As a net source of nutrients fuelling global primary production, changes in Southern Ocean productivity are expected to influence biological carbon storage across the global ocean. Following a high-emission, low-mitigation pathway (SSP5-8.5), we show that primary productivity in the Antarctic zone of the Southern Ocean is predicted to increase by up to 30 % over the 21st century. The ecophysiological response of marine phytoplankton experiencing climate change will be a key determinant in understanding the impact of Southern Ocean productivity shifts on the carbon cycle. Yet, phytoplankton ecophysiology is poorly represented in Coupled Model Intercomparison Project phase 6 (CMIP6) climate models, leading to substantial uncertainty in the representation of its role in carbon sequestration. Here we synthesise the existing spatial and temporal projections of Southern Ocean productivity from CMIP6 models, separated by phytoplankton functional type, and identify key processes where greater observational data coverage can help to improve future model performance. We find substantial variability between models in projections of light concentration (>15 000 (mu E m(-2) s(-1))(2)) across much of the iron- and light-limited Antarctic zone. Projections of iron and light limitation of phytoplankton vary by up to 10 % across latitudinal zones, while the greatest increases in productivity occurs close to the coast. Temperature, pH and nutrients are less spatially variable - projections for 2090-2100 under SSP5-8.5 show zonally averaged changes of +1.6 degrees C and -0.45 pH units and Si* ([Si(OH)(4)]-[NO]) decreases by 8.5 mu mol L-1. Diatoms and picophytoplankton and/or miscellaneous phytoplankton are equally responsible for driving productivity increases across the subantarctic and transitional zones, but picophytoplankton and miscellaneous phytoplankton increase at a greater rate than diatoms in the Antarctic zone. Despite the variability in productivity with different phytoplankton types, we show that the most complex models disagree on the ecological mechanisms behind these productivity changes. We propose that a sampling approach targeting the regions with the greatest rates of climate-driven change in ocean biogeochemistry and community assemblages would help to resolve the empirical principles underlying the phytoplankton community structure in the Southern Ocean.