In January 2010, the UK carried out the annual groundfish survey in CCAMLR Sub-Area 48.3 on the FV Sil. Seventy-five random hauls were completed with good coverage across the shelf, representing an increase in effort overall compared with surveys carried out in 2008 and 2009 on the same vessel. Catch-weighted length frequencies of mackerel icefish indicated that although 3+ sized fish dominated the population, 2+ and 1+ sized fish were also present, and in larger proportions than in 2009. A mean biomass of 52,329 tonnes was estimated for mackerel icefish, with a lower 1-sided 95% CL of 24,334 tonnes, an increase on the biomass estimate for 2009. Considering the anomalous environmental conditions and low availability of krill to mackerel icefish in the area (evidenced by dietary analysis) in 2009, this increase in biomass was unexpected. The importance of krill in the mackerel icefish diets was greater in 2010 than in 2009 but still remained low compared to historical data, and Themisto gaudichaudii dominated diets in Northern area strata where the largest numbers of mackerel icefish were caught. As seen during the 2009 survey, mackerel icefish remained close to the bottom and very few acoustic marks were observed in the water column during the survey. Catches of Patagonian toothfish were greater than in 2009, but still low compared with historical survey catches. There was however, the first evidence since 2003/04 of a recent recruitment of evident at Shag Rocks, with presence of putative age 2+ fish (29-39cm). Very large catches of marble rockcod were experienced during the 2010 survey, with increased catches across all areas of shelf surveyed around South Georgia and Shag Rocks compared with previous surveys. The mean biomass estimate was 173,223 tonnes, more than ten-fold the estimate for 2008. Catches and biomass estimates for Scotia Sea icefish and South Georgia icefish were also greater than estimates for 2008 and 2009. Exploration of historical marble rockcod catch data and environmental variables is underway, but careful monitoring of all three species should be carried out to determine whether these estimates represent population recovery.