A regional study of landslide hazards and related features in the Upper Indus River Basin, northern Pakistan

"A spatial landslide inventory is a fundamental step preparatory to developing an understanding of regional landslide hazards. The techniques honed for this regional landslide mapping study began with compiling a stitched shaded topographic map of the Upper Indus River watershed; using remotely-sensed topographic maps from Russia combined with ASTER DEM 30m resolution data. Topographic recognition keys were then employed to aid in the visual identification and delineation of past landslides, rockslide avalanches, and landslide dam sites. The first order reconnaissance level inventory excerpted 2,254 landslides (mostly deep seated translational and complex slides) along the main stem of Indus River, land area of ~17000km², extending more than 700 km upstream of Tarbela Dam and Reservoir. Knickpoints in the channel's thalweg profile may prove useful in identifying old landslide debris dam sites because these features often create flow obstructions. To test this hypothesis, 251 knickpoints were identified and marked along the channel profile of the Indus River, then compared with the mapped and known landslide dam sites. The comparison revealed that the channel's longitudinal profiles were a useful discriminator of past channel blockages, most of which appear to have been caused by pre-historic landslides and rockslide avalanches along the Indus River and its principal tributaries. These analyses also helped to validate the interpretations made in our landslide inventory. Another new concept, introduced to regional landslide hazard mapping was the consideration of "wind driven" rainfall. This process modified the regional rainfall data and coupled it with directional Monsoon (slope aspect) in order to estimate a more realistic idea of the actual rainfall distribution over slopes of varying inclination and shape possessing different aspect directions. Two approaches for landslide susceptibility analysis were then employed so that their results could be compared: heuristic (qualitative index based) and fuzzy logic. The landslide susceptibility maps produced by incorporating the incident rainfall map combined with slope aspect appeared more reasonable and detailed than any others yet produced of such a large land area (~75,000 km²)"--Abstract, page iii.

Name thesis:degree_name

Ph. D. in Geological Engineering

Grantor

Missouri University of Science and Technology

Year dc:date.available

2016

Author dc:creator

• Ahmed, Muhammad Farooq

• Susceptibility mapping
• Geological Engineering

Repository record dc:identifier

https://scholarsmine.mst.edu/doctoral_dissertations/2109

OAI identifier oai:identifier

oai:scholarsmine.mst.edu:doctoral_dissertations-3111