MODIS-Based Estimation of Terrestrial Latent Heat Flux over North America Using Three Machine Learning Algorithms

Terrestrial latent heat flux (LE) is a key component of the global terrestrial water, energy, and carbon exchanges. Accurate estimation of LE from moderate resolution imaging spectroradiometer (MODIS) data remains a major challenge. In this study, we estimated the daily LE for different plant functional types (PFTs) across North America using three machine learning algorithms: artificial neural network (ANN); support vector machines (SVM); and, multivariate adaptive regression spline (MARS) driven by MODIS and Modern Era Retrospective Analysis for Research and Applications (MERRA) meteorology data. These three predictive algorithms, which were trained and validated using observed LE over the period 2000–2007, all proved to be accurate. However, ANN outperformed the other two algorithms for the majority of the tested configurations for most PFTs and was the only method that arrived at 80% precision for LE estimation. We also applied three machine learning algorithms for MODIS data and MERRA meteorology to map the average annual terrestrial LE of North America during 2002–2004 using a spatial resolution of 0.05°, which proved to be useful for estimating the long-term LE over North America.

[1]  S. Liang,et al.  Estimating turbulent fluxes through assimilation of geostationary operational environmental satellites data using ensemble Kalman filter , 2011 .

[2]  Jiemin Wang,et al.  Intercomparison of surface energy flux measurement systems used during the HiWATER‐MUSOEXE , 2013 .

[3]  Xiaotong Zhang,et al.  Review on Estimation of Land Surface Radiation and Energy Budgets From Ground Measurement, Remote Sensing and Model Simulations , 2010, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[4]  Peter S. Sephton,et al.  Forecasting recessions: can we do better on MARS? , 2001 .

[5]  A-Xing Zhu,et al.  Prediction of Continental-Scale Evapotranspiration by Combining MODIS and AmeriFlux Data Through Support Vector Machine , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[7]  Ahmed El-Shafie,et al.  Multi-lead ahead prediction model of reference evapotranspiration utilizing ANN with ensemble procedure , 2012, Stochastic Environmental Research and Risk Assessment.

[8]  C. Tucker,et al.  A Global 9-yr Biophysical Land Surface Dataset from NOAA AVHRR Data , 2000 .

[9]  Muhammad Adnan,et al.  Estimating Evapotranspiration using Machine Learning Techniques , 2017 .

[10]  C. Priestley,et al.  On the Assessment of Surface Heat Flux and Evaporation Using Large-Scale Parameters , 1972 .

[11]  S. Running,et al.  Regional evaporation estimates from flux tower and MODIS satellite data , 2007 .

[12]  S. Seneviratne,et al.  Global intercomparison of 12 land surface heat flux estimates , 2011 .

[13]  Shaomin Liu,et al.  Validation of remotely sensed evapotranspiration over the Hai River Basin, China , 2012 .

[14]  J. Freidman,et al.  Multivariate adaptive regression splines , 1991 .

[15]  S. Shukla,et al.  Support vector machine based modeling of evapotranspiration using hydro-climatic variables in a sub-tropical environment , 2015 .

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

[17]  A. Huete,et al.  Overview of the radiometric and biophysical performance of the MODIS vegetation indices , 2002 .

[18]  Qiming Qin,et al.  Simple method to determine the Priestley–Taylor parameter for evapotranspiration estimation using Albedo-VI triangular space from MODIS data , 2011 .

[19]  Eric F. Wood,et al.  Global estimates of evapotranspiration for climate studies using multi-sensor remote sensing data: Evaluation of three process-based approaches , 2011 .

[20]  Donald F. Specht,et al.  A general regression neural network , 1991, IEEE Trans. Neural Networks.

[21]  Maosheng Zhao,et al.  Sensitivity of Moderate Resolution Imaging Spectroradiometer (MODIS) terrestrial primary production to the accuracy of meteorological reanalyses , 2006 .

[22]  Fei Zhang,et al.  Monitoring Soil Salinization in Keriya River Basin, Northwestern China Using Passive Reflective and Active Microwave Remote Sensing Data , 2015, Remote. Sens..

[23]  S. Seneviratne,et al.  Recent decline in the global land evapotranspiration trend due to limited moisture supply , 2010, Nature.

[24]  E. Wood,et al.  Projected changes in drought occurrence under future global warming from multi-model, multi-scenario, IPCC AR4 simulations , 2008 .

[25]  J. Adamowski,et al.  Comparison of multiple linear and nonlinear regression, autoregressive integrated moving average, artificial neural network, and wavelet artificial neural network methods for urban water demand forecasting in Montreal, Canada , 2012 .

[26]  R. W. McClendon,et al.  Estimating daily pan evaporation with artificial neural networks , 2000 .

[27]  Richard D. De Veaux,et al.  Multicollinearity: A tale of two nonparametric regressions , 1994 .

[28]  S. Liang,et al.  Improving Predictions of Water and Heat Fluxes by Assimilating MODIS Land Surface Temperature Products into the Common Land Model , 2011 .

[29]  Hossein Tabari,et al.  Estimation of daily pan evaporation using artificial neural network and multivariate non-linear regression , 2010, Irrigation Science.

[30]  Yi Lin,et al.  Differences in estimating terrestrial water flux from three satellite-based Priestley-Taylor algorithms , 2017, Int. J. Appl. Earth Obs. Geoinformation.

[31]  D. Baldocchi.  ‘Breathing’ of the terrestrial biosphere: lessons learned from a global network of carbon dioxide flux measurement systems , 2008 .

[32]  J. Norman,et al.  Source approach for estimating soil and vegetation energy fluxes in observations of directional radiometric surface temperature , 1995 .

[33]  A. Huete,et al.  Evaluation of optical remote sensing to estimate actual evapotranspiration and canopy conductance , 2013 .

[34]  L. S. Pereira,et al.  Crop evapotranspiration : guidelines for computing crop water requirements , 1998 .

[35]  Eric F. Wood,et al.  Multi‐model, multi‐sensor estimates of global evapotranspiration: climatology, uncertainties and trends , 2011 .

[36]  Alan H. Strahler,et al.  Global land cover mapping from MODIS: algorithms and early results , 2002 .

[37]  Seung Oh Lee,et al.  Validation of MODIS 16 global terrestrial evapotranspiration products in various climates and land cover types in Asia , 2012 .

[38]  Maurice R Puyau,et al.  Validation of cross-sectional time series and multivariate adaptive regression splines models for the prediction of energy expenditure in children and adolescents using doubly labeled water. , 2010, The Journal of nutrition.

[39]  Shaohua Zhao,et al.  onitoring global land surface drought based on a hybrid vapotranspiration model , 2011 .

[40]  Maosheng Zhao,et al.  Improvements to a MODIS global terrestrial evapotranspiration algorithm , 2011 .

[41]  P. Ciais,et al.  Assimilation of global MODIS leaf area index retrievals within a terrestrial biosphere model , 2007 .

[42]  J. Monteith.  Evaporation and environment. , 1965, Symposia of the Society for Experimental Biology.

[43]  Vladimir Vapnik,et al.  The Nature of Statistical Learning , 1995 .

[44]  Maosheng Zhao,et al.  Development of a global evapotranspiration algorithm based on MODIS and global meteorology data , 2007 .

[45]  Zhanqing Li,et al.  A simple method to estimate actual evapotranspiration from a combination of net radiation, vegetation index, and temperature , 2007 .

[46]  Shaomin Liu,et al.  A comparison of eddy-covariance and large aperture scintillometer measurements with respect to the energy balance closure problem , 2011 .

[47]  Maosheng Zhao,et al.  Improvements of the MODIS terrestrial gross and net primary production global data set , 2005 .

[48]  S. Liang,et al.  Improving global terrestrial evapotranspiration estimation using support vector machine by integrating three process-based algorithms. , 2017 .

[49]  R. Deo,et al.  Estimation of monthly evaporative loss using relevance vector machine, extreme learning machine and multivariate adaptive regression spline models , 2016, Stochastic Environmental Research and Risk Assessment.

[50]  J. Randerson,et al.  Continental-scale net radiation and evapotranspiration estimated using MODIS satellite observations , 2011 .

[51]  J. Norman,et al.  Correcting eddy-covariance flux underestimates over a grassland , 2000 .

[52]  R. Dickinson,et al.  A review of global terrestrial evapotranspiration: Observation, modeling, climatology, and climatic variability , 2011 .

[53]  M. Taillefer,et al.  Highly efficient and mild copper-catalyzed N- and C-arylations with aryl bromides and iodides. , 2004, Chemistry.

[54]  Xiaoliang Lu,et al.  Evaluating evapotranspiration and water-use efficiency of terrestrial ecosystems in the conterminous United States using MODIS and AmeriFlux data , 2010 .

[55]  Dimitri P. Solomatine,et al.  Model Induction with Support Vector Machines: Introduction and Applications , 2001 .

[56]  D. Baldocchi,et al.  Global estimates of the land–atmosphere water flux based on monthly AVHRR and ISLSCP-II data, validated at 16 FLUXNET sites , 2008 .

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

[58]  Jia Xu,et al.  Satellite-Derived Spatiotemporal Variations in Evapotranspiration over Northeast China during 1982-2010 , 2017, Remote. Sens..

[59]  William P. Kustas,et al.  Use of remote sensing for evapotranspiration monitoring over land surfaces , 1996 .