Tenuous Correlation between Snow Depth or Sea Ice Thickness and C- or X-Band Backscattering in Nunavik Fjords of the Hudson Strait

Sophie Dufour-Beauséjour,Monique Bernier,Saeid Homayouni,Anna Wendleder,Jérome Simon,Achim Roth,Juupi Tuniq,Véronique Gilbert,Yves Gauthier

doi:10.3390/rs13040768

Abstract

Radar penetration in brine-wetted snow-covered sea ice is almost nil, yet reports exist of a correlation between snow depth or ice thickness and SAR parameters. This article presents a description of snow depth and first-year sea ice thickness distributions in three fjords of the Hudson Strait and of their tenuous correlation with SAR backscattering in the C- and X-band. Snow depth and ice thickness were directly measured in three fjords of the Hudson Strait from 2015 to 2018 in April or May. Bayesian linear regression analysis was used to investigate their relationship with RADARSAT-2 (C-band) or TerraSAR-X (X-band). Polarimetric ratios and the Cloude–Pottier decomposition parameters were explored along with the HH, HV and VV bands. Linear correlations were generally no higher than 0.3 except for a special case in May 2018. The co-polarization ratio did not perform better than the backscattering coefficients.

Highlights

Seasonal snow and ice covers in Nunavik are affected by the impacts of climate change: Kangiqsujuammiut have reported later sea ice freeze-up in the fall [1], as well as less snow on the ground, earlier sea ice breakup in spring, changes in travel routes and more variable winds [2]
We present the linear correlation coefficients between these variables and the results of the Bayesian linear regression analysis
The apparent disparity between the Bayesian and frequentist approaches, for example in case A, can be traced back to the fact that the Bayesian linear regression model evaluates the probability that a linear relationship exists considering the data provided

Summary

Introduction

Seasonal snow and ice covers in Nunavik are affected by the impacts of climate change: Kangiqsujuammiut (people of Kangiqsujuaq, Nunavik, in Canada) have reported later sea ice freeze-up in the fall [1], as well as less snow on the ground, earlier sea ice breakup in spring, changes in travel routes and more variable winds [2]. Inuit lives are embedded in the climate change context, and its impacts on sea ice have practical and immediate consequences on personal safety and access to travel and marine wildlife [3]. 2064 climate projections for the region showing shorter snow cover periods [5], sea ice conditions and their impact on land-use and marine transport will continue to evolve. This work is relevant due to land use and shippingrelated operations by communities and industries in the area and for the sea ice remote sensing community. It is part of the Ice Monitoring project, a collaboration between the Remote Sens.

Objectives

Methods

Results

Discussion

Conclusion