The detection of trace gas plumes allows us to improve attribution of pollutant emissions and photochemical processing in the global troposphere. Data collected by the ESA TROPOspheric Monitoring Instrument (TROPOMI) has resulted in a growing number of case studies that have used ad hoc methods to detect plumes for science applications. Developing a more comprehensive understanding of TROPOMI data will help to identify new research avenues and support the development of new applications. However, this is difficult because of the associated data volumes, a challenge that will only grow with time. We address this challenge by using artificial intelligence methods, underpinned by domain-level expertise, to develop plume reference datasets for TROPOMI.
Sulphur dioxide hotspots
We will develop our plume identification algorithm to study the entire TROPOMI SO2 record and build an up-to-date database of the time and location of each plume we identify. We anticipate, based on recent work, we will find the location of volcanoes, powerplants, smelting facilities, and shipping routes. These facilities can mostly be evaluated using existing inventories, although we expect that some new coal-fired power plants will be missing from the inventories due to a lag between national emission reports and inventory compilation. We will also use our new SO2 plume reference dataset to examine the spatial and temporal variations in the SO2 columns.
Elevated surface ozone levels are detrimental to human health and to the growth of a range of agricultural crops. Understanding the sensitivity of surface ozone to changes in emissions of nitrogen oxides (=NO+NO2) and volatile organic compounds (VOCs) is therefore an important scientific and policy-relevant quantity to understand. We will use collocated plumes of formaldehyde (HCHO), a high-yield product of VOC oxidation, and nitrogen dioxide (NO2) from our TROPOMI plume reference datasets to examine spatial and temporal variations in photo-chemical environments. The resulting HCHO:NO2 ratio plume reference data will help us to study changes in the photo-chemical environment in urban areas across the world.