Sensitivity of alpine glacial change detection and mass balance to sampling and datum inconsistencies

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.