Surface solar radiation (SSR) is an important factor in surface energy flow, which can accurately capture long-term climate change and understand the energy balance of the Earth's atmospheric system
The globally integrated and homogenized solar surface radiation dataset includes the homogenized SSR dataset (SSRIHgrid) and the artificial intelligence reconstructed SSR dataset (SSRIH20CR). SSRIHgrid is a 5 ° × A global monthly SSR anomaly dataset with 5 ° homogenization, SSRIH20CR is based on a 20CRv3 artificial intelligence model with 5 ° × 2.5 ° full coverage monthly land (excluding Antarctica) SSR anomaly reconstruction dataset.
The original SSR data mainly comes from two sources: the GEBA data from the Federal Institute of Technology Zurich comes from monthly data from 2445 globally distributed sites from 1922 to 2020; The WRDC data comes from monthly data from 1136 globally distributed sites since 1964.
Based on the original observation SSRs of GEBA and WRDC, combined with other existing homogenized SSR datasets, global station observations were integrated. In addition, the RHtestV4 software package was also used to homogenize the globally distributed site data. Subsequently, an improved CNN deep learning algorithm was used to reconstruct SSR anomalies. Based on the training set (20CRv3), obtain the SSR anomaly reconstruction dataset SSRIH2ocR from 1955 to 2018 with a resolution of 5 ° × 2.5 °.
The data quality is good.
This work is licensed under a
Creative
Commons Attribution 4.0 International License.
CSTR:11738.11.NCDC.FIGSHARE.DB4155.2023
10.6084/m9.figshare.21625079.v1