A machine learning approach to estimation of downward solar radiation from satellite-derived data products: An application over a semi-arid ecosystem in the U.S.

Shortwave solar radiation is an important component of the surface energy balance and provides the principal source of energy for terrestrial ecosystems. This paper presents a machine learning approach in the form of a random forest (RF) model for estimating daily downward solar radiation flux at the land surface over complex terrain using MODIS (MODerate Resolution Imaging Spectroradiometer) remote sensing data. The model-building technique makes use of a unique network of 16 solar flux measurements in the semi-arid Reynolds Creek Experimental Watershed and Critical Zone Observatory, in southwest Idaho, USA. Based on a composite RF model built on daily observations from all 16 sites in the watershed, the model simulation of downward solar radiation matches well with the observation data (r2 = 0.96). To evaluate model performance, RF models were built from 12 of 16 sites selected at random and validated against the observations at the remaining four sites. Overall root mean square errors (RMSE), bias, and mean absolute error (MAE) are small (range: 37.17 W/m2-81.27 W/m2, -48.31 W/m2-15.67 W/m2, and 26.56 W/m2-63.77 W/m2, respectively). When extrapolated to the entire watershed, spatiotemporal patterns of solar flux are largely consistent with expected trends in this watershed. We also explored significant predictors of downward solar flux in order to reveal important properties and processes controlling downward solar radiation. Based on the composite RF model built on all 16 sites, the three most important predictors to estimate downward solar radiation include the black sky albedo (BSA) near infrared band (0.858 μm), BSA visible band (0.3–0.7 μm), and clear day coverage. This study has important implications for improving the ability to derive downward solar radiation through a fusion of multiple remote sensing datasets and can potentially capture spatiotemporally varying trends in solar radiation that is useful for land surface hydrologic and terrestrial ecosystem modeling.

[1]  Min Chen,et al.  An Efficient Method of Estimating Downward Solar Radiation Based on the MODIS Observations for the Use of Land Surface Modeling , 2014, Remote. Sens..

[2]  Clayton L. Hanson,et al.  Long‐Term Soil Water Content Database, Reynolds Creek Experimental Watershed, Idaho, United States , 2001 .

[3]  G. Likens,et al.  Evaluation of an integrated biogeochemical model (PnET‐BGC) at a northern hardwood forest ecosystem , 2001 .

[4]  M. Seyfried,et al.  Simulation of long‐term soil water dynamics at Reynolds Creek, Idaho: implications for rangeland productivity , 2016 .

[5]  Maosheng Zhao,et al.  A Continuous Satellite-Derived Measure of Global Terrestrial Primary Production , 2004 .

[6]  Bernard Pinty,et al.  Determination of land and ocean reflective, radiative, and biophysical properties using multiangle imaging , 1998, IEEE Trans. Geosci. Remote. Sens..

[7]  C. Tucker,et al.  Climate-Driven Increases in Global Terrestrial Net Primary Production from 1982 to 1999 , 2003, Science.

[8]  A. Angstrom.  Solar and terrestrial radiation. Report to the international commission for solar research on actinometric investigations of solar and atmospheric radiation , 2007 .

[9]  Shunlin Liang,et al.  Mapping daily snow/ice shortwave broadband albedo from Moderate Resolution Imaging Spectroradiometer (MODIS): The improved direct retrieval algorithm and validation with Greenland in situ measurement , 2005 .

[10]  Steven W. Running,et al.  Reconciling satellite with ground data to estimate forest productivity at national scales , 2012 .

[11]  T. Sullivan,et al.  Factors influencing critical and target loads for the acidification of lake–watersheds in the Adirondack region of New York , 2015, Biogeochemistry.

[12]  Z. Li,et al.  Towards a local split window method over land surfaces , 1990 .

[13]  B. Liepert,et al.  Observed reductions of surface solar radiation at sites in the United States and worldwide from 1961 to 1990 , 2002 .

[14]  Alan H. Strahler,et al.  Using a multikernel least-variance approach to retrieve and evaluate albedo from limited bidirectional measurements , 2001 .

[15]  A. Angstroem.  Solar and terrestrial radiation , 1924 .

[16]  Toshio Koike,et al.  A general model to estimate hourly and daily solar radiation for hydrological studies , 2005 .

[17]  Z. Wan.  New refinements and validation of the MODIS Land-Surface Temperature/Emissivity products , 2008 .

[18]  Zong-Liang Yang,et al.  Technical description of version 4.5 of the Community Land Model (CLM) , 2013 .

[19]  Johannes R. Sveinsson,et al.  Random Forest Classification of Remote Sensing Data , 2006 .

[20]  T. Sullivan,et al.  Responses of 20 lake-watersheds in the Adirondack region of New York to historical and potential future acidic deposition. , 2015, The Science of the total environment.

[21]  Thomas H. Painter,et al.  Time-space continuity of daily maps of fractional snow cover and albedo from MODIS , 2008 .

[22]  P. Heuberger,et al.  Calibration of process-oriented models , 1995 .

[23]  Gautam Bisht,et al.  Estimation of the net radiation using MODIS (Moderate Resolution Imaging Spectroradiometer) data for clear sky days , 2005 .

[24]  Alan H. Strahler,et al.  An algorithm for the retrieval of albedo from space using semiempirical BRDF models , 2000, IEEE Trans. Geosci. Remote. Sens..

[25]  Xiaotong Zhang,et al.  Generating Global LAnd Surface Satellite incident shortwave radiation and photosynthetically active radiation products from multiple satellite data , 2014 .

[26]  Gerald Stanhill,et al.  Global dimming: a review of the evidence for a widespread and significant reduction in global radiation with discussion of its probable causes and possible agricultural consequences , 2001 .

[27]  D. Marks,et al.  Long‐term snow, climate, and streamflow trends at the Reynolds Creek Experimental Watershed, Owyhee Mountains, Idaho, United States , 2010 .

[28]  D. K. Butt.  Solar and Terrestrial Radiation , 1978 .

[29]  W. Russell Hamon.  Estimating Potential Evapotranspiration , 1960 .

[30]  D. Lettenmaier,et al.  A simple hydrologically based model of land surface water and energy fluxes for general circulation models , 1994 .

[31]  Tingjun Zhang.  Influence of the seasonal snow cover on the ground thermal regime: An overview , 2005 .

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

[33]  Hongyu Huang,et al.  A Comparison of Two Open Source LiDAR Surface Classification Algorithms , 2011, Remote. Sens..

[34]  Jeff Dozier,et al.  A generalized split-window algorithm for retrieving land-surface temperature from space , 1996, IEEE Trans. Geosci. Remote. Sens..

[35]  N. C. Strugnell,et al.  First operational BRDF, albedo nadir reflectance products from MODIS , 2002 .

[36]  Rachel T. Pinker,et al.  Modeling shortwave radiative fluxes from satellites , 2012 .

[37]  G. Powers,et al.  A Description of the Advanced Research WRF Version 3 , 2008 .

[38]  Steven W. Running,et al.  Modeling and Monitoring Terrestrial Primary Production in a Changing Global Environment: Toward a Multiscale Synthesis of Observation and Simulation , 2014 .

[39]  John D. Aber,et al.  Variation among solar radiation data sets for the Eastern US and its effects on predictions of forest production and water yield , 2000 .

[40]  David C. Robertson,et al.  Long‐Term Snow Database, Reynolds Creek Experimental Watershed, Idaho, United States , 2001 .