A modeling study of shelf circulation off northern California in the region of the Coastal Ocean Dynamics Experiment 2. Simulations and comparisons with observations

This is the second part of a modeling study of wind‐forced flow on the continental shelf off northern California in the region (37°–40°N) of the Coastal Ocean Dynamics Experiment (CODE). Gan and Allen [2002] analyzed the shelf flow response to idealized wind stress forcing in a process‐oriented study. The study here applies forcing from observed winds and heat flux for April–May 1982 and compares the model results with moored current and temperature measurements. The Princeton Ocean Model (POM) is used in a three‐dimensional limited area domain with a high‐resolution curvilinear grid (approximately 1 km horizontal spacing, 60 vertical levels) and realistic coastline and bottom topography. The objectives of the study are to simulate the response of the shelf circulation field to time‐varying observed wind stress and heat flux, to compare model results with oceanographic observations to establish confidence in the model, and to subsequently analyze the model fields and the model dynamical balances to help understand the behavior of the observed flow. The model variables show overall good agreement with corresponding observations. Similar to the conclusions by Gan and Allen [2002] , it is found that the alongshore variability of upwelling is mainly controlled by the interaction of the wind‐forced shelf flow with the coastline and bottom topography. Different dynamical regimes in the regions north and south of the coastal capes formed by Pt. Reyes and Pt. Arena and in the more uniform region between these capes are identified and investigated. The results demonstrate that the coastal capes play a dominant role in causing alongshore variability of the upwelling flow, including the setup of an alongshore pressure gradient that forces northward currents during relaxation of southward upwelling favorable winds. An analysis of the balance of terms in the equation for potential temperature indicates that across‐shore temperature advection is the major contributor to the cooling of coastal water during upwelling, with a larger magnitude to the south of the capes. To the north of the capes, however, alongshore temperature advection is the dominant contributor to the colder water near the coast.