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.

Wolkowinsky and Granger (2004) analyze gravels deposited by the San Juan River near Mexican Hat and Bluff, Utah, for their abundances of 26Al and 10Be and the ratio 26Al/10Be as a function of depth. Using primarily the Bluff data because of the greater depth of sampled section (11.7 m), and assuming that this section is a single depositional unit, Wolkowinsky and Granger find that these gravels were deposited at 1.36 ± 0.20 Ma and have eroded at a rate of 14 ± 4 m/m.y. From a sample apart from the depth profile, Wolkowinsky and Granger also determine an “effective surface exposure age” for the Bluff deposit to be 660 ± 84 ka, assuming no surface lowering. Finally, Wolkowinsky and Granger determine an average incision rate for the San Juan River since 1.36 Ma to be 110 ± 14 m/m.y.

The Wolkowinsky and Granger “burial age” calculations employ a χ2-minimization procedure to determine the age of the deposit, its surface erosion rate, its bulk density, additional surface cover on the deposit, and the nuclide inheritance of each sample; a manifold of nonunique solutions should exist in this complex parameter space. Our principal disagreement with Wolkowinsky and Granger, however, lies in their assumption that the river gravels at Bluff are a single depositional unit.

Evidence against the one-stage deposition model lies in Figure 3 of Wolkowinsky and Granger, redrafted here as Figure 1A to exclude, for clarity, the nondefinitive Mexican Hat data and the Bluff 1.5-m outlier. In contrast to the Wolkowinsky and Granger model curve calculated with uniform inheritance, the 26Al/10Be data for both Bluff and Mexican Hat are indistinguishable from a constant value down to a depth of 4.3 m. The Wolkowinsky and Granger one-stage model crosses this “vertical line” with a very different slope, suggesting to us that the material above ~4.3 m is significantly younger than that below it. Moreover, a significant unconformity exists in the Bluff section at ~4 m depth: coarse, bedded river gravels lie above it with a massive sand unit below (D.E. Granger, 2004, personal commun.). Finally, we find the results of Wolkowinsky and Granger to be inconsistent with their assumption of one-stage deposition. The great variation in the model inheritances found by Wolkowinsky and Granger for the six Bluff data (a factor of ~20) suggests multiple episodes of deposition of materials experiencing very different exposure histories. Neither does it seem likely that at least 31 m of river gravels and sands, about the height of a ten-story building, were deposited by the San Juan River as a single unit.

Our two-stage depositional model consists of a basal unit deposited at 1.5 Ma subject to an erosion rate of 16 m/m.y. for the next 0.84 m.y. At 0.66 Ma, the exposure age of the lag deposit dismissed by Wolkowinsky and Granger, the upper unit is deposited and subjected to an erosion rate of 18 m/m.y., such that now it is 4.5-m thick. With respect to other parameters used in Wolkowinsky and Granger, we use a density of 1.8 g/cm3 (1.5 g/cm3 in Wolkowinsky and Granger), no added surface material (6.0 cm in Wolkowinsky and Granger) and inherited sample abundances of 10Be ranging from 80 to 1200 kiloatom/gm SiO2 (55–1000 kiloatom/gm SiO2 in Wolkowinsky and Granger). These inheritance values are given for the time of deposition, not for today as presented in Wolkowinsky and Granger's Table 2; the Table 2 data have also been corrected for transcription errors (D.E. Granger, 2004, personal commun.). We follow Wolkowinsky and Granger in using the equations of Granger and Muzikar (2001) to calculate the 10Be and 26Al production from neutron spallation and from slow and fast muon reactions.

We compare predicted 26Al/10Be ratios and 10Be abundances for both of these models, together with the Wolkowinsky and Granger data in Figure 1. Figure 1A shows model predictions for both uniform inheritance (continuous curves) and for point-by-point determinations of sample nuclide inheritance. For uniform inheritance, the two-stage depositional model fits the data somewhat better than the one-stage model of Wolkowinsky and Granger. For point-by point inheritance, both models fit the data reasonably well. The 10Be abundances that correspond to the point-by-point inheritances are shown for both deposition models in Figure 1B.

We do not claim that the two-stage deposition model presented here correctly portrays the history of deposition and incision by the San Juan River at Bluff over the past ~1.5 Ma; indeed, the model inheritance data, ours as well as Wolkowinsky and Granger's, suggest multiple stages of deposition, not just two. Neither do we claim that the two-stage model presented here is the best two-stage model that can be constructed; our exploration of its entire parameter space is preliminary. We do claim, however, that this two-stage deposition model is a viable alternative to the one-stage model assumed by Wolkowinsky and Granger. That different models with different implications can fit the same data set is hardly news in the earth sciences; even so, we are surprised how poorly constrained burial-age calculations can be if the number of depositional units is itself a variable.

Specifically, the two-stage deposition model allows for the San Juan River to be depositing river gravels at the Bluff site at least as recently as 660 ka, ~140 m above its present elevation. The minimum average incision rate of the San Juan since that time exceeds 200 m/m.y. Finally, while Wolkowinsky and Granger correctly state that their results are incompatible with the results of Hanks et al. (2001) and Garvin (2004), we are convinced that it is not the data of Wolkowinsky and Granger that are incompatible with Hanks et al. (2001) and Garvin (2004), only the assumption of a single-stage deposition model.

We appreciate several communications with D.E. Granger as we developed the ideas and calculations of this study and the constructive reviews of it that we received from A. Matmon and R.H. Webb. This work was supported in part by U.S. Department of Energy/Lawrence Livermore Laboratory Contract No. W-7405-Eng-48.