Glacier Lake Outburst Floods (GLOFs)—Mapping the Hazard of a Threat to High Asia and Beyond - Impact of Global Changes on Mountains: Responses and Adaptation - page 331

Table 14.1. Overview of the utilized NDWI thresholds for different sensors (Bolch et al. 2011).

Sensor Landsat MSS ASTER Landsat ETM+

Threshold 0.45-0.9 0.3-0.7 0.3-0.9

drawback of the NDWI is that cast shadows are also included in the mask
and must be excluded in a post-processing step. A shadow mask may be
used for this purpose (cf. Huggel et al. 2002) or the classifi cation needs to
be done separately for shadowed and non-shadowed areas (Chen et al.
2012). In a last step, visual checks and manual improvement are necessary
especially for lakes in shadow and turbid lakes.

Hazard assessment

It is obvious and has already been mentioned in the literature repeatedly
(cf. i.a. Haeberli et al. 2010, ICIMOD 2011) that the hazardous processes
can interact with each other and with less climate-sensitive parts of the
involved geomorphic system. Disasters have mostly resulted from cascading
processes rather than single phenomena (Haeberli et al. 2010). Therefore
it is necessary to apply an integrated system approach to avoid missing
important processes (Huggel et al. 2004, Bolch et al. 2011a, 2011b, Mergili
and Schneider 2011). Uncertainties at all stages of the process cascade are
considerable. Actually, uncertainties also cascade, i.e., they increase along
the process chain. Hence, systematic and detailed observations are essential
for objective and sound assessments (Kääb et al. 2005).
The design of appropriate observation systems is thereby greatly
facilitated by modelling scenarios of possible developments including the
following ones (after Haeberli et al. 2010): Landslides in bedrock (1a) and
moraines (1b) due to debuttressing effects from glacier retreat since the Little
Ice Age, ice avalanches from polythermal steep glaciers due to increasing
fi rn and ice temperatures (2a), and from dissected temperate steep glacier
parts (2b), rock fall and avalanches in relation with permafrost degradation
(3a), regular rock fall from areas of warm bedrock permafrost (3b) fuelling
debris fl ow (4) initiation zones, lake formation (5) due to glacier retreat
or down-wasting, posing outburst fl oods hazards that are aggravated by
potential impacts from multiple mass movement processes.
Several factors need to be taken into account when assessing the
hazard potential of a glacial lake outburst. Bolch et al. (2011a) adapted
a methodology developed by Huggel et al. (2002) and Bolch et al. (2008)
introducing a higher number of variables for the hazard assessment.
The latter can be summarized by four major parameter groups: (a) lake
characteristics, (b) characteristics of the lake surroundings, (c) characteristics