LEE- Atmosfer Bilimleri-Doktora
Bu koleksiyon için kalıcı URI
Gözat
Yazar "Deniz, Ali" ile LEE- Atmosfer Bilimleri-Doktora'a göz atma
Sayfa başına sonuç
Sıralama Seçenekleri
-
ÖgeAir-Sea Interactions in the Formation of Thunderstorms over Marmara Region: Physical Processes and Modelling(Graduate School, 2022-10-28) Yavuz, Veli ; Deniz, Ali ; 511162006 ; Atmospheric SciencesThunderstorm and sea-effect snow (SES) are meteorological events that adversely affect both daily life and the transportation sector, and occur as a result of air-sea interaction. Thunderstorm, is a meteorological phenomenon that usually contains strong winds and causes strong precipitation (e.g., rain, hail) accompanied by lightning. Intense atmospheric instability, high humidity, vertical triggering mechanisms, and wind shears are essential components for thunderstorm formation. Thunderstorms caused by cumulonimbus clouds occur both with precipitation and without precipitation. On the other hand, they occur not only with rain, but also with snow and hail. This phenomenon, known as a severe weather event, occurs as a result of convectional movements. For this reason, the increasing differences between the upper atmospheric level air parcel temperature and the sea or lake surface temperatures play an intensifying role in thunderstorm activities. SES occurs as a result of atmospheric instability, which occurs as a result of dry and cold air masses gaining heat and moisture fluxes as they pass over warmer water bodies. As a result of the instability of very cold and dry air parcels of arctic or polar origin during their passage over the Black Sea located in the north of our country, the Marmara Region and the Black Sea Region are exposed to SES in the winter months. Although SES cause heavy snowfalls in regions where they are effective as severe weather events, they do not need synoptic-scale systems to occur. SES that occur in a region where synoptic-scale systems are currently dominant are called as snowfall with increased sea effect. This type of snowfall causes more intense snow on the regions where the SES bands pass. In our country, there are studies on both severe weather events mentioned above. There are both case studies and climatological analyses in the literature regarding the thunderstorm event. However, climatological analyses and short-term forecast models are limited in terms of their effects on the Marmara Region and especially on the aviation sector. On the other hand, there are a very limited number of studies in the literature regarding the SES events for our country. There is no climatological comprehensive analysis or a prediction mechanism established for certain sectors for this event, where only a few assessments and case studies are included in the international literature. Both a knowledge and a scientific base are needed for this severe weather event, which is very effective in the northern parts of our country during the winter months and directly affects both daily life and transportation activities. In this context, in this thesis, a total of nine articles, two on thunderstorm events and seven articles on SES, were published in six different internationally refereed journals. In the first article, a short-term forecast of the thunderstorm that occurred at Istanbul Atatürk International Airport on February 2, 2015 was made. It has been determined that the thunderstorm and its severity can be predicted in the range of 42-57 minutes in advance by analysing radar images, Lightning Detection Tracking System (LDTS), sea surface temperature information, and surface/upper level atmospheric information. It has been revealed that this result is important for the decision makers working in the meteorology office. In the second article, various 15-year long-term statistics of thunderstorm events were investigated at 11 airports in the Marmara Region. Atmospheric conditions in which thunderstorm events occur and the effects of air-sea interaction were examined. Intraday, monthly, and annual distributions were analszed. No trend was determined on an annual basis, and the most observations were made in September on a monthly basis. In the daytime thunderstorm distribution, the most observations occurred between 1100-1900 hours. 72% of thunderstorms occurred with rain and 22% without any precipitation. One of the most important results of the study is that the Convective Available Potential Energy (CAPE) is zero in approximately 50% of the events. In the third article, the morphological classification of SES bands was made for the Danube Sea Area (western Black Sea) in the Black Sea. Between 2009 and 2018, SES bands were observed over the region in a total of 83 days, and these bands caused snowfall within the borders of the Marmara Region in 75 days. Various satellite and radar images obtained from the Turkish State Meteorological Service (TSMS) were used for the detection of SES bands. In total, five different types of SES bands have been defined for the region. The most common type of band among all events was Type-2 SES band with 85%. The most important reason for this is that this band type consists of parallel bands in the longitudinal direction. The north-south movements of polar and arctic air masses along the long fetch distance of the western Black Sea have created suitable atmospheric conditions for the Type-2 band. On average, the lowest sea surface temperatures (SESs) occurred in the northernmost part of the Danube Sea Area. This region is the region where SES bands mostly begin to form and are of weak intensity. The intensity of the bands increased due to the increasing SST towards the south. In the fourth article, the analysis of vortex (Type-5) SES bands that occurred in the western Black Sea on January 30-31, 2012 was carried out. The vortex band, which was detected with various images of two different satellites (Terra and MSG satellites), continued its effect on the region for approximately 24 hours. The structure of the vortex was analysed by analysing the surface level meteorological data obtained from 12 airports, 16 meteorological stations, and 5 radiosonde stations belonging to the countries surrounding the western Black Sea and various charts of the upper atmospheric levels. The 24 cm of snow depth was measured at the İnebolu meteorology station, where the SES bands connected to the vortex were effective, and a total of 102 flights were canceled across the countries surrounding the western Black Sea. In addition to these, many traffic accidents have also occurred due to heavy and sudden snowfalls. The surface and upper level atmospheric conditions that create the vortex were found as follows: presence of a strong/deep inversion layer between the surface and 700 hPa, the wind direction change being limited to a maximum of 50° in the same vertical range, the temperature difference reaches 25.8 °C in between the sea surface and 850 hPa at the point where the vortex core is located. In addition, in the analyses made with the synoptic charts, it was determined that the vortex was formed during the two-day period and there was a low pressure centre in the region where the core is located. In the fifth article, the statistical characteristics of SES events for the western Black Sea were analysed for the years 2009-2018. The main purpose of the study is to determine the meso- and synoptic-scale structures of SES events for the region in general and to create knowledge for nowcasting and forecasting applications to be made later. A total of 95 events were detected using satellite and radar images. These events were determined according to four different scale categories. As a result, 36 events (38%) are determined as Black Sea (BS) Events, 24 (25%) as Synoptic-scale (SYNOP) Events, 23 (24%) as Over Sea Convergence (OSC) Events, and 12 (13%) as Transition (TRANS) Events. While the average duration of SES events in four different categories was 15.9 hours, the longest time was 59 hours in the SYNOP Events. Except for the OSC Events, the prevailing wind direction under the inversion layer was northerly in all three categories. Inversion layer was detected in most of the BS Events and SYNOP Events, and the sea surface and upper level temperature difference was 4 °C to 6 °C higher on average compared to the other two categories. The highest number of events occurred in January with 51, and the lowest in March with 2. In the sixth article, the effect of short-wave troughs on the formation and development of SES bands in the western Black Sea was investigated. Between 2010 and 2018, a total of 48 short-wave troughs and long waves were detected in the presence of SES bands throughout the region. The most important role of long waves has been realized as a result of determining the direction of short-wave troughs. Afterwards, short-wave troughs and long waves were classified based on their direction of motion. Five different types of short-wave troughs and five different types of long waves were determined, and their monthly and annual statistics, and their durations were revealed. In addition, the temperature differences between sea surface and upper atmospheric levels were analysed for each type, and inversion conditions were examined. The average duration of short-wave troughs was found to be 27.8 hours, and the longest time was observed in LWT type with 60 hours. In 77% and 79% of all events, respectively, short-wave troughs and long waves were found to have the same direction of motion as the current SES bands. In the seventh article, thundersnows occurring in the Marmara Region within a 22-year period were investigated. For these events, the formation mechanisms, suitable surface, sea, and upper level atmospheric conditions were investigated. Based on the reports of 11 airports in the region, a total of 19 incidents were identified. In 17 of these events, the SES mechanism was found to be effective. After statistical temporal analysis, the predictability of these events was investigated with atmospheric stability indices. Accordingly, the most successful stability indices were Total Total Index and TQ Index. In the eighth article, a comparison of SES and non-SES for two international airports in Istanbul is made. A SES event prediction algorithm was developed for both airports using 10 years of atmospheric data. At the same time, suitable atmospheric conditions were determined according to the snow depths. Finally, in the ninth article, historical extreme winters in Istanbul were investigated. The events that took place within the scope of the 17-century period are shown, and based on the events that took place in the last few centuries, heavy snowfall and harsh winter forecasts are made for the period up to 2050 and 2100.
-
ÖgeAnalysis of the effects of air pollution on respiratory system diseases in İstanbul(Graduate School, 2021-06-12) Çapraz, Özkan ; Deniz, Ali ; 511142004 ; Atmospheric SciencesBreathing is the most basic human function required to sustain life. Humans are exposed to different types of air polluting substances present in air originating form various emission sources like industry, heating and traffic in cities and large urbanised areas. Breathing air contaminated with toxic substances produces health risks for individuals. Sensitive and vulnerable groups such as pregnant women, children, the elderly and those already suffering from respiratory and other serious illnesses are especially affected from air pollution. In the meantime, there is strong evidence that reductions of air pollution make a positive effect on public health. The relationship between air pollutants and its health effects has not been studied extensively so far in İstanbul and Turkey. The scope of this thesis is to assess the relationship between air pollutants and respiratory hospital admissions in İstanbul in order to better understand the association between air pollution and respiratory health in the largest city of Turkey. In this thesis, the relationship between air pollution and respiratory hospital admissions in İstanbul was investigated for the period of 2013 – 2017 using single-pollutant Poisson generalized linear model (GLM) while controlling for time trends and meteorological factors. Hourly air pollution data, including PM10, PM2.5 and NO2, were obtained from the database of Ministry of Environment and Urbanization, the government agency in charge of collection of air pollution data in Turkey. The daily concentrations for each pollutant were averaged from the available monitoring results of fixed-site stations of Air Quality Monitoring Network in İstanbul under Republic of Turkey Ministry of Environment and Urbanization. To allow adjustment for the effect of weather on hospital admissions, hourly meteorological data (temperature and relative humidity) were obtained from the Air Quality Monitoring Stations where meteorological measurements are also made. We have used daily means of the pollutants and weather variables calculated from the hourly data to represent the daily reading for İstanbul. Data of daily respiratory hospital admissions of the public hospitals in İstanbul from March 1, 2013 to December 31, 2017 (1767 days) were obtained from the database of the hospitals which are coordinated by the Republic of Turkey Ministry of Health. This thesis study is based on 3 SCI articles. In the first article, the associations between the daily variations of air pollutants and hospital admissions for respiratory diseases in İstanbul, the largest city of Turkey was examined. A time series analysis of counts of daily hospital admissions and outdoor air pollutants was performed using single-pollutant Poisson generalized linear model (GLM) while controlling for time trends and meteorological factors over a 3-year period (2013 - 2015) at different time lags (0 - 9 days). Effects of the pollutants (Excess Risk, ER) on current-day (lag 0) hospital admissions to the first ten days (lag 9) were determined. Data on hospital admissions, daily mean concentrations of air pollutants of PM10, PM2.5 and NO2 and daily mean concentrations of temperature and humidity of İstanbul were used in the study. The analysis was conducted among people of all ages, but also focused on different sexes and different age groups including children (0 - 14 years), adults (35 - 44 years) and elderly (≥65 years). Significant associations between air pollution and respiratory related hospital admissions were found in the city. Our findings showed that the relative magnitude of risks for an association of the pollutants with the total respiratory hospital admissions was in the order of: PM2.5, NO2, and PM10. The highest association of each pollutant with total hospital admission was observed with PM2.5 at lag 4 (ER = 1.50; 95% CI = 1.09 - 1.99), NO2 at lag 4 (ER = 1.27; 95% CI = 1.02 - 1.53) and PM10 at lag 0 (ER = 0.61; 95% CI = 0.33 - 0.89) for an increase of 10 mg/m3 in concentrations of the pollutants. In conclusion, this study showed that short-term exposure to air pollution was positively associated with increased respiratory hospital admissions in İstanbul and women and elderly people were more sensitive to respiratory risk of air pollution. In the second article, the Saharan dust particulate matter (PM10 and PM2.5) episode on İstanbul in February 2015 by using air quality and meteorological data with NASA satellite images and Aqua/Modis Satellite aerosol products was examined. The aim of this study was to better understand the effect of dust transportation on İstanbul's air quality. Although the effect of Saharan dust transportation on PM10 concentrations in Turkey was examined many times, its effect on PM2.5 concentrations has not been studied yet sufficiently. In February 2015, İstanbul experienced a Saharan dust episode and during this event the concentrations of particulate matter rose to very high levels. Satellite products, and air quality monitoring data from ground observations were utilized. On 01 February 2015, a cyclone centered on Adriatic Sea with a 990-hPa low pressure center caused a southerly wind event on the eastern Mediterranean. Desert sands lifted by strong winds hovered off the coasts of North Africa and spanned the Aegean Sea, passing over Istanbul and reaching as far north as the Black Sea. The dust storm hit the Marmara, Aegean, Black Sea, andMediterranean Sea regions of Turkey. Dust-laden weather was accompanied by low atmospheric pressure, warm air, and strong winds during the episode. The daily average air temperature on the day of the event was 17.4 °C which is well above the average (6.1 °C) and maximum temperature (9 °C) values of February of İstanbul. The daily average wind speed (6.5 m/s) was also remarkably high compared with the average wind speed value (2.6 m/s) of the city. On 01 February, very high AOD values (>1.0) were observed due to atmospheric dust transportation starting from the northern part of Libya, passing through the Aegean Sea and reaching to Black Sea over İstanbul. PM10 concentrations climbed to 325.1 μg/m3 on 01 February. However, PM2.5 concentrations did not increase considerably, only a slight increase occurred. This study showed that the PM10 concentrations increased significantly during the dust episode while PM2.5 concentrations did not increase considerably. There was only a slight rise in the values of PM2.5. The significant increase for the PM10 values can be explained by the higher gravitational settling velocities of coarse particles in the atmosphere. Another result of this study is the dust storm period was not significantly associated with respiratory hospital admissions. In the third article, the effects of air pollutants (PM10, PM2.5, and NO2) on hospital admissions for asthma, chronic obstructive pulmonary disease (COPD), and acute bronchitis to better understand the association between air pollution and respiratory diseases were assesed in the city. In order to investigate the health effects of air pollutants (Excess Risk, ERR), a time-series analysis of daily respiratory hospital admissions and outdoor air pollutants was performed using single-pollutant Poisson generalized linear model (GLM) over a 5- year period (2013–2017) at different time lags (0–9). Our results show that air pollutants have significant immediate and delayed effects on hospital admissions depending on different diseases. NO2 and PM2.5 have the highest risk effects on the hospital admissions. According to the results, a 10 μg/m3 increase in PM10 was associated with a 2.0 % (95%CI: 0.63–6.30) increase at lag 5, PM2.5 with a 2.15 % (95%CI: 1.57–2.96) increase at lag 1 and NO2 with a 2.3 % (95%CI: 1.33–3.96) increase at lag 7 in the number of hospital admissions for asthma, respectively. A 10 μg/m3 increase in PM10 was associated with a 1.62 % (95%CI: 0.42–6.32) increase at lag 0, PM2.5 with a 1.78 % (95 %CI: 0.28–11.3) increase at lag 8 and NO2 with a 1.41 % (95%CI: 0.50–3.96) increase at lag 8 in the number of hospital admissions for COPD, respectively. A 10 μg/m3 increase in PM10 was associated with a 0.84 % (95%CI: 0.56–1.26) increase at lag 0, PM2.5 with an 8.06 % (95%CI: 3.36–19.4) increase at lag 9 and NO2 with a 0.73 % (95%CI: 0.68–0.77) increase at lag 0 in the number of hospital admissions for acute bronchitis, respectively. Compared to PM10 and NO2, the risk effect of PM2.5 on acute bronchitis is much higher. This study showed that air pollution is associated with increased hospital admissions for some of the most common and serious respiratory diseases in Turkey.
-
ÖgeShort-term wind power generation forecasting by coupling numerical weather prediction models and machine learning algorithms(Graduate School, 2023-06-29) Özen, Cem ; Deniz, Ali ; 511162001 ; Atmospheric SciencesRenewable energy has a crucial place in ensuring the security of the energy supply and achieving energy independence for the countries. Furthermore, the transition to renewable energy which is an eco-friendly alternative to the conventional power generation methods with fossil fuels has a very influential role in preventing global climate change. In addition to all these motives; wind energy has become a primarily preferred energy source for countries and investors thanks to having one of the cheapest levelized cost of energy among both renewable and other energy sources in recent years. The power generation in wind power plants is directly associated with the wind, which is an atmospheric variable that is difficult to predict with its dynamic structure and chaotic nature. Moreover, forecasting the wind which is an intermittent energy source becomes very important by considering the increasing ratio of the wind in the total energy share in terms of stability and reliability of the electricity grids. In order to ensure the energy supply security and keep the electricity grid in balance, wind power plant owners like all other power plants are required to undertake their power generation forecasts to the institutions responsible for the energy markets and/or transmission of the electricity of the countries. Any deviation between the observation and the forecast results in energy imbalance and it causes energy imbalance penalties for the power plant owners. Therefore, increasing the accuracy of the power generation forecasts not only prevents the large financial penalties it also contributes to energy supply security by facilitating the control of the electricity grid. In this thesis study, short-term power generation forecasts of wind power plants were covered in detail and three articles prepared about the topic have been published in the international peer-reviewed journal, Wind Engineering. In the first article, a novel hybrid day ahead wind power forecasting model that couples numerical weather prediction (NWP) model and gradient boosting machines have been proposed. While the Weather Research and Forecasting (WRF) model is used as NWP, two different WRF models have been run in the study. The first model has been run in low spatial resolution and their outputs were directly used in machine learning model training. Global Data Assimilation System (GDAS) data with a 0.25-degree spatial resolution has been used as the initial and boundary condition data for the low-resolution model. The reason for using the outputs of WRF models instead of using GDAS data directly is to increase the temporal resolution up to 10 minutes with a dynamical model instead of statistical methods. While the outputs extracted from the surrounding four grid points were used for the training of the model, a high-resolution WRF model with 333 meters of spatial resolution has also been run to compare the results of the proposed model with a well-configured WRF model. Since the study has been focused on the day-ahead wind power forecast, day-ahead forecasts of Global Forecast System (GFS) data were used in the testing of the proposed model and used as initial and boundary condition data for the WRF model. The proposed model has shown its superiority to the WRF model according to the statistical performance metrics, and improvement of 28.86%, 28.47%, and 14.8% has been reached in mean absolute error, root mean squared error, and Pearson correlation respectively. Besides its superiority in statistical metrics, the proposed model could also produce its forecasts in just 28.75 seconds after a training process which is done only once, while the WRF model requires 2.9 hours. Therefore, computational time in the operational stage of the model has also outperformed the WRF model. In the second article, a country-based wind power generation (WPG) forecast model was proposed using the CatBoost model with atmospheric variables of surface level and 700 hPa, 500 hPa, and 300 hPa pressure levels are extracted from the ERA5 data, which has 1-hour temporal and 2.5-degree spatial resolution. Twenty-six out of thirty-six different grid points which is the total grid number with 2.5 degrees to cover the entire country have been selected considering Turkey's spatial distribution of wind power plants. Besides the atmospheric variables, virtual wind turbines (VWT) have been cited on each grid point based on the wind class so that the power generation output of each VWT is calculated and used in training. Day-ahead forecasts of High Resolution (HRES) data of the European Centre for Medium-Range Weather Forecasts's (ECMWF) have been used as the test subset in this study since ERA5 and HRES resulted with the same model which is the Integrated Forecast System (IFS) of ECMWF. This also leads to a better understanding of the accuracy of the proposed forecast model. On the other hand, due to the continuous increase in Turkey's installed wind power, Turkey's hourly wind energy production was not directly used as the outcome of the model; instead, hourly production divided by total installed power was used. As in the first study, a decision tree-based machine learning algorithm, Catboost was used so that the importance of each variable was also presented. On the other hand, while feature selection (FS) methods were also included in the study; the effects of each of these methods on the model were also examined. After applying the collinearity detection in all data, Lasso, two different principal component analysis (PCA), recursive feature elimination (RFE), generalized orthogonal matching pursuit (gOMP), and forward variable elimination methods with early dropping (FBED) were used. While these methods reduce the complexity of the model by reducing the number of variables; they were also used to increase the accuracy. In addition, using the results of these FS methods; five different hybrid FS methods have also been proposed. The first of these is created by choosing the variables of the grid point that has been selected mostly by the FS methods; the remaining four select the variables selected by at least three, four, five, and six of these abovementioned six different FS methods, respectively. The fourth hybrid method has outperformed all the other methods in the study, and the normalized root mean square error and R2 were calculated as 7.6% and 0.8989, respectively. Besides, the energy production of the VWTs is selected as the most important variable, followed by wind speed and direction. In the third article, a short-term wind speed forecasting model which can predict the wind speeds of the six wind turbines of a wind farm located in the western part of Turkey from 10 minutes to 1 hour is proposed. Since this study is not focused on day-ahead forecasts and differs from the first and the second, GFS or HRES data were not used so that the forecasting has been done with the CatBoost model by using the System and Supervisory Control and Data Acquisition (SCADA) based data of the wind turbines, and the outputs of the two different WRF models have been used. While the first WRF model is configured in a single domain and National Centers for Environmental Prediction/Final (NCEP/FNL) data with a 0.25-degree spatial resolution has been used as initial and boundary condition data in that model, the outputs of the model have also 0.25-degree spatial resolution and 10 minutes time-frequency. On the other hand, the second WRF model was run to obtain the weather patterns affecting the wind farm. A VWT algorithm that has been used in the first and second studies was not used in the third article since it is aimed to forecast wind speed. Since SCADA data has outliers and missing data within, data preprocessing techniques like outlier detection, data treatment, and missing data imputation have been applied to the SCADA data before feeding the data into the model. First of all, a method in which k-means and isolation tree applications were combined to detect outliers in the data. Therefore, statistical models have been used to treat those predetermined outliers. On the other hand, the CatBoost model was used to build the relationship between WRF model outputs and SCADA data. This model has been used to impute the missing data afterward. In the study, the effects of three different data, namely SCADA, weather pattern, WRF model outputs, and the three data preprocessing techniques applied to SCADA data which are outlier detection, data treatment, and missing data imputation, on the wind speed forecast model were examined separately. Since it is aimed to forecast the wind speed of each wind turbine at 10-minute time intervals from 10 minutes to 1 hour, there were 36 variables in total to be predicted. While the best model has been chosen as the model in which all data preprocessing was performed and all different data types were used considering the statistical performance metrics, each proposed model has outperformed the simple persistence model which uses the previous time step for the next time step. On the other hand, while the air pattern that most affects Urla was calculated as purely advective with 50.76% relative frequency, the best mean absolute percentage error was obtained with 14.534% in this weather pattern. According to R2, the highest performance was seen in hybrid weather patterns with 0.9161; The lowest root mean square error and mean absolute error were observed in the pure anticyclonic weather pattern, which is usually associated with low wind speeds.