Publication details
Event: EGU General Assembly 2021 ()
Date: April 19th 2021 – April 30th 2021
Fulltekst: meetingorganizer.copernicus.org/EGU21/EGU21-9126.html
Summary:
Svalbard is a near pristine Arctic environment, where long-range transport from mid-latitudes is an
important air pollution source. Thus, several previous studies investigated the background
nitrogen oxides (NO x ) and tropospheric ozone (O 3 ) springtime chemistry in the region. However,
there are also local anthropogenic emission sources on the archipelago such as coal power plants,
ships and snowmobiles, which may significantly alter in situ atmospheric composition.
Measurement results from three independent research projects were combined to identify the
effect of emissions from various local sources on the background concentration of NO x and O 3 in
Svalbard. The hourly meteorological and chemical data from the ground-based stations in
Adventdalen, Ny-Ålesund and Barentsburg were analysed along with daily radiosonde soundings
and weekly data from O 3 sondes. The data from the ERA5 reanalysis were used to evaluate the
prevailing synoptic conditions during the fieldwork. Although the correlation between the NO x
concentrations in the three settlements was low due to dominant influence of the local
atmospheric circulation, cases with common large-scale meteorological conditions increasing the
local pollutant concentration at all sites were identified. In colder and calmer days and days with
temperature inversions, the concentrations of NO x were higher. In contrast to NO x values, O 3
concentrations in Barentsburg and at the Zeppelin station in Ny-Ålesund correlated strongly, and
hence the prevailing synoptic situation and long-range transport of air masses were controlling
factors for them. The Lagrangian models HYSPLIT and FLEXPART have been used to investigate air
mass transport and transformations during the large scale O 3 depletion and enrichment events.
The factors affecting Arctic springtime photochemistry of O 3 have been investigated thoroughly
using Lagrangian and Eulerian numerical weather prediction model data and Metop GOME-2
satellite observations.