Bill- My understanding with dating speleothems is that they are matching curves of U-Th isotopic ratios (my guess would be that they are using ICP-MS analysis for everything these days). It's been a few years, but my experience seeing one of my grad school friends trying to match deepwater coral U-Th records would lead me to believe there is a lot of uncertainty with speleothem dating. Trying to determine where your record gaps are must be a really tricky exercise even with the increased resolution offered by ICP-MS - are there worthwhile results coming out of the research these days? Any papers out lately that you recommend reading? I'm not really in touch with dating/paleoclimate literature these days.
Simon > ---------- Forwarded message ---------- > From: Mixon Bill <[email protected]> > To: Cavers Texas <[email protected]> > Date: Wed, 23 Jan 2008 21:27:42 -0600 > Subject: speleothem dating > When scientists are trying to date the layers in an entire stalagmite, > they take samples from bottom to top, so coring in from the side not a > practical alternative. They pretty much have to saw the thing in half > lengthwise to know they are sampling down the middle, in chronological > order. If one just wanted to date a single sample, say from the bottom > to date something it is growing on, then a core might work. > Speleothem dating, and subsequent stable-isotope analysis of the > dated layers, is a big thing. Starting maybe twenty-five years ago, > Science and Nature magazines both became suckers for this sort of > thing as a paleoclimate study, so of course lots of people jumped on > the bandwagon. There's been an awful lot of awful garbage published in > the area, even in good journals. Initially, radioisotope dating was > done by alpha counting, very imprecise when samples are minuscule. > I've seen papers proudly displaying the dates from various layers in a > stalagmite that had error bars so large there seemed a good chance > that the thing had grown from the top down. The group at Macmaster > University in Canada was a pioneer in that sort of dating. They > managed to date enough speleothems from western Canada to show that > speleothems don't grow during ice ages (surprise!). That is, they > showed it by making a histogram of dates without taking into account > the error bars, and then helpfully adding to the figure the already > known dates of the ice ages. > More recently, technology has improved, mainly through > accelerator mass spectroscopy, to the point where scientists are able > to get data that actually look pretty real. Still, as far as I've been > able to see, nothing has come out of speleothem paleoclimate studies > that wasn't already known or easily conjectured from other sources. > (One wonders how many studies that _didn't_ confirm accepted wisdom > didn't get published.) True, speleothems give a local picture, unlike > ice cores or seafloor cores, but that is actually the problem. The > growth of a single speleothem over time is subject to all sorts of > purely local (like one acre) geological effects having nothing to do > with climate. One groundwater path might clog up and another open, > changing the ratio of stormwater drips to baseflow drips on the same > formation, for example. Scientists are way too quick to overinterpret > data they get from a single stalagmite. One paper I saw had data from > three stalagmites and claimed that they showed some correlation with > something or other (probably an ice age). Well, to my eye, two of the > three did. But if one regards those three stalagmites as a random > sample of a larger collection, that amounts to something like 2 +/- 1 > out of 3. Big whoop! The only exception to this problem I can think of > is a study on the underwater mammalary crusts in Devil's Hole, Nevada. > There the deposit really was sampling regional groundwater, not just a > drip. > It is true that in regions remote from ice caps or stable ocean > floor deposits, speleothems in principle could be one of the few > sources of paleoclimate data, if the study is done right. But even in > such cases, people publish unremarkable findings like it was cooler in > Texas during an ice age. > Oh, well. Professors have to suppport their graduate students > somehow. -- Bill Mixon > ---------------------------------------------- > You may "reply" to the address this message > came from, but for long-term use, save: > Personal: [email protected] > AMCS: [email protected] or [email protected] > --------------------------------------------------------------------- Visit our website: http://texascavers.com To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
