Numerical simulation of waves in the Caspian Sea: calibration and verification of the observation-based source terms

Third-generation models employ a host of parameterization schemes to consider the input wind forcing and the wave energy dissipation under different physical settings and environmental conditions. To evaluate the performance of such models in large-scale enclosed water bodies, WAVEWATCH-III model has been employed for the Caspian Sea which is a land-locked water body, located in the north of Iran. The Caspian Sea consists of three interconnected regions, called the northern, middle, and southern Caspian Sea. The division is based on the bathymetry and morphologic features, and the rapid variability in depth provides a unique modeling challenge. Recently implemented “ST6” physics in the WAVEWATCH-III model is calibrated and evaluated for the Caspian Sea. Moreover, both bottom friction methods, viz., JONSWAP and SHOWEX, in combination with calibrated ST6 were employed to fine-tune the model for this fetch-limited basin. Model performance was assessed using recent coastal measurements along the southern Caspian Sea as well as altimeter data via statistical parameters. Using ST6 calibration, the HH index for model performance against altimeter data was decreased by 7.7%. For the ADCP data, this improvement was ~17% at Noshahr station, and ~7% at Roudsar station, both along the southern boundary of the sea. The bias was decreased by 19.3% by calibrating the ST6 source term, specifically in the deeper areas. It should be noted that the bias against altimeter data has been improved by 19.5%. Due to the steep slope of the surf zone, the model was not so sensitive to the bottom friction formulation; however, the 1D propagation tests confirmed a slightly better performance of the SHOWEX formulation than that of the JONSWAP formulation.