Sea ice parameters operate at different spatial scales ranging fr.om millimetres to hundreds of kilometres. The Synthetic Aperture Radar (SAR) aboard the operational satellites ERS and RADARSAT has been successful in capturing some of those parameters, with limitati.ons linked to their spatial scale. In this study, results from in-situ measurements of ice physical properties are used with coincident SAR images to evaluate the utility of the images in retrieving ice parameters. The parameters are grouped into the following scale-based categories: sub-pixel, intermediate, and mesoscale. The sub-pixel scale parameters (e.g. brine volume and brine pockets geometry) are difficult to retrieve from single channel SAR data. The intermediate scale parameters (e.g. ridges and floe definition) can be identified when the incidence angle of the radar beam is relatively large. The mesoscale parameters (e.g. ice types and concentrations) can be retrieved using an empirical approach that links them to the observed backscatter.

INTRODUCTION

Since the beginning of this decade, the spaceborne Synthetic Aperture Radar (SAR) has been the main source of information for sea ice monitoring. At present, there are two operational satellite platforms with SAR on board: the European ERS and the Canadian RADARSAT. Sea ice information originally envisaged as being retrievable from SAR data encompassed a wide range of parameters which operate at scale levels ranging from a few millimetres to hundreds of kilometres. During four years of experience with ERS-I SAR images (1992–1995) in the operational ice monitoring program of the Canadian Ice Service (CIS), and through several field validation programs designed to relate observed backscatter to ice and snow parameters, more realistic perception of retrievable parameters has been realized. The scales of the geophysical processes at which ice parameters operate vary from a few millimetres to hundreds of kilometres. For the purpose of this study, the parameters are grouped into three categories according to their relevant scale.

This content is only available via PDF.
You can access this article if you purchase or spend a download.