Share of Discontinuities in the Ozone Concentration Data from Three Reanalyses

0548518 - ÚFA 2022 RIV CH eng J - Journal Article
Križan, Peter - Kozubek, Michal - Laštovička, Jan - Lán, Radek
Share of Discontinuities in the Ozone Concentration Data from Three Reanalyses.
Atmosphere. Roč. 12, č. 11 (2021), č. článku 1508. E-ISSN 2073-4433
R&D Projects: GA ČR(CZ) GA21-03295S
Grant - others:ESA - The European Space Agency(XE) ESA 4000133567/20/I-BG
Institutional support: RVO:68378289
Keywords : discontinuity in ozone concentration data * trend analysis * data quality
OECD category: Meteorology and atmospheric sciences
Impact factor: 3.110, year: 2021
Method of publishing: Open access
https://www.mdpi.com/2073-4433/12/11/1508

Ozone is a very important trace gas in the stratosphere and, thus, we need to know its time evolution over the globe. However, ground-based measurements are rare, especially in the Southern Hemisphere, and while satellite observations provide broader spatial coverage generally, they are not available everywhere. On the other hand, reanalysis data have regular spatial and temporal structure, which is beneficial for trend analysis, but temporal discontinuities might exist in the data. These discontinuities may influence the results of trend studies. The aim of this paper is to detect discontinuities in ozone data of the following reanalyses: MERRA-2, ERA-5 and JRA-55 with the help of the Pettitt, the Buishand, and the Standard Normal Homogeneity tests above the 500 hPa level. The share of discontinuities varies from 30% to 70% and they are strongly layer dependent. The share of discontinuities is the lowest for JRA-55. Differences between reanalyses were found to be larger than differences between homogeneity tests within one reanalysis. Another aim of this paper is to test the ability of homogeneity tests to detect the discontinuities in 2004 and 2015, when changes in versions of satellite data took place. We showed the discontinuities in 2004 are better detected than those in 2015.
Permanent Link: http://hdl.handle.net/11104/0324651