| dc.description.abstract | Avalanches are among the most frightening hazards that people face in mountainous regions of the
Alps, killing around 100 Alpine residents and tourists each year. Austria welcomes the most tourists in
wintry months, making itself the most avalanche-sensible country in the region and Tirol, - Tyrol in
English – is the leading province when it comes to winter tourism. Therefore, the Lawinenwarndienst
Tirol zones the danger for avalanches for warning purposes. Its warning system relies mostly on
temporary weather and snow conditions and is thus updated daily. However, large areas tend to be
given the same danger level while they vary heavily in geographic terrain. Therefore, this research sets
out to zone avalanche danger leaving the temporary conditions empty, so that an all-year-valid template
with more geographical detail can be developed. Tyrol is obviously a place that could benefit from this
research and the model is thus designed for a part of Tyrol, the area of Landeck-Imst. The rest of Tyrol is
used to validate the model; to what extent is it applicable to other, similar areas?
The key to success for this model to understand the role of avalanche-causing factors. These are derived
from theoretical sources and tested on dangerous situations and weight. Elevation and slope steepness
turned out to be the most influential ones, followed by the presence of skiers and other winter’s sports
tourists. Less important are the slope shape (convex, concave or planar), a slope’s direction to the sun,
the type of land cover and the characteristics of a ski resort if nearby. The most dangerous situations per
factor are the highest located areas, slopes with a steepness around 38 degrees, areas within or nearby
ski resorts, north-facing slopes and terrain of bare rocks and glaciers.
The model is constructed by discovering under which circumstances the factors are getting dangerous.
All situations are reclassified to new values, depending on how dangerous they are. This is done by
assessing historical avalanche data; which terrain produces the most avalanches? Together with each
factor’s importance, these values are making up a place’s danger. When comparing with the real
avalanche locations, the model turns out to be quite successful. Almost every avalanche from the past
five years took place in an area that is marked by the model as dangerous above average. Fortunately,
the model’s success does not stop at Landeck-Imst’s boundaries. Applying it to the rest of Tyrol results
in meaningful outcomes, although success rates are obviously higher in the original development area.
To conclude, it is reasonable to believe that this model could be a useful contribution to the forecasting
of the Lawinenwarndienst. | |
