Improved estimation of surface solar insolation using a neural network and MTSAT-1R data

Solar surface insolation (SSI) provides information on how much solar radiance reaches the Earth's surface at a specified location during the daytime. The amount of insolation reaching the surface is a critical parameter for climate change estimation and numerical weather prediction (NWP). We calculated SSI from MTSAT-1R data using a neural network (NN) model to obtain more accurate results than obtained using empirical and physical methods. The use of retrieved SSI data depends on the accuracy of the output results. Thus, before adding the input parameters to the NN, principal component transformation was performed using the eigenvectors and normalized input data to eliminate data redundancy. An NN model with one hidden layer was then used to simulate SSI using early-stop and Levenberg-Marquardt back-propagation (LMBP) methods. We separated the NN architecture into two parts according to cloudy or clear-sky conditions, which have different processes because of complicated cloud physical characteristics. The SSI estimates from the NN model were compared with pyranometer measurements and showed better agreement with ground-truth values than did estimates obtained using conventional methods, especially under the clear-sky condition.

[1]  Robert Frouin,et al.  A technique for global monitoring of net solar irradiance at the ocean surface. I - Model , 1992 .

[2]  Hiroshi Kawamura,et al.  A system to distribute satellite incident solar radiation in real-time , 2001 .

[3]  Elie Bienenstock,et al.  Neural Networks and the Bias/Variance Dilemma , 1992, Neural Computation.

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

[5]  S. Fritz,et al.  Satellite Measurements of Reflected Solar Energy and the Energy Received at the Ground , 1964 .

[6]  Ferdinand Hergert,et al.  Improving model selection by nonconvergent methods , 1993, Neural Networks.

[7]  Edit J. Kaminsky,et al.  Textural neural network and version space classifiers for remote sensing , 1997 .

[8]  Man-Wai Mak,et al.  Application of MLP and RBF networks to cloud detection , 2001, Proceedings of 2001 International Symposium on Intelligent Multimedia, Video and Speech Processing. ISIMP 2001 (IEEE Cat. No.01EX489).

[9]  P. Green,et al.  Analyzing multivariate data , 1978 .

[10]  Lutz Prechelt,et al.  Automatic early stopping using cross validation: quantifying the criteria , 1998, Neural Networks.

[11]  Martha C. Anderson,et al.  Validation of GOES-Based Insolation Estimates Using Data from the U.S. Climate Reference Network , 2005 .

[12]  George R. Diak,et al.  A note on first estimates of surface insolation from GOES-8 visible satellite data , 1996 .

[13]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[14]  Olivier Jourdan,et al.  Mesoscale cloud pattern classification over ocean with a neural network using a new index of cloud variability , 2006 .

[15]  John N. Tsitsiklis,et al.  Neuro-Dynamic Programming , 1996, Encyclopedia of Machine Learning.

[16]  W. James Shuttleworth,et al.  Solar radiation and evapotranspiration in northern Mexico estimated from remotely sensed measurements of cloudiness , 2001 .

[17]  Hiroshi Kawamura,et al.  Validation and Improvement of Satellite-Derived Surface Solar Radiation over the Northwestern Pacific Ocean , 2005 .

[18]  S. Kizu,et al.  A study on thermal response of ocean surface layer to solar radiation using satellite remote sensing , 1995 .

[19]  Christopher J. Watts,et al.  Use of satellite data to estimate radiation and evaporation for northwest Mexico , 1999 .

[20]  C. Gautier,et al.  A Simple Physical Model to Estimate Incident Solar Radiation at the Surface from GOES Satellite Data , 1980 .

[21]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[22]  J. C. Klink,et al.  An Evaluation of Satellite-based Insolation Estimates for Ohio. , 1986 .

[23]  Catherine Gautier,et al.  An investigation of the effects of spatially averaging satellite brightness measurements on the calculation of insolation , 1984 .

[24]  Hiroshi Kawamura,et al.  Estimation of insolation over the Pacific Ocean off the Sanriku coast , 1998 .

[25]  Kyung-Soo Han,et al.  A land cover classification product over France at 1 km resolution using SPOT4/VEGETATION data , 2004 .

[26]  Bernard Bobée,et al.  Daily reservoir inflow forecasting using artificial neural networks with stopped training approach , 2000 .

[27]  Kyung-Soo Han,et al.  Neural network determination of cloud attenuation to estimate insolation using MTSAT‐1R data , 2008 .

[28]  Timothy Masters,et al.  Advanced algorithms for neural networks: a C++ sourcebook , 1995 .

[29]  Ronald M. Welch,et al.  A neural network approach to cloud classification , 1990 .