Here, by the way, is the link to Thermit’s report:
www.chemtrailcentral.com...
As I mentioned to Thermit when I provided him a critique of the report (which critique he had requested), the study is an extremely ambitious one,
and, as such, involves several measurements which are not relevant to the goals of the study.
I’d mentioned that a simpler approach would be to ask the following questions:
1. Can we correlate contrails with atmospheric conditions?
In order to do this, of course, we’d have to have the altitude for each aircraft which creates a contrail, which means it’d have to show up as
identified with altitude, vector, etc on the FE screen. Now you might say that this would eliminate from the study any of then “unidentified”
aircraft, most of which are purported to be military. I know; but we’ll get back to them in a minute.
The next step would be to build a table of observed aircraft contrails. The number of days which it is overcast is irrelevant, since we wouldn’t be
taking any measurements that day anyway. The table header would look like this:
Aircraft********Contrail Duration********Altitude********Temperature********Humidity
The time of the aircraft flight, its vector, speed, etc. are irrelevant, because all you want to do is to correlate the persistent contrail to the
temperature/humidity
You’d get the flight number and altitude from FE, contrail duration from your own timed observation, and Temperature/Humidity from GOES atmospheric
soundings. Noter that Thermit splits the difference in many cases by interpolating temperatures from neighboring regions; this is not the greatest
thing around, since the atmosphere is very dynamic even over a 5000-feet altitude delta, but it’s the best we can do, and no one would fault Thermit
for his interpolation.
What would this show? Well, you’d probably use a binomial distribution (or maybe a Poisson distribution, if the incidence is very low) to predict
whether a contrail would be found at a particular temperature/humidity regime. But the actual measurement itself would be a simple correlative of
“contrail duration of x or greater seconds” and “temperature is below minus 40 deg and RH is 100%” If it’s statistically significant
within, say, the first sigma, you’ve pretty much determined what’s causing the contrails that persist for (plug in your number) of seconds.
2. Can you correlate persistent contrails with unidentified aircraft?
This is a lot easier, because all you have to do is to see an aircraft which doesn’t show up on the FE log, and time the persistence of its
contrails (if it has any). Let’s say that, in a given week, you see 171 identifiable aircraft of which 134 produce contrails of interest
(“interest” being persisting for a predetermined number of seconds). During that same period, you see 16 unidentifiable aircraft of which 14
produce contrails of interest.
Simple math shows that the commercial aircraft have an
n of 0.78 and the unidentified aircraft have an
n of 0.875 which, given the
small sample size is statistically insignificant. So you could conclude that there isn’t any difference between a military or a civilian aircraft;
they’re both equally likely, using binomial distribution, to produce contrails of interest.
But what if the commercial aircraft have an
n of 0.1 and the military aircraft have an
n of 0.85, and you have a significant sample?
Well, you can make several observations:
1. the military aircraft are flying at a higher altitude.
2. the military aircraft are using different fuel additives, such as Prist, which results in a different combustion characteristic.
3. The military aircraft are flying using different flight rules for their engines than are the commercial aircraft.
4. The military aircraft are emitting something other than the typical combustion byproducts and water vapor.
Now which one of those reasons is the real one? Well, you can’t tell; there’s just not enough information available.
But, unfortunately, despite all the graphs, Thermit never provides the basic correlation. He shows a lot of data, but not all of it, and it’s
simply not organized right. But the good thing is that at leat he’s doing some real research. What needs to be done is to design the study with a
lot more rigor, and do it again – and again – and again.
[edit on 2-3-2005 by Off_The_Street]
