Interpolation of Instantaneous Air Temperature Using Geographical and MODIS Derived Variables with Machine Learning Techniques

Several methods have been tried to estimate air temperature using satellite imagery. In this paper, the results of two machine learning algorithms, Support Vector Machines and Random Forest, are compared with Multiple Linear Regression and Ordinary kriging. Several geographic, remote sensing and time variables are used as predictors. The validation is carried out using two different approaches, a leave-one-out cross validation in the spatial domain and a spatio-temporal k-block cross-validation, and four different statistics on a daily basis, allowing the use of ANOVA to compare the results. The main conclusion is that Random Forest produces the best results (R2 = 0.888 ± 0.026, Root mean square error = 3.01 ± 0.325 using k-block cross-validation). Regression methods (Support Vector Machine, Random Forest and Multiple Linear Regression) are calibrated with MODIS data and several predictors easily calculated from a Digital Elevation Model. The most important variables in the Random Forest model were satellite temperature, potential irradiation and cdayt, a cosine transformation of the julian day.

[1]  J. Cristóbal,et al.  Modeling air temperature through a combination of remote sensing and GIS data , 2008 .

[2]  Emilio Chuvieco,et al.  Air temperature estimation with MSG-SEVIRI data: Calibration and validation of the TVX algorithm for the Iberian Peninsula , 2011 .

[3]  Marion Schroedter-Homscheidt,et al.  Parameterization of air temperature in high temporal and spatial resolution from a combination of the SEVIRI and MODIS instruments , 2009 .

[4]  Gerard B. M. Heuvelink,et al.  About regression-kriging: From equations to case studies , 2007, Comput. Geosci..

[5]  Guo Yu Qiu,et al.  An evapotranspiration product for arid regions based on the three-temperature model and thermal remote sensing , 2015 .

[6]  Gautam Bisht,et al.  Estimation of net radiation from the MODIS data under all sky conditions: Southern Great Plains case study , 2010 .

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

[8]  Shaofeng Jia,et al.  Retrievals of all-weather daytime air temperature from MODIS products , 2017 .

[9]  Jane Elith,et al.  blockCV: an R package for generating spatially or environmentally separated folds for k-fold cross-validation of species distribution models , 2018, bioRxiv.

[10]  G. Heuvelink,et al.  A generic framework for spatial prediction of soil variables based on regression-kriging , 2004 .

[11]  P. Ceccato,et al.  Evaluation of MODIS land surface temperature data to estimate air temperature in different ecosystems over Africa , 2010 .

[12]  J. S. Long,et al.  Using Heteroscedasticity Consistent Standard Errors in the Linear Regression Model , 2000 .

[13]  Andreas Ziegler,et al.  ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and R , 2015, 1508.04409.

[14]  Xianzhou Zhang,et al.  Elevation-dependent temperature change in the Qinghai–Xizang Plateau grassland during the past decade , 2014, Theoretical and Applied Climatology.

[15]  Kyung-Soo Han,et al.  Remotely sensed retrieval of midday air temperature considering atmospheric and surface moisture conditions , 2013 .

[16]  R. O’Brien,et al.  A Caution Regarding Rules of Thumb for Variance Inflation Factors , 2007 .

[17]  Renhua Zhang,et al.  A Remote Sensing Method for Estimating Surface Air Temperature and Surface Vapor Pressure on a Regional Scale , 2015, Remote. Sens..

[18]  A. McBratney,et al.  Further results on prediction of soil properties from terrain attributes: heterotopic cokriging and regression-kriging , 1995 .

[19]  V. M. Chowdary,et al.  Reconstruction of time series MODIS EVI data using de-noising algorithms , 2018 .

[20]  S. Goward,et al.  Estimation of air temperature from remotely sensed surface observations , 1997 .

[21]  D. Legates,et al.  Evaluating the use of “goodness‐of‐fit” Measures in hydrologic and hydroclimatic model validation , 1999 .

[22]  Edzer J. Pebesma,et al.  Real-time automatic interpolation of ambient gamma dose rates from the Dutch radioactivity monitoring network , 2009, Comput. Geosci..

[23]  Carsten F. Dormann,et al.  Cross-validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure , 2017 .

[24]  Itai Kloog,et al.  Predicting spatiotemporal mean air temperature using MODIS satellite surface temperature measurements across the Northeastern USA , 2014 .

[25]  Jürgen Vogt,et al.  Mapping regional air temperature fields using satellite‐derived surface skin temperatures , 1997 .

[26]  Renhua Zhang,et al.  Estimating the Surface Air Temperature by Remote Sensing in Northwest China Using an Improved Advection-Energy Balance for Air Temperature Model , 2016 .

[27]  A. Sabziparvar,et al.  Estimation of instantaneous air temperature using remote sensing data , 2018 .

[28]  Jian Yang,et al.  Evaluation of MODIS Land Surface Temperature Data to Estimate Near-Surface Air Temperature in Northeast China , 2017, Remote. Sens..

[29]  J. Roning,et al.  A study on the differences in the interpolation capabilities of models , 2005, Proceedings of the 2005 IEEE Midnight-Summer Workshop on Soft Computing in Industrial Applications, 2005. SMCia/05..

[30]  Hongbo Su,et al.  Evaluating Four Remote Sensing Methods for Estimating Surface Air Temperature on a Regional Scale , 2017 .

[31]  Nuno Carvalhais,et al.  Estimating air surface temperature in Portugal using MODIS LST data , 2012 .

[32]  Robert E. Dickinson,et al.  Interpolation of surface radiative temperature measured from polar orbiting satellites to a diurnal cycle: 1. Without clouds , 1999 .

[33]  Y. Radhika,et al.  Atmospheric Temperature Prediction using Support Vector Machines , 2009 .

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

[35]  Jiaping Wu,et al.  valuation of estimating daily maximum and minimum air temperature with ODIS data in east Africa , 2012 .

[36]  J. Seibert,et al.  On the calculation of the topographic wetness index: evaluation of different methods based on field observations , 2005 .

[37]  Zhenxi Shen,et al.  Estimating air temperature of an alpine meadow on the Northern Tibetan Plateau using MODIS land surface temperature , 2011 .

[38]  Aifeng Lv,et al.  A Universal Ts‐VI Triangle Method for the Continuous Retrieval of Evaporative Fraction From MODIS Products , 2017 .

[39]  Menas Kafatos,et al.  Diurnal temperature range over the United States: A satellite view , 2006 .

[40]  Markus Metz,et al.  A New Fully Gap-Free Time Series of Land Surface Temperature from MODIS LST Data , 2017, Remote. Sens..

[41]  Lorenzo Busetto,et al.  MODIStsp: An R package for automatic preprocessing of MODIS Land Products time series , 2016, Comput. Geosci..

[42]  Tomislav Hengl,et al.  Improving performance of spatio-temporal machine learning models using forward feature selection and target-oriented validation , 2018, Environ. Model. Softw..

[43]  Miquel Ninyerola,et al.  Improving Mean Minimum and Maximum Month-to-Month Air Temperature Surfaces Using Satellite-Derived Land Surface Temperature , 2017, Remote. Sens..

[44]  Fulgencio Cánovas-García,et al.  Optimal Combination of Classification Algorithms and Feature Ranking Methods for Object-Based Classification of Submeter Resolution Z/I-Imaging DMC Imagery , 2015, Remote. Sens..

[45]  Richard J. Telford,et al.  Technical note: Estimating unbiased transfer-function performances in spatially structured environments , 2015 .

[46]  A. Cescatti,et al.  Descriptor : A global dataset of air temperature derived from satellite remote sensing and weather stations , 2018 .

[47]  Chunlin Huang,et al.  A simplified data assimilation method for reconstructing time-series MODIS NDVI data , 2009 .

[48]  Tomislav Hengl,et al.  Spatio-temporal interpolation of soil water, temperature, and electrical conductivity in 3D + T: The Cook Agronomy Farm data set , 2015 .

[49]  François Anctil,et al.  Neural network estimation of air temperatures from AVHRR data , 2004 .

[50]  Richard H. Waring,et al.  Ecological Remote Sensing at OTTER: Satellite Macroscale Observations , 1994 .

[51]  Samuel N. Goward,et al.  Biospheric environmental monitoring at BOREAS with AVHRR observations , 1997 .

[52]  S. Running,et al.  Estimation of regional surface resistance to evapotranspiration from NDVI and thermal-IR AVHRR data , 1989 .

[53]  Stephen J. Connor,et al.  Estimating surface air temperatures, from Meteosat land surface temperatures, using an empirical solar zenith angle model , 1999 .

[54]  Yongming Xu,et al.  Estimating daily maximum air temperature from MODIS in British Columbia, Canada , 2014 .

[55]  Joseph H. A. Guillaume,et al.  Characterising performance of environmental models , 2013, Environ. Model. Softw..

[56]  Saeid Hamzeh,et al.  Diurnal Air Temperature Modeling Based on the Land Surface Temperature , 2017, Remote. Sens..

[57]  Roman M. Balabin,et al.  Support vector machine regression (LS-SVM)--an alternative to artificial neural networks (ANNs) for the analysis of quantum chemistry data? , 2011, Physical chemistry chemical physics : PCCP.