Attribution: You must attribute the work in the manner specified by the author or licensor ( but no in any way that suggests that they endorse you or your use of the work).Noncommercial ‒ you may not use this work for commercial purpose.No Derivative works ‒ You may not alter, transform, or build upon this work.Sharing ‒ Individual scientists are hereby granted permission, without fees or further requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in other subsequent works and to make unlimited photo copies of items in this journal for noncommercial use in classrooms to further education and science.

Pederson et al. (2002) present bedrock incision rates of the Colorado River at three sites in the Grand Canyon. From west to east, these sites and rates (averaged over the past 300 to 500 k.y.) are Granite Park, 72–92 m/m.y.; Toroweap, 133 ± 16 m/m.y.; and Eastern Canyon, 135 ± 17 to 144 ± 18 m/m.y. Granite Park is located ~10 fault-normal km west of the Hurricane fault, which in turn is located ~15 km west of the Toroweap fault. Toroweap is located just east of the Toroweap fault, where it crosses the Colorado River, while Eastern Canyon is ~110 km east of the Toroweap fault, at the mouth of the Little Colorado River (Pederson et al., 2002, their Figs. 1 and 2). The Hurricane and Toroweap faults are down-to-the-west normal faults, with mid to late Quaternary slip rates of 70–170 and 70–180 m/m.y., respectively (Fenton et al., 2001), which Pederson et al. (2002) propose as the cause of the east-west difference in bedrock incision rates across these faults. We agree that these faults can affect the nearby Granite Park and Toroweap incision rates; we disagree that this is possible at the Eastern Canyon locality.

More generally, the Hurricane and Toroweap faults can do little to perturb erosion rates anywhere along the Colorado River more than several tens of kilometers away from these faults, either upstream or downstream. On an earthquake-by-earthquake basis, the vertical displacements arising from crustal normal faults attenuate with a horizontal (away-from-the-fault) distance metric determined by the depth of faulting or equivalently, in most cases, the seismogenic thickness. Typically, this is ~15 km in the western United States. Elastic dislocation models of elevation data for the Borah Peak, Idaho, earthquake (18 October 1983; M = 7.3), the best documented case of crustal normal faulting in this region, clearly show this effect (Barrientos et al., 1987), as do the 1959 Hebgen Lake and 1954 Dixie Valley–Fairview Peak earthquakes. Over the course of many earthquakes spanning hundreds of thousands of years or more, these elastic dislocation solutions are additive, until such time as the strength of the locally deformed crust is exceeded. New faults are then formed, or reactivated, and the Hurricane and Toroweap faults are probably related in this way. The spacing between these faults will also be measured in terms of a seismogenic thickness or two, as is the spacing of the basins and ranges to the west.

Figure 1 shows the topography of the Hurricane and Toroweap faults just south of their intersection with the Colorado River, in both plan (top) and in elevation (bottom), and clearly reveals the very local effect these faults have on the topography. This topography bears little resemblance to either of the essentially block-motion tectonic models proposed in Pederson et al.'s (2002) Figure 3; neither does it provide any rationale for the nearly coincident incision rates at Toroweap (which should be affected by faulting) and at the Eastern Canyon locality (which should not be).

Any of these faulting models, however, provide for a steepened Colorado River gradient in the vicinity of the Hurricane and Toroweap faults, so it is a surprise to discover that just the opposite is happening: Pederson et al.'s (2002) Figure 2 reveals a slight but significant flattening of the river gradient in the vicinity of these two faults. Better seen in Leopold (1969, their Fig. 97), this reach of the river is the rising limb of a pronounced bulge in the river profile that extends for some 90 river miles (160 to 250). This local river-profile convexity requires an active and powerful process capable of not only overprinting any signature of faulting but also capable of defeating the river's otherwise monotonic trajectory to base level. Leopold (1969) shows not only the profile bulge in the western canyon, but a second in the eastern canyon, between river miles 40 and 80, possibly extending to river mile 100. The eastern canyon profile bulge is far removed from the tectonic and volcanic processes that should be affecting the western canyon profile bulge. Table 5 of Webb et al. (1999) reveals that all but one of the 14 largest debris flow fans in the Grand Canyon, as ranked by area, occur within one of these two reaches of the Colorado River, and it is likely that debris-flow activity is a principal cause of the two river-profile bulges: as a matter of chance alone, this would be a coincidence with probability of <1 in 105.

An important implication of the preceding paragraph is that all three of the Pederson et al. (2002) incisionrates sites are in places where the riverbed would seem to be aggrading, not incising. The amplitudes of these river-profile convexities are substantial, ~15 to 30 m, comparable to the elevation differences used in Pederson et al. (2002) to determine incision rates. These convexity amplitudes, however, are not accounted for by the Pederson et al. (2002) “pool-depth” corrections, which leave the thickness of debris fill between river bottom and channel bottom unknown, at times both past and present. The temporal durability of these river-profile convexities is also unknown, adding further uncertainty to incision-rate estimates in these reaches of the river.

We appreciate the commentary of G. Billingsley, R. Hereford, P.A. Pearthree, J. Pederson, F.F. Pollitz, H.D. Stenner, R.H. Webb, I.J. Winograd, and M.L. Zoback on earlier versions of this manuscript.