The baseline image on which our study is based is the Natural Resources Canada (NRCan) image mosaic of the CAA, compiled from Landsat 7 scenes between 1999 and 2002, and orthorectified using ground control and conjugate points (40). The seamless mosaic is considered accurate to ~20 m, or less than one image pixel (30 m) (40). We downloaded the image tiles for our region of interest from the Canadian Open Government portal ( The glacier basin outlines were downloaded from the Global Land Ice Measurements from Space (GLIMS) database (41), with manual checking and correction of gross errors (e.g., coregistration of the basin outlines to the NRCan Landsat 7 mosaic). The glacier frontal positions for the period 1999–2002 were digitized from the mosaic and joined to the glacier basin outlines to produce closed polygons, with relevant attributes (such as glacier length, type, and ID) from the GLIMS database. Marine-terminating glaciers were identified as those that terminated in the ocean at the time of the Landsat 7 mosaic (1999–2002). In many cases, it was not possible to identify from the images the extent to which the glacier descended below sea level, and so, we used the general term “marine-terminating” rather than “tidewater” glacier. In total, we mapped all 300 marine-terminating glaciers in the CAA with basin size greater than 2 km2 (in 1999–2002), excluding those on the northern coast of Ellesmere Island (Fig. 1).

Glacier frontal positions were manually digitized, using Esri ArcGIS10.5 software from a range of sources (table S1). Aerial photographs, from survey missions between 1956 and 1961, were the first observations that covered the whole region. These were accessed from the NRCan National Air Photo Library. The glacier frontal positions were mapped from the aerial photographs by on-screen digitizing from scans of the photographs georeferenced to the Landsat 7 mosaic base image. Where original photographs were not available, the glacier fronts were digitized from scans of National Topographic System map sheets, which were compiled in the 1960s from the aerial photographs. An average of seven frontal positions, at approximately decadal intervals, were mapped per glacier. Post-1960, most glacier frontal positions were digitized from Landsat satellite images. These were accessed and downloaded using the U.S. Geological Survey Earth Explorer portal ( Image resolutions ranged from 5 m (aerial photographs) to 60 m (Landsat Multispectral Scanner), with all post-2000 images (Landsat Enhanced Thematic Mapper Plus and Landsat 8) at 15-m resolution. High relative positional accuracy was maintained by ensuring that the frontal positions were georegistered to the geospatially accurate Landsat 7 mosaic. The digitization methodology follows standard procedures when mapping glacier changes in previous work (42). Uncertainties in the glacier change results stem from the reliability of the source material, measurement error, and methods used to calculate change, as documented in Cook et al. (42). Similarly, in this study, it is estimated that the digitization uncertainties are less than the largest image pixel size (60 m). Uncertainty in the data analysis methods is described in the following section.

All glacier frontal data were assigned the relevant source material attributes including date, year, and source ID. In total, 2039 frontal positions were digitized for 211 glaciers in the QEI and 89 in BBI (fig. S1).

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.