Abstract
This study investigates the predictability of the dense advection fog over Istanbul on February 19, 2014, which significantly halted international as well as local transportation. Sensitivity simulations were conducted using the Weather Research and Forecasting (WRF) model forced by the ERA-Interim reanalysis data. A hierarchical approach was adopted. The first group of sensitivity simulations involving different microphysics schemes (WSM6, Morrison, Thompson-aerosol, NNSL, NNSL-CCN, and Milbrandt) indicated that the simulation with Milbrandt reproduced slightly better results for the fog event. Further sensitivity tests involving different planetary boundary layer (PBL) schemes (ACM2, BouLac, MYJ, MYNN2.5, MYNN, and YSU) were conducted. The YSU PBL scheme provided better diurnal air and dew point temperature variations compared to the observations at Ataturk and Sabiha Gokcen airports. We further investigated the performances of RRTMG, RRTMG-fast and Dudhia shortwave radiation schemes, and RRTMG and RRTM longwave radiation schemes. Our analyses revealed that simulation of the fog was very sensitive to radiation scheme. Although all PBL schemes were able to generate fog, a configuration with the YSU PBL scheme with Dudhia shortwave and RRTM longwave schemes produced comparatively low RMSE for temperature depression, 0.31 °C (0.23 °C), during the fog hours at Sabiha Gokcen (Ataturk) Airport. The model simulated the onset time of the afternoon fog well; however it reproduced the onset and dissipation times of the morning fog earlier than the observations. It is also found that the use of high-resolution initial and boundary condition data did not provide a significant improvement in the advection fog simulation.
Original language | English |
---|---|
Pages (from-to) | 617-641 |
Number of pages | 25 |
Journal | Theoretical and Applied Climatology |
Volume | 148 |
Issue number | 1-2 |
DOIs | |
Publication status | Published - Apr 2022 |
Bibliographical note
Publisher Copyright:© 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature.
Funding
This study is funded by Istanbul Technical University Scientific Research Program (grant number MGA-2017-40771). Computing resources used in this work were provided by the National Center for High Performance Computing of Turkey (UHeM) under grant number 5005092018). We would like to thank the Turkish State Meteorological Service for providing us with observation data. We also thank Mahmut Muslum for his help during the research. The WRF model simulations were initialized using ERA-Interim and ERA5 data downloaded from ECMWF and FNL data from NCEP.
Funders | Funder number |
---|---|
ECMWF | |
Istanbul Technical University Scientific Research Program | MGA-2017-40771 |
NCEP | |
National Center for High Performance Computing of Turkey | |
Turkish State Meteorological Service | |
Ulusal Yüksek Başarımlı Hesaplama Merkezi, Istanbul Teknik Üniversitesi | 5005092018 |
Keywords
- Advection fog
- Physics parameterizations
- Sensitivity simulations
- WRF