Analysis of 5.5 years of atmospheric CO2, CH4, CO continuous observations (2014–2020) and their correlations, at the Observatoire de Haute Provence, a station of the ICOS-France national greenhouse gases observation network

Abstract

Since 2014, a 100 m tall tower measures continuously greenhouse gases at the Observatoire de haute Provence (OHP) located in the southeast of France (43° 55′ 51″ N, 5° 42′ 48″ E) as a monitoring station of the French National Greenhouse Gases Observation network (ICOS-Fr). This rural station allows to study the short, mid, and long terms variability of atmospheric CO2, CH4 and CO concentrations at the continental, regional and local scales in a region characterized by a Mediterranean climate. Measurements are performed using cavity ring-down spectroscopy at three levels above ground level (AGL); 10 m, 50 m and 100 m. Using the ICOS European Infrastructure procedure to calibrate and ensure the data quality control, the precision of our datasets matches the international WMO/GAW recommendations. Time series from July 2014 to February 2020 were analysed. We inferred a mean annual growth rate at 100 m AGL of 2.7 ppm/year for CO2 (7.8 ppb/year for CH4) over the period of study, whereas no significant annual growth rate was found for CO. These growth rates are comparable to other remote ICOS and WMO/GAW sites. A seasonal amplitude of 13 ppm, 30 ppb, 45 ppb was found for atmospheric CO2, CH4 and CO, respectively. As expected, the amplitude of the diurnal cycle of these three species varies in function of the season, from 2.6 (1.6) ppm in winter and 10.7 (6.6) ppm in summer for CO2, 3.7 (5.1) and 7.7 (7.1) ppb for CH4, and contrary to CO2 and CH4 smaller amplitude in summer with 2.15 (2.5) ppb and 9.3 (8.9) ppb in winter for CO at 10 m (100 m) AGL. Significant correlations (R2 between 0.67 and 0.91) between the three species have been detected, especially in the winter season. Using thresholds on wind speed and on the standard variation of hourly concentrations, more than 16% of the data were identified to be enriched either: 1/by regional anthropogenic plumes; 2/during stable synoptic conditions inducing the accumulation of anthropogenic emissions in the atmospheric boundary layer (13%); and 3/by local's sources inducing short pollution events (3%). On average, ΔCO/Δ CO2 ratios of 3.72 ± 0.06 ppb/ppm and 0.8 ± 0.2 ppb/ppm for Δ CH4/ΔCO were inferred during local pollution events in winter and are typical of traffic and residential heating as given by the local bottom-up emissions inventory delivered by the regional air quality agency ATMOSUD. Adding specific tracers or isotopic measurements would be very interesting to distinguish anthropogenic sources and monitor the evolution of their characteristics as emission ratios at the OHP station. Filtering out these conditions, about 84% of the data are not undergoing the influence of local and regional anthropogenic plumes, and are thus representative of “background” CO2, CO and CH4 concentrations at the local to the regional scales. These background conditions are shown to be dependent on wind speed and direction.