A (nearly) forgotten constraint
Over the past decade or so, a fairly common event has been the publication of a paper taking on a piece of evidence for large magnitude extension in the Basin and Range. For the most part, this has been going after the various individual constraints used to make Wernicke et al.’s (1988) estimate of 247 ± 56 km of WNW-oriented extension. For instance, there are papers attacking the displacement on (or existence of) the Mormon Peak detachment, on the reconstructions of the fold-and-thrust belt across the Death Valley region, and on the age and amount of extension across Panamint Valley. The net result some would infer is that extension in the Basin and Range here was actually a lot less than 247 km (yes, GG has been told “a lot less”).
Rather than wade into these multiple controversies, GG would prefer to step back and ask, are there other ways of coming at this? There are a couple of approaches. For instance, one can try to improve the somewhat circular logic of Coney and Harms (1984) and try to use the modern and estimates of the pre-extensional thickness of the crust to get at total extension. A problem is that the crust has been an open system and constraining the amount of magmatic additions limits this approach.
The other approach comes from a rather unexpected quarter: plate tectonics. Or more precisely, plate tectonic reconstructions, and is has maybe been overshadowed by these other arguments. It is rather clever, but to see its power, we have to take a moment to understand what is going on.
The boundary between two plates can be a mid ocean spreading ridge, a subduction zone, or a transform (strike-slip) fault. All you really know about two plates separated by a subduction zone is that a pair of points, one on each plate, used to be farther apart. While there is more hope for two plates separated by a transform fault, in practice this means identifying and dating offset features, which proves to be challenging. In contrast, the separation of two plates across a spreading center is recorded by the new crust created between the two plates. The age of that new crust is inferred from magnetic anomalies created as the earth’s magnetic field flips north-to-south or south-to-north every so often. In a sense, you can slice out the new crust, move one plate back towards another and now know their relative positions when the youngest remaining crust was formed.
We are interested in the position of the Pacific and North American plates back about 30 million years ago. But there is only a small and relatively young piece of spreading ridge between these two plates, in the Gulf of California. Aside from not going back all the way to 30 million years, the magnetic anomalies and associated transform faults are poorly defined. To figure out the relative position of these two plates, we have to travel a circuit, crossing from plate to plate, moving each back as we cross a spreading center. The circuit goes from North America through Africa through Antarctica (usually) and then to the Pacific plate. It is crudely illustrated by an animation at this link, where each piece of seafloor younger than a certain age is first highlighted and then removed, the remaining plates pushed back towards an unmoving North America. We pull old crust back out of subduction zones (purple) as we do this. A pair of arrows highlights positions on either side of the San Andreas Fault in California; at the end, we can see that we recover the motion of the San Andreas Fault over the past ~10 million years.
OK, so we can do this for times like 30 million years ago; so what? Here is the other trick: some parts of the Pacific plate were just being created then at a mid-ocean ridge in very close proximity to North America–but you cannot create ocean floor underneath a continent. So these points have to project back to a spot off of North America. One such point is now just off the northern California coast and was restored to its original position by Stock and Molnar in 1988:
The reproduction of the figure leaves a little to be desired, but the shaded ellipse represents the position of the northern end of anomaly 10 crust and its uncertainty (~30 Ma as understood in 1988, now estimated to be closer to 28 Ma). The thick black lines are the active spreading centers at this time from the best fit reconstruction. But the spreading center reconstructs on top of North America–and keep in mind that the contour on the seaward side of the coast is the edge of continental crust. But you cannot create seafloor under a continent!
The solution is simple: the continent had to have expanded since 30 Ma. Stock and Molnar wrote
“If all of this overlap were due to extension within western North America, and the other plates in the plate circuit had behaved rigidly, this overlap implies a minimum of 340±200 km of extension perpendicular to the coast (azimuth N60°E) at the latitude of northern Sonora (30°N) since the time of anomaly 10 (30 Ma). Note that if extension at this latitude were not oriented parallel to N60°E, a greater amount of extension would be implied.”
Now there are some alternatives in the 1988 paper owing to uncertainties in deformation in Antarctica; much of this was resolved in a later paper by Atwater and Stock, although the focused on slightly younger times:
Again, the eastern edge of the Pacific plate cannot be under North America. Although they don’t show a basic reconstruction for the opening of the Gulf of California, closing up that deformation still doesn’t get you off North America. Overall they estimate a total of 310 km of extension trending S60°W, only 60 km of which is due to the San Andreas/Gulf of California deformation and 235 km of which is in the Basin and Range. Uncertainties in the plate reconstruction are now much smaller, about +/- 40 km for the S60°W component. Atwater and Stock draw the resulting net picture of deformation like this:
A 2016 paper by DeMets and Merkouriev refines this a bit, reducing the extensional overlap by 75 km, still leaving a minimum of 235 km of extension to exist in North America (a minimum because the reference points used might have been somewhat distant from North America).
So advocates of small amounts of extension in the southern and central Basin and Range are going to have to do more than just question some of the geological interpretations within that area, they would also need to come to grips with this constraint derived from magnetic anomalies measured in oceans around the world. You never know what will rise up to tell you what happened….