An assessment is made of the ability of the Coupled Model Intercomparison Project 5 (CMIP5) models to represent the seasonal cycles of biogeochemistry of the Ross Sea over the late twentieth century. In particular, sea surface temperature, sea ice concentration, surface chlorophyll a, nitrate, phosphate and silicate, and the depth of the seasonal thermocline (measuring vertical mixing) are examined to quantify the physical-biogeochemical capabilities of each model, and to provide for ‘ranked’ model ensembles. This permits critical assessment of modelled Ross Sea biogeochemical cycling, including less well observed variables such as iron and vertically integrated primary production. The assessment enables determination of model output confidence limits; these confidence limits are used to examine future model scenario projections for consideration of potential ecosystem changes. The future scenarios examined are the representative concentration pathways rcp4.5 and rcp8.5. Our study suggests that by the end of the twenty-first century under rcp4.5 and/or rcp8.5 that there will be average increases in sea surface temperature, surface chlorophyll a, integrated primary production and iron, average decreases in surface nitrate, phosphate and silicate, and relatively large decreases in the depth of the seasonal thermocline and percentage coverage by sea ice in the Ross Sea.