Türkiye Yagis Siddeti Serilerinin Homojenlik Analizi
[ X ]
Tarih
2022
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Hakan ÇAGLAR
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Bu çalismada, Meteoroloji Genel Müdürlügü tarafindan isletilen 103 istasyondan 30 ile 78 yil arasinda degisen periyotlar için yillik maksimum yagis siddeti serilerinin homojenlik kontrolü yapilmistir. Mutlak homojenlik testleri olarak bilinen Standart Normal Homojenlik Testi (SNHT), Buishand Aralik Testi (BRT), Pettitt Testi (PT) ve Von Neumann Oran Testi (VNRT) %95 güven araliginda uygulanmistir. Standart süre verilerinden en az biri süpheli veya sorunlu ise, istasyon homojen degil olarak siniflandirilmistir. Homojenligi bozan faktörün muhtemel trend bilesenleri oldugu varsayilarak, trend bilesenleri ayrilmistir. Daha sonra, yagis siddeti serilerin mutlak homojenlik testleri yeniden uygulanmistir. Analizler sonucunda 103 istasyonun 49'u 14 standart sürenin tamaminda kullanilabilir olarak siniflandirilmistir. Kalan 54 istasyonun 45'i, trend bilesenlerinden sonra tüm standart süreleri kullanilabilir olarak siniflandirilmistir. 103 istasyondan geriye kalan 8’inde ise trend bilesenleri ayrildiktan sonra homojen olmayan degerlere sahip oldugu tespit edilmistir.
In this study, homogeneity checking the annual maximum rainfall intensity series for periods ranging from 30 to 78 years were obtained from 103 stations operated by the Turkish State Meteorological Service. Absolute homogeneity tests namely Standard Normal Homogeneity Test (SNHT), Buishand Range Test (BRT), Pettitt Test (PT), and Von Neumann Ratio Test (VNRT) were applied at a confidence level of 95%. Stations were classified inhomogeneous if at least one of the standard durations data classified suspect or doubtful. Assuming that the factor destabilizing the homogeneity is the trend, a detrended methodology (DFA) was performed. After the implementation of DFA, absolute homogeneity tests were reapplied to the series of rainfall intensity. As a result of this study 49 of 103 stations called useful all 14 standard durations. 45 of the remaining 54 stations classified useful all standard durations after trend components separated from rainfall intensity series. As a result of the analysis, it was determined that the remaining 8 of 103 stations had inhomogeneous values after the trend components were separated.
In this study, homogeneity checking the annual maximum rainfall intensity series for periods ranging from 30 to 78 years were obtained from 103 stations operated by the Turkish State Meteorological Service. Absolute homogeneity tests namely Standard Normal Homogeneity Test (SNHT), Buishand Range Test (BRT), Pettitt Test (PT), and Von Neumann Ratio Test (VNRT) were applied at a confidence level of 95%. Stations were classified inhomogeneous if at least one of the standard durations data classified suspect or doubtful. Assuming that the factor destabilizing the homogeneity is the trend, a detrended methodology (DFA) was performed. After the implementation of DFA, absolute homogeneity tests were reapplied to the series of rainfall intensity. As a result of this study 49 of 103 stations called useful all 14 standard durations. 45 of the remaining 54 stations classified useful all standard durations after trend components separated from rainfall intensity series. As a result of the analysis, it was determined that the remaining 8 of 103 stations had inhomogeneous values after the trend components were separated.
Açıklama
Anahtar Kelimeler
Homojenlik analizi, Türkiye, Yagis Siddeti
Kaynak
Journal of Sustainable Engineering Applications and Technological Developments
WoS Q Değeri
Scopus Q Değeri
Cilt
5
Sayı
2
