Satellite-based assessment of national carbon monoxide concentrations for air quality reporting in Finland

Abstract

Carbon monoxide (CO) has negative health effects, especially on the respiratory system, when present in large concentrations in ambient air. Therefore, it is one of the 12 air pollutants with monitoring and assessment obligations set by the European Union Ambient Air Quality Directives. With low CO concentrations in Finland, continuous surface measurements are not compulsory but the monitoring and assessment obligation remains. Currently, there are no air quality stations equipped with CO measurements in Finland, and the assessment can only be based on emission estimates.In this work, we demonstrate the use of satellite-based CO measurements from the TROPOspheric Monitoring Instrument (TROPOMI) on-board Sentinel-5 Precursor satellite (S5P) to strengthen the assessment with actual measurements. Averaging satellite measurements over several years reveals the spatial distribution of CO with nationwide coverage in city-scale resolution. Individual satellite measurements, in turn, are used to estimate the surface concentrations and their annual variability in different regions in Finland.For the surface estimates, we used a simple and robust linear model between the satellite measurements and ground-level in-situ measurements. We found that the linear model produces reasonable surface concentration estimates to be used in air quality assessments where the objective is to evaluate whether the monitoring threshold set by the EU is exceeded on a yearly scale. It is noted that cloudiness can cause local data gaps of several days and the low solar elevation angles prevent satellite measurements from 9 to 18 weeks during the winter time in Finland.In Finland, the average total columns show low spatial variability, with only a few high-concentration areas identified. For the last three years, the estimated surface concentrations have stayed well below the lower assessment threshold set by the Air Quality Directive.