Regionally coupled modelling of the Tropical Atlantic

Currently, Global Coupled Models (GCMs) have difficulty capturing key phenomena and achieving accurate climate projections on regional and local scales because limitations in computer po wer do not allow them to reach the necessary horizontal resolutions. Regional climate models (RCMs) provide dynamically downscaled climate information within the region of interest, improving this drawback of current GCMs. At this point, naturally raises the question of how much, if any, the RCM can improve the GCMs results. It has been argued that regional models can reproduce an observed climatology but are not able to predict the change of the climatology in response to a changing climate (e.g. Kerr, 2013). However, Feser et al. (2011) could demonstrate an added value in those parameters that exhibit high spatial variability such as near surf ace temperature in different regional atmospheric models. They show that the added value originates mainly from the higher resolved orography in the regional models.