Mesoscale modeling of aerosol levels in Istanbul using a high resolution MM5/CMAQ air quality modeling system

Abstract

İstanbul is the largest and most populated city of Turkey and one of the biggest megacities in the world that frequently experiences high air pollution levels. Statistical approaches have been widely used to study these episodes. However, air quality modeling has not been much used to understand the nature and the meteorological and chemical backgrounds of these episodes. This study aims to evaluate the aerosol levels in İstanbul by establishing a mesoscale air quality modeling system using the PSU/NCAR MM5 meteorological model and CMAQ chemistry and transport model on a high temporal and spatial resolution. A high resolution emission inventory is developed for the modeling purposes for the first time for the city of İstanbul. The inventory covers most of the important anthropogenic source sectors including energy, residential and industrial combustion, traffic and shipping on 2 km horizontal resolution and hourly temporal resolution. The results shows that on-road traffic emissions are the main source for most of the pollutants such as CO, PM and NOx, industrial combustion is responsible for a high portion of SO2 emissions and solvent use and traffic are the main contributors for NMVOC emissions. However large uncertainties may be introduced due to the activity data, emission factors and temporal profiles used in the study. The modeling study is conducted for a 5-day period from January 13 to January 17, 2008. The period included a 5-day winter episode of PM10, with 24-hour averaged concentrations reaching at least twice as high as the EU legislation of 50 μgm-3. The results from the MM5 simulation are compared with observations from Kandilli and Finokalia (Crete) meteorological stations. Statistical measures show that the model captured the surface temperature and wind profiles successfully. On the other hand, better agreement can be achieved by optimization studies that focus on the physics option used for the parameterizations.|Keywords: İstanbul, Aerosol, Mesoscale meteorological modeling, Emission modeling, Chemistry and transport modeling.