On the Spatial and Temporal Characterization of Motion Induced Sea Ice Internal Stress

In April 2007 an array of buoys was deployed in the Beaufort Sea with one aim (among others) of examining the relationship between internal ice stress and ice pack strain-rate or deformation. Here we present preliminary analysis of stress data from this experiment. This analysis is discussed in the context of strain-rate analysis that has been performed previously. In order to identify ice motion induced stress from stress measurements recorded at a point in the ice pack, we first need to remove the thermal stress signal from the measurement time series. We introduce a conceptual model of thermal stresses to support a method of extracting ice motion induced stress from stress buoy data. The model will require independent verification, which we outline, however is useful for understanding our results. In this paper we focus on spectral and scaling analysis of ice motion induced stresses, and compare these to similar analysis of sea ice strain-rate. By comparing spectral properties of stress and divergence we estimate that dynamic stress events (such as ridge building) may be felt at a stress sensor up to 45km from the site of deformation. Ice motion induced stresses demonstrate fractal scaling properties, and are anti-persistent. This echoes similar results that have been identified for sea ice strain rate across spatial scales from 10 to 1000 km. Ice motion induced stress and sea ice strain rate can not be described by Gaussian statistics, and have “fat tailed” probability distribution functions. These findings provide insight into how to model risk of large deformation, with large ice motion induced stress, events impacting any given place in the Arctic ice pack.