Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Master → Autriche → Comprehensive evaluation of WRF-Chem simulations for a Saharan dust episode in spring 2018 with earth observation data

Universität Wien (2020)

Comprehensive evaluation of WRF-Chem simulations for a Saharan dust episode in spring 2018 with earth observation data

Perny, Katharina

Titre : Comprehensive evaluation of WRF-Chem simulations for a Saharan dust episode in spring 2018 with earth observation data

Auteur : Perny, Katharina

Université de soutenance : Universität Wien

Grade : Masterarbeit 2020

Résumé
The capability of the Weather Research and Forecasting model coupled with chemistry (WRF-Chem) to represent Saharan dust outbreaks has been investigated by applying two different settings to a case study of a dust episode in March/April 2018 in the Mediterranean Region. Two model configurations in WRF-Chem regarding their ability to simulate PM10 surface concentration and aerosol optical depth (AOD) were compared : The Goddard Chemistry Aerosol Radiation and Transport (GOCART) model with simple GOCART dust emission and the Modal Aerosol Dynamics Model for Europe (MADE) - Secondary Organic Aerosol Model (SORGAM) model with GOCART dust emission with Air Force Weather Agency (AFWA) modification. \\ The PM10 concentrations were validated against ground-based measurements of the European Environmental Agency (EEA) provided by the Zentralanstalt f\"ur Meteorologie und Geodynamik (ZAMG). Simulated AOD values were compared with Aerosol Robotic Network (AERONET) in Athens, Finokalia and Lampedusa, as well as European Aerosol Research Lidar Network (EARLINET) data from Athens. This data was collected from the respective online databases. Additionally the ZAMG provided Moderate Resolution Imaging Spectroradiometer (MODIS) data for the evaluation period, which was interpolated to allow a point evaluation. \\ Both model configurations tend to overestimate the observed PM10 concentrations with significantly larger errors shown by the GOCART configuration. The simulated PM10 pattern of the different models does not vary much. The major difference in magnitude between the models can be found over the desert area, hence, the emission source region.\\ The AOD observation datasets show many missing values due to the cloudy conditions over the Mediteranean Basin on this occasion and thereby complicate the evaluation. An overestimation of the models regarding their simulated AOD peaks and an underestimation of the background AOD when compared to AERONET was found. The background MODIS AOD values fit better to the model values while the peaks are also overestimated by both models. Due to the data availability issues the result should be interpreted with caution

Présentation

Version intégrale

Page publiée le 9 mars 2022