Inferring Glacier Mass Balance and Detecting Climate Change using RADARSAT

Inferring Glacier Mass Balance and Detecting Climate Change using RADARSAT

Michael N. Demuth

National Glaciology Programme

mdemuth@nrcan.gc.ca

Alain Pietroniro

Aquatic Ecosystems Impacts Branch

al.pietroniro@ec.gc.ca

In 1996, the capability of RADARSAT SAR for the purpose of inferring glacier mass balance for a temperate alpine glacier and detecting climate change was examined. The following summarises the findings of the field and laboratory research conducted at Peyto Glacier, Alberta and the NWRI Geomatics Laboratory at the National Hydrology Research Centre in Saskatoon.

At the close of the summer ablation period RADARSAT can discriminate between firn and bare ice facies.
The interpretation of the electromagnetic variability of the ice facies in the ablation area is complicated by the influence of remnant weathering crust and macro-scale periodic surfaces (melt-water runnels), the later modulating the effective incidence angles relative to smaller-scale roughness constituents of the former.
The uncertainties of late-summer weather and their possible impact on the dielectric and scattering properties of the glacier surface must be reconciled using data from automated weather stations and related geophysical sensors.
For glaciers where superimposed ice may form, defining the equilibrium line and the lower extent of the accumulation area may be difficult. The paucity of data describing the dielectric and scattering properties of superimposed ice suggests a future research area for the radar remote sensing of high latitude and polar glaciers where this type of ice is most prevalent.
A reconnaissance-order SAR estimate of the of equilibrium line altitude (ELA) will be impacted by the accuracy of the reference digital elevation information. Moreover, for many glacier configurations where mass balance variations due to altitude are influenced or in some cases completely masked by local balance variations, defining the ELA may pose an irreconcilable problem.
A reconnaissance-order SAR estimate of the accumulation area ratio (AAR) will be impacted by difficulties in defining the upper bounds of the glacier. For an alpine glacier, these bounds are often co-located along large terrain-slope transitions where digital elevation information may be limited and have a deleterious influence on the ortho-modelling process.
The net mass balance - ELA, AAR regression model residuals for Peyto Glacier are not insignificant. It is determined that the total error inherent in the SAR reconnaissance method would seriously impact the confident estimation of mass balance normals and climate-related trends if the method were to be utilized over the longer term. As such, it is recommended that the technique be adopted only when traditional ground measurements used in the direct glaciological method are absent.

Figure 1. September 20, 1996 RADARSAT SAR late-summer interpretation image of Peyto Glacier after radar brightness processing and ortho-rectification.

Satellite path is left of image, ascending orbit looking east (C-band, HH polarization, nominal incidence angle at scene centre = 42º)

Figure 2. November 7, 1996 RADARSAT SAR early-winter reference image after radar brightness processing and ortho-rectification.

Figure 3 Glacier facies radar-brightness signatures and their temporal shift between image dates (diagonal lines represent the one-to-one correspondence).

All Figures by permission Geografiska Annaler.

Please visit:

Workshop on Methods of Mass Balance Measurements and Modelling, Tarfala, Sweden August 10-12, 1998.

A workshop on methods and modelling of mass balance measurements was run at Tarfala Research Station in August 1998. The workshop was limited to 25 participants but the total number of abstracts received for the proceedings of the workshop exceeded 40, indicating large interest. At present roughly 30 papers are under review for the proceedings. Accepted papers will be published as issue 4 of volume 81A of Geografiska Annaler.