| ABSTRACT | Several extreme weather events took place in Istanbul in July 2017. The heavy hail event on July 27 damaged hundreds of buildings and thousands of vehicles. The cost of this hazardous event was estimated to be around 300 Million US Dollars. As a result of these weather events, the total precipitation was recorded as of 30-40 kg. This study investigates this hail event using the state-of-the-art Weather Research and Forecasting (WRF) model. The model domain is set up with 4 nested domains (27, 9, 3 and 1 km resolutions from outer to inner) and Istanbul, located in northwestern Turkey, was used as the central point (41.96°N 20.06°E). Model simulations are performed for 30 hours starting from 18:00 UTC on 26 July 2017, and this time range includes 12-hour spin-up time. ERA-Interim Reanalysis dataset with 0.75°x0.75° spatial and 6-hour temporal resolution is used as the initial and lateral boundary conditions. Sensitivity tests were performed with different combinations of the parameterization. The performance of the model in simulating the hail event was assessed by comparing the model outputs with radar, satellite and meteorological station data which are obtained from Turkish State Meteorological Service. It is found that the hail event is best simulated when the model is run with Milbrandt 2-moment microphysics scheme, Kain-Fritsch cumulus scheme, MYNN2 planetary boundary layer scheme, RRTMG radiation scheme. As a result of the best model outputs, accumulated precipitation is 40 mm from 27 July to 28 July, hail event started at 14:15 UTC and the cloud top temperature on Istanbul is about -50°C in the same time. Deep convective clouds reached about 12 km height. Maximum hail concentration is about 400/kg at 14:15 UTC and occurred about 500mb pressure level. Reflectivity is about 50 dBZ when hail event occurred. |