All Climate Events

Arctic Sea Ice Predictability and Prediction

Download as iCal file

Tuesday, 30 January 2018, 10:30

Tuesday, January 30, 2018. 10:30AM. Arctic Sea Ice Predictability and Prediction. Chao-Yuan Yang, University at Albany. Sponsored by Geophysical Fluid Dynamics Laboratory. More information here.


Arctic sea ice has experienced a dramatic change for the past few decades. This drastic change and its associated impacts have led to increasing demands on sea ice predictions to a wide scope of stakeholders from seasonal to decadal timescales. This talk will focus on Arctic sea ice predictability in CMIP5 decadal hindcasts and seasonal Arctic sea ice prediction using a newly-developed regional coupled model in combination with the assimilation of satellite-based sea ice data. First, we examined to what extent present-day coupled climate models can predict Arctic sea ice at longer timescales by analyzing CMIP5 decadal hindcast/prediction simulations. The results show that for most models, the areas showing significant predictive skill become broader associated with increasing lead times. Sea ice in the Atlantic side has lower predictability than that of the Pacific side, particularly at a lead time of 3–7 years, but the Atlantic side show reemerging predictive skill at a lead time of 6–8 years. Our analysis also suggested that initialized decadal hindcasts show improved predictive skill compared to uninitialized simulations. Second, we have developed a regional coupled atmosphere-sea ice-ocean model by coupling the Los Alamos Sea Ice Model (CICE) into the framework of the Coupled-Ocean-Atmosphere-Wave-Sediment Transport system (COAWST) for seasonal sea ice prediction. To better predict sea ice at seasonal timescale, accurate sea ice initialization is required. Here we implement the Parallel Data Assimilation Framework (PDAF) into the new model and use the localized error subspace transform ensemble Kalman filter to assimilate SSMIS sea ice concentration and CyroSat-2 and SMOS sea ice thickness to improve sea ice initial conditions. The sea ice prediction results during the melting season in 2017 will be discussed. 

 

Location  NOAA GFDL, Smagorinsky Seminar Room, Princeton, NJ.