遇见数据集

KAPPA Arc V5 — GFS κ Validation Dataset (Paper 4)

收藏
Zenodo2026-08-02 更新2026-08-13 收录
官方服务:

资源简介:

Raw validation datasets supporting the manuscript 'Real-Time Atmospheric Refraction Correction for Surveying Using GFS Numerical Weather Prediction', submitted to the Journal of Surveying Engineering (ASCE). Files: gfs_validation.json (28 quality GFS–IGRA2 comparisons, 4 tropical stations), gfs_validation_v2.json (29 quality comparisons, expanded run including Nottingham), nottingham_gfs.json (6 quality Nottingham comparisons, altitude-corrected, unaffected by any correction below), saudi_gfs_validation.json (90 GFS-vs-ERA5 comparisons, Saudi Arabia; unaffected by the correction below, carrying forward a prior boundary-value correction — see README.md), paper4_reproducibility_script.py (worked example with real raw profile data), and Paper4_JSE_Supplementary.docx (Sections S1–S6, corrected). This version corrects a station-identity error: Pohnpei radiosonde comparisons across gfs_validation.json and gfs_validation_v2.json used IGRA2 station FMM00091334 (Truk/Chuuk, ~706 km from Pohnpei) rather than the correct station, FMM00091348. GFS and ERA5 values were unaffected — both were always queried at Pohnpei's correct coordinates; only the radiosonde reference values changed. 19 dates were recomputed against the correct station; 2 were excluded as physically implausible. See README.md for full detail, including the resulting change to headline accuracy figures. These are the primary data underlying Tables 3–4 and supplemental Sections S1–S6. This is a standalone deposit dedicated to Paper 4, separated from an earlier shared multi-paper record for clarity.

提供机构:
Zenodo
创建时间:
2026-08-02
二维码
社区交流群
二维码
科研交流群
商业服务