Wind Data Quality Control Method Leveraging Instantaneous Image-Structure Characteristics
Journal of Atmospheric and Oceanic Technology · 2026-11 · 43 (11) · 1341-1356
DOI: 10.1175/jtech-d-26-0084.1 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Atmospheric and Oceanic Technology, 2026-11. Verfasst von Wanchen Li und 3 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Hainan Li’an Education, Science and Technology Innovation, Nanjing University of Information Science and Technology.
Abstract (Originalsprache)
The 10-m wind is critical for boundary layer analysis and severe weather research. However, its poor spatiotemporal continuity and drastic instantaneous fluctuations increase the difficulty of quality control. Traditional quality control (QC) methods based on long-term statistics have obvious defects and easily misjudge transient wind fields caused by extreme weather as erroneous observations. This study introduces curvelet transform to develop a new QC method relying only on the spatial structure of single-time step wind fields. The instantaneous wind field is decomposed into spatially structured components and stochastic components, and differentiated QC targeting each component effectively identifies erroneous observations under severe weather such as typhoons. Hourly 10-m wind field observations (standard operational height) from national automatic surface stations in China, with one representative month chosen for each season of 2024, are used for practical QC tests. The results show that the rejection rate of the new QC method remains steadily below 8%, and the spatial distribution of rejected observations is barely disturbed by weather and presents favorable random features consistent with error characteristics. Comparisons with China Meteorological Administration Land Data Assimilation System (CLDAS) wind products show spatial correlation coefficients rise for over 90% time steps after QC, with a maximum increase of 7.7%. Wind speed errors at all time steps are effectively optimized, with the standard deviation reduced by 2.3% on average and the absolute error decreased by 3.6%. These results fully prove that the proposed QC method improves the quality of wind observations from automatic surface stations. Significance Statement Wind measurements from ground automatic weather stations support weather forecasting and disaster warning, yet raw wind data often contain noisy abnormal signals. Current operational quality control uses fixed universal thresholds for format, missing value, climate limit, and consistency checks, which only detect obvious extreme outliers with rigid rules. This study develops a curvelet-transform-based wind field quality control method. It separates large-scale circulation and small-scale disturbances to capture subtle hidden observational errors within normal value ranges, effectively compensating the inherent defects of traditional routine quality control (QC) schemes.
Förderung
- National Natural Science Foundation of China
- Hainan Li’an Education, Science and Technology Innovation
- Nanjing University of Information Science and Technology
Zitieren
APA
Li, W., Qin, Z., Zhang, B., & Li, X. (2026). Wind Data Quality Control Method Leveraging Instantaneous Image-Structure Characteristics. Journal of Atmospheric and Oceanic Technology, 43(11), 1341-1356. https://doi.org/10.1175/jtech-d-26-0084.1
Vancouver
Li W, Qin Z, Zhang B, Li X. Wind Data Quality Control Method Leveraging Instantaneous Image-Structure Characteristics. Journal of Atmospheric and Oceanic Technology. 2026;43(11):1341-1356. doi:10.1175/jtech-d-26-0084.1
BibTeX
@article{Li2026,
author = {Li, Wanchen and Qin, Zhengkun and Zhang, Bing and Li, Xin},
title = {Wind Data Quality Control Method Leveraging Instantaneous Image-Structure Characteristics},
journal = {Journal of Atmospheric and Oceanic Technology},
year = {2026},
volume = {43},
number = {11},
pages = {1341--1356},
doi = {10.1175/jtech-d-26-0084.1}
}
EUVIMED