Measuring coastal cliff erosion using terrestrial photogrammetry in
the Kongsfjorden area, Svalbard – First results
B. Wangensteen¹, T. Eiken ¹, R. S. Ødegård² and J.L Sollid ¹
1.Department of Geosciences, University of Oslo, P.O.Box 1047 Blindern, NO-0316 Oslo, Norway
2. Gjøvik University College, Norway. P.O. Box 191, NO-2802 Gjøvik, Norway
E-mail: bjorn.wangensteen@geo.uio.no
Abstract
Four sites for measuring coastal cliff erosion in the Kongsfjorden
area on Svalbard (79°N, 12°E) were established in August 2002 and
revisited in August 2004. Sites with cliffs in unconsolidated material
and cliffs in bedrock where chosen. Photos were taken at distances
ranging from 7 to 15 meters from the cliff walls with a Hasselblad
camera. At each site photographs were taken from two or three
different camera positions to be able to create 3D digital terrain
models of the cliffs. The photos were scanned and digital terrestrial
photogrammetry applied to construct digital terrain models (DTMs).
The erosion rate can then be estimated by taking the difference
between two DTMs of different years. Due to the short distance
between camera and cliff, the accuracy is in the millimeter to
centimeter range. Results from site 1 and 4 show a generally stable
rock wall with spot like erosion of up to 11.9 cm/y and 3.8 cm/y
respectively.
Location of Svalbard and the sites in the Kongsfjorden area
Technique and precision
At all sites a fixed point for global reference were established; a
bolt was drilled into the bedrock and its position measured by
GPS. The positions of the camera stations were measured by
GPS and surveyed from the fixed point to create both a global and
a local reference. The photos were acquired with a stereo overlap
by using a Hasselblad camera with a 60 mm lens.
Photogrammetric control points were established using bolts
drilled into the cliff wall.
Results
The automatically generated DTM from 2002 for site 4 is shown
below as a shaded relief model. Above the erosion rate for site 1
is shown as the difference between the 2002 and 2004 DTMs. At
site 1 the measurements reveal an overall stable cliff wall, with
patches of erosion of up to 11.9 cm/y. Below the erosion rate of
site 4 is shown and it also reveals a spot like erosion of up to 3.8
cm/year in the otherwise stable rock wall. At both sites there are
also areas of sedimentation. This is probably due to actual
sedimentation, and to some extent also small inaccuracies of the
technique. The photo distance for site 1 is 14 meters and for site
4 it is 7 meters, giving a theoretical accuracy of some 3.0
mm/year for site 1 and some 1.5 mm/year for site 4 for the DTM
differencing over the two year time interval. The mean values for
the erosion rates are 5 mm/y at site 1 and 0.5 mm/y at site 4.
Meaning the overall erosion rate at site 4not being significant
compared to the accuracy.
The camera was mounted on top of a theodolite with a special
device making it possible to measure the exact position of the
camera and to ensure that the photo stations are on a line
parallel to the cliff wall. Photo distances in the range of 7-14 m
gives a scale of 1:117 to 1:233 and a XY-resolution in the order
of 1.2 – 2.3 mm. The theoretical elevation accuracy (Z-direction)
is in the range 1.4 – 5.6 mm for the given photo distances (0.200.40 ‰ of the photo distance).
Conclusions and future work
The method of close up digital terrestrial photogrammetry seems suitable
for creating accurate digital terrain models of coastal cliffs and hence
promising for detecting changes in the order of mm to cm/year. André
(1997) reports of Holocene rock wall retreat in the range 0 to 1.58 mm/y
in the same area.
Reference: André, M.-F. (1997) Holocene Rockwall Retreat in Svalbard: A triplerate evolution. Earth Surface and Processes and Landforms 22.
The overlapping photos of site 4 from 2002. The erosion rate is
calculated for the rock wall of this site.
Acknowledgement: The fieldwork was financially supported by the INTAS-project
“Arctic coasts of Eurasia: dynamics, sediment budget and carbon flux in connection
with permafrost degradation” (INTAS-2001-2329) and by two scholarships from the
Norwegian Research Council on behalf of the Norwegian Polar Committee .