Calibration and evaluation of Wavewatch III under extreme conditions in eastern mediterranean and aegean sea basin


Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Fen Bilimleri Enstitüsü, Türkiye

Tezin Onay Tarihi: 2019

Tezin Dili: İngilizce

Öğrenci: CEM BİNGÖL

Danışman: Gülizar Özyurt Tarakcıoğlu

Özet:

In this study, performance of WAVEWATCH III under extreme waves in Eastern Mediterranean Sea and Aegean Sea is assessed using Climate Forecast System Reanalysis wind data. Best source term is selected from all five source terms that is available in WAVEWATCH III for representation of wind waves in the region. Calibration of the selected source term is conducted for Aegean Sea and Turkish Coast of Mediterranean Sea. Three new Turkish State Meteorological Service buoys are utilized in the study which were never used for the calibration of extreme events in addition to Buoy 61277, Athos, Lesvos, Mykonos, Santos, Saronic and Skyros buoys from POSEIDON network. Two of these new buoys are located in the Turkish Coast of Mediterranean Sea and these buoys provide insight into wave characteristics in the region for the first time. Calibration for wind-wave input term parameters, swell dissipation parameters and wave-wave interaction parameters is conducted in the Aegean and Turkish Coast of Mediterranean Sea for eight case studies. It has been observed that in some regions which are close to land in Aegean Sea, wind interpolation performed by the model reduced the quality of input data. Four addition storms were modelled with the calibrated model setup. Overall, the calibrated model performs better than default parameterization of WAVEWATCH III, ST4 default source term for those cases where the wind input data underestimates the field conditions. These validation results are in a good agreement with buoy data for Antalya and Silifke Buoys. However, for the Aegean Sea, although the calibrated model setup performs significantly better, further improvement is still required both in wind input data as well as a better understanding of the spatial interpolation of the model.