Application of RACMO to the world’s largest Ice Sheet masses of Greenland and Antarctica is only useful after adaption or extension of the model to be capable of simulating the specific conditions met in these regions. Substantial modifications include the implementation of a roughness formulation for snow and ice covered surfaces (Reijmer et al., 2004); a snow melt – refreezing module to simulate (sub)surface phase transitions in the marginal Greenland ablation zones (Ettema et al., 2009); and a snow drift module bring into account the release of extra latent heat in regions of wind induced blowing snow (Lenaerts et al. 2011).

Various polar versions of RACMO have been extensively evaluated and shown to reliably simulate the climate and surface mass balance (SMB) of the Antarctic Ice Sheet (AIS) and the Greenland Ice Sheet (GrIS) (Reijmer et al., 2005; Van de Berg et al., 2006; Lenaerts et al., 2011c; Ettema et al., 2009; 2010a; Agosta and others, 2011). Based on its performance, output of RACMO2 was used to:

• Quantify the forcing of katabatic and barrier winds in Antarctica and Greenland (Van den Broeke and others, 2002; Van den Broeke and Van Lipzig, 2003a; Van Angelen and others, 2011a); this is illustrated for Greenland in Fig. 1
• Identify interactions between the large scale atmospheric circulation and poleward moisture and heat transport (Van Lipzig and Van den Broeke, 2002; Van den Broeke and Van Lipzig, 2003b);
• Assess the coupling between regional atmospheric circulations and sea ice transport through Fram Strait (Van Angelen and others, 2011b);
• Assess atmospheric blocking in the Antarctic Peninsula region (Van Lipzig et al., 2008);
• Create new maps and time series of the SMB of the AIS (Van Lipzig et al. 2004; Van de Berg et al., 2006; Lenaerts et al., 2011c) and GrIS (Ettema et al., 2009), identifying data-sparse regions on the AIS with very high and low accumulation rates (Van den Broeke et al., 2006a; 2006b), see Fig. 2.

Fig. 1. Vector average 10-m wind speed (arrows) and directional constancy (colours) over the Greenland ice sheet and surrounding oceans. Areas with high directional constancy are indicative of persistent, thermally driven regional circulations. Vectors are plotted every 9 model grid points. From Van Angelen and others (2011).

In combination with ice discharge data, RACMO2 provided SMB data to calculate ice mass balance time series for the AIS and GrIS and their recent contributions to sea level rise, including a partitioning of the mass fluxes (Van den Broeke et al., 2009; Rignot et al., 2008). High-resolution SMB results from RACMO2 are used for mass balance studies of individual glacier basins, as well as separate larger ice caps (Flade Isblink in northeast Greenland, Austfonna in Svalbard). RACMO2 is presently upgraded to the latest cycle of ECWMF physics and preparations are ongoing for coupling to a finite-element ocean model.