Improved ozone monitoring by ground-based FTIR spectrometry

Abstract

Accurate observations of atmospheric ozone (O3) are essential to monitor in detail its key role in atmospheric chemistry. The present paper examines the performance of different O3 retrieval strategies from FTIR (Fourier transform infrared) spectrometry by using the 20-year time series of the high-resolution solar spectra acquired from 1999 to 2018 at the subtropical Izaña Observatory (IZO, Spain) within NDACC (Network for the Detection of Atmospheric Composition Change). In particular, the effects of two of the most influential factors have been investigated: the inclusion of a simultaneous atmospheric temperature profile fit and the spectral O3 absorption lines used for the retrievals (the broad spectral region of 1000–1005 cm−1 and single micro-windows between 991 and 1014 cm−1). Additionally, the water vapour (H2O) interference in O3 retrievals has been evaluated, with the aim of providing an improved O3 strategy that minimises its impact and, therefore, could be applied at any NDACC FTIR station under different humidity conditions. The theoretical and experimental quality assessments of the different FTIR O3 products (total column (TC) amounts and volume mixing ratio (VMR) profiles) provide consistent results. Combining a simultaneous temperature retrieval with the optimal selection of single O3 micro-windows results in superior FTIR O3 products, with a precision of better than 0.6 %–0.7 % for O3 TCs as compared to coincident NDACC Brewer observations taken as a reference. However, this improvement can only be achieved provided the FTIR spectrometer is properly characterised and stable over time. For unstable instruments, the temperature fit is found to exhibit a strong negative influence on O3 retrievals due to the increase in the cross-interference between the temperature retrieval and instrumental performance (given by the instrumental line shape function and measurement noise), which leads to a worsening of the precision of FTIR O3 TCs of up to 2 %. This cross-interference becomes especially noticeable beyond the upper troposphere/lower stratosphere, as documented theoretically as well as experimentally by comparing FTIR O3 profiles to those measured using electrochemical concentration cell (ECC) sondes within NDACC. Consequently, it should be taken into account for the reliable monitoring of the O3 vertical distribution, especially over long-term timescales.