Going with the floe: insights from the CESM Large Ensemble on the MOSAiC expedition

by Patricia DeRepentigny (University of Colorado Boulder)

In October 2019, scientists trapped the icebreaker RV Polarstern in sea ice with the intention of drifting around the Arctic Ocean for a full year. The international Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) experiment aims to collect valuable observations on coupled air-sea-sea ice processes in a new regime of young, thin, seasonal Arctic sea ice. In recent years, there has been increasing awareness of the impact of internal climate variability on the possible range of sea ice conditions and the resulting representativeness of a single year of observations. But what insights can a climate model bring to a field expedition such as MOSAiC?

The Community Earth System Model (CESM) Large Ensemble (CESM-LE) is a fully coupled global climate model with well-represented Arctic sea ice mean state and variability. As an initial condition ensemble, the CESM-LE is a tool designed to explore the effects of internal climate variability and forced change. We use a Lagrangian ice tracking system to track virtual sea ice floes under satellite-derived historical conditions, perennial Arctic modeled conditions (characterized by old, thick perennial sea ice), and seasonal Arctic modeled conditions (characterized by young, thin, more seasonal sea ice). Using the likely starting location determined by the MOSAiC planners, we can put into context a single year’s observations, provide guidance about observations that can be used to improve climate models and demonstrate how free-running climate models might assist with future campaign planning. Note however that this study is not intended to provide a forecast for the campaign, but rather to contextualize the range of possible ice floe tracks, sea ice conditions and variability it may experience.

As the Polarstern searches for an initial floe from which to establish camp, the CESM-LE indicates that there is likely to be widespread ice cover with a mix of predominantly new, thin ice and some old, thick ice, but there is wide variability in the spatial ice coverage. Starting from the assumed likely starting location, the CESM-LE indicates that in seasonal conditions a path following the Transpolar Drift Stream is likely (47%) but would not have been as likely in perennial conditions (27%). The increase in likelihood of a Transpolar Drift path is consistent with satellite-derived tracks, which show the frequency of this type of trajectory increasing from 14% in the first half of the satellite record to 79% in the second half. The CESM-LE tracks show that in seasonal conditions, as compared to perennial conditions, thinner ice will drift more quickly. There is even a small (17%) chance the floe may melt out in August or September before a full calendar year, which is not the case for any observed or perennial floes. The variability in melt season conditions is higher for the seasonal tracks than for perennial tracks. As a result, it is more difficult to predict the initial sea ice state from which the campaign will start well in advance in seasonal conditions.

Future campaigns could use climate model ensembles to better understand the likely conditions contributing to outlier, hazardous paths. These simulated paths can also be used to coordinate airborne or surface measurements with acquisition of satellite imagery. Ultimately, the path that MOSAiC takes can be used to validate and improve modeled sea ice motion. Finally, observations of processes occurring during the autumn freeze-up taken during MOSAiC will be beneficial for understanding the formation and evolution of the sea ice thickness distribution and how this, in turn, affects sea ice predictability throughout the year.

DuVivier, A. K., DeRepentigny, P., Holland, M. M., Webster, M., Kay, J. E., & Perovich, D. (2020). Going with the floe: tracking CESM Large Ensemble sea ice in the Arctic provides context for ship-based observations. The Cryosphere14(4), 1259-1259. doi.org/10.5194/tc-14-1259-2020