotranspiration estimation using feed-forward neural networks , 2006 .

[16]  D. Budikova,et al.  Eastern U.S. summer streamflow during extreme phases of the North Atlantic oscillation , 2013 .

[17]  Yutong Chen,et al.  Spatio-temporal Changes and Frequency Analysis of Drought in the Wei River Basin, China , 2014, Water Resources Management.

[18]  Vladimir Vapnik,et al.  Support-vector networks , 2004, Machine Learning.

[19]  Aris Psilovikos,et al.  Forecasting of Remotely Sensed Daily Evapotranspiration Data Over Nile Delta Region, Egypt , 2013, Water Resources Management.

[20]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[21]  F. Anctil,et al.  Teleconnections and interannual variability in Canadian groundwater levels , 2011 .

[22]  Soroosh Sorooshian,et al.  Developing reservoir monthly inflow forecasts using artificial intelligence and climate phenomenon information , 2017 .

[23]  Shengzhi Huang,et al.  Spatial-temporal changes of maximum and minimum temperatures in the Wei River Basin, China: Changing patterns, causes and implications , 2018 .

[24]  Jun Yu Li,et al.  Long‐term trend of precipitation in China and its association with the El Niño–southern oscillation , 2007 .

[25]  P. Willems,et al.  Possible influences of North Atlantic Oscillation on winter reference evapotranspiration in Iran , 2014 .

[26]  Guy Fipps,et al.  Deployment of artificial neural network for short-term forecasting of evapotranspiration using public weather forecast restricted messages , 2016 .

[27]  Fraser,et al.  Independent coordinates for strange attractors from mutual information. , 1986, Physical review. A, General physics.

[28]  Quan J. Wang,et al.  Evidence for Using Lagged Climate Indices to Forecast Australian Seasonal Rainfall , 2012 .

[29]  O. Kisi,et al.  SVM, ANFIS, regression and climate based models for reference evapotranspiration modeling using limited climatic data in a semi-arid highland environment , 2012 .

[30]  Ronald Harley,et al.  A random forest method for real-time price forecasting in New York electricity market , 2014, 2014 IEEE PES General Meeting | Conference & Exposition.

[31]  Ozgur Kisi,et al.  Generalizability of Gene Expression Programming-based approaches for estimating daily reference evapotranspiration in coastal stations of Iran , 2014 .

[32]  F. Meza.  Variability of reference evapotranspiration and water demands. Association to ENSO in the Maipo river basin, Chile , 2005 .

[33]  Turgay Partal,et al.  Comparison of wavelet based hybrid models for daily evapotranspiration estimation using meteorological data , 2016 .

[34]  Alberto de la Fuente,et al.  Discovery of meaningful associations in genomic data using partial correlation coefficients , 2004, Bioinform..

[35]  Lakshman Nandagiri,et al.  Performance Evaluation of Reference Evapotranspiration Equations across a Range of Indian Climates , 2006 .

[36]  Daily Reference Evapotranspiration Estimation using Linear Regression and ANN Models , 2012 .

[37]  Hossein Tabari,et al.  ENSO teleconnection impacts on reference evapotranspiration variability in some warm climates of Iran , 2011 .

[38]  Yoshihide Tominaga,et al.  Air flow around isolated gable-roof buildings with different roof pitches: Wind tunnel experiments and CFD simulations , 2015 .

[39]  Jan Adamowski,et al.  Bootstrap rank‐ordered conditional mutual information (broCMI): A nonlinear input variable selection method for water resources modeling , 2016 .

[40]  Dawei Han,et al.  Model data selection using gamma test for daily solar radiation estimation , 2008 .

[41]  Shan Suthaharan,et al.  Support Vector Machine , 2016 .

[42]  Ramón Díaz-Uriarte,et al.  Gene selection and classification of microarray data using random forest , 2006, BMC Bioinformatics.

[43]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[44]  S. Alexandris,et al.  Solar radiation and relative humidity based, empirical method, to estimate hourly reference evapotranspiration , 2015 .

[45]  Johannes R. Sveinsson,et al.  Random Forests for land cover classification , 2006, Pattern Recognit. Lett..

[46]  Ashish Sharma,et al.  Seasonal to interannual rainfall probabilistic forecasts for improved water supply management: Part 1 — A strategy for system predictor identification , 2000 .

[47]  Yoshiki Saito,et al.  Interannual and seasonal variation of the Huanghe (Yellow River) water discharge over the past 50 years: Connections to impacts from ENSO events and dams , 2006 .

[48]  D. Jato-Espino,et al.  Prediction of Evapotranspiration in a Mediterranean Region Using Basic Meteorological Variables , 2016 .

[49]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[50]  Holger R. Maier,et al.  Non-linear variable selection for artificial neural networks using partial mutual information , 2008, Environ. Model. Softw..

[51]  J. Holden,et al.  Comparison of soil erosion models used to study the Chinese Loess Plateau , 2017 .

[52]  Vahid Nourani,et al.  Application of Entropy Concept for Input Selection of Wavelet-ANN Based Rainfall-Runoff Modeling , 2016 .

[53]  Soroosh Sorooshian,et al.  Simulating California reservoir operation using the classification and regression‐tree algorithm combined with a shuffled cross‐validation scheme , 2015 .

[54]  Richard G. Allen,et al.  Estimating Reference Evapotranspiration Under Inaccurate Data Conditions , 2002 .

[55]  Slavisa Trajkovic,et al.  Comparative analysis of 31 reference evapotranspiration methods under humid conditions , 2011, Irrigation Science.

[56]  Chong-Yu Xu,et al.  Possible influence of ENSO on annual maximum streamflow of the Yangtze River, China , 2007 .

[57]  S. Sorooshian,et al.  Effective and efficient global optimization for conceptual rainfall‐runoff models , 1992 .

[58]  Jie Yang,et al.  Optimal sizing of utility-scale photovoltaic power generation complementarily operating with hydropower: A case study of the world’s largest hydro-photovoltaic plant , 2017 .

[59]  Guohe Huang,et al.  A stepwise-cluster forecasting approach for monthly streamflows based on climate teleconnections , 2015, Stochastic Environmental Research and Risk Assessment.

[60]  Shengzhi Huang,et al.  Identification of the non-stationarity of extreme precipitation events and correlations with large-scale ocean-atmospheric circulation patterns: A case study in the Wei River Basin, China , 2017 .

[61]  Airfares 2002Q,et al.  MULTIPLE LINEAR REGRESSION , 2006, Statistical Methods for Biomedical Research.

[62]  S. Sorooshian,et al.  Shuffled complex evolution approach for effective and efficient global minimization , 1993 .