Space-Time Contour Models for Sea Ice Forecasting
The amount of sea ice (frozen ocean water) found in the Arctic is declining rapidly as a result of climate change. This has increased the need for accurate forecasts of where sea ice will be located. Of particular interest is predicting the sea ice edge contour, or the boundary of the region where at least 15% of the area is ice-covered. Current sea ice forecasts are issued from deterministic numerical prediction systems. While these physics-based models have skill in anticipating the location of the sea ice edge, they are also affected by systematic errors with complicated spatiotemporal structure. In particular, bias in sea ice edge forecasts is location-dependent, varies by season, and is non-stationary due to the climate change trend. In this talk, I will introduce contour-shifting, a new statistical method that anticipates and corrects spatiotemporal bias in contours. For a test set of predictions from the Geophysical Fluid Dynamics Laboratory CM2.5 Forecast-Oriented Low-Ocean Resolution model for 2001-2013, contour-shifting reduced the area incorrectly forecast by an average of 21%. I will also propose future work on a Bayesian framework for probabilistic contour modeling and its application to sea ice forecasting.