기상청 WMO/GAW 세계표준센터 (WCC-SF6) 비교실험 활동을 통한 육불화황 관측 정확도 향상 연구

• Published in: 한국대기환경학회지(국문) Vol. 37; no. 3; pp. 512 - 522
• Main Authors: 이해영(Haeyoung Lee), 이정순(Jeongsoon Lee), Brad Hall, Ed Dlugokencky, 김수민(Sumin Kim), 김연희(Yeon-Hee Kim)
• Format: Journal Article
• Language: Korean
• Published: 한국대기환경학회 01.06.2021
• Subjects: 대기과학; Measurement; 6< WMO/GAW; Greenhouse Gas; SF< Calibration; SUB&gt
• ISSN: 1598-7132; 2383-5346
• DOI: 10.5572/KOSAE.2021.37.3.512

Under the World Meteorological Organization (WMO)/Global Atmosphere Watch (GAW) Programme, there are central facilities for quality control and quality assurance. Among them, World Calibration Centres organize inter-comparisons for scale traceability, a necessary step to meet the compatibility goal of ±0.02 ppt (extended goal: ±0.05 ppt) within the GAW network. Korea Meteorological Administration (KMA)/National Institute of Meteorological Sciences (NIMS) has been the World Calibration Centre for SF6 (WCC-SF6) from 2012 and has organized inter-comparison experiments using WCC’s travelling cylinders (with measurement uncertainty of 0.01~0.02 ppt). Here we present our system and analysis methods for those cylinders. The 1st SF6 inter-comparison experiment (SICE) was conducted from 2016 to 2017 with 11 labs from 7 countries. Laboratories that calibrated the circulated cylinders on the same scale with WCC-SF6 (WMO-X2014) meet the extended compatibility goal for the high or low levels cylinders, and 75% of laboratories among them meet the compatibility goal for both levels. When laboratories calibrated the cylinders with their own standard scales, differences between WCC-SF6 and laboratories increased to 2.3% depending on their scales. This suggested the importance all labs adopting a single standard scale for long-term measurements and in the same network. Through inter-comparison experiments conducted with the Central Calibration Laboratory (CCL) on the same scale, drift and non-linear instrument characteristic affected the analysis results. When instrument drift corrections were applied, the differences from the CCL decreased from [(-0.25) to 0.43 ppt] to [(-0.07) to 0.02 ppt]. Non-linear analyzers underestimate high level SF6, e.g., by 8% at 20 ppt (even after instrument drift correction), so that a calibration strategy with a minimum of two standards (bracketing the expected range of observations) is necessary for accurate measurements. KCI Citation Count: 3