Consistency of tropospheric ozone observations made by different platforms and techniques in the global databases

Abstract

A large quantity of tropospheric ozone observations are conducted all over the world using different platforms and techniques for different purposes and goals. These observations are commonly used to derive seasonal cycles, interannual variations and long-term trends of ozone in the troposphere. In addition, they are used for comparison with three-dimensional chemistry-transport models to evaluate their performance and hence to test our current understanding of the tropospheric ozone variability. It is still challenging to provide robust tropospheric ozone trends throughout the world because of the great variability of ozone, its complex photochemical reactions, the rarity of long-term records, the diversity of measurement techniques and platforms, and the issues with data quality. In this work, we evaluated, with emphasis on the lower troposphere, the consistency of tropospheric ozone observations made by means of multiple platforms, including surface sites, sondes and regular aircraft, that are publicly available in the global databases, but excluding space-borne platforms. Concomitant observations were examined on an hourly basis (except for ±3 hours for sonde versus aircraft) for pairs of locations at less than 100-km distance. Generally, we found good agreement between sonde and surface observations. We also found that there was no need to apply any correction factor to ozonesonde observations except for Brewer–Mast sondes at Hohenpeissenberg. Because of a larger distance between the site pairs, the correlations found between the aircraft and surface measurements were poorer than those between sonde and surface measurements. However, a relatively simple wind segregation improved the agreement between the aircraft versus surface measurements. We found also that due to diurnal cycles, the sonde launching at a fixed local time led to positive or negative biases against the surface observations, suggesting that great attention should be paid to local time and diurnal variations when using ozonesonde in the analysis of seasonal cycles, long-term trends and interannual variations of lower tropospheric ozone. The comparison of surface data at Mt. Happo to regular aircraft data over Tokyo/Narita showed a relatively reasonable agreement, ensuring regionally representative ozone data sets in this region for trend analysis.

Keywords

• tropospheric ozone
• MOZAIC
• IAGOS
• WDCGG
• WOUDC