Multivariate model-based investigation of the temperature dependence of ozone concentration in Finnish boreal forest. (15th November 2022)

• Record Type: Journal Article
• Authors: Isokääntä, Sini; Mikkonen, Santtu; Laurikainen, Maria; Buchholz, Angela; Schobesberger, Siegfried; Blande, James D.; Nieminen, Tuomo; Ylivinkka, Ilona; Bäck, Jaana; Petäjä, Tuukka; Kulmala, Markku; Yli-Juuti, Taina
• Is Part Of: Atmospheric environment. Volume 289(2022)
• Publication Date: 2022-11-15
• Subjects: Tropospheric ozone -- Statistical modelling -- Temperature dependence -- OVOCs; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2022.119315
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: Tropospheric ozone (O3 ) concentrations are observed to increase with temperature in urban and rural locations. We investigated the apparent temperature dependency of daytime ozone concentration in the Finnish boreal forest in summertime based on long-term measurements. We used statistical mixed effects models to separate the direct effects of temperature from other factors influencing this dependency, such as weather conditions, long-range transport of precursors, and concentration of various hydrocarbons. The apparent temperature dependency of 1.16 ppb °C −1 based on a simple linear regression was reduced to 0.87 ppb °C −1 within the canopy for summer daytime data after considering these factors. In addition, our results indicated that small oxygenated volatile organic compounds may play an important role in the temperature dependence of O3 concentrations in this dataset from a low-NOx environment. Summertime observations and daytime data were selected for this analysis to focus on an environment that is significantly affected by biogenic emissions. Despite limitations due to selection of the data, these results highlight the importance of considering factors contributing to the apparent temperature dependence of the O3 concentration. In addition, our results show that a mixed effects model achieves relatively good predictive accuracy for this dataset without explicitly calculating all processes involved in O3 formation and removal. Highlights: Surface data differentiated apparent and direct temperature dependence of [O3 ]. Volatile organic compounds contribute vitally to [O3 ] in this low-NOx environment. Statistical model reproduces the [O3 ], without exactly stating all involved processes.