Avulsion threshold and planform dynamics of the Kosi River in north Bihar (India) and Nepal: A GIS framework R. Sinha a, ⁎, K. Sripriyanka a , Vikrant Jain b , Malay Mukul c a Department of Earth Sciences, Indian Institute of Technology Kanpur, Kanpur 208016, India b Division of Earth Sciences, Indian Institute of Technology Gandhinagar, Ahmedabad 382424, India c Department of Earth Sciences, Indian Institute of Technology Bombay, Mumbai 400076, India abstract article info Article history: Received 4 January 2013 Received in revised form 19 March 2014 Accepted 24 March 2014 Available online 4 April 2014 Keywords: Alluvial rivers Planform dynamics Avulsion models Threshold condition Avulsion prediction Kosi River Models for river avulsions have identified the ratio between down-valley and cross-valley slopes of channels as the triggering factors for the sudden channel shift but have remained untested in the field. The August 2008 avul- sion of the Kosi River at Kusaha, 12 km upstream of the Kosi barrage in Nepal, provided an opportunity to study a large-scale avulsion (~120 km) for its causal factors and driving mechanisms. We used the SRTM-based digital elevation model and remotely sensed data coupled with field topographic mapping with a kinematic GPS and a Total Station to characterise a ~50-km-long stretch of the Kosi River. We have computed reach-scale avulsion threshold index (ATI) integrating SRTM-derived slopes and planform dynamics on a GIS platform. We show that several reaches along the Kosi River are avulsion-prone, including the Kusaha point that is consistent with the August 2008 avulsion. We suggest that apart from cross-valley and down-valley slopes, planform dynamics such as thalweg shift, sinuosity variation, and channel multiplicity significantly influence the avulsion threshold in alluvial reaches of the rivers such as the Kosi. © 2014 Elsevier B.V. All rights reserved. 1. Introduction Avulsion has been defined as a rapid and spatially discontinuous shift of a river or a distributary channel to a new course on a lower part of a floodplain (Allen, 1965) and is considered a major fluvial haz- ard in large population centres (Jain and Sinha, 2004; Sinha, 2009). Avulsion commonly occurs when a reach of the river is at or near an ‘avulsion threshold’ (Jones and Schumm, 1999) and such shifts in river course significantly influence its morphology, the water and sediment distribution in rivers, and the architecture of fluvial deposits (Törnqvist and Bridge, 2002; Jain and Sinha, 2003; Slingerland and Smith, 2004; Aslan et al., 2005; Stouthamer and Berendsen, 2007). Al- though avulsions have a strong impact on river morphology and present a major natural hazard, surprisingly little is known about the factors that control avulsion. The avulsion process can be studied through iden- tification and quantitative characterisation of threshold condition(s) and the controlling factors that can help in predicting avulsion (Richards and Clifford, 2011; Jain et al., 2012). The trigger for an avul- sion largely depends upon the regional slope conditions and the lowest elevation available in the region. Therefore, topographic analysis is one of the most important components in avulsion studies. In particular, the relationship between the channel slopes in the cross-sectional (cross- valley slope, S cv ) and longitudinal direction (down-valley slope, S dv ) determines the critical points at which avulsion is likely to occur. A large number of studies on the assessment of avulsion threshold have therefore been based on the examination of longitudinal and cross- sectional morphology of river channels (Bryant et al., 1995; Mackey and Bridge, 1995; Slingerland and Smith, 1998; Ethridge et al., 1999; Mohrig et al., 2000; Karssenberg and Bridge, 2008). Some of these stud- ies have proposed physical and mathematical models for a better under- standing of threshold conditions and mechanisms of avulsion and various causal factors, but most of these models have yet to be tested or validated for natural river systems. Apart from topographic analysis, channel movements and temporal changes in planform characteristics influence the avulsion process. For example, an increase in sinuosity results in the decrease in the down- valley gradient of the channel with respect to cross-valley gradient, which in turn may trigger avulsion (Jones and Schumm, 1999). Changes in channel width and increase in bar area indicate dominance of aggra- dation processes (Ethridge et al., 1999). Similarly, variation in bar area or braid-channel ratio (Friend and Sinha, 1993) may reflect changes in river behaviour in terms of aggradation and degradation processes, which will have a significant bearing on avulsion process. Therefore, an integration of both morphological and topographic data should pro- vide better insights to avulsion processes. The Kosi River in eastern India and Nepal has been documented as a highly dynamic river (Geddes, 1960; Gole and Chitale, 1966; Wells and Dorr, 1987; Gohain and Parkash, 1992), and several engineering inter- ventions in terms of a barrage and embankments on both sides of the Geomorphology 216 (2014) 157–170 ⁎ Corresponding author. Tel.: +91 512 6797317; fax: +91 512 679260. E-mail address: rsinha@iitk.ac.in (R. Sinha). http://dx.doi.org/10.1016/j.geomorph.2014.03.035 0169-555X/© 2014 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Geomorphology journal homepage: www.elsevier.com/locate/geomorph