Bayes Model Ortalaması Yöntemiyle Kavramsal Yağış-akış Modeli Çıktılarının Değerlendirilmesi
Loading...
Files
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
Hidrolojik modeller yağış-akış ilişkisinin eldesi, akım tahmini, meteorolojik değişkenliğin akışa olan etkilerinin ölçümünde etkili bir rol oynamaktadır. Herhangi bir havzada tek bir hidrolojik model kullanmak diğer modellere kıyasla daha iyi sonuçlar vermeyebilir. Hazırlanan çalışmada Gediz Havzası'nın toplam drenaj alanını büyük ölçüde temsil eden Gediz Nehri üzerindeki Muradiye akım gözlem istasyonunda gözlenen akımlar altı farklı kavramsal yağış-akış modeli ile modellenmiştir. Elde edilen çoklu model çıktıları Bayes Model Ortalaması (Bayesian Model Averaging (BMA)) yöntemi ile değerlendirilmiş ve havzayı temsil edebilme performanslarına göre ağırlıklandırılmışlardır. Ağırlıklandırma işlemi ile her bir modelin katkısı değerlendirilip yeni akış serileri türetilmiştir. BMA yöntemi ile elde edilen çıktılar tekil kavramsal yağış-akış modeli çıktıları ile karşılaştırıldığında, BMA yöntemiyle elde edilen sonuçların daha güvenilir olduğu tespit edilmiştir
Hydrologic models play an effective role on acquisition of rainfall-runoff relationship, prediction of flow and measuring of the impacts of meteorological variability on runoff. Using single hydrologic model in any basin may not give satisfied results when it is compared to other models. In the study prepared, the flows observed at Muradiye streamflow gauging station located at the Gediz River, which represents the drainage area of Gediz Basin considerably, were modeled by means of six different conceptual rainfall-runoff models. The acquired multi-model outputs were assessed by means of Bayesian model averaging (BMA) method, and then weighted depending on representability performances of basin. By weighting process the contribution of each model was evaluated and the new runoff series were reproduced. When the outputs derived from BMA application was compared to those of single conceptual rainfall-runoff models, it was observed that the results related to BMA method were more reliable
Hydrologic models play an effective role on acquisition of rainfall-runoff relationship, prediction of flow and measuring of the impacts of meteorological variability on runoff. Using single hydrologic model in any basin may not give satisfied results when it is compared to other models. In the study prepared, the flows observed at Muradiye streamflow gauging station located at the Gediz River, which represents the drainage area of Gediz Basin considerably, were modeled by means of six different conceptual rainfall-runoff models. The acquired multi-model outputs were assessed by means of Bayesian model averaging (BMA) method, and then weighted depending on representability performances of basin. By weighting process the contribution of each model was evaluated and the new runoff series were reproduced. When the outputs derived from BMA application was compared to those of single conceptual rainfall-runoff models, it was observed that the results related to BMA method were more reliable
Description
Keywords
Bayes model ortalaması, Gediz Havzası, Kavramsal yağış-akış modelleri
Fields of Science
Citation
WoS Q
Scopus Q
Source
Volume
0
Issue
121
Start Page
41
End Page
57
