Some Au Gravel Thoughts
So having survived a few too many vehicle adventures on the recent field forum, GG is trying to consolidate some thoughts. Many of which are still in flux, so this is a snapshot of thinking and not necessarily a final product…. Some of this is kind of trivial but some has some major implications. Others are rehashes of older thoughts that haven’t yet been banished.
First up, the age of the gravels and the meaning of Eocene zircon grains. Reexamining the detrital zircons from the Malakoff-Alpha system, it seems this was entirely an intra-Sierra drainage and the Eocene zircons are airfall. The still-unpublished bigger samples of Tye and Niemi at Malakoff seem to make this pretty clear as we are now seeing 8 and 10 Eocene zircons in two samples instead of the 1 or 2 from older work, and those two samples seem to have very distinct and separate peaks, which is probably hard to do with fluvial grains but perfectly understandable from airfall. Toss in the very robust results from Haskell Peak of a lot of Cretaceous grains no matter where you sample in the section and the previously drawn connection of Haskell Peak and its eastern tributaries to the Alpha-Malakoff drainage is clearly wrong.
So then we face the issue with the K-absent samples from Malakoff and the K-present samples from adjacent North Columbia. There are three options GG can see: they are different ages (North Columbia being older), Malakoff’s river did not flow into North Columbia’s, and Malakoff’s zircons were overwhelmed by whatever else was flowing into North Columbia. It seems hard to make these significantly different in age, in part because there is some overlap in elevation and in part because of one miserable zircon in the Cecil et al. measurement at North Columbia. While it is quite plausible that the lower parts of North Columbia are older than the accessible gravels at Malakoff, it does seem that the upper North Columbia and lower Malakoff gravels are nearly coeval.
Because there are so many gravels hanging out in this area, you can steer Malakoff’s channel around some, probably most plausible to the southwest towards Nevada City. But you have to cross the Blue Tent-Enterprise-Spring Creek exposures, and even if you get to the Manzanita channel, you still see that large change in zircons. So GG’s guess is that the most likely scenario is that Malakoff was a relatively minor stream while the main river that fed into North Columbia dominated the sediment volume. The absence of early Paleozoic Bowman Lake batholith zircons at North Columbia from existing samples would speak to the kind of dilution of the Malakoff material; presumably larger samples of North Columbia would yield a few of these zircons. Such samples might also yield better age control. Finding a downstream deposit with the early Pz grains could be quite helpful.
This brings us to the Oligocene channels. Most of the geologic maps out there hide the Oligocene rhyolitic material, but these rocks filled the upper extents of the old paleovalleys and can be correlated through these drainages. When you allow these rocks to stand out (in yellow on map below), questions emerge. Why is there so little of this material near both North Columbia and Malakoff, and yet so much not far to the south at Scotts Flat and Nevada City? Is the exposure mapped northeast of North Columbia really these rhyolites (this is Tvs of Yeend (1974)), and if so, which ashflows are present, and why only on that side of the ridge?
It seems likely that if North Columbia was the destination for gravels from the south in the Eocene, that this relationship ended by the Oligocene, with that upper channel captured by a drainage heading southwest on the north side of Nevada City. There seems to be a clear major channel running from near Donner Pass through French Meadow Reservoir on down towards Foresthill. Did it really turn north as suggested by earlier workers’ interpretation of the Eocene gravels? And given the Cretaceous zircons seen at Dutch Flat, is there another channel that somehow taps into Cretaceous zircons from the east? Might this be the channel that has the rhyolitic material to the east of Alpha and Omega?
Then there are the rhyolites near Jackson Meadow reservoir in the NE corner of the map. These have been left out of the channels drawn by, say, Garside et al. (2005), yet this appears to represent a fairly sizable deposit.
So there are issues still to be solved in assembling the paleodrainages. But a lot of the attention on these is related to the possible tilt of the range. It is pretty obvious from the elevations on the map above that if those reflect the elevations of a river bottom at a single moment in time, then the southwest flowing segments are far steeper than the north-northwest flowing segments. There are two arguments that would negate such an interpretation as reflecting tilt of the range: that the points are from different times, and that structural anisotropy from the dominant metamorphic rocks will generate these different grades.
So let’s stand on the old river bottom near You Bet, where flow direction indicators from multiple investigators indicates flow was to the north-northwest. Basement rock is above you to both the east and west. So the river had to flow north when it was touching these rocks. If you now go downstream some distance, the deepest the river could have been would be the basement surface at that spot. So the gradient measured in these areas is the maximum gradient possible; it could have been even shallower had the downstream stretch already been covered by sediment or not yet eroded out. The story is a bit different if we are looking at some of the steeper segments. In those places where the channel is again constrained to the sides, we might stand and look upstream. The channel there could have been higher, but it could not be lower, so that grade is the minimum gradient. The result is that the gradients inferred by connecting old thalwegs are probably telling us about actual gradients of rivers. There is the step of constraining rivers of the same age out there, but again you run into limits as the top of the gravels where marked by rhyolites puts a different limit on gradients (that, for now, GG leaves as an exercise for the reader).
[Wait a minute, GG hears you cry, if looking downstream is a minimum and upstream is a maximum, then if you look at the same segment from both ends, the value is precisely determined? Well, yes, except that the times need not be the same. Say you are standing at point A in the past at time T1, right at a knickpoint separating a steeper upstream reach and a shallower downstream reach. Let’s pretend you are looking at a point B 500m upstream. At that moment, point B 500m away is, say, 10m higher–a grade of 2%. Now the knickpoint migrates upstream past this other point to create the modern river-bottom surface at time T2, and in the process point B has been lowered by 5 m. And let’s say that 2.5 m of sediment accumulated on point A, so standing at B looking back at A at time T2 yields a gradient of 10-5-2.5 = 2.5m over 500m or 0.5%. The gradient we see on bedrock is 5 m over 500m or 1%–which the river never had in this case–the downstream maximum at time T2 was 1%, which is indeed higher than the 0.5% we’ve proposed existed at time T2 in this example. The upstream minimum at time T1 was also 1%, and at T1 indeed the grade proposed was 2%, which is greater than 1%. So time matters… Pretty much the magnitude of a knickpoint would determine the maximum spread away from that measured value]
How about the basement anisotropy argument? Certainly this can and does happen, though it cannot make former downhill stretches of river trend uphill in modern coordinates. So the negative grades have to be dealt with as being deformed, misinterpreted or non-synchronous (but note the discussion above). If those are in error (there were five such points identified by Jones et al. (2004), Fig. 3a), what of the other systematic variations?
Rather than try to do battle over this again, GG points to an earlier post that basically suggested that a horizon within the sediments would get around any issues with basement rock. And just to be clear, a more careful examination of the basement rock hypothesis is also possible (e.g., plotting grade vs. difference in azimuth of basement strike and river strike as opposed to just river azimuth). Basically, as there is significant sediment on many of the southwest-flowing segments (e.g., Malakoff), you find sediment accumulating on fairly steep grades when trending southwest but very shallow grades when heading NNW. Lacking both a significant temporal argument and any influence from bedrock, these kind of point right back to some kind of tilt to make everything behave consistently.
Sitting at the very bottom of all this is the peculiar transition from net eroding to net aggrading over long periods of time. In a simple system, you’d expect that as sediment was removed from a mountain range, that the range would rise up isostatically. This should keep rivers from aggrading, yet the Sierra did have rivers go from cutting down to accumulating sediment. Now if there is a huge input of sediment, this could overwhelm the system–but was that the case in the Eocene? Well, if we were seeing this, we’d expect that fans debouching from the range would prograde out into the ocean. Looking at the Ione Formation, which certainly seems equivalent to at least the upper gravels, we don’t see that. The shoreline kind of waxed and waned a few times. Yes, sediment was emerging from the Sierra, but not in such volume as to force shorelines well out to sea. In contrast, the Oligocene Valley Springs shoreline was well to the west, a product probably of both lowering sea levels and a greater influx of sediment.
So if it wasn’t a huge influx of sediment in the Eocene, why did the rivers aggrade all the way to the modern Sierra crest in some drainages? Something had to suppress the rebound of the range; most likely it would have been the continued refrigeration of the range that is now so evident in the very low heat flow values in the bulk of the range. Whether the range actually started to tilt back or simply stalled out rebounding is hard to say, but this would have opened the door to rivers widening their valleys rather than to continuing to cut down.
A corollary to aggradation in the rivers is that relief had to be declining once aggradation began (and actually before then, once the rivers had cut down as far as they were going to cut down). How much relief was removed? That some stream reorganization was going on suggests it wasn’t insignificant. How far east might this was extended? You’d expect less of an effect farther east, so it could be that the Sierra kind of lowered a bit while the hinterland remained at its elevation, which might somehow have led to the rivers headwaters moving east (it is kind of backwards from a typical stream capture environment, but it depends on what the rivers to the east were doing).
And if the rivers were stable and relief was going down, mean elevation was probably declining. Not very dramatically–there was clearly still some significant topography in the Sierra–but still going declining. Note that this might have been a decrease in surface elevation (so surface downfall?) while rock elevations could have been neutral (near zero rock uplift). This is a curious condition for a mountain range…