THALWEG PROFILES AND KNICKPOINTS AS USEFUL DISCRIMINATORS OF PREHISTORIC LANDSLIDE DAMS IN NORTHERN PAKISTAN

The incised gorge along the upper Indus River in northern Pakistan appears to have been pinched repeatedly by large landslides that have temporarily dammed the river. Clusters of landslides are most often observed in areas with fractured strata, tectonic uplift/offset, high rates of bedrock incision, and intense precipitation. The geomorphology along the main stem of Indus River was analyzed by using ASTER DEM data and from visual inspection of longitudinal profiles generated by ArcGIS and Matlab. This study sought to explore some of the indirect methods and criteria that might be employed to verify the likely locations of prehistoric landslide dams. These included analyses of geomorphological indicators, such as evaluations of channel cross-sections, thalweg profiles, and knickpoints, which can be diagnostic of prehistoric landslide dams.

After breaching and debris removal the landslide dam sites are typically occupied by rapids that include large bedrock blocks and boulders which serve to constrict flow and maintain the knickpoints, depending upon bedrock lithology and other climatic factors. This study found that morphological traits of landslides are contingent with the type of slope failure and exhibit distinct physical features that are anomalous with surrounding terrain that has not experienced landslippage. Numerous knickpoints were noted in the thalweg profiles of the river channel. Most of these appear to be influenced by changes in lithology, tectonics, fault offset, and mass wasting. The remnants of known landslide debris dams also exhibit significant knickpoints, most frequently near the confluence of major tributaries.

This study suggests that knickpoints can be used as a useful discriminator in identifying old landslide dam sites, even when the debris of such features has all but been swept away. It also provides a check on the validity of the landslide inventory mapping, which is important for regions that are bereft of both physical and historical data.