Sea ice is an important driver of climate patterns and polar marine ecosystem dynamics. In particular, primary production by microalgae in sea ice has been postulated as a sink for anthropogenic CO2, and as a critical resource in the life cycle of Antarctic krill Euphausia superba, a keystone species. Study of the sea ice ecosystem is difficult at regional and global scales, however, because of the expense and logistical difficulties in accessing such a remote and hostile environment. Consequently, models remain valuable tools for investigations of the spatial and temporal dynamics of sea ice and associated ecology and biogeochemistry. Recent advances in model representations of sea ice have called into question the accuracy of previous studies, and allow the creation of new tools to perform mechanistic simulations of sea ice physics and biogeochemistry. To address spatial and temporal variability in Antarctic sea ice algal production, and to establish the bounds and sensitivities of the sea ice ecosystem, a new, coupled sea ice ecosystem model was developed. In the vertical dimension, the model resolves incorporated saline brine, macronutrients concentrations, spectral shortwave radiation, and the sea ice algae community at high resolution. A novel method for thermodynamics, desalination, and fluid transfer in slushy, high-brine fraction sea ice was developed to simulate regions of high algal productivity. The processes of desalination, fluid transfer, snow-ice creation, and superimposed ice formation allowed the evolution of realistic vertical profiles of sea ice salinity and algal growth. The model replicated time series observations of ice temperature, salinity, algal biomass, and estimated fluid flux from the Ice Station Weddell experiment. In the horizontal dimension, sub-grid scale parameterizations of snow and ice thickness allow more realistic simulation of the ice thickness distribution, and consequently, sea ice algal habitat. The model is forced from above by atmospheric reanalysis climatologies, and from below by climatological ocean heat flux and deep-water ocean characteristics. Areal sea ice concentration and motion are specified according to SSM/I passive microwave satellite estimates of these parameters. Sensitivity testing of different snow and ice parameterizations showed that without a sub-grid scale ice thickness distribution, mean ice and snow thickness is lower and bottom sea ice algal production is elevated. Atmospheric forcing from different reanalysis data sets cause mean and regional shifts in sea ice production and associated ecology, even when sea ice extent and motion is controlled. Snow cover represents a first-order control over ice algal production by limiting the light available to bottom ice algal communities, and changes to the regional, rather than mean, snow thickness due to the use of different ice and snow representations are responsible for large differences in the magnitude and distribution of sea ice algal production. Improved convective nutrient exchange in high-brine fraction (slush) sea ice is responsible for up to 18% of total sea ice algal production. A continuous 10-year model run using climatological years 1996-2005 produced a time series of sea ice algal primary production that varied between 15.5 and 18.0 Tg C yr-1. This study represents the first interannual estimate of Antarctic sea ice algal production that dynamically considers the light, temperature, salinity, and nutrient conditions that control algal growth. On average, 64% of algal production occurred in the bottom 0.2 m of the ice pack. Production was spatially heterogeneous, with little consistency between years when examined at regional scales; however, at basin or hemispheric scales, annual production was fairly consistent in magnitude. At a mean of 0.9 g C m-2 yr-1, the magnitude of carbon uptake by sea ice algae will not significantly affect the Southern Ocean carbon cycle. Light availability was the dominant control on sea ice algae growth over the majority of the year; however, severe nutrient limitation that occurred annually during late spring and summer proved to be the largest control over sea ice algal productivity.