High-resolution modeling of the mean flow and meso-scale eddy variability around the Grand Banks of Newfoundland

Abstract

The mean flow and meso-scale eddy variability in the region around the Grand Banks of Newfoundland (GBN) are quantified by analyzing surface drifter observations, mean dynamic topography (MDT), along-track satellite altimeter observations, and the solutions of two high-resolution ocean models. By increasing the horizontal resolution from 6.5 km (CREG12) to 2.2 km (GBN36), the modeled mean kinetic energy (MKE) decreases and the eddy kinetic energy (EKE) increases in the study area. GBN36 obtains the MKE of surface geostrophic currents and total currents similar to that derived from MDT and drifter data, whereas CREG12 overestimates these quantities by 40-53 %. CREG12 and GBN36 underestimate the EKE of surface geostrophic currents by 46 and 30 %, respectively, with respect to the EKE derived from along-track altimeter data. The models do not reproduce the strong eddy activity in the Gulf Stream and its downstream region, possibly related to a northward shift of the Gulf Stream position by 0.5°in latitude compared with observations. Both models obtain wavenumber spectra of sea level anomaly in close agreement with the spectrum derived from along-track altimeter data, with a slope of −5 at wavelengths near 100 km on logarithmic spectral density scales.

Impacts of model resolution on simulation of meso-scale eddies in the Northeast Pacific Ocean

“…The models are integrated for one year under realistic atmospheric and lateral forcing. Unlike the region around the Grand Banks of Newfoundland (GBN) where strong western boundary currents are present (e.g., Zhai et al, 2015), the Northeast Pacific Ocean (NEP) features an eastern boundary current and a strong seasonal variation influenced by atmospheric forcing. Similar with the results for GBN, in NEP the models produce sea level anomaly (SLA) variations similar with altimeter observations, in terms of rms, skewness and wavenumber spectra.…”

“…Vertical eddy diffusivity and viscosity uses a 1.5 turbulence closure scheme (Gaspar et al, 1990;Blanke and Delecluse, 1993;Mellor and Blumberg, 2004;Axell, 2002 Zhai et al (2015) in analysis of the GBN36 model. They pointed that this model behavior is related to a numerical issue noted by Hollingsworth et al (1983).…”

“…The study of area of Zhai et al (2015) presents strong meso-scale eddies generated by instabilities associated with the confluence of strong western boundary currents, i.e., the cold and fresh Labrador Current from north, and the warm and salty North Atlantic Current from the southwest. In this study, we examine the impacts of model resolution on the simulation of meso-scale eddies in the Northeast Pacific Ocean, an eastern boundary current system with much weaker mean currents than the northwest Atlantic.…”

“…It simulates fine spatial scale structure of the seasonal hydrography and circulation, as well as the influence of stratification on tides represented by a striation pattern of alternating highs and lows in the M2 surface summer maximum speed in the Gulf of Maine (Katavouta et al, 2016). GBN36 covers the region around the Grand Banks of Newfoundland (Zhai et al, 2015). Compared with the solution of a model covering the North Atlantic and Arctic Oceans with a nominal horizontal resolution of 1/12 (CREG12; Dupont et al, 2015), GBN36 obtains increased eddy kinetic energy (EKE) and decreased mean kinetic energy (MKE).…”