Summary: | The observed increase in Antarctic sea ice area over time is not reproduced by Earth System Models. One proposed reason for this discrepancy is that these models do not realistically represent ice shelves and the associated freshwater flux into the Southern Ocean due to basal melting. Previous work on the artificial addition of fresh water to the Southern Ocean has produced conflicting results depending on the model used. In this thesis results are presented from new experiments artificially enhancing the freshwater to the Southern Ocean in the Community Earth System Model version 1 (Community Atmosphere Model version 5) CESM1(CAM5) Earth System Model, building on previous experiments with the same model. Results were compared to the CESM1(CAM5) Large Ensemble (LENS), an available set of control runs of CESM1(CAM5). Experiments have been conducted to test the response of the Southern Ocean and Antarctic sea ice to seasonally varying freshwater input, and to determine the residence time of the artificial freshwater signal after the forcing has been turned off. We have also tested the response to freshwater input that increases linearly over time, both with and without the effect of the latent heat required to melt the ice that is entering the ocean. The amount of freshwater input is much larger than present observations, in an effort to isolate the response from the variability of the system. The seasonal freshwater enhancement experiments showed no significant difference in response from constant freshwater input at the same annual mean rate, due to the residence time of the freshwater signal being much longer than the period of the artificial freshwater input. Experiments with linearly increasing freshwater input over time without latent heat uptake resulted in a small positive trend in sea ice area in the austral summer, winter and spring, although the response was not significantly different from the LENS in autumn. The experiments with linearly increasing freshwater enhancement and latent heat uptake resulted in positive trends in sea ice area that were significantly higher than the LENS, and sea ice area magnitude up to 2.1 × 106 km2 greater than the LENS mean. This response is attributed to a combination of the indirect cooling effect of the stratification-induced reduction in vertical heat advection from depth and the direct cooling effect of latent heat uptake. The enhanced sea ice melt/freeze cycle in the experiments with latent heat uptake resulted in less freshening near the continent and greater freshening further north. This reduced ocean stratification meant that the direct cooling effect of the latent heat uptake from the ocean was the dominant mechanism in determining the sea ice response to freshwater input from ice shelves.
|