Türkiye'deki akarsu havzalarında taşkın mevsimselliğinin belirlenmesi
Determination of flood seasonality in Turkey river basins
- Tez No: 507178
- Danışmanlar: PROF. DR. BİHRAT ÖNÖZ
- Tez Türü: Yüksek Lisans
- Konular: Mühendislik Bilimleri, Engineering Sciences
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2018
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: İnşaat Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Hidrolik ve Su Kaynakları Mühendisliği Bilim Dalı
- Sayfa Sayısı: 417
Özet
Bu çalışmada, Türkiye'deki 25 akarsu havzasında taşkın mevsimselliği incelenmiş, bu 25 akarsu havzasına ait 211 akım gözlem istasyonunda ölçümü yapılmış yıllık maksimum akımlar ve bunların gerçekleşme tarihleri ile çalışılmıştır. Bu çalışmanın sonucunda, Türkiye'deki 25 akarsu havzasında çalışılan istasyonlarda ölçümü yapılan taşkın görülme tarihleri kullanılarak, Türkiye genelinde, taşkınların mevsimselliği hakkında genel bir yorum getirebilmek, Mann-Kendall trend testi ile akım değerleri kullanılarak, incelenen istasyonlardaki akımın trendi ve akımın yönü hakkında bilgi edinmek amaçlanmıştır. Bu çalışmada, çalışılan istasyonların, karakteristik özellikleri ve coğrafi konumları gösterilmiş, çalışılan istasyonların, istatistik parametreleri hesaplanmış, yönsel istatistik metodu, göreceli frekans analizi, bu yöntemlerle benzerlik çalışması ve Mann-Kendall testi ile trend analizi metodları kullanılmıştır. Yönsel istatistik metodu ile mevsimsellik analizinde taşkının ölçümünün yapıldığı tarihin yılın kaçıncı gününe denk geldiği tespit edilip, ölçümü yapılmış her yılın taşkın gözlem tarihi, takvim günü olarak elde edildikten sonra bu tarihin açısal değeri radyan cinsinden hesaplanır. Bu sayede birim daire üzerine işaretlenebilen mevsimsellik parametreleri belirlenir ve ortaya çıkan birim daire üzerinde görsel bir hidrolojik model oluşturulur. İstasyonun taşkın verilerinin vektörel parametreleri hesaplanıp istasyonların birbiri arasında öklid mesafeleri hesaplanarak istasyonlar arası benzerlik değerleri hesaplanır. Yönsel istatistik metodu ile, çalışılan tüm istasyonlardaki taşkın görülme tarihleri, iki farklı yıl periyoduna bölünüp, bu veriler ile çalışılarak, ortalama taşkın tarihleri hesaplanır ve tüm istasyonlar için, taşkın görülme tarihlerinin, ileri ötelendiği yada geri çekildiği hesaplanıp istasyonlardaki farklı yıl periyotlarındaki taşkın görülme tarihlerindeki değişim incelenir. Göreceli frekans analizi yönteminde, istasyonlarda gözlemlenen yıllık görülen maksimum akımların tarihleri, ay bazında incelenmiş olup, aylık dönemler, 30 günlük periyotlara dönüştürülerek ayarlanmış frekans değerleri hesaplanıp, bu değerler neticesinde istasyonlar arasındaki benzerlik parametreleri belirlenip, bu işlemler neticesinde havzadaki homojenlik karşılaştırmaları bu analiz ile yenilenmiştir. Trend, bir zaman serisindeki rastgele değişkenin değerlerinin zaman içerisinde giderek artması yada azalmasıdır. Rastgele değişkenin zaman içerisinde seyrinin incelenmesinin amacı, hidrolojik süreçlerdeki düzensizliklerin belirlenip, ileriye yönelik tahminlerin yapılmasıdır. Bu çalışmanın sonucu olarak, Türkiye genelinde taşkın mevsimselliği ve mevsimsellik ölçütlerinin kullanılması ile benzer istasyonların belirlenmesi için kullanılan iki yöntemin sonuçları karşılaştırılmış olup, istasyonlardaki veriler 2000 yılı öncesi ve sonrası şeklinde iki farklı periyoda ayrılıp, farklı yıl periyotlarındaki taşkın tarihlerindeki değişimler incelenmiş olup, Mann-Kendall testinin sonucuna göre ileriye yönelik tahminler yapılmasına olanak vermiştir. Yapılan çalışma neticesinde, Türkiye'deki akarsu havzalarında taşkınların yoğun olarak kış ve ilkbahar aylarında gerçekleştiği görülmektedir.
Özet (Çeviri)
In this study, 25 river basins in Turkey, work was undertaken to examine the flood seasonality. At the 211 current monitoring stations of these 25 river basins, the measured annual maximum flows and their observed dates were studied. The mean flood dates, station similarities and data separated by different periods were analyzed by means of directional statistical method using the measured flood dates, average flood date by using directional statistical method. With the relative frequency analysis, the observed frequencies of the maximum currents observed at the monthly basis were calculated and the station similarities were determined at the current monitoring stations. In the study, directional statistic method, relative frequency analysis and trend test with Mann-Kendall methods were used. This study aimed to get an overview about the maximum flow characteristics in rivers in Turkey. As a result of the investigated current values and the calculations made, a general interpretation of all the rivers could be made. This work is based on the idea of creating a general resource for future planning and design of river structures. In this study, the characteristics and geographical locations of the studied current monitoring stations are shown. Statistical parameters are calculated. With the directional statistical method, the dates at which the maximum currents are observed in the current monitoring stations are calculated and each flood is shown on the unit circle. For each station, one average flood date is determined and similarity values are calculated with the parameters of this date. By using the relative frequency analysis method, the maximum currents observed at the current observation stations and the frequency at which the observations are made at the month-base are calculated and the similarity values of the stations are calculated as a result of these calculations.In the method of relative frequency analysis, the dates of annual maximum flows observed at stations were examined on a monthly basis. The frequency values adjusted by converting them into monthly periods and 30-day periods were calculated. Similarity parameters between stations were determined after these values and the homogeneity comparisons in the basin were renewed with this analysis. In specifying the station characteristics, the name of the current observation station, the time of observation, the precipitation area, and the elevation of the current observation station are used. When selecting the stations to be used in the study, stations with at least 15 years of observation time, no stream structure in the main, and no intermittent measurements were selected. As a result of this study, Turkey similar stations through the use of wide flood seasonality and seasonality criteria has been compared two methods results used to sectoring, according to the results of Mann-Kendall test has allowed forward for making predictions. While the flood parameters are calculated in the working current observation stations, the average of the currents observed in the station, the highest value is used. Central tendency measures, propagation measures and autocorrelation coefficients were calculated as statistical parameters. Result of these calculated values, the flood characteristics are investigated. With the directional statistical method, it is calculated that the observed dates of the maximum currents measured in the current observation stations correspond to the day of the year and are converted to an angular value. The coordinates of the flood day are calculated to obtain the visual parameter of the hydrological regime, the date when the maximum current is seen with this angular value. With the calculated coordinates, unit circles which are the visual parameters of the current observation station are obtained. The general coordinates of the flood dates are calculated in order to make comments on the general of the current observation station. After these coordinates are calculated, the angular value of the station's flood date is recalculated, the flood date is returned, and the flood time is calculated. At each station, the parameter that shows the diversity of the maximum currents seen in different years is calculated. With these calculated values, the intervals of the maximum currents in the stations are marked on the unit circle. Then, similarity values between the stations are calculated by using the calculated station coordinates. As the computed similarity value approaches zero, the similarity of the current observation stations increases, and the similarity of the current observation stations decreases as the individual approaches. This similarity values generated by homogeneous regions to comment on flood onset dates in Turkey. The average flood day for the basins is determined using data from all stations that are run for each basin. This determined flood day is obtained using the date of occurrence of maximum currents in the basin. The average flood day determined for each basin was determined and comments were made about the reasons for the floods in the basin. By relative frequency analysis, monthly relative frequencies are calculated for each month. According to this method, each day of the month is converted to an angular value of 1 °. According to this method, the 31-day month is multiplied by 31/30, 30-month 30/30, and February 28/30 by 29/30. In this way the year is reduced to 360 days. Monthly relative frequencies of flood formation provide detailed information on flood seasonality. After these values are calculated, the similarity of the flow monitoring stations is calculated by the root mean square differences method using the frequency values set for each month of the year. As the computed similarity value approaches zero, the similarity of the current observation stations increases, and the similarity of the current observation stations decreases as the individual approaches. This similarity values generated by homogeneous regions to comment on flood onset dates in Turkey. Relative frequency analysis tables were used to determine the frequency of floods observed on a monthly basis, and on the basis of these calculations, the number of floods observed more intensively was determined and comments were made on the reasons for the floods in the basin. In a time series, the value of the random variable increases gradually over time, and the trend is called the trend. The purpose of determining the variation of the random variable over time is to determine the order in the hydrological series to make forward estimates. Some parametric and nonparametric tests are used in practice with the trend specified. Prametric trend tests are more powerful than nonparametric tests, but such tests are not very reliable because they are assumed to be linear and independent in normal distribution. Nonparametric tests, on the other hand, are independent of the distribution of the random variable, and the outliers can be deduced. The most common non-parametric test used to test the trend in the time series is the Mann-Kendall test. When the Mann-Kendall trend test is performed, the time series is divided into two groups. After this divisor is calculated, the test statistic is calculated. The test statistic and the self correlation coefficient are calculated, then the standard normal distribution of the statistic is calculated. If this value is less than the selected significance level, the null hypothesis is accepted and the result is that the time series is not trendy. The results of the two seasonality methods studied were examined and the similarity of the methods with each other was examined. seen the maximum current date and current status of the review and similarities observation stations, made a general comment about the river basins, stations and flood similar to that seen by looking at the history, Turkey has been made public in a regionalization study. When data are analyzed in different periods, the data are divided into two different periods. The currents measured at the current observation stations are divided into two different periods, before and after the year 2000. Once this is done, the directional parameters of the stations calculated by the directional statistical method are discarded for the data divided into different periods. The general coordinates of the flood dates are calculated in order to make comments on the general of the current observation station. After these coordinates have been calculated, the angular value of the station flood date is recalculated and the flood date is calculated by calculating the flood time. As a result of this process, the data divided into different periods are examined and the changes in the maximum currents observed are examined. As a result of this process, the changes in the observed values of the current values are examined as they approach the current date. As a result of the study, 25 river basins in Turkey, studied in 211 gauging stations, the overall flood onset dates were determined by the method of directional statistics we have. Relative frequency analysis was used to determine frequency of floods on a monthly basis. This research studies, on the basis of both basins, and in some stations, were made both in throughout Turkey. Similar stations were identified and necessary data and information were obtained to complete the missing data. This operation results in 25 river basins in Turkey on studies of 211 gauging stations, onset dates of the maximum current calculated by the directional statistical method, the streams inspection station working with relative frequency analysis, and the distribution by the month of overflow seen frequency. The similarities of the stations studied with these calculations were calculated, and the similarity ratios of the current observation stations with each other were determined. Mann-Kendall trend test with all river basins in Turkey, the current trend has been observed. The trends in the current observation stations and the connections between the measured data in the stations have been determined. As a result of examining the data separated into different periods, the changes in the observed maximum currents were examined. The direction of this change and the changes in different year intervals have been determined. The results of this study, with the flood parameters calculated studied 211 gauging stations on 25 river basins in Turkey, in Turkey, could be made a general comment about the onset dates of floods. As a result of these investigations, it is made a general comment relating to the floods observed in Turkey. It determined that the predictions based on the maximum current observed in Turkey, this study is such that it references in the planning and design of future structures streams to do. This calculation results shows us, in Turkey observed flood seens in the winter and spring seasons. Winter floods are thought to be due to the heavy rainfall observed in winter, while spring floods are due to the increase in air temperatures and snow melt factors. The calculation result in Turkey's western coastal areas of flooding in winter, middle, northern and easters parts, is believed to be effective in the spring floods. It is observed that there is no trend in most of the stations studied, and most of the stations studied have withdrawn in large scale calendar dates on flood dates when they are divided into periods before and after year 2000.
Benzer Tezler
- 1253 Sohu deresi Fındıklı araştırma havzasına jeomorfolojik anlık birim hidrograf uygulamaları
Geomorphologic instantaneous unit hydrograph applications on 1253 Sohu stream Fındıklı research basin
ARİF YUSUF IŞIK
Yüksek Lisans
Türkçe
2001
İnşaat Mühendisliğiİstanbul Teknik Üniversitesiİnşaat Mühendisliği Ana Bilim Dalı
PROF. DR. BEYHAN OĞUZ
- Küçükdoğanca (Keşan-Edirne) bölgesi ve civarındaki kömür yataklarının yüzey ve yer altı sularına etkisi
Effects of coal mines near Kucukdoganca region (Edirne-Kesan) on surface and subsurface water
CEMİLE ERARSLAN
Yüksek Lisans
Türkçe
2010
Jeoloji Mühendisliğiİstanbul Teknik ÜniversitesiJeoloji Mühendisliği Ana Bilim Dalı
PROF. DR. YÜKSEL ÖRGÜN
- The influence of instream formations on hydraulic characteristics in natural lowland rivers
Doğal ova nehirlerinde akım ortamındaki oluşumların hidrolik özellikler üzerindeki etkisi
NAGHMEH HEIDARI
Doktora
İngilizce
2023
İnşaat Mühendisliğiİstanbul Teknik Üniversitesiİnşaat Mühendisliği Ana Bilim Dalı
PROF. DR. ORAL YAĞCI
DR. ÖĞR. ÜYESİ MURAT AKSEL
- Türkiye'de akarsu havzalarında kişi başına düşen su miktarının coğrafi bilgi sistemleriyle analizi
The analysis of per capita water amount in the river basins in Turkey by geographical information systems
CANSU KARŞILI
Yüksek Lisans
Türkçe
2011
CoğrafyaAnkara ÜniversitesiCoğrafya Ana Bilim Dalı
YRD. DOÇ. DR. NECLA TÜRKOĞLU
- Exploring spatial patterns and hotspots of hepatitis A and amoebic dysentery using GIS and geostatistical analysis in Turkey
Türkiye'de amipli dizanteri ve hepatit A hastalıklarının mekansal doku ve sıcak noktalarının CBS ve geoistatistik analizler ile incelenmesi
RUUSA-MAGANO DAVID
Yüksek Lisans
İngilizce
2017
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiBilişim Uygulamaları Ana Bilim Dalı
YRD. DOÇ. DR. AHMET ÖZGÜR DOĞRU