The Climatology , precipitation types and atmospheric conditions of extreme precipitation events in western Turkey

This paper investigates the climatology, precipitation types and background physical mechanisms of extreme precipitations events (EPEs) over western Turkey during the period 2006-2015. The EPEs are described as the precipitation values above the 90 15 percentile obtained from the hourly precipitation dataset having high spatial resolution. Precipitation types associated with EPEs are identified by using radar outputs and Lamb Weather Type (LWT) approach. It is found that EPEs occurred more frequently in the Marmara and Aegean regions during autumn and winter months. In Marmara, mainly 21%, 17% and 15% of total autumn EPEs are observed as convective (E circulation 20 types (CTs)), cyclonic (C), and sea-effect (NE) extreme precipitations (EPs), in order. While convective EPEs are generally more active in the southern portions having rugged topography, cyclonic and sea effect EPs are more effective in the southwest and northeastern parts of Marmara. Among these three precipitation types, convective CTs produce more intense daily precipitation in the Marmara region with daily average 25 value of 66.1 mm. Based on the hourly observations, convective types of EP which developed by the interaction between high pressure center over Balkan Peninsula and low pressure center over eastern Mediterranean, show two peak values during afternoon Nat. Hazards Earth Syst. Sci. Discuss., https://doi.org/10.5194/nhess-2018-29 Manuscript under review for journal Nat. Hazards Earth Syst. Sci. Discussion started: 20 March 2018 c © Author(s) 2018. CC BY 4.0 License.


Introduction
The occurrence of extreme precipitation events (EPEs) and background physical mechanisms triggering these episodes become a fundamental issue in the last decade due to its great impacts on agriculture, health, energy, and tourism.From this perspective, many researchers investigated the role of large-scale circulations (e.g., Madden-Julian Oscillation, ENSO, PDO) on EPEs over the selected regions in US 10 (Jones, 2000;Higgins et al. 2000;Deflorio et al. 2013).Afterwards, the climatology and the synoptic characteristics of the extreme precipitations (EPs) are investigated in detail (Schumacher and Johnson 2006;Warner et al. 2012;Moore et al. 2015).
Owing to the spatial complexity, rugged topography, and land-sea interactions of the Mediterranean Basin, many devastating flash floods occurred in the various part of 15 the region in the last decade.Therefore, researchers have analyzed the atmospheric conditions that cause these extraordinary events by focusing on the selected flood days (e.g., Ferretti et al. 2000;Nuissier et al. 2008;Pastor et al. 2010).Only a few researchers analyzed the climatological and general synoptic behaviors of the EPEs for this large territory (Ricard et al. 2011;Reale and Lionello 2013).Ayamama creek in Istanbul (NW of Turkey, most populated city in Europe) was overflowed as the consequence of the dense daily precipitation episodes, which produced more than 250 mm rainfall over 3-day, and 32 people died together with millions of dollars of economic losses (Kömüşçü and Çelik 2013).During 9 October 2011, 238 mm rainfall total was measured during an 6 hour time-period at the province 5 of Antalya (south of Turkey) and damaged the infrastructure of the tourism center of the country (Demirtaş, 2016).During August 2015, torrential rainfalls ended up with a devestating landslide in Hopa district (NE Turkey, sloppy domain of the country) and 11 people died during this natural hazard (Baltaci, 2017).Turkey and its sub-basins are mainly influenced by these EPEs in all seasons in the variety of the atmospheric 10 conditions such as baroclinic waves and cyclones, mesoscale convective systems, landsea interactions and orographic forcing.
In literature, numerous studies investigated the influence of large-scale circulation patterns or synoptic weather types on precipitation mechanism over Turkey and its sub-regions (Karabörk and Kahya 2003;Karabörk et al. 2005;Unal et al. 2012; 15 Baltaci et al. 2015Baltaci et al. , 2017)).Only a limited number of these studies explored the atmospheric conditions that caused extreme precipitation over Turkey for a set of selected episodes (Kömüşçü et al. 1998;Kömüşçü and Çelik 2013;Demirtaş 2016).In Section 2, description of the precipitation characteristics of EPEs, along with the data and methods used, are described.Results of the EPEs and related discussion are presented in Section 3. The last part, Section 4, is devoted to the summary and 5 conclusions.

Precipitation dataset
Values of meteorological parameters in Turkey had been recorded manually from the late 1920s to the beginning of the 21 st century.After the year 2003, starting 10 from the western regions, existing meteorological stations were replaced by automatic ones (Automatic Weather Observing Systems, AWOS) and also virgin land was covered with AWOS stations by the support obtained from large projects.These projects can be explained in four parts as follows: 3) AWOS 246 and 350: To expand the spatial density of the meteorological stations, 246 and 350 AWOS stations have been started to be used in the following years.By the 246 stations, new districts were aimed to be 10 meteorologically covered, which did not have any active meteorological stations before (blue stars in Fig. 1a).Later, due to high topographical differences of the country, 350 new automated meteorology stations were mainly located at the higher elevation points and started to operate since 2016 (black squares in Fig. 1a).

15
In our study, for the first time, we aimed to obtain the climatological perspectives of EPEs in Turkey with high resolution and coverage.Therefore, we have chosen longterm hourly precipitation dataset of AWOS stations.For this reason, hourly precipitation records of 206 AWOS stations were selected for the investigation for the environmental characteristics of EPEs.Firstly, daily total precipitation amounts 20 (00-24 UTC) were calculated from hourly precipitation records.Quality controls of data were done by RCLIMDEX method which was explained by Baltaci et al. (in press).The years having more than 10 % missing data days and stations that are subjected to relocation were eliminated from the study.As a consequence of the quality control and assurance of precipitation data in the period 2006-2015, we selected 97 stations densely located in the west Turkey (Fig. 1b).From 97, 51 stations are found to be located in the Marmara (NW Turkey, pink points), and 46 to be located in the Aegean (W Turkey, light brown points) regions of Turkey.

Lamb Weather Type (LWT) methodology
The subjective version of the Lamb's work (Lamb 1972) was developed in 1972 and the objective methodology for the British Isles was applied by Jones et al. (1993).
According to the methodology, vorticity and directions of the geostrophic flows are  2015) and (2017) using the 5 daily mean sea level pressure data of NCEP/NCAR Reanalysis for the selected period.
For the Aegean region, six circulation parameters were calculated by using different 16 grid points and coefficients (due to latitudinal difference).

Identification of EPEs and precipitation characteristics
An extreme precipitation event is generally defined as a daily amount exceeding 10 a certain threshold.For example, Karl et al. (1996) used 50.8 mm to define extreme precipitation events for the United States.For our country, this and other threshold limits was not suitable because of the large topographical difference and irregularity of the precipitation distribution.For this reason, we developed a methodology that defines the threshold levels of each station according to its own precipitation characteristics.15 Thus, this relative technique is based on considering the largest 10% of the daily precipitation amounts of each station separately as its own extreme.Then, the annual contribution of EPs for each station is determined by a standardized total that is the division of the cumulative totals of EPs for each station by 10.
We defined EPE types by using the LWT technique and the radar outputs.For The physical mechanisms behind the EPs were investigated by using NCEP/NCAR Reanalysis products (Kalnay et al. 1996).For this purpose, sea level pressure (SLP) and temperature data of 850-hPa were examined on a 2.5° X 2.5° grid 10 resolution of the reanalysis data.For the sea surface temperature (SST) distribution over the neighboring sea areas around Turkey, NOAA High Resolution SST data provided by the NOAA/OAR/ESRL PSD, Boulder, CO, USA, from their website at http://www.esrl.noaa.gov/psd,(Reynolds et al. 2007) were used in the study.

Climatology of EPEs on the west Turkey
For the first time, different daily precipitation threshold limits of 97 stations were constructed from a 10-year dataset (Fig. 3a).According to the results, highest daily precipitation rates exceeding 100 mm are observed on the southern Aegean region where it can be classified as 'rich' in terms of extreme amounts of precipitation.This 20 suggests that if the daily precipitation amount of a station located in the south Aegean exceed this limit, that day is recorded as an EPE for that station.Daily precipitation threshold ranging from 60 to 100 mm is shown to be mainly located on the coastal regions of the west Turkey.When one move towards interior continental areas, daily EP The annual contribution of EPs for each station (cumulative totals of EPs for 5 each station divided by 10) is shown in Fig. 3b.We observe that the largest normalized annual amounts of EPEs is located mainly on the southwest of Aegean, middle-south and northeast of the Marmara region with values larger than 60 mm.It is interesting to see that the interior continental areas of the Aegean and Marmara region that was characterized as poor in terms of extreme amounts of precipitation (Fig. 3a), now 10 exhibit a better picture in their normalized value as generally having a better value between 40-60 mm.The reason of this can be the convective precipitation, generating intensified rain that can accumulate higher amounts of precipitation during a single rainstorm.On the other hand, western regions of Marmara that exhibited considerably larger threshold value with precipitation totals larger than 60 mm (Fig. 3a Seasonal distribution of the EPE frequencies can provide important information to understand the physical mechanisms forcing these extreme events.For this reason, 5 we analyzed total counts of EPEs for four seasons and the results are depicted in Fig. 5. It can be stated from Fig. 5 that winter (DJF) and autumn (SON) are more significant than the other seasons (Figs.5a, d).During winter, two cores over Aegean result in more than 6 extreme precipitation days (Fig. 5a).Spring is mainly characterized as having EPEs between 2 to 4 days on the eastern portions of the Aegean region (Fig. 5b). 10 During summer, highest count of the EPEs with 3 days is shown to be located over the Black Sea effected areas of the Marmara region (Fig. 5c).Seasonally, second highest frequency of EPEs can be found in the autumn.In this season, an area extending from northeast to south of Marmara receive a frequency considerably higher than 6 days (Fig. 5d).From this point of view, a detailed analysis of the atmospheric systems generating 15 EPEs and effecting Aegean region mainly during winter and Marmara region during autumn becomes important.Next section is focusing on this aim.

Regional features of the seasonal EPEs
In this section, we carried out frequency analysis of the seasonal EP events for Marmara region and documented the results in Table 1.Marmara basin seems to be the 20 most sensitive to these occasional precipitation events during the autumn season.It is clear in the Table 1 that 53% of the EPEs (sea-effect, cyclonic, and convective originated) occurred a total of 43 (addition of 14, 18, and 11) days in this season, and followed by 27 (addition of 2, 13, 11, 1), 15 (addition of 2, 9, 3, 1), and 22 (addition of  2).It can be seen from the table that cyclonic EPEs represent 61% of the winter-time extreme precipitation events belonging to the region.Second highest frequency belong to the autumn with a value of 43% corresponding to 28 events.

Precipitation characteristics of EPEs over Marmara with its 10 background synoptic-scale atmospheric conditions
As discussed in the previous section, we mainly focused on the months of autumn to analyze the spatial distribution of daily mean precipitation, to determine the counts of station based EPEs in the Marmara region and investigate the synoptic-scale atmospheric conditions responsible from the development of these extreme precipitation 15 events.In this respect, 2006-2015 period autumn mean precipitation values, counts of EPEs and their associated average weather maps are illustrated in Figure 6.
During cyclonic CTs, highest daily mean precipitation amounts exceeding 8 mm are shown to exist on the southwestern parts of the region.Similarly, the count of EPEs is higher on this portion of the Marmara region (Fig. 6a).When the synoptic composite 20 maps are analyzed, one can see the low-pressure center that probably came from west (Karaca et al. 2000) and located over Aegean Sea and west of Marmara.Sea surface temperature varies between 19 and 20 °C and temperature in the low level of the Peak times of extreme precipitation in a day can give us important information about understanding the possible causes triggering this event.For this reason, we investigated hourly behavior of the mean EPs for the Marmara and Aegean regions belonging to the autumn and winter months (Fig. 9).In Marmara, highest daily mean extreme precipitation is shown to occur under the convective types, and followed by 5 Black-Sea effected and cyclonic CTs.During convective activity, we showed a peak during afternoon hours of the day.Main reason of this event can be the diurnal heating and this is further investigated in the next section.For the Black Sea effected EPs, we observe an hourly peak of the precipitation close to noontime and this suggests that when maximum solar radiation reaches the sea surface, significant amount of moisture 10 and heat are transferred by northerly flows to Marmara, generating a considerable amount of precipitation.During the cyclonic CT, the region takes dense hourly precipitation at the mid-afternoon of the day.In regard to Aegean in winter, cyclones generally release dense precipitation potentials from night to noon times.• At Aegean region, 61% of the total EPs occur from the cyclonic activity during winter and torrential rainfall is found to be experienced at the majority of the 20 stations, especially those located in the south.This condition can be explained by cold air transfer from north that meet with the relatively warm Aegean Sea and thus, convergence of warm air above the cold air generates cyclogenesis which results in heavy precipitation.

20 Turkey
is located at the east Mediterranean and EPEs there, in general, cause sudden flash floods resulting with deaths and economic losses in infrastructure and agriculture.As a result of the EPEs in the last decade, numerous flash floods and landslides occurred in some particular regions of Turkey.During September 2009, Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.
Although a number of prior studies have focused on the synoptic characteristics of the EPEs ending up with life or economic losses over Turkey, climatological characteristics 20 of these EPEs and underlying causes were not studied in detail.To overcome this deficit, we utilized high-resolution precipitation datasets to construct ten-year (2006-2015) climatology of EPEs in the western Turkey.Therefore, the goal of this study is to Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.document the spatio-temporal and environmental characteristics of the EPEs, and investigate the synoptic-scale patterns associated with EPEs.
1) AWOS 206: Excessive rainfalls on May 21-25, 1998, which also triggered 15 landslides, resulted in many flash floods over the western Black Sea region of Turkey.In order to eliminate damages originated from floods, TEFER project (Turkey Emergency Flood and Earthquake Recovery) was introduced that was financially supported by International Public Works and Development Bank (World Bank) by a fund of 369 million dollars, to strength the emergency early 20 warning systems in the west Turkey.As a part of this project, 206 AWOS stations as a total have been started to be operational during 2003 and 2004 (red points in Fig. 1a).With this project, 120 classical meteorology stations were Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.replaced by the new automated ones and additional 86 AWOS stations were installed into new areas.2) AWOS 151: After the setup of 206 AWOS, extra 151 AWOS stations were installed in the central and eastern parts of Turkey during 2009 (brown triangles in Fig. 1a).Out of 151, 120 manual meteorology stations were transformed into 5 new tech ones and the remainder 31 were located in the maiden areas of the country.

5 First
meteorology radar over Turkey was installed in Ankara for nowcasting purposes during the year of 2000 (Fig. 2).Afterwards, Istanbul, Zonguldak, and Balikesir radars were installed during 2003 by the TEFER project.Later, to detect EPEs that can be effective over Mediterranean and Black Sea, another six C band radars were setup to Izmir, Mugla, Antalya, Hatay, Samsun, and Trabzon cities during 2007.Due to 10 the forecast difficulties of convective precipitation by the numerical weather prediction (NWP) models, another four C band radars (Bursa, Afyon, Karaman, and Gaziantep) were located in the inner parts of the country during 2013.From the network of these 14 radar stations, we used Istanbul, Zonguldak, Ankara and Balikesir radar 8-min PPI (Plan position indicator), and Max (Maximum) products, which are provided by Turkish State 15 Meteorological Service (TSMS), to check the characteristics of EPEs (cyclonic, convective or sea-effect originated) in combination with Lamb Weather Type (LWT) technique (e.g., Hellström 2003; Burt and Ferranti 2012; Moore et al. 2015).Additionally, data of the other radars, Izmir and Mugla were analyzed in the study from the beginning of 2007, and data of Bursa and Afyon from the year of 2013.

20
Nat. Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.calculated using sea level pressure fields over a predetermined central point.As a consequence of the six parameters and certain thresholds for the defined region, totally 27 different circulation types (CTs) were defined (16 hybrid, 8 directional, cyclonic, anticyclonic, and unclassified types).In this study, same methodology and equations were adapted for the Marmara region from Baltaci et al. (

20 LWT
technique, basic air flows coming from the sea and generating extreme precipitation, defined as sea-effect precipitation.The cyclones, which generate severe precipitation over a defined region, are also characterized with cyclonic EPEs.We identified the convective EPEs as the precipitation bands coming from the terrestrial Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.areas.As a result of these definitions, EPEs having northerly (N) and northeasterly (NE) CTs for the Marmara region were considered as having Black-Sea effect.For the seaeffect EPEs over Aegean, westerly (W) and southwesterly (SW) CTs were selected.For the convective EPEs, easterly (E), southeasterly (SE), and southerly (S) CTs for Marmara and E and SE types for Aegean were chosen.In terms of cyclonic (C) EPEs, 5 low-pressure center over Marmara and Aegean was selected as cyclonic CT in accordance to LWT methodology.
Nat. Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.threshold decrease from 60 to 20 mm level.The lowest limits are observed in the semiarid continental areas of the Aegean and Marmara region as having threshold value lower than 40 mm, as illustrated with blue color in Fig. 3a and can be classified as 'poor' in terms of extreme amounts of precipitation.
), show a 15 worse image with the normalized values as having precipitation values between 40 to 60 mm.On October 28, 2010 intense daily rainfalls and associated many flash floods occurred on the western parts of Turkey.As a result of this extraordinary event, daily precipitation amounts exceeded 70 mm in the Bandirma province (south seaside station 20 of Marmara Sea, in Fig. 4).During this day, daily precipitation totals exceeding 50 mm is shown in yellow color on Fig. 4 that is extending from coastal Aegean region towards Marmara as an enlarging region and reaching up to Black Sea passing over Gulf of Izmit and Silivri.This squall line affected majority of the Marmara region and to a Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.lesser extends the Aegean region.However, many stations located outside of this critical yellow region also had rainfall totals above their extreme daily precipitation limits.
Nat. Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License. 1, 16, and 5) days in the summer, spring and winter months, respectively.During autumn, convective, cyclonic and Black Sea-affected types of EPs are most influential over Marmara with the percentages of 21%, 17% and 15%, respectively.At Aegean region that is represented by a total of 46 stations, EPs are more frequent during the winter months and totally 35 different winter days ended up with 5 cyclonic EPs during the 10-yr period (Table Nat. Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.

15 temperature
From the previous studies, it can be said that the link between precipitation intensity and temperature was explained by Clausius-Clapeyron (C-C) relation.C-C relation presents the moisture-holding capacity of the atmosphere to temperature, hinting a roughly 7% increase in atmospheric moisture storage per degree Celcius.Pall 20 et al. (2007) found a high agreement between the C-C relation and the changes in the rainfall extremes at midlatitudes.Lenderink and van Meijgaard (2008) found for Netherlands that changes in hourly and daily precipitation intensity generally increased at the 7% °C-1 rate anticipated by the C-C at temperatures below 10 °C, but that hourly Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.• While highest EP threshold limits are shown to exist at the seaside stations of western Turkey (above 80 mm), the lowest limits are observed at the semi-arid continental areas of the Aegean and Marmara regions.Seasonal numbers of the EP days showed that Marmara and Aegean areas of Turkey are more influenced from these intense rainfall episodes during autumn and winter months, 5 respectively.• During autumn, convective, cyclonic and sea-effect originated EPEs represent 21%, 17%, and 15% of total extreme precipitation numbers occurring in the stations of Marmara.If the region has the proper synoptic conditions (HPC over Balkan Peninsula and LPC over eastern Mediterranean) and diurnal heating, 10 convective types of EP mainly occur at the south of Marmara during afternoon and evening times of the day.Daily extreme precipitation amounts are more common in the southwestern parts of Marmara when the cyclone is located over Marmara.Additionally, as a consequence of the interaction between HPC over eastern Europe and LPC over central Anatolia, strong moisture can be 15 transferred by the northeasterly flows and this can result in higher daily precipitation records that was sea-effect originated are shown to develop at the northeast parts of Marmara.

Table 1 .
Total extreme precipitation numbers over Marmara considering 51 stations and their percentage frequency distribution according to Black Sea-effect, cyclonic and convective precipitation types for the period 2006-2015.The total numbers of the days causing EPEs are shown in parenthesis.Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.

Table 2 .
Seasonal total extreme precipitation numbers over Aegean considering 46 stations and their percentage frequency distribution according to Aegean Sea-effect, cyclonic and convective precipitation types for the period 2006-2015.The total numbers of the days causing EPEs are shown in parenthesis.Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.

Table 3 .
Daily mean temperature and extreme hourly precipitation records and their temporal correlations during the 10 convective extreme precipitation days in the southern (Bursa) and eastern (Kocaeli) stations.Nat.Hazards Earth Syst.Sci.Discuss., https://doi.org/10.5194/nhess-2018-29Manuscript under review for journal Nat.Hazards Earth Syst.Sci. Discussion started: 20 March 2018 c Author(s) 2018.CC BY 4.0 License.