Sensitivity of alpine glacial change detection and mass balance to sampling and datum inconsistencies T. Goulden1, C. Hopkinson1,2, and M. N. Demuth3 1Process Engineering and Applied Science, Dalhousie University, 1360 Barrington Street, Halifax, Nova Scotia, Canada 2Cold Regions Research Centre, Wilfrid Laurier University, University Ave, Waterloo, Ontario, N2L 3C5, Canada 3Geological Survey of Canada – Cryosphere Geoscience, 601 Booth Street, Ottawa, Canada
Received: 18 October 2012 – Accepted: 12 November 2012 – Published: 02 January 2013
Abstract. Glacial mass balance estimated through the geodetic method requires glacial
surface coordinate observations from historical and contemporary sources.
Contemporary observations and historical topographic maps are typically
referenced to separate horizontal and vertical datums and observed with
different sampling intervals. This research demonstrates the sensitivity of
glacial change detection to the datum considerations and sampling schemes
through case studies of Andrei, Bridge and Peyto glaciers in Western Canada.
To simulate the procedure of observing the glacial surfaces, profile lines
were sampled from Digital Elevation Model (DEMs) on contour intervals for
historical data and horizontal intervals for contemporary data. Profile
lines from the following scenarios were compared: (1) different horizontal
and vertical sampling schemes; (2) the horizontal datum was correctly
reconciled but the vertical datum was not; (3) the vertical datum was
correctly reconciled but the horizontal datum was not; (4) both the
horizontal and vertical datums were correctly reconciled; and (5) both the
horizontal and vertical datums were incorrectly reconciled. Vertical errors
of up to 6.9 m, 6.0 m and 5.0 m were observed due to sampling effects and
vertical errors of 22.2 m, 9.9 m and 55.0 m were observed due to datum
inconsistencies on Bridge, Andrei and Peyto glacier respectively. Horizontal
datum inconsistencies manifested as erratic levels of growth or downwasting
along the glacial surface profile and vertical datum errors manifested as a
consistent vertical offset. Datum inconsistencies were identified to
contribute errors of up to 257.2 × 106 m3 (or 87%) and 54.6 × 106 m3 (or 580%) of estimated volume change below and above the
equilibrium line respectively on Peyto Glacier. The results of this study
provide an estimate of typical errors due to sampling constraints or datum
inconsistencies as well as guidance for identifying where these error
sources have contaminated mass balance results.
Citation: Goulden, T., Hopkinson, C., and Demuth, M. N.: Sensitivity of alpine glacial change detection and mass balance to sampling and datum inconsistencies, The Cryosphere Discuss., 7, 55-101, doi:10.5194/tcd-7-55-2013, 2013.