Quantification of shoreline movements along the Yukon Territory mainland coast between 1951 and 2011

Abstract:

The Arctic is warming, but the impacts on its coasts are not well documented. To better understand the reaction of Arctic coasts to increasing environmental pressure, shoreline position changes along a 210 km length of the Yukon Territory mainland coast in north-west Canada were investigated for the time period from 1951 to 2011. Shoreline positions were extracted from georeferenced aerial photographs from 1951, 1953, 1954, 1972, 1976, 1992, 1994, and 1996, and from WorldView and GeoEye satellite imagery from 2011. Shoreline change was then analyzed using the Digital Shoreline Analysis System (DSAS) extension for ESRI ArcGIS. Shoreline change rates decelerated to a mean rate of -0.5 m/a between the 1970s to 1990s, which was followed by a significant increase in coastal erosion to -1.3 m/a in the 1990s to 2011 time period. These observation indicate that the current rate of coastal retreat along the Yukon coast is higher than at any time before in the 60 year long observation record.

Coverage: 

Median Latitude: 69.300000 * Median Longitude: -139.140000 * South-bound Latitude: 68.950000 * West-bound Longitude: -141.000000 * North-bound Latitude: 69.650000 * East-bound Longitude: -137.280000

Description:

Total and annual shoreline movements along a 210 km long stretch of the Yukon Territory mainland coast were quantified, using a combination of air photos and satellite imagery. Aerial black and white photographs were obtained from the Canadian National Air Photo Library [NRCan, 2016] for the 1950s (i.e., 1951, 1952, 1953, 1954), the 1970s (i.e., 1972, 1976) and the 1990s (i.e., 1992, 1994, 1996). The usage of pictures from different years within the respective decades assured a complete coverage of the shoreline of the study area in the 1950s and the 1970s and coverage of seven short shoreline sections in the 1990s. The most recent (i.e. 2011) shoreline position was mapped using satellite imagery derived from the GeoEye-1 and WorldView-2 satellites [Digital Globe 2014, 2016].

Orthorectification of all aerial photos was performed by geo-coding all imagery to the 2011 satellite images using PCI Geomatic's Geomatica Orthoengine© software (2014) in the UTM map projection WGS84 Zone 7 North. The Yukon Digital Elevation Model (DEM) (30.0 m ground resolution) [Environment Yukon, 2016], airplane-based LiDAR (Light Detection and Ranging) elevation data (1.0 m ground resolution) [Obu et al., 2016], and the TanDEM-X intermediate DEM (12.0 m ground resolution) [Huber et al., 2012, doi:10.5194/isprsarchives-XXXIX-B7-45-2012 ] datasets were used to reduce terrain relief displacement. Aerial photographs from the 1950s and 1970s have a ground resolution of 3.5 m and 3.0 m, respectively, meaning that the smallest distinguishable objects are 3.5 m, or 3.0 m apart. Assuming that landscape changes which occurred between the overflights from 1951 until 1954 and from 1972 and 1976 are within the range of the ground resolution, photos from the 1950s and 1970s were treated as two coherent series. The 1990s were treated differently because the ground resolution of the photographs was higher (0.3 m). Photographs taken within one year in the 1990s each display a single site which was spatially separated from pictures from other years.

After geocoding of all pictures, the shoreline was digitized at a scale of 1:1 000. A set of indicators like (i.e., vegetation line, wet-dry line, cliff top line and cliff toe line) was used to digitize the shoreline. The usage of different shoreline types was necessary in order to capture the wide variety of landforms along the Yukon coast. The landforms were classified in five geomorphological classes, being: 1) Beach, barrier island, spit, 2) Inundated tundra, 3) Tundra flats, 4) Tundra slopes, and 5) Active tundra cliff.

Shoreline change rates were calculated using the Esri ArcGIS extension Digital Shoreline Analysis System (DSAS) version 4.3 [Thieler et al., 2009]. The rates were computed along transects perpendicular to the shoreline, with a transect spacing of 100 m. The data table contains DSAS transect intersection points with each shoreline in UTM Map projection WGS84 Zone 7 North. For each transect and each time step, the total shoreline movement as well as yearly shoreline change rates were extracted. DSAS analyses were conducted for the time steps from the 1950s to the 1970s, the 1970s to 2011 and for the 1950s to 2011 for the whole study area, and for the time steps from the 1970s to the 1990s and from the 1990s to 2011 for the areas for which shoreline position data from the 1990s was available.