GG has been piddling along though the Sierra (ostensibly to give a campfire talk in Mineral King) and in doing so stared a bit longer at a recent paper on the age of a pediment in the Sierran foothills by Sousa et al. in Geosphere in 2017. In a way this is a callback to concepts from far back in the geologic literature, namely the significance of an “Eocene erosion surface.”
Here, to be brief, low-temperature thermochronology from a low-elevation pediment in the western foothills of the Sierra yields very old ages–in fact, overlapping with the emplacement of plutons in the Sierran crest [this was not a unique observation; Cecil et al., 2006, had a pretty old point in their collection]. Sousa and coauthors model these data and get a cooling to surface conditions by about 40 Ma. Because these pediments abut noticeable topography, this means there was at least that much local relief in the ancient Sierra. While the pediments had been noticed by others, many suspected a far more recent age.
In some ways, this is old news. The Eocene sediments in the northern Sierra have long made clear the presence of significant local relief, and many workers had inferred that such relief was probably higher in the southern Sierra (e.g., Wakabayashi and Sawyer, 2001). But the southern Sierra lacked the Eocene sediments necessary to know what the Eocene landscape might have looked like, so this paper opens up a new window for us.
Where does this lead us? Kind of down a rabbit hole only to come up with no strong and useful statement–though perhaps future work could nail things down. This is more a personal attempt to try and grasp what is going on, so profound errors might exist and insights are few. So, proceed at your own risk….
How should one read a scientific paper? As presenting conclusions one should take as our best estimate of truth? Or as information one can use to test competing hypotheses? You might think it must be one or the other, but that is rarely the case.
Consider the just-published paper by Bahadori, Holt and Rasbury entitled “Reconstruction modeling of crustal thickness and paleotopography of western North America since 36 Ma”. From the abstract you might be tempted to say that this paper is solving a problem, in this case the Late Cenozoic paleoelevation history of the western U.S.:
Our final integrated topography model shows a Nevadaplano of ∼3.95 ± 0.3 km average elevation in central, eastern, and southern Nevada, western Utah, and parts of easternmost California. A belt of high topography also trends through northwestern, central, and southeastern Arizona at 36 Ma (Mogollon Highlands). Our model shows little to no elevation change for the Colorado Plateau and the northern Sierra Nevada (north of 36°N) since at least 36 Ma, and that between 36 and 5 Ma, the Sierra Nevada was located at the Pacific Ocean margin, with a shoreline on the eastern edge of the present-day Great Valley.
There is one key word in that paragraph that should make you careful in accepting the results: “model”. What is the model, and how reliable is it?
In the previous post, we discussed how Occam’s Razor is of little use in some arguments, leading to the principle of least astonishment. But here GG would like to suggest that the shear immensity of geologic time means that Occam sometimes cuts us off from explanations we need to consider.
In this case, let’s talk Laramide. Orogeny, that is, the creation of the Southern Rocky Mountains between something like 75 and 45 million years ago. The prevailing explanation is that the subducting ocean floor only went down to about 100 km or so and turned flat, interacting with the continent in a way to make mountains far from the plate edge. It is a nice compact explanation.
The thing is, there are a lot of places where slabs today are flat and none of them produce anything of the scale of the Laramide Orogeny. Closest are the Sierras Pampeanas in Argentina, which are far closer to the trench than the Laramide ranges were, among other difficulties. Even looking over past orogenies yields few plausible rivals–maybe the Alice Springs orogeny in Australia, or if you push things hard, perhaps the Atlas ranges in northern Africa. Or, of course, the Ancestral Rockies in almost the same place as the Laramide. But these are just as cryptic and far less common than all the events that created the Appalachians, or the Urals, or the Caledonides, or the bulk of the Alpine-Himalayan system.
Perhaps, when we encounter oddities in the past, we need to recognize that something unusual happened, meaning that Occam’s bias for parsimony might in fact be precisely the wrong bias. For instance, somebody walks up and says they will flip a coin ten times and it will come up heads. He asks a passerby for a coin and then does as he says. Parsimony says this was luck, but perhaps a better explanation is that it is a trick either involving an accomplice or sleight-of-hand [scientists are suckers for sleight-of-hand, as the Amazing Randi often showed].
Given the number of times slabs probably have been flat and given the far rarer production of mountain ranges far from the trench, maybe our bias for parsimony should be relaxed–odd and unusual results might demand more than a single cause. Maybe things were a bit Rube Goldberg-ish for awhile. In a similar vein, some workers are arguing that the impact ending the Cretaceous was so effective not just because of its size but because of the sulfur-rich rocks it hit (this in part a response to the absence of other impacts in causing extinction events and other extinction events seemingly lacking a coincident impact). Arguably something like this has or will emerge in explaining how one branch of the great apes led to humans despite lots of earlier evolutions of animals failing to reach a similar end. We often focus on the positive outcome–the mountains made, the extinction that happened–and miss how often the simple explanation predicts something that didn’t happen (kind of like the old quip that the stock market predicted nine of the past five recessions). We don’t ask, why are there no mountains in Iowa, for instance; we ask, why are there mountains in Colorado? But perhaps we need to ask both.
Occam reminds us to be distrustful of overly-complex explanations, but maybe we need to be careful not to demand too much simplicity. All theories will conflict with some observations in some way; there are always strange things that happen that are coincidences or results of unrelated phenomena. This reality means that no theory will fit every possible observation; what’s more, we tend to accept more misfits for simpler theories (for instance, the half space cooling model for ocean floor topography is widely accepted despite all the oceanic plateaus and seamounts one has to ignore to get a decent fit). Given that, we should wield the Razor more carefully least we cut off our theoretical nose to spite our parsimonious face….
A favorite shortcut employed by many in trying to decide between hypotheses is to enlist Occam’s Razor–that the simplest explanation for something is most probably right. Now this has strength because humans are pretty good at rationalizing notions they put forward, adding in new ingredients to keep a favored explanation from collapse. But a theory that has probably passed its must-use-by date will have enough extra bells and whistles to discourage Rube Goldberg from trying to get it to work.
However, there is nothing that says Mother Nature had to be supremely parsimonious. In a complex system like Earth, there can be odd coincidences that are meaningless (like the Moon and Sun sharing the same apparent diameter from Earth’s surface) and outcomes that might be highly improbable (taking over 500 million years to get intelligence after making complex animals with hard parts seems like dawdling, especially when burning most of that time on dinosaurs). Even so, Occam can be a help if used with care.
But lots of times you can face competing hypotheses that lack Occam-style clues. For instance, which is simpler: that post-5 Ma erosion of the High Plains of the U.S. was caused by an eastward tilt, or that this was the product of a changing climate? Both are pretty easy to describe; both have issues. Yet many earth scientists feel pretty comfortable arguing that one is correct; what is the basis of such assurance?
Arguably the most common discriminator used by earth scientists is the principle of least astonishment. What surprises you least feels, in an Occam kind of way, like the interpretation that is most likely. The problem is, we all are astonished differently.
If you are a sedimentologist, you might look at the problem of the High Plains as one of depositing the Ogallala Group in the Miocene as crucial. Could you possibly deposit something like that on a slope like that we have today? This seems so astonishing that if can’t be right; the original slope had to be lower.
But maybe you are a geophysicist looking at the ways to create a tilt about 5 million years ago over something like 1000 km. That looks really hard to do, especially if dynamic topography from flow under the lithosphere is ruled out. It would be astonishing if that happened; it must be that the grade was already there much longer ago.
Skepticism from both geoscientists is warranted; either of these seems really hard to do. Data is gathered by both sets of experts. Margaret McMillan and colleagues measure paleogradients in the Ogallala using a widely applied approach and find there must have been a lot of tilting. Will Levandowski and colleagues (including GG) look at geophysical measurements and find support for the elevations comes from within the crust, where changes over the past 5 million years seem exceptionally implausible.
Could these be resolved? Well, you could posit that prior to 5 Ma there was dynamic subsidence holding the western end down and once that was released, the crustal buoyancy expressed itself. But now Occam detectors are flashing red–this feels ad hoc. Of course, there could be mistakes in the measurements of paleogradient, or in relating seismic wavespeeds to densities–each side has poison darts to shoot at the other side.
What makes this frustrating is what makes this interesting. After all, in the end somebody will be astonished–the earth did something they didn’t expect.
And a funny thing, shoes flip feet in the Sierra, where those studying the sediments argue for no tilting despite deposition at an even steeper grade than modern-day Ogallala, while geophysicists feel they have good evidence for a very recent change in the buoyancy structure of the region.
Are you astonished yet?
It used to be when we thought what the southwestern U.S. looked like at 85 million years ago, back in the Cretaceous, we thought things were pretty simple. There was a nice volcanic arc running from the Sierra down through the Mojave into the (restored) Peninsular Ranges. To the east was a fold-and-thrust belt extending nicely from well up in Canada down to the Las Vegas area in continuous fashion before taking a left turn to angle more erratically into more complex geology across Arizona and New Mexico. East of that was the foreland, its western edge bowed down under the weight of those fold-thrust mountains and the torrent of sediment washed off those ranges toward the inland sea that stretched up from Texas towards the Arctic Ocean. The only real wrinkle in this had to do with where the troublesome exotic terranes now in British Columbia were at the time.
Now it feels like 85 Ma is instead a pivot for everything about the western U.S. A number of puzzling things seem to be going on around this time.
- Emplacement of the first of the subducted oceanic/forearc schists (the Catalina and San Emigdio Schists) appears to occur about 85-90 Ma [Jacobson et al., 2011]
- The Sierra Nevada sees the culmination of the most massive plutonic episode in its history just prior to a complete end to plutonism [e.g., Ducea, 2001]
- The Mojave Desert will continue to see both peraluminous and metaluminous magmatism for another 10 million years despite the inferred emplacement of most of the Rand Schist over that time. [e.g., Barth et al., 2013]
- Shortening in the hinterland appears to be a short-lived interval between extensional episodes [Wells et al., 2012]
- The Sevier fold-and-thrust belt in the vicinity of Las Vegas is dead or dying–some put its end before 90 Ma [e.g., Fleck et al., 1994]
- Perhaps a northwest-trending thrust belt has instead invaded the area about this time. [Pavlis et al., 2014]
- Canyons start to form in the southwesternmost Colorado Plateau [Flowers and Farley, 2012].
- In contrast, the simple foredeep geometry of sediment accumulation in the foreland begins to broaden out, suggesting a “dynamic subsidence”. [Liu and Nummedal, 2004]
By 75 Ma, it is really clear that things have gone super strange. The Sierran arc is dead with no real replacement. Magmatism at that latitude is limited to smaller intrusions in the Colorado Mineral Belt and some peraluminous (two-mica) granites in Nevada, eastern California and parts of Arizona. In contrast, the Mojave continues to see plutonic activity with both peraluminous and metaluminous magmatism, seeming to be the southwestern end of a long linear belt of unusual magmatism. The Sevier hinterland has several examples of significant decompression accompanied by more limited evidence of extension. The foredeeps in front of the fold-thrust belt seem to have ceased to accumulate much sediment, which is now captured in broader basins farther east. Laramide uplifts are certainly underway.
Although 75 Ma is often taken to be the start of the Laramide orogeny [Dickinson et al., 1988], it increasingly seems like the orogeny’s origins lurk in the preceding 10 million years. What exactly was going on?
Some of the things we don’t really have a solid handle on:
- Where was the Insular Superterrane? [We might have a more decent view of the Intermontane Superterrane due to work in Idaho over the past decade].
- What really was the plate geometry? The Kula plate dates back to about this time; were there other plates to the east of the Kula-Farallon-Pacific triple junction?
- Do we have a good handle on where plutonism was active in the Mojave/Sonoran deserts?
- How much of the Cretaceous decompression events are from extension vs. some kind of intra-crustal pseudo-convection?
- How could the southern Nevada part of the fold-thrust belt be inactive?
- How can schists be emplaced against the middle continental crust immediately below fresh plutonic rocks?
The Laramide Orogeny made North America look the way it does today; this time period holds the keys to understanding how it could happen. Some way-out ideas are kicking around [e.g., Hildebrand, 2009, 2013; Sigloch and Mihalynuk, 2013], though some are moderating a bit [Sigloch and Mihalynuk, 2017]. Conversely, most earth scientists have gravitated to some flavor of subduction of an oceanic plateau as the cause of all the misadventures near the start of the Laramide orogeny. That all of these have pretty substantial flaws shows that the community is struggling to understand just what was going on at the start of the Laramide. Hopefully over the next few months we can explore some of these topics in some detail…
Having just remembered the 1906 San Francisco Earthquake brings to mind Harry Fielding Reid’s model of elastic rebound for earthquakes developed from observations of that 1906 quake. The idea that the earth’s surface was slowly moving in opposite directions across a fault over a long time period, straining the rocks near the fault until a critical point was reached when the strained rocks would cause the fault to rupture, allowing each side of the fault to “catch up” with the more distant parts of the earth’s surface farther away.
Much later, when plate tectonics was developed, earth scientists could tell what the average velocity of plates were over a couple million years from analysis of magnetic anomalies on the seafloor. When space-based geodesy came along, first with VLBI and then with GPS, geodesists found that the plates were moving today at a rate equal to that seen over millions of years. It seemed as though the earth ran at a smooth and even pace.
The combination of ideas would suggest that one hope expressed about a hundred years ago was that faults would be triggered like clockwork. Every so many years, termed the recurrence interval, a fault would rupture with what would be called a characteristic earthquake. Ideally you could then predict the next earthquake if you knew when the last couple had happened.
This ideal view of the earthquake world has gradually unravelled, with a couple of observations in the past decade indicating that there really is something more variable in how geologic strain is created than the elastic rebound model and smooth plate motions would have suggested.
It seems like the Fall AGU meeting brings some new wrinkle to the GPS measurements in the Sierra. In the past we’ve seen suggestions that the Sierra were going up tectonically, then that they were going up because of water removal from the Central Valley, then they were still going up even with water removal in the Central Valley, and now we have the Sierra going up because of water removal in the Sierra itself. This latest missive is from Don Argus and several colleagues at JPL deserves a look; their paper on this was published about at the same time in the Journal of Geophysical Research (though that paper doesn’t have the coda from the AGU talk about the loss of elevation in the wet winter of 2016-2017)
Basically they wrote that during the California drought from 2011 to 2015 that the Sierra lost 48 km³ of water and so rose at least 17 mm from that loss while also rising an additional 5 mm from water loss in the adjacent Central Valley and then might have risen no more than 2 mm from tectonics for a total elevation gain of 24 mm (or just about an inch). That is a lot of uplift for a few years. This interpretation means that the Sierra actually stores a lot more water within its granites than is typically thought to be the case, which aligns with earlier work by Argus and colleagues.