In the last decade, terrestrial laser scanning has been increasingly
applied for continuous snow depth mapping (e.g.,
Deems, 2013; Schirmer et al., 2011; Prokop, 2008; Prokop et
al., 2008). To calculate snow depth, an elevation model of the
bare ground and another of the snow-covered winter surface
is produced. Snow depth is then obtained by subtracting the
two surfaces from each other. In this study, we use the Riegl
LPM-321 device operating at 905 nm. This device has been
proven to accurately measure snow depth in alpine terrain
(Prokop, 2008; Prokop et al., 2008). Grünewald et al. (2010)
compared TLS measurements to tachymeter measurements
and found a mean vertical deviation of 4 cm with a standard
deviation of 5 cm at a distance of 250m using the LPM-
321. To assure the quality of the laser scans, we additionally
performed reproducibility tests. A laser scan acquired in
a coarse resolution (three points per square meters at a distance
of 300 m) was compared with the full-resolution acquisition
(eight points per square meters at a distance of 300 m).
This allows detecting misalignments between the two data
sets due to an instable scan setup (unstable tripod, wind influence,
etc). Scans which showed a mean difference larger
than 10 cm were excluded. The upper end of the Steintaelli
was scanned once in summer 2011 and a second time on
20 March 2012 during the ADS80 data acquisition (Fig. 2c).
Fixed installed reflector points were used to match the summer
and winter TLS data sets.