Assessment of the quality of ACE-FTS stratospheric ozone data

Abstract

For the past 17 years, the Atmospheric Chemistry Experiment Fourier Transform Spectrometer (ACE-FTS) instrument on the Canadian SCISAT satellite has been measuring profiles of atmospheric ozone. The latest operational versions of the level 2 ozone data are versions 3.6 and 4.1. This study characterizes how both products compare with correlative data from other limb-sounding satellite instruments, namely MAESTRO, MLS, OSIRIS, SABER, and SMR. In general, v3.6, with respect to the other instruments, exhibits a smaller bias (which is on the order of 1/43%) in the middle stratosphere than v4.1 (1/42%-9%); however, the bias exhibited in the v4.1 data tends to be more stable, i.e. not changing significantly over time in any altitude region. In the lower stratosphere, v3.6 has a positive bias of about 3%-5% that is stable to within ±1% per decade, and v4.1 has a bias on the order of -1% to +5% and is also stable to within ±1% per decade. In the middle stratosphere, v3.6 has a positive bias of 1/43% with a significant negative drift on the order of 0.5%-2.5% per decade, and v4.1 has a positive bias of 2%-9% that is stable to within ±0.5% per decade. In the upper stratosphere, v3.6 has a positive bias that increases with altitude up to 1/416% and a significant negative drift on the order of 2%-3% per decade, and v4.1 has a positive bias that increases with altitude up to 1/415% and is stable to within ±1% per decade. Estimates indicate that both versions 3.6 and 4.1 have precision values on the order of 0.1-0.2ppmv below 20km and above 45km (1/45%-10%, depending on altitude). Between 20 and 45km, the estimated v3.6 precision of 1/44%-6% is better than the estimated v4.1 precision of 1/46%-10%.