An integrated and homogenized global surface solar radiation dataset and its reconstruction based on an artificial intelligence approach
Abstract. Surface solar radiation (SSR) is an essential factor in the flow of surface energy, enabling accurate capturing of long-term climate change and understanding the energy balance of Earth's atmosphere system. However, the long-term trend estimation of SSR is subjected to significant uncertainties due to the temporal inhomogeneity and the uneven spatial distribution of the in-situ observations. This paper develops an observational integrated and homogenized global-terrestrial (except for Antarctica)) stational SSR dataset (SSRIHstation) by integrating all available SSR observations, including the existing homogenized SSR results. The series are then interpolated in order to obtain a 5°×5° resolution gridded dataset (SSRIHgrid). On this basis, we further reconstruct a long-term (1955–2018) global land (except for Antarctica) SSR anomalies dataset with a 5°×2.5° resolution (SSRIH20CR) by training improved partial convolutional neural network deep learning methods based on the reanalysis 20CRv3. Based on this, we analysed the global land (except for Antarctica) /regional scale SSR trends and spatiotemporal variations: the reconstruction results reflect the distribution of SSR anomalies and have high reliability in filling and reconstructing the missing values. At the global land (except for Antarctica) scale, the decreasing trend of the SSRIH20CR (-1.276±0.205 W/m2 per decade) is slightly smaller than the trend of the SSRIHgrid (-1.776±0.230 W/m2 per decade) from 1955 to 1991. The trend of SSRIH20CR (0.697±0.359 W/m2 per decade) from 1991 to 2018 is also marginally lower than that of the SSRIHgrid (0.851±0.410 W/m2 per decade). At the regional scale, the difference between the SSRIH20CR and SSRIHgrid is more significant in years and areas with insufficient coverage. Asia, Africa, Europe and North America cause the global dimming of the SSRIH20CR, while Europe and North America drive the global brightening of the SSRIH20CR. Spatial sampling inadequacies have largely contributed to a bias in the long-term variation of global/regional SSR. This paper's homogenized gridded dataset and the Artificial Intelligence reconstruction gridded dataset (Jiao and Li, 2023) are all available at https://doi.org/10.6084/m9.figshare.21625079.v1.