Diagnostic sea ice predictability in the pan-Arctic and U.S. Arctic regional seas

Abstract

This study assesses sea ice predictability in the pan-Arctic and U.S. Arctic regional (Bering, Chukchi, and Beaufort) seas with a purpose of understanding regional differences from the pan-Arctic perspective and how predictability might change under changing climate. Lagged correlation is derived using existing output from the Community Earth System Model Large Ensemble (CESM-LE), Pan-Arctic Ice-Ocean Modeling and Assimilation System, and NOAA Coupled Forecast System Reanalysis models. While qualitatively similar, quantitative differences exist in Arctic ice area lagged correlation in models with or without data assimilation. On regional scales, modeled ice area lagged correlations are strongly location and season dependent. A robust feature in the CESM-LE is that the pan-Arctic melt-to-freeze season ice area memory intensifies, whereas the freeze-to-melt season memory weakens as climate warms, but there are across-region variations in the sea ice predictability changes with changing climate.