Monday, November 08, 2004

SVP

SVP was great this year. As usual I missed a few of the morning talks because I was sleeping off some of the drinks from the night before. Paleo conferences are most certainly not places to get some sleep. Or rather, they aren't if you're a paleontologist. I could imagine a lot of people falling fast asleep in the dimmed halls while someone spouted jargon way above their heads. Thankfully though I was quite enthralled by the lectures and now have a much clearer idea of what I'm doing with all of this dino and bird data that I've collected. It boils down to: I have so much more to do. There was a couple of fellows who modeled the craniofacial ontogeny of the allosaurs using simple linear regression of measurements that they took using calipers. They took alot more measurements than I have, but I think that I'll take their 50 some odd different variables and apply it across taxa in order to get a better picture of the evolution of these animals. Either that or I could just use image analysis software to plot the change in shape of these skulls. In either case, I'll then use Finite element analysis to plot the amount and area of strain on the bones in the skull in a 3-D computer model based on ct scans, and see how the forces are transmitted and how it changes with a change in each of the measured variables. Then, using the concept of functional domains in the skull, I'll determine how those changes in force distribution effect each of the domains. Then if I figure out how growth rates are involved (within a phylogenetically independent context), I'll have a pretty good Idea how these animals are evolving and about the multitudinous tradeoffs involved during evolution. It'll boil down to a hell of a lot of multiple linear regressions, resulting in a model that will probably have about 50 dimensions. This way, if I ever get a partial skull, I'll be able to predict all kinds of crap about the rest of it, without ever seeing it. People with think I'm crazy and say that no-one can possibly predict all of that because of a cumulative effect of all of the deviations in the data away from an exact line. Then I'll avoid making sense out of my statements by referring them to information that they would have already known or predicted on their own and stroll out amidst a hail of questions, angry shouts and calls from starry eyed undergrads.
The last part I embellished on a bit, but I don't think that it's an unreasonable prediction.

No comments: