Abstract
Surface shortwave net radiation (SSNR) flux is essential for the determination of the radiation energy balance between the atmosphere and the Earth’s surface. The satellite-derived intermediate SSNR data are strongly needed to bridge the gap between existing coarse-resolution SSNR products and point-based measurements. In this study, four different machine learning (ML) algorithms were tested to estimate the SSNR from the Landsat Thematic Mapper (TM)/ Enhanced Thematic Mapper Plus (ETM+) top-of-atmosphere (TOA) reflectance and other ancillary information (i.e., clearness index, water vapor) at instantaneous and daily scales under all sky conditions. The four ML algorithms include the multivariate adaptive regression splines (MARS), backpropagation neural network (BPNN), support vector regression (SVR), and gradient boosting regression tree (GBRT). Collected in-situ measurements were used to train the global model (using all data) and the conditional models (in which all data were divided into subsets and the models were fitted separately). The validation results indicated that the GBRT-based global model (GGM) performs the best at both the instantaneous and daily scales. For example, the GGM based on the TM data yielded a coefficient of determination value (R2) of 0.88 and 0.94, an average root mean square error (RMSE) of 73.23 W∙m-2 (15.09%) and 18.76 W·m-2 (11.2%), and a bias of 0.64 W·m-2 and –1.74 W·m-2 for instantaneous and daily SSNR, respectively. Compared to the Global LAnd Surface Satellite (GLASS) daily SSNR product, the daily TM-SSNR showed a very similar spatial distribution but with more details. Further analysis also demonstrated the robustness of the GGM for various land cover types, elevation, general atmospheric conditions, and seasons
Highlights
Surface energy fluxes profoundly affect our ability to understand how Earth’s climate responds to increasing concentrations of greenhouse gases [1]
The observations from six AmeriFlux sites were used for validation, and the results showed an root mean square error (RMSE) of 77.5 W·m-2 and 36.1 W·m-2 for the instantaneous and daily Surface shortwave net radiation (SSNR), respectively, which demonstrated the performance of this method
Some previous studies demonstrated that Landsat BT data have been widely used for land cover classification and assessment of land use change [72,73], it was reasonable to be used as input to estimate the SSNR
Summary
Surface energy fluxes profoundly affect our ability to understand how Earth’s climate responds to increasing concentrations of greenhouse gases [1]. Wang [32] developed a modified hybrid model (WangLUT, hereinafter) to estimate SSNR from Landsat5/TM and Landsat7/ETM+ TOA data. This model is based on the study conducted by Kim et al [18] for retrieving SSNR using Moderate Resolution Imaging Spectroradiometer (MODIS) TOA and surface reflectance data. In the WangLUT method, the radiative transfer model MODTRAN5 is used firstly to establish the simulations of SSNR and TOA spectral radiance under various atmospheric and surface conditions base on Landsat data. New empirical models were developed with machine learning algorithms to replace the LUT of the WangLUT method, separately for the estimation of SSNR at instantaneous and daily scales.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
