|DEETZ, K - Free University Of Berlin|
|KLOSE, MARTINA - University Of Cologne|
|KIRCHNER, I - Free University Of Berlin|
|CUBASCH, U - Free University Of Berlin|
Submitted to: Atmospheric Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/21/2015
Publication Date: 2/1/2016
Publication URL: https://handle.nal.usda.gov/10113/5859843
Citation: Deetz, K., Klose, M., Kirchner, I., Cubasch, U. 2016. Numerical simulation of a dust event in northeastern Germany with a new dust emission scheme in COSMO-ART. Atmospheric Environment. 126:87-97.
Interpretive Summary: On 8th April 2011, a severe dust event hit the Autobahn A19 in northeastern Germany, leading to a massive pile-up. According to a local news agency, visibility was still below 50m during the rescue phase at the Autobahn. The dust was emitted from an adjacent agricultural area which was at the time prepared for the cultivation of potatoes and had a bare surface. The event was further favored by above-average temperatures and below-average rainfall in spring 2011. The study presented in this paper aims at reproducing this spring-time dust event in Germany using numerical modeling. The model is also tested for its sensitivity to a variation of soil moisture, vegetation cover, and dust source area. Model results are compared to measurements of the regional air quality monitoring network.
Technical Abstract: The dust emission scheme of Shao (2004) has been implemented into the regional atmospheric model COSMO-ART and has been applied to a severe dust event in northeastern Germany on 8th April 2011. The model sensitivity to soil moisture and vegetation cover has been studied. Soil moisture has been found to be relatively high in the model during the investigation period and has been reduced by different degree to investigate the resulting changes in dust emissions. Two different vegetation datasets have been tested as model input: the climatological vegetation cover data of COSMO-ART (ECOCLIMAP) and the SPOT5 remote sensing vegetation cover data for the time of the event. By varying soil moisture, vegetation cover and by restricting the potential emission area, a set of eleven simulations was generated. Vegetation cover during the event was about 24% lower on average compared to the climatological mean. Thus, dust emissions modeled with SPOT5 vegetation exceeded that modeled with climatological data by a factor of about 5. The modeled dust concentrations were compared with in-situ measurements of aerosol concentration. The temporal evolutions of simulations and observations have significant correlations (0.42 - 0.85) especially in rural backgrounds. The lower correlations at urban sites are attributed to anthropogenic PM10 sources, which are not included in the model. However, a verification of the magnitude of modeled dust concentrations is not possible due to the uncertainty in soil moisture and emission area.