[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Precision

Tim and Mike (and possibly others) "collaborated" to write the meta-post I'd like to comment on:

Certain estimates are less trustworthy - like Colbert's 78-tonne
uber-brachiosaur, for example.  It all depends upon the method(s)
used to estimate body mass, including the accuracy of the 3-D
("flesh-and-blood") reconstructions.

... and indeed the accuracy of computer models.  Here I am thinking of
Gunga et al's (1995) Colbert-like mass estimate of 74.4 tonnes, which
is based on a computer model built out of tubular sections ... that's
circular, rather than elliptical, in cross-section.  Using this method
they managed to estimate 11 tonnes for the neck alone(!), i.e. three
quarters of the mass that Russell et al. (1980) came up with for the
whole animals.  So the moral is that more recent estimates, such as
that of Gunga, are not necessarily better.

Agreed! To further editorialize, mass estimates taken from say, femoral circumfrence simply are not very accurate, even amongst extant animals. It may seem "less arbitrary", but it's more wrong none the less. While volumetric mass estimates rather obviously depend on the accuracy of the model used, this isn't really a bad thing, it just means that we need to work to make our models as accurate as possible. Having spent some time dipping models (and calculating 3d scans volumes in a computer) I ahve to second Greg Paul's DInofest volume in that "artistic judgements", e.g. adding a dewlap, or making a limb more or less scrawny, don't have nearly as large of an impact on a models volume as simply getting the basic proportions wrong (e.g. making the trunk far too long, or not being able to fit a skeleton inside the model). This does not mean that all of Greg's estimates should be taken as gospel (sorry Greg), but at the least they are good attempts based on actual anatomical interpretations. If those interpretations are in error in some way (e.g. if the chest is too wide, as David has suggested) then we can fix those errors and get more precise. The problem with mass estimates isn't that too many people have produced disperate estimates, it's that not enough people are trying with a more standardized method.

Curtice and Stadtman (2001) made a pretty convincing case for the
_Dystylosaurus_ specimen being an anterior dorsal of _Supersaurus_
(not least because it was found between the two _S._ scapulae :-) but
I noticed that Upchurch et al. (2004) still lists _D._ as a
brachiosaurid, which was the assessment made by McIntosh (1990).
I've not spoken to Paul about this, so I don't know whether this
constitutes a disagreement with Curtice and Stadtman's assessment; but
I think it's more likely that their manuscript was submitted before
C&S's became available.

Dave Lovelace and myself are also supporting synonymizinf Dystylosaurus into Supersaurus. Frankly, we think Curtice and Stadtman did a good job with that, but we went ahead an re-examined it; it's pretty clearly Supersaurus when, you know, you have more Supersaurus bones to compare it with.

By the way, _Supersaurus_ is generally considered to
be closest to _Barosaurus_, not _Diplodocus_ -- for good reasons I
think.  I know there is another hypothesis out there, but that
hypothesis doesn't bring _Supersaurus_ any closer to _Diplodocus_.

Indeed, this has traditionally been the case. Almost entirely because of the caudal series (which we believe to actually be Barosaurus) that was ascribed to it by Curtice. The tail attached to our Supersaurus (hurray for single taxon sites!) doesn't look remotely likely Diplodocus, and is also clearly not Barosaurus. There are some other Morrison sauropods that it looks like though...

Scott Hartman Science Director Wyoming Dinosaur Center 110 Carter Ranch Rd. Thermopolis, WY 82443 (800) 455-3466 ext. 230 Cell: (307) 921-8333


Check out the new AOL. Most comprehensive set of free safety and security tools, free access to millions of high-quality videos from across the web, free AOL Mail and more.