Abstract

Although data-driven methods including deep neural network (DNN) were introduced, there was not enough assessment about spatial characteristics when using limited ground observation as reference. This work aimed to interpret the feasibility of several machine learning approaches to assess the spatial distribution of solar radiation on Earth based on the Communication, Ocean, and Meteorological Satellite (COMS) Meteorological Imager (MI) geostationary satellite. Four data-driven models were selected (artificial neural network (ANN), random forest (RF), support vector regression (SVR), and DNN), to compare their accuracy and spatial estimating performance. Moreover, we used a physical model to probe the ability of data-driven methods, implementing hold-out and k-fold cross-validation approaches based on pyranometers located in South Korea. The results of analysis showed the RF had the highest accuracy in predicting performance, although the difference between RF and the second-best technique (DNN) was insignificant. Temporal variations in root mean square error (RMSE) were dependent on the number of data samples, while the physical model showed relatively less sensitivity. Nevertheless, DNN and RF showed less variability in RMSE than the others. To examine spatial estimation performance, we mapped solar radiation over South Korea for each model. The data-driven models accurately simulated the observed cloud pattern spatially, whereas the physical model failed to do because of cloud mask errors. These exhibited different spatial retrieval performances according to their own training approaches. Overall analysis showed that deeper layers of networks approaches (RF and DNN), could best simulate the challenging spatial pattern of thin clouds when using satellite multispectral data.

Highlights

  • The energy required to drive terrestrial processes is mostly provided by solar radiation, which is, an important factor of influence for agriculture, forest science, hydrology, and meteorology.solar radiation powers photosynthesis in terrestrial ecosystems, and drivesSensors 2019, 19, 2082; doi:10.3390/s19092082 www.mdpi.com/journal/sensorsSensors 2019, 19, 2082 evaporation from the surface and is a variable that connects land–atmosphere fluxes [1,2].In land surface and hydrological modeling, solar radiation incident on a given surface is one of the indispensable driving factors controlling both water and heat exchanges between land and atmosphere [3]

  • The Communication, Ocean, and Meteorological Satellite (COMS) Meteorological Imager (MI) geostationary satellite developed by the Korea Aerospace Research Institute (KARI) was mainly used to estimate the spatiotemporal distribution of incoming solar radiation on surface

  • The criteria that separating training and test datasets is based on locations of ground sites randomly for both validation approaches to evaluate the spatial assessment of solar radiation

Read more

Summary

Introduction

The energy required to drive terrestrial processes is mostly provided by solar radiation, which is, an important factor of influence for agriculture, forest science, hydrology, and meteorology.solar radiation powers photosynthesis in terrestrial ecosystems, and drivesSensors 2019, 19, 2082; doi:10.3390/s19092082 www.mdpi.com/journal/sensorsSensors 2019, 19, 2082 evaporation from the surface and is a variable that connects land–atmosphere fluxes [1,2].In land surface and hydrological modeling, solar radiation incident on a given surface is one of the indispensable driving factors controlling both water and heat exchanges between land and atmosphere [3]. The direct use of pyranometer data from ground sites is one of the simplest ways to estimate on-surface solar radiation, providing mostly accurate estimates of incoming solar radiation with high temporal resolution over established ground points [9]. This approach suffers from many technical and financial issues such as high costs and the need for highly skilled labor, periodical maintenance, cleaning, and calibration of solar sensors [10,11], which means that ground networks of pyranometers are typically not available in sufficiently high spatial coverage to resolve spatial patterns [12]. The utilization of ground pyranometers for directly estimating the spatial distribution of solar radiation is subject to fundamental limitations

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call