We are grateful to Zheng et al. (2011) for their interest in our work, and welcome the opportunity to explore some of the issues surrounding the origin of the Three Gorges in more detail. Zheng et al. raise three primary concerns with our interpretation that incision of the Three Gorges likely began in the Eocene. First, they argue that Paleogene rocks in the Jianghan Basin, immediately downstream of the Three Gorges, are dominated by redbed sandstone, mudstone, and evaporites, and that there is no sedimentological evidence for a Yangtze-sized trunk stream. Depositional ages for the Jianghan Basin appear to be derived from dated volcanic rocks, although no details on this are provided. In response, we note that we interpreted our cooling ages as indicating an Eocene onset of incision in the Three Gorges region—this does not mean that all of the present-day relief was constructed during the Paleogene, and indeed thermal modeling published independently by Richardson et al. (2010) and Hu et al. (2006) shows monotonic cooling at moderate rates (1–5 °C m.y.−1) for samples in the gorge region. Our data are thus consistent with slow but steady incision and base-level lowering. Our model of progressive capture of Sichuan Basin drainage area would have resulted in gradual enlargement of the proto-Yangtze basin, so it would not be appropriate to expect immediate, rapid establishment of a Yangtze-sized river downstream of the gorges. Zheng et al. supply no information on the location or density of stratigraphic sections and other pertinent data in the Jianghan Basin, so it is difficult to evaluate their claim that there is no evidence of major Paleogene fluvial systems. It is also not clear why the clastic lithologies present in the Jianghan Basin would rule out progressive capture of the Sichuan Basin, as the majority of the sediment supplied by basin erosion would have been fine-grained, reworked continental deposits from the Sichuan foreland basin (Li et al., 2003). The intercalated evaporites described by Zheng et al. suggest that the Paleocene Jianghan Basin was characterized by low rates of accommodation generation, indicating an increased tendency toward sediment bypass and making widespread preservation of proto-Yangtze river deposits unlikely in any event.
Second, Zheng et al. argue that folds in Jurassic and Cretaceous rocks in the Sichuan Basin are parallel to topography and have low relief, consistent with little post-folding erosion. While we agree with these observations, we do not see how they can be used as evidence against denudation of the Sichuan Basin. As Richardson et al. (2008) clearly documented, a range of low-temperature thermochronometers indicate that the Sichuan Basin underwent widespread denudation beginning in the Eocene. The thin-skinned deformation cited by Zheng et al. appears to have begun in late or post-Oligocene time (Jia et al., 2006; Burchfiel et al., 2008) and is still active, albeit at low rates (Richardson et al., 2008; Hubbard and Shaw, 2009). Importantly, that deformation has clearly folded the Eocene unconformity identified by Richardson et al. (2008), indicating that deformation certainly postdates the onset, and probably postdates the majority, of major erosion in the Sichuan Basin. Thus, the observation of little post-folding erosion is not relevant for determining the timing of incision in the Three Gorges.
Finally, Zheng et al. propose that Three Gorges incision most likely occurred during an Oligocene–Miocene “regional tectonic adjustment,” characterized in this case by rock uplift and erosion of the Jianghan Basin. While this is certainly a plausible explanation, it appears to be only an opinion of the authors, who neither provide any evidence for such a causal link, nor propose a physical mechanism by which rock uplift in the (downstream) basin might drive (upstream) incision in the gorges and Sichuan Basin. It is also not clear why cessation of normal faulting along the basin margin would necessarily be associated with enhanced erosion in the fault footwalls to the west. Any appeal to such a model must also (separately) explain the increase in cooling rates in the Three Gorges region at 45–40 Ma, the driving mechanism behind coeval denudation in the Sichuan Basin, and the fate of sediments eroded from the basin. Our proposed model of headward incision from the Three Gorges region, progressive capture of the Middle Yangtze River, and base-level fall in the Sichuan Basin is certainly not a unique scenario, but does explain all available thermochronological data from both the Sichuan Basin and Three Gorges areas. The true test of the origin and development of the Yangtze River system, however, may only come from provenance and stratigraphic analysis of the offshore sedimentary record.