Investigation sensitivity of extreme precipitations to blacksea sea surface temperature in Artvin/Hopa case specialization with produce ensemble simulations

The increasing temperatures in many parts of the world and the consequent increase in the intensity and frequency of precipitation are very important for humanity and the future of the world. The importance of investigating the global and local effects of temperature changes as a result of the global warming and climate change is increasing every day. The mean global temperature has been risen 0.7 °C since the 1890 and anthropogenic sources like consumption of carbon derivatives are the first reason of climate change. Changing of global air circulations and increasing amount of extreme precipitations have been caused by global warming. Some regions are influenced more by extreme weathers and these areas are known as "hot-spot region". In additionally, sea surface temperature (SST) is also affected by climate change. Sea surface temperature causes changing of evaporation rates, wind speed and direction, near surface temperature and most importantly extreme precipitations. In this study, we investigated the extreme precipitation case over the Eastern Black Sea region of Turkey by using weather forecast model, WRF-ARW 3.8. The flood is triggered by excessive precipitation in August 23-24, 2015 killed 11 people and the landslides in Hopa, Artvin province have damaged many buildings. It is known that the SST is higher than a critical value of 27 ° C in the east of the Black Sea on 23-24 August 2015. The station based two days total precipitation exceeds 200 mm. One of the usual reason for this extreme incident is positive anomaly of sea surface temperature (SST) over the Black Sea where the significant warming trend is clear in the last three decades. In August 2015, the monthly mean SST is higher than 2 °C with respect to the period of 1981-2010. It is thought that the production of ensemble simulations with the high resolution model results of the relationship of extreme precipitation with the sea surface temperature will be more healthy to predict which extreme values will be released in the future. In this study, we designed six sensitivity simulations with WRF-ARW to describe the influence of the Black Sea as a moisture source. The inner simulation domain with 3-km horizontal resolution covers all the countries bordering the Black Sea and, initial conditions have been customized to produced ensemble simulations. The simulations have been driven by the ECMWF ERA ¬Interim and the NCEP NOAA SST data Nine different time simulations are generated by modifying the initialization date which are 1, 3, 7, 15, 25, 25.5, 26, 26.5 and 27 ¬days before the precipitation event. It is also noticed that the spatial variability of the precipitation produced by the reference simulation (SimRef) is consistent with the GPM data. Firstly, the simulations have been forced in the 3 days, 7 days, 14 days and 25 days time period for the August 23 and 24, 2015 heavy precipitation case for tuning. The simulation results have been produced by YSU PBL and Kessler microphysics which have compatible results compared to the satellite data. We analyzed the vertical cross section to investigate the model success of simulated atmospheric variables, such as the temperature, cloud mixing ratio, relative humidity and vertical wind. For 24.08.2015 06Z, the formation of cloud caused by strong updraft and vigorous vertical development over the coastal zone and steep mountains. Reference simulation results show that, model is successful that capture the precipitation pattern, moisture area and transporting these moisture source to hotspot region. In terms of analysis of the sensitivity to SST, we forced the simulations by subtracting 1 °C (Sim-1), 2 °C (Sim-2), 3 °C (Sim-3) from and adding 1 °C (Sim+1), 2 °C (Sim+2), 3 °C (Sim+3) to the NCEP NOAA Global SST data, considering only the Black Sea. Reference simulations (Ref¬Sim), sea surface temperatures unchanged simülations, consistent with GPM data and 3¬days simulation produced 185 mm for two days rainfall. The sensitivity subtracting simulations show that daily total precipitation for all these simulations gradually decreased based on the reference simulation and these results are important for Black Sea and its coastline where is extreme precipitation frequency is higher comparing to other Turkey locations. In addition, the latent heat over the Black Sea is decreased directly when the sea surface temperature has been reduced. However the rainfall intensity does not decrease in a same over the hotspot region for all simulations forced by modified SST. 24-hourly ensemble average precipitation rates over the all domain area for Sim+3, Sim+2, Sim+1, SimRef, Sim-1, Sim-2 and Sim-3 are 46.68, 40.21, 35.11, 31.45, 27.37, 21.47, 14.10 mm respectively over the hot-spot region. Despite the fact that the simulations signal indicates the same direction, degradation and increasing of the precipitation intensity does not point out the same magnitude for all simulations. Differences appear especially in the precipitation patterns. In additionally, modifying the sea surface temperature affects skin temperatures higher coastline than inner side of lands. Latent heat parameter is one of the usual suspects of extreme precipitations and in the -2° C simulations, latent heat amounts over the eastern Black Sea decrease dramatically, it occurs decreasing of total precipitation rates, so it is revealed that 2 °C (Sim-2) threshold is critical for SST sensitivity. 2 - 3 ° C SST changes cause more effective changes over the skin temperatures on land near sea areas. We have also calculated the humidity and evaporation differences between the simulations and these differences become more apparent over the eastern Black Sea. The latent heat amount, which is thought to be a triggering factor for rainfall, has changed in parallel with the SST change. Minus ensemble simulation results (except 2-days simulations) show that total precipitation for a day could not exceed 270 mm in each hotspot region grid .The one of the most significant result of this study is that 2°C temperature change is critical threshold for extreme precipitation over the hotspot region and the average of precipitation totals increases in parallel with the increase in SST. According to these results, the decreasing in SST not change only extreme rainfall but also cause a significant decrease in total precipitation. Another significant result is split of the precipitation pattern in +3°C ensemble simulations and a part of the pattern affects Georgia coastline, another part of the pattern affects Rize and Trabzon coasts. It was determined that the regional precipitation distribution shifted towards the extreme values on the hotspot area with the increase in sea surface temperature. Especially, the area covered by the rainfall higher than 100 mm on the selected region was determined as 7.6% for +1, 9.6% for +2 and 12.7% for +3. This value is 7% for reference simulation.. Based on the results obtained, it is seen that the models represent extreme rainfall very well. The ensemble¬ based sensitivity simulations have been analyzed with using other extreme precipitation in the past extensively to define the critical SST threshold and investigate extreme precipitation reaction to changing sea surface temperature.