A manuscript titled “A Fair Assessment of Sea Ice Age Reduces Bias and Gives New Insight to Arctic Sea Ice Dynamics”, led by Richard Davy (NERSC, and member of WP500), has recently been published in Geophysical Research Letters (GRL).

It presents the application of an ice-tracking algorithm to diagnose sea-ice age from daily concentration and drift data in a climate model and compares it to satellite-derived estimates using the same method.

It is shown that the derived sea ice age retains the spatial pattern and long-term trends of the reported model age, but with less than half the bias. Changes in sea-ice age highlight shifts in sea-ice dynamics more clearly than thickness or volume. This shows that using a common algorithm enables both a fair model validation and improves insight into sea-ice dynamics.

Figure 3 from the manuscript is shown below to highlight significant differences between the Model Derived and Model Default data sets and their respective biases relative to observations.

Figure highlighting significant differences between the Model Derived and Model Default data sets and their respective biases relative to observations.
a) Time series of mean sea ice age for NorESM2-MM Model Default, Model Derived, and observations (Korosov data set). The time series of the area covered by 1st, 2nd, 3rd, and 4th+ year old ice for (b) the observations, (c) the Model Default output, and (d) the Model Derived age. Dashed lines indicate the initial spin-up period that is excluded from analysis (see Methods). The insets show the climatology of the areas covered by each age category during the period 2000–2010.
Meer leren