Modes and rates of horizontal deformation from rotated river basins: Application to the Dead Sea fault system in Lebanon

Partitioning of horizontal deformation between localized and distributed modes in regions of oblique tectonic convergence is, in many cases, hard to quantify. Here we use the geometry of river basins and numerical modeling to evaluate modes and rates of horizontal deformation associated with the Arabia-Sinai relative plate motion in Lebanon. We focus on river basins that drain Mount Lebanon to the west and are bounded by the Yammouneh fault, a segment of the Dead Sea fault system that transfers left-lateral deformation across the Lebanese restraining bend. We quantify a systematic counterclockwise rotation of these basins and evaluate drainage area disequilibrium using the c metric. The analysis indicates a systematic spatial pattern whereby tributaries of the rotated basins appear to experience drainage area loss or gain with respect to channel length. A kinematic model reveals that since the late Miocene, 23%–31% of the relative plate motion parallel to the plate boundary has been distributed along a wide band of deformation to the west of the Yammouneh fault. Taken together with previous, shorter-term estimates, the model indicates little variation of slip rate along the Yammouneh fault since the late Miocene. Kinematic model results are compatible with late Miocene paleomagnetic rotations in western Mount Lebanon. A numerical landscape evolution experiment demonstrates the emergence of a similar pattern of drainage area disequilibrium in response to progressive distributed shear deformation of river basins with relatively minor drainage network reorganization. INTRODUCTION Theory (e.g., Sanderson and Marchini, 1984; Teyssier et al., 1995) and observations (e.g., Vallage et al., 2014) of regions of oblique tectonic convergence have demonstrated that deformation in such settings is commonly partitioned into localized (faults) and distributed zones of pure or combined convergence and shear in variable ratios. In such situations, knowledge of long-term relative plate velocity is insufficient for estimating geological slip rates on the main faults, which hinders our understanding of the related regional tectonics. Rivers are commonly used as offset markers across strike-slip faults (e.g., Klinger et al., 2000), and entire river basins have been suggested to record large-scale distributed horizontal deformation (Hallet and Molnar, 2001; Ramsey et al., 2007; Castelltort et al., 2012). Here we use morphometric analysis of river basins and modeling to constrain rates and modes of localized and distributed deformation along the Lebanese segment of the Dead Sea fault system (DSFS). THE DEAD SEA FAULT SYSTEM IN LEBANON The DSFS accommodates the left-lateral motion between Sinai and Arabia plates (Freund et al., 1970). Along Lebanon, the DSFS bends 30°E, forming a 180-km-long restraining bend (Fig. 1). The left-lateral Yammouneh fault (YF) and the Serghaya fault are the two main structures that run along the sides of Bekaa Valley (Fig. 1). By contrast, the Beirut-Tripoli thrust system offshore of Lebanon accommodates mainly dip-slip motion (Elias et al., 2007; Carton et al., 2009). The YF, the sole continuous structure across the Lebanese bend, has been suggested to accommodate the fastest slip in the area (Daëron et al., 2004). However, there are gaps and uncertainties in our understanding of the kinematic history of the Arabia-Sinai plate boundary along the Lebanese restraining bend. For example, how much of the relative plate velocity is taken up by the YF is still debated. This is partly reflected in the relatively wide range of slip rates, 3.8–6.4 mm/yr, that have been proposed for the YF (Daëron et al., 2004; Gomez et al., 2006, 2007). Furthermore, these rates are considered relevant for Holocene/Quaternary times, but the onset of relative motion has been suggested to date back to the late Miocene (Quennell, 1984; Ben-Avraham et al., 2008). Moreover, paleomagnetic measurements of ~30° counterclockwise rotation along an ~30kmwide region west of the YF (Van Dongen et al., 1967; Gregor et al., 1974; Ron, 1987; Henry et al., 2010) suggest distributed deformation off the major faults (McKenzie and Jackson, 1983). However, the timing of the rotation and its relation to the DSFS activity remain poorly constrained (Gomez et al., 2007).