After looking at hundreds of trees and measuring the marks on quite a few, I'm just not convinced I could tell PIWO scaled trees from IBWO scaled trees. Paul Sykes and I thought we were onto something with the measurements, but there is just too much variability in individual trees, degrees of decay, and frankly individual measurement error to make any groove size difference meaningful. I guess we'll just have to find some Ivory-bills scaling a tree and THEN take some measurements ;^)Additional related information is here.
Friday, September 15, 2006
Holzman on bark scaling
Steve Holzman has posted this on BirdForum:
Subscribe to:
Post Comments (Atom)
Posts
14 comments:
It seems as though there is some precedent for a university policing its own ... in this case (linked above) where Columbia university took measures to recall a fraudlent study that appeared in a peer reviewed journal.
Could it happen here?
Is CLO a source or a sink for Cornell?
Oh man, one of these "Tree Groove" searchers came and visited me to ask if I knew of any sites in LA where there might be IBWO work he could measure to "confirm" the measurements he'd taken in AR... I was at a loss! Why waste time trying to confirm IBWOs by such a proximate means when the most obvious thing would be to find a real, live one and not worry about grooves at all! What a waste of effort. It's not like IBWOs are dark matter (can only be detected by gravity effects, can't be seen directly), although CLO certainly seems to think they are... sigh.
My Two Cents
But how does Holtzman explain the fact that ca. 60% of the grooves measured in the Cache/White River NWRs fall well beyond the range of grooves measured elsewhere? See here and here. Can these all be written off as regional variability or observer error, or does it provide circumstantial evidence of the existence in this area of a population of woodpeckers larger than Pileateds? I wonder . . .
Does anybody have any genuine ivory billed woodpecker poop frozen and stored away in liquid nitrogen somewhere?
We should thaw out it out and train bloodhounds to detect it. Then set like 500 or 1000 of those dogs loose in the swamps to see if they can find any.
But how does Holtzman explain the fact that ca. 60% of the grooves measured in the Cache/White River NWRs fall well beyond the range of grooves measured elsewhere?
Most likely a tribe of nomadic Swamp Yetis trying to scare us away from their buried treasure.
Can these all be written off as regional variability or observer error, or does it provide circumstantial evidence of the existence in this area of a population of woodpeckers larger than Pileateds?
This is what I consider to be psuedo-scientific claptrap. We don't even know the variability of Pileated scraping across tree types and regions. Much less what IBWO scrapings look like. We don't even know if all the scrapings so far measured are even by woodpeckers.
Dark Matter! hahahahahah
That was good!
But how does Holtzman explain the fact that ca. 60% of the grooves measured in the Cache/White River NWRs fall well beyond the range of grooves measured elsewhere?
He mentions an easy explanation:
individual measurement error
It's just another example of how this is all about the human mind and hope rather than the reality of living birds.
Get a bunch of believers looking for larger than normal grooves in an area where they "know" there are IBWOs, in a situation where said grooves are indistinct and subject to observer bias, and you're going to find lots of IBWO grooves.
ca. 60% of the grooves measured in the Cache/White River NWRs fall well beyond the range of grooves measured elsewhere?
People can come up with statistics to prove anything. 14% of all people know that.
When Paul and I saw our first scaled tree in the White River NWR the weekend after the announcement, sure our adrenaline was up. Paul's been a hard core birder for probably 50 years, and he said at the time that he's never seen anything like it. So instead of saying, 'yep, that must be Ivory-billed work'. He decided to measure the marks and measure bills and see where that left us. As he started looking in other southeastern states he became more aware that these marks were pretty much everywhere, but they were still smaller than the eastern AR marks. We talked with forest entomologists and they said that the marks weren't from insects, the marks were not beaver and any other mammal. Okay it could have been a hungry man with a screwdriver looking for grubs but that seemed unlikely. It looked really good until we started doing some blind tests and discovered some significant measurement error. Now you would think the error would be the same across the region, so maybe that wasn't the only factor at work. Some of the big marks were mostly in Green Ash...which was mostly in the White River NWR...but we submitted the data to some sophisticated statistical analysis (we used the talents of a UGA biometrician, I can't even begin to explain the analysis myself). Anyway after all that it really looked like there was something going on in eastern Arkansas that wasn't going on anywhere else. I'm still at a loss to explain it.
But of course you are right, EVERYONE wants to find the damn birds and document them. Measuring marks is no substitute at all.
BTW, thanks for sticking up for us, methinks.
I didn't read any of the comments as attacking Holzman and Sykes. We were attacking the Anonymous that was subtly trying to convince us that IBWO exist in Arkansas because of their analysis.
A view of their analysis which Holzman and Sykes apparently do not support.
That is all and that is everything. That Dark Matter comment was still a good one though.
Anyway after all that it really looked like there was something going on in eastern Arkansas that wasn't going on anywhere else. I'm still at a loss to explain it.
Seriously? You can not come up with a SINGLE rational explanation which could explain the difference in the marks, that you have not already tested?
Ok, I was trying to be nice to them. But Amy brought me back to reality.
Come on, Methinks. Look at Holtzman's quote that Amy repeats. It's always there with these IBWO scientists. Always a "yeah but" that they end with. Always trying to leave the impression that their data still might suggest IBWO's live out there.
Yes, and crop circles might be made by IBWOs too. At least in these peoples minds. Amy is right. Holzman, you can easily come up with one reasonable reason that your data's mean has such a wide variance that an area might just stand out in any random analysis.
Eastern Arkansas was your variance. It's that easy. The only question is whether you were sampling insects, Pileateds, Red_headed, other woodpeckers, ring-tailed cats, or Aliens.
It's Pileated, stupid!
Yes, methinks, I agree with everything I said before and everything you just said.
But I am thinking of the next stage. Where there is no "yeah but, it could still be IBWOs" to us to ridicule. Back to the good old days, I say. Off with there heads.
And if your report, Methinks, of graduate students with 7 pairs and no pictures is truly presented at AOU, then I dare say the return to the good old days is not far off.
In my expert, non-humble, opinion. Commenter #2 said it all. (and he had the funniest comments)
Obviously, he/she is a level headed, knowlegeable, well-spoken, genius.
Anyone, who doesn't accept Commenter #2's analysis needs a mentor.
Post a Comment