Comparison of Wintertime Mesoscale Winds Over the Ocean Around Southeastern Greenland in WRF and ERA-Interim
Use the self organizing map (SOM) artificial neural network clustering algorithm to identify the southeastern Greenland oceanic wind regimes in regional climate model (WRF) and reanalysis (ERA-Interim) data. Assess if these patterns are correlated with the North Atlantic Oscillation and compare the strength of these patterns in WRF and ERA-I.
This analysis revealed the details of various barrier wind and tip jet wind regimes around the coast of southeastern Greenland. It was found that the North Atlantic Oscillation is not well correlated with barrier winds during tip jet events suggesting that this large-scale climate index is insufficient for assessing mesoscale atmosphere / ocean coupling in this region. The analysis also indicated that ERA-I underestimates the strength of these strong mesoscale winds highlighting potential shortcomings in reanalysis forced ocean-only simulations of these events.
Strong, mesoscale tip jets and barrier winds that occur over the ocean near southern Greenland have the potential for impacting deep convection in the ocean. The self-organizing map (SOM) training algorithm was used to identify and classify the range of 10 m wind patterns present during ten winters (1997–2007, NDJFM) in the ECMWF interim reanalysis (ERA-I) and from a regional simulation using the weather research and forecasting (WRF) model at 50 km into a SOM. The SOM is used to identify differences in the manifestation of westerly tip jets, easterly tip jets, and barrier flow. The North Atlantic Oscillation (NAO) index is well correlated with the type of tip jet present at Cape Farewell, but the NAO was not well correlated with the absence or presence of barrier flow. WRF simulated patterns with strong barrier-parallel flow more frequently than ERA-I, and WRF also had faster coastal winds than ERA-I during all types of strong wind events. The difference in coastal winds is likely related to model resolution and the resulting ability of each model to simulate strong mesoscale winds that are driven by Greenland’s steep terrain.