بخشی از مقاله انگلیسی:
River evolution is a major driver of morphological landscape development. Rivers follow topographic gradients; they adjust and move their bed by removing or depositing material in bedrock and unconsolidated rock channels and export sediment from their catchments. River beds in active orogens are dominated by bedrock channels that are considered key agents in forming mountain geomorphology (Hancock et al., 1998; Whipple et al., 2000; Whipple, 2004; Jansen, 2006). Bedrock channel width, depth and slope, bed roughness, bedrock exposure and sediment size distribution interact under dynamically varying discharge and sediment flux by means of several erosional processes (Wohl, 1998; Johnson and Whipple, 2007; Yanites and Tucker, 2010). Identifying and understanding the rates of the physical processes driving erosion, and how they are affected by material and energy input at highly resolved spatial and temporal scales is fundamental for channel morphodynamics and landscape evolution modelling. In particular, such understanding is essential for transferring these processes to larger spatial scales (e.g. Sklar and Dietrich, 2006). Erosion of bedrock surfaces can be classified into three main process mechanisms: corrosion (chemical weathering), corrasion (abrasion and plucking/macro-abrasion by impacting sediment particles), and cavitation (implosion damage induced by the collapse of gas bubbles generated by turbulence in the stream) (Hancock et al., 1998; Wohl, 1998; Whipple, 2004; Chatanantavet and Parker, 2009; Whipple et al., 2013). Bedrock morphology is generally linked to these processes: It is thought that plucking is dominant when knickpoints and inner channels are formed, while abrasion is responsible for sculpting the rock and for creating undulations (e.g. Tinkler and Wohl, 1998; Lamb and Fonstad, 2010; Wilson et al., 2013). Except where rocks are relatively soluble, corrosion may be of secondary importance in many bedrock channels (Turowski, 2012). Cavitation is thought to be important only in pothole formation, as it requires high flow velocities that are rare in natural streams (Barnes, 1956; Hancock et al., 1998; Whipple et al., 2000). Besides climate, channel incision rates are strongly dependent on the lithology of the channel bed and of the transported material (e.g. Whipple et al., 2000; Jansen, 2006; Lamb and Fonstad, 2010). Actual detachment of solid rock, or erosional efficiency, seems to depend on the sediment’s tools and cover effects (Gilbert, 1877; Sklar and Dietrich, 1998, 2004; Turowski et al., 2007; Turowski and Rickenmann, 2009). The shielding of bedrock by a sediment layer (the cover effect) limits erosion under high rates of sediment supply (e.g. Turowski et al., 2008; Johnson et al., 2009), whereas the availability of impacting grains to abrade bedrock (the tools effect; e.g. Foley, 1980) fosters erosion under lower rates of sediment supply (e.g. Finnegan et al., 2007; Cook et al., 2013). Theoretical and laboratory investigations of the process physics have led to the saltation–abrasion model (Sklar and Dietrich, 2004). In this model, erosion rates depend on bed shear stress, on the erodibility of the bedrock (quantified by its tensile strength, its elastic modulus and a dimensionless rock resistance parameter), on the size and impact energy of saltating bedload grains, and on bedload supply relative to transport capacity. Erosion rates thus are explicitly dependent on local sediment transport rates. Peak bedrock erosion rates are thought to occur at moderate bedload supply relative to transport capacity. This is due to the competition of the tools and cover effects (Sklar and Dietrich, 2001; Nelson and Seminara, 2011), and also depends on the maximum saltating grain size (Sklar and Dietrich, 2001, 2004) and on saltation velocity respectively (Chatanantavet et al., 2013), both conditioned by local channel morphology. Field measurement of bedload transport is a difficult and challenging task due to strong spatial and temporal fluctuations as well as due to the destructive effect of bedload on equipment. Bedload transport rates can be measured with direct methods like hand-held box samplers (e.g. Helley and Smith, 1971), which are mainly used at lower transport rates and deliver data at-a-point in space and time. Temporarily and spatially integrated rates and sediment budgets can be derived from topographic change detection analysis based on repeated surveying (e.g. Lane et al., 1995). In addition, there are surrogate techniques for continuous bedload monitoring that can cover the full range of discharge conditions, but have to be calibrated by direct methods (e.g. Gray et al., 2010; Rickenmann et al., 2014). All available methods have their advantages, disadvantages and restrictions in application and lack of accuracy in space and time, due to extrapolation, interpolation and calibration problems. Geophone-based methods are the most developed surrogate techniques for coarse bedload monitoring (Gray et al., 2010); however, as with all surrogate methods, their field-calibration is challenging. Similar to bedload transport, fluvial bedrock erosion is difficult to measure, since generally it is a slow process, particularly in resistant substrates (Wohl, 1998). At-a-point, erosion over time has been studied in nature by monitoring (i) borehole depths (Hancock et al., 1998), (ii) heights of erosion pins like nails (Stock et al., 2005) or expansion bolts (Johnson et al., 2010), (iii) by repeated individual point measurements based on fixed benchmarks (Hartshorn et al., 2002; Stephenson, 2013) and (iv) by traditional (Chatanantavet and Parker, 2011) or global positioning system (GPS) survey (Johnson et al., 2009). Field monitoring of whole bedrock surfaces at promising sites have been conducted over several spatial scales using different techniques like aerial photogrammetry and terrestrial laser scanning (TLS; e.g. Cook et al., 2013) for kilometre to centimetre and even to sub-millimetre resolution and accuracy (Rieke-Zapp and Nichols, 2011; Rieke-Zapp et al., 2012; Wilson et al., 2013). However, such measurements have not been paired with bedload transport observations. Due to the difficulties of obtaining high-quality data, field evaluation of erosion models has so far relied on simplifying assumptions, using long-term erosion rates and sediment yields. In an early paper Foley (1980) inferred the long-term abrasion rate of a glacially diverted stream (Dearborn River, Montana) by dating moraines and estimating corresponding discharge and sediment transport rates by calculations from river geometry and sediment deposits. At the South Fork Eel River, California, basin-averaged paleo-erosion rates were calculated using beryllium-10 (10Be)-concentrations of strath terrace sediments, whose burial age was determined independently by optical-stimulated-luminescence (OSL) dating (Fuller et al., 2009). Strath formation due to extensive sediment cover and lateral erosion could be assigned to elevated sediment supply combined with increasing discharge. Tomkin et al. (2003) inferred average erosion rates from strath terrace incision of the Clearwater River, Washington State, assuming long-term steady state conditions and found that none of the six tested erosion models could account for their data. In a comparable attempt using five erosion equations, van der Beek and Bishop (2003) modelled stream evolution from mapped palaeo-channel profiles in the Upper Lachlan catchment, southeast Australia, and concluded that with individual suitable model parameters sets each of the tested models was able to reproduce the current stream profile. Currently available field data do not allow process and model analysis at high temporal resolution, e.g. on the basis of individual events. In addition, upscaling process-based model formulations to the cross-section, reach or catchment scale is problematic (Turowski and Rickenmann, 2009; Lague, 2010; Turowski, 2012). There is no dataset available of simultaneous measurements of hydraulics, sediment transport, bed forces and bedrock erosion rates for a natural stream (Wohl, 1998) to study their interactions. To fill this gap, we have constructed the ‘erosion scales’ measuring setup to provide accurate, spatially and temporally high-resolved field data. The aims of this article are (i) to describe the new erosion scale instrumentation, (ii) to evaluate the quality of the data recorded (discharge, bedload transport, erosion), and (iii) to discuss first measurement results and the potential of the equipment for quantitative process studies and evaluation of fluvial bedrock erosion models (cf. Hancock et al., 1998).