Gravel Grumpiness

So GG is a co-leader of the 2022 Thompson Field Forum in the Sierra Nevada next June and this has him refocused on evidence related to Eocene(?) gravels and topography. So let’s review how these have been interpreted, some issues we’d like to see solved, and some speculation from a somewhat different angle. We’re going deep into the weeds, so hang on…

One reference point for these river gravels has been their bed, the basement thalweg. Why? Well, for one thing, it is quite recognizable. For another, as gold was concentrated in the lowest gravels, a lot of attention was paid to finding this interface. And so we have a fairly large number of locations of the old thalweg and the elevations of that as well as the local gradient. And so it turns out that the southwest-trending segments are steeper today than the northwest-trending segments, e.g., this plot from Jones et al., 2004:

Figure 1. Gradient versus azimuth of segments of paleo channel derived from the distance and elevation difference between adjacent segments of basement thalweg.

Or, if it makes more sense, here is a profile along the river channel of the paleo-Yuba:

Figure 2. Modern day elevations of the base of an inferred paleo-Yuba River based on Hudson (1955). The steeper segments from about 0-30 miles and 60-120 miles are trending mainly to the southwest (bottom panel); the flatter part in the middle trends largely to the north-northwest.

So the classic interpretation of this is that the Sierra were tilted after deposition of these gravels to the west-southwest. That would make the north-northwest trending segments flatten while steepening the southwest trending segments. But objections have been raised. One is that the connections that previous workers made (Lindgren, 1911, Hudson, 1955, and Yeend, 1974) are incorrect, in which case there never was a river with a path like that in the second figure above (e.g., Gabet, 2014). Or the different basement thalwegs were only preserved from different times, meaning that connections might be incorrect and that gradients would certainly be incorrect (Cassel and Graham, 2011). Or the different gradients might simply be a function of anisotropy in the underlying rock (Gabet, 2014). So perhaps this evidence isn’t really compelling. (A partial response to these complaints is in Wakabayashi, 2013).

[One objection in the literature–well, in supplemental material–was made by Mulch et al. (2006): “The argument for recent surface uplift of the northern Sierra Nevada uses geometric reconstruction of river slopes of the Eocene Yuba River to assess the amount of post-Eocene tilting of the Sierra Nevada (S21).  The Eocene Yuba River forms a Z-pattern, with the central segment trending parallel to the range (Fig. 1) (S1, S2).  The tilting reconstructions (S21) are based on the assumption that the Eocene river segments perpendicular to the range had identical slopes to this range parallel section.  However, the range parallel segment of the Eocene Yuba River where the most voluminous river deposits are found is aggradational, in contrast to the downcutting east-west trending segments of the river.  This indicates that the range parallel segment of the Eocene Yuba River initially had a shallower slope than the east-west segments and only has limited use in reconstructing the Eocene river gradient. Variations in slope between aggradational and downcutting river segments are observed in many modern (e.g. Yarlung Tsangpo and Brahmaputra River) and ancient rivers (S22).”  [S22 is G.E. Hilley. M. Strecker, Geol. Soc. Am. Bull. 117, 887 (2005)]. First up, the cited example is a river cutting across active faults, which is not the issue in the Sierra; it is hardly a shock when there is incision of a hanging wall and deposition on the footwall of a thrust. And there are gravels on the southwest-trending segments in the Sierra, so while they might well have been incising longer that the NNW-trending segments, they did reverse to aggrading before the Oligocene, it seems. This particular objection doesn’t seem to hold water.]

Now there are other approaches to this broader problem, but here GG would like to find a way to resolve this within these sediments themselves. And one approach that hasn’t been attempted yet would seem the clearest. This would be to measure the elevations of some marker bed well up in the section rather than the basement thalweg. A marker bed–probably one of the recognized Oligocene tuffs that flowed down channels–would represent a moment in geologic time and thus would be comparing points along a river. And such a river would be flowing not on bedrock but on gravels from earlier time, so any bedrock effects would not affect the continued deposition along the river. And if the tuff was limited to a single drainage on the west side of the Sierra, questions about connecting pieces along a single river would be addressed. In sum, this seems like a solid alternative to the previous approach. While this is a promising line of attack, it isn’t trivially done with available information. (There is such an attempt farther south, in the San Joaquin drainage, where the San Joaquin table mountains are the remnants of a Miocene flow from near the Sierra crest; this work strongly suggests ~2 km of post-Miocene uplift of the range crest at this latitude).

For the moment then, let’s consider two other ways of looking at this. First up, sedimentation. The reason there are all those points in Figure 2 is that there are gravels at each of these sites. While the initiation of deposition could well be diachronous, it seems unlikely that the upper gravels were not being deposited at the same time along this whole system (though this perhaps could be checked if good age control emerges). Deposition is some combination of sediment being transported and loss of ability by the stream to continue to carry that full load. Loss of ability to continue to transport coarse material like that in the gravels is usually reflecting a decrease in stream velocity, which in turn is a combination of a decrease in slope or a decrease in water depth or increase in width. This is all complicated a bit by seasonality and extreme events, but we can get a crude idea from those three variables. To first order we might expect that the slope of deposits would increase going upstream, so take as a lower bound that the increase going upstream is constant. This yields rather quickly a pretty risible result:

Figure 3. Extrapolating a slope of sedimentation from the lower paleo-Yuba upstream. Yellow is a minimal accumulation along the southwestward segment of the paleo-Yuba. Green is then extrapolating at the same grade back into the flat part of the paleo-Yuba. This predicts a couple thousand feet of sediment; the maximum observered is under 600′ and is closer to the 30 mile mark than the 60 mile mark.

The area in yellow is plausible for the western parts of the old channel. But the extrapolation in green is crazy: there are not thousands of feet of sediment at these sites (the thickest sediments are near mile 40 of about 500′ or so). So what is going on? Obviously an assumption would be violated. One, certainly, is that the reconstruction is wrong; we’ll come back to that. Maybe a lot of sediment originated near mile 30 where the tributaries from Malakoff Diggings come in–but an awful lot of sediment moved into the drainage before that point. GG is not a sedimentologist/geomorphologist, so maybe there is a way to do this, but it sure seems odd. A key point is that sediment is accumulating on the steep segments as well as the shallow ones by the time the first volcanic debris came downstream. Here is the same plot with the published estimates of the pre-volcanic thickness of gravels at those spots with estimates.

Figure 4. Hudson’s proposed reconstruction of the Eocene Yuba with gravel thicknesses. Gravel thickness from multiple sources (Lindgren, 1911; Hudson, 1955; Yeend, 1974; Clark, 1970). In general, Lindgren’s estimates were not formally measured sections. Non-Lindgren estimates are at distances of 3.2, 29, 53, 67 and 109 miles.

The other approach uses the sedimentology of the coarser of the upper (bench gravel) deposits. These rocks are interpreted to be from braided streams. The grade of braided rivers are largely thought to be reflected in the ratio of clast size to water depth. This was used by Cassel and Graham, (2011) to infer the grade of channels trending dominantly southwest as having been between 0.4% and 5.5%, which kind of brackets the modern slope of strath terraces of 2-6.3% but also includes at the low end values comparable to restored slopes of 0.2-0.6% inferred from untilting the profiles shown above. Water depth seems unlikely to change much as braided streams tend to just add more channels in response to increased volume in the river overall. So to have a much gentler slope on the north-trending leg, the clast size should drop dramatically. So the prediction is that the SW trending legs should be quite coarse and the NW trending legs be very fine. To first order, this is not what is seen; pretty much at most sites there is a deeper coarse gravel and a higher finer sand and gravel.

Assuming this style of analysis isn’t stupid (and maybe it is, comments are welcome), it points to the importance of two assumptions. One is that this reconstruction of the paleo-Yuba is correct. There is definitely room for improvement on this point; the weakest link in the reconstructions is roughly between the modern South Yuba and near You Bet (roughly mile 35 to 45 in the plots above); in this area exposures are infrequent and considerable amounts of area are covered by the Mio-Pliocene volcanogenic materials. There are some other spots where this hypothesis could be challenged. The second is that we can identify coeval sedimentary rocks. At present, there is very little control for the age of deposits as a whole, let alone within the deposits. This seems to be most true of the deep blue gravels that sat above the basement and less likely for the highest gravels that start to contain some Eocene or even Oligocene volcanic material.

So if there is anything in this approach, it depends on what the paleo-Yuba really looked like. Now other drainages have patterns like Figure 1, but they are not nearly so dramatic (in part, they haven’t been examined nearly as closely as the Yuba). With the focus of the Field Forum on the Yuba area, it will be interesting to see how the group will end up viewing this drainage.

Tags: ,

Trackbacks / Pingbacks

  1. Some Au Gravel Thoughts | The Grumpy Geophysicist - July 4, 2022

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: