Wednesday, September 30, 2009

The Bishop Tuff

There's been a lot of hubbub in the geoblogosphere over the past couple of days about caldera-forming eruptions. The trigger for all the discussion was a report about a really old (Permian) caldera-forming eruption in Italy. This invokes discussion of our modern worry-inducing "supervolcanos," like Yellowstone. There's also a lesser-known place in California, where the USGS maintains a volcano observatory there just like they do with Yellowstone: Long Valley Caldera. This caldera formed ~760,000 years ago, and deposited a lot of ash, collectively called the Bishop Tuff. The Long Valley Caldera's true area is about 300 km2, although central sagging has generated ring faults that give it a topographic area of ~350 km2. The Bishop Tuff is thickest right around the caldera, but ash from this eruption can be found as far away as Nebraska.

I got to see the Bishop Tuff firsthand the week before last when I spent a week in the Owens Valley as part of a Geological Society of America Field Forum. I was lucky to be introduced to the tuff by Wes Hildreth, a volcanologist at the USGS's Menlo Park office. Wes "wrote the book" on the Bishop Tuff, and shared an immense amount of information and perspective with the Field Forum participants. I am indebted to him for all the information I'm sharing here. Maggie Mangan just took over the reins of the Long Valley Observatory, and she also participated in the Field Forum. I also really benefitted from talking to her about the eruption. (Any errors that you may find here, of course, are my own.)

The Bishop Tuff is the most striking of many volcanic eruptions along this same system. It's the only one that has produced a caldera. It was preceded by dacite and basalt eruptions at 3.5 to 2.5 Ma, and then by rhyolite and obsidian during the appropriately-named Glass Mountain Interval, from 2.1 to 0.8 Ma. (The Glass Mountain Interval is pretty cool in its own right: at least 60 eruptive units, each high-silica rhyolite!) The focus of both of these was further to the northeast. That area is also home to some post-Bishop eruptions, the youngest of which is at Mono Lake (only 250 years ago). In 1989, a dike came within a few km of the surface, and degassed a CO2 "burp" which killed trees near Mammoth Mountain, which lies on the caldera boundary.

The Bishop Tuff is compositionally similar from bottom to top: it's all rhyolitic pyroclastics, whether it's welded (fused together) or not. Some went north from Long Valley Caldera under the Mono Lake area, while the bulk of it went south towards Bishop, forming the Volcanic Tableland. it has a density of about 1.5 g/cm3.

In this photo, Kim Bishop (yes, that's really his last name) and Peter Lovely (yes, that's really his last name) check out the first of the ashfall deposits, dumped atop lake sediments in a cool outcrop on the southern margin of the Volcanic Tableland, north of Bishop and the Owens River:

A close-up of this contact:
The ashfall portion of the Bishop Tuff has 9 subunits, and you can see the first (F1) and the base of the second (F2) here, overlying the silty lake sediments.

Here's another outcrop, in the Owens River Gorge, where you can see the welded ashflow "caprock" up top, and down below, and outcrop that showcases nonwelded ashfall and ashflow deposits. I've put a box around the area that I'll zoom into in the next photo:

The ashfall deposits are finely stratified and well-sorted, with no reworking. Overlying them, the first of the ashflow (ignimbrite) units shows characteristic poor sorting: big blobs of pumice mixed in with the finer pyroclastics. Most of the ashflow is pinkish in color, but you can see here that the first of it is white, same as the ashfall:
Why pink in the ashflow portion? It's hot when it gets deposited, and heat retention promotes oxidation. The earliest ashflows were dumped atop ashfall (which gets deposited cold), and so likely lost much of its heat downward; hence less oxidation. The entire eruptive seqence is preserved in the Volcanic Tableland north of Bishop. Here, at the southern rim of the Tableland, we're getting the latest flows. The earlier flows didn't make it this far south.

Here's a close-up of that basal ashflow, from the first outcrop. My field notebook is 18 cm "tall," for scale. Note the white color and all the large pumice clasts:
The iron to titanium ratio in these clasts suggests that they erupted at 770 to 800 degrees C. The temperature of the eruption increased as it progressed. This corresponds with an increase in the mafic content of the tephra over the course of the eruption. In the early layers, there's about 77.7% silica, but when you get towards the end of the eruption, you see that number drop to 74%, as well as a doubling of iron content, a quadrupling of the Ca content, and ten times as much magnesium as in the earliest strata.

Here's a close-up of some semi-welded material. This is float, so I don't know precisely where in the sequence it fits, but I would guess the "Ig1" layer, the lower of the two welded ashflows.

And another. One thing I noticed about a lot of the included pumice blobs is that their vesicles were all stretched out into cigar-shaped tubes (prolate), like an L-tectonite. Anyone have any idea what's up with that? I would expect oblate strain ellipsoids (pancake-shapes) here due to post-depositional compaction, but that's not what I noticed...

We made a trip to the lip of the Owens Gorge to look down on the upper ignimbrite (ashflow tuff) layers of the Bishop Tuff:

The first half of the eruptive sequence, dubbed "Ig1" (for "ignimbrite 1") is below the sharp line. In the upper half of the sequence, Ig2, you'll find rhyolite lithics that can be sourced to the earlier Glass Mountain Interval, as well as pyroxene-bearing pumice. You can see here some abortive cooling columns in Ig1:
Likely these don't extend very far down because as soon as they started forming, Ig1 was buried underneath piping hot Ig2 ashflow. This addition of heat disrupted the cooling front and truncated the fracturing process. Sorry I don't have a sense of scale in this photo: it's hard to do when you're photographing the opposite side of a deep gorge. I'd guess these columns are a meter or so across. In one spot, a little downstream (southeast) of here, you can actually see a little ashfall intercalated with these ashflows (it's the F9 ashfall subunit). This, Wes Hildreth told us, is most unusual and quite handy for interpreting the stratigraphy of the Bishop Tuff. The only other place he's seen such a thing is in the Valley of 10,000 Smokes in Katmai, Alaska.

Some close-ups of the Ig2 unit, which is classic "welded" tuff with nice pumice blobs and rhyolite lithics, as well as pyroxene-bearing pumice:

Rhyolite lithic clast in "Ig2" welded Bishop Tuff ashflow deposit:
That's likely from the earlier Glass Mountain Interval, through which the Bishop Tuff erupted.

The "Ig2" layer wasn't the last part of the Bishop Tuff eruptive sequence, but the stuff deposited on top of it was unwelded, and has since been eroded away. In order for a tuff to weld, it needs to be close to 600 degrees C when it stops (this temperature is for rhyolite: it's actually composition and H2O dependent). But the welding process (essentially superhot glass fragments warp around one another and lock into place) has made for a resistant layer atop the modern Volcanic Tableland, and this layer preserves the weaker layers beneath, preventing them from being eroded (except, say, where a river incises downward through the caprock). It's a nice example of differential weathering. Cosmogenic 10Be measurements on the upper welded tuff suggest a modern weathering rate of 2 mm/1000 years.

Now here's the thing that I thought was most interesting about the Bishop Tuff: it's big, and it erupted quickly. There are about 200 km3 of ignimbrite (ashflow tuff), another 100 km3 of fallout (ashfall tuff) out to Utah and Nebraska, as well as 300 to 350 km3 of welded tuff that filled the downdropping caldera (2.5 km of subsidence). That's a lot of magma fluffed out and ejected onto the surface! But Wes calculated that this whole sequence, from the first puff of ash descending from above to the last of the sizzling nuee ardentes, lasted a mere 6 days: a single huge eruption! And, Wes added, "on the seventh day, it rested."

Further reading (I particularly recommend taking a look at Figure 5d!):
Hildreth, Wes, and Wilson, Colin, 2007. Compositional Zoning of the Bishop Tuff. Journal of Petrology 48 (5):951-999.

Labels: , , ,


Blogger Silver Fox said...

The shape of glass shards thrown out in ash-fall eruptions from volcanoes such as Mt. St. Helens, Mt. Mazama, and the Long Valley Caldera is distinctive and diagnostic of particular eruptions - and can include stretched out shards and pumice with lots of tubular vesicles. Possibly what you are seeing in particular pumice fragments is related to the actual eruption rather than to welding and cooling.

I don't have a photomicrograph of the Bishop Ash (air-fall tephra) from Jonathan O. Davis' PhD dissertation (it's not online but I have a copy) - but he describes the morphology of the glass shards as including pipes, spindles, vesicles, shatter, and chunks. The pipes, while smaller in size in photomicrographs of other tephra containing pipes, are similar in appearance to the stretched-looking vesicles in the pumice frags you show.

I should probably work on putting this dissertation online.

September 30, 2009 12:39 PM  
Anonymous Boris Behncke said...

That's a very nice and informative article here, and I loved the bit about the names of some of the researchers in the images - Bishop and Love. The photos are very useful for teaching purposes, so I hope you don't mind if I use them in my volcanology class ... obviously naming the source.

September 30, 2009 3:06 PM  
Blogger Callan Bentley said...

Go for it!
If you want larger/higher res copies of any of these, just e-mail me and give me the specific images you want.

Other educators:
That goes for any image on this blog. If it's mine, I'm happy to give you a bigger copy for teaching purposes.


September 30, 2009 7:05 PM  

Post a Comment

<< Home