2026-10-10T06:49:35Z https://www.scidb.cn/oai

10.57760/sciencedb.012rf 2026-09-28T15:58:41Z

2026-09-28 Analysis code for assessing observation history and solar and geomagnetic activity in low-latitude TEC models doi:10.57760/sciencedb.012rf en Author-owned Python analysis and figure-reconstruction code supporting "Solar and geomagnetic information for background TEC and short-scale fluctuations over Hainan: an explainable machine-learning analysis", based on 2015-2020 GNSS observations at Danzhou, Hainan.Version 2 adds complete-activity-range geomagnetic information-group comparisons and disturbance-onset and continuing-amplitude analyses. P05_ScienceDB_Code_UPDATE_v16.zip contains 30 files, including author-owned analysis and plotting scripts, configurations, dependency and execution guidance, and licenses. Together with the companion data archive it supports reconstruction of four main-text and six supplementary figures. Any retained earlier code archive belongs to the previous manuscript version and is not required for the current figures.The companion CC BY 4.0 aggregate data record is https://doi.org/10.57760/sciencedb.012qg. Combine the current archives while preserving relative paths and follow the README instructions. Full model refitting and attribution require upstream hourly tables, event products and principal-run predictions that are not distributed. Main local history uses values at lags 1 and 24 h. Centered dTEC windows are used for retrospective estimation rather than operational forecasting.No raw GNSS or ionosonde observations, manual Es parameters, spread-F labels, extracted E/F-layer parameters, individual event catalogs, row-level inputs/predictions/SHAP values, or fitted model weights are included. No third-party library source or binary is bundled. The MIT license applies to included author-owned software and code documentation only; it does not relicense companion data or grant access to excluded research assets. Source-observation access remains subject to provider terms: Chinese Meridian Project Danzhou vertical absolute TEC (DOI 10.12176/01.06.00588-V01) and NASA SPDF OMNI hourly (DOI 10.48322/1shr-ht18). Ionosphere; XGBoost; total electron content (TEC) Zheng Wang Pengdong Gao PUBLIC https://api.github.com/licenses/mit dataset Science Data Bank