Thursday, March 04, 2010

Jumping the Gun

We’ve recently seen another example of scientists jumping the gun by making absurd claims which only serve to backfire upon science as a whole. In this instance, the claims were made so quickly after the events that it is impossible for the claims to have even been peer-reviewed. Yet they have become “scientific truth” because of the authority of the scientists who made the claim.

Of what do I speak? I speak of the claim made by a NASA scientist (and apparently backed up by “NASA officials”) that the recent earthquake in Chile was responsible for moving the Earth’s axis by 8 cm and increased the rotation of the Earth such that our days have permanently become 1.26 milliseconds shorter, as documented in this article. We are told in very scientific terms that this axis is the “figure axis” and not the north-south axis, and that this figure axis “is the axis around which the Earth's mass is balanced.”

I’ve mentioned in the past how science relies on both accuracy and precision, and what the distinction between the two is. Accuracy refers to how closely a measurement corresponds to reality; precision is the degree to which we are certain of our measurement. Precision is measured in science by the number of zeros after the decimal point in a number (as expressed in Scientific notation). So, 1.0 meters is not as precise as a measurement of 1.00 meters, because 1.0 meters could have been rounded from 0.9 meters while 1.00 could NOT have been rounded from 0.9.

When we deal with an object the size of the Earth, 8 cm is a very precise measurement. Let’s put this in perspective. Wiki says the Earth has an average diameter of 12,742 km. That's precise to the nearest kilometer...but if we only have the measurement of the diameter of the earth to the nearest kilometer, we do not have the precision available to claim the movement of something by mere centimeters. (Note: I assume NASA has more precise measurements, but this is just to demonstrate my point regarding precision.)

Now here’s the thing. The rules of precision still apply, and they cross between types of measurement too. That is, if you want to know what temperature water boils at a given air pressure, you’re looking at two different measurements that require precision. First, you need to know the precision of how you measure air pressure (barometer); secondly, you need to know the precision of how you measure temperature (thermometer). Since your answer relies on both, even if you have an uber-precise thermometer, if your barometer is imprecise your final answer is likewise imprecise. Your answer can only be as precise as your least precise measurement.

So let’s move on to the claim by NASA that the Earth’s axis shifted 8 cm. Centimeters are units of length, but this axis is related to the mass of the Earth too. So we need to know a) precisely what the mass of the Earth was and precisely how it was distributed before the Earthquake and b) precisely what the mass of the Earth is and precisely how it is distributed now. And the precision of our measurement of mass must be precise enough to justify the claim of a move of a mere 8 cm. Thus, we need extremely precise mass measurements, and extremely precise length measurements. If we do not have both, we do not have a precise answer.

Just to show you the impossibility of the measurement, this requires NASA to know precisely the mass of the material that was moved in the Earthquake and precisely how far away from/toward the original figure axis this material moved.

What about the shorter day then? Is it possible to verify a shortening of 1.26 milliseconds in our day? Well, that has its own problems. If we measure sunrise and sunset, we have to account for such things as air temperature and humidity when determining how long light travels before it reaches our sensor, as well as the location of the horizon (i.e., perhaps we measure sunrise from the ocean—in which case, wave height comes into play—or from a hill, in which case plate tectonics and erosion come into play). All of these things cause miniscule precision errors, and we haven’t even talked about the inherent precision level of the sensor itself. All of which renders it impossible to verify that the days are 1.26 milliseconds shorter now then they would have been otherwise.

But in reality, the situation is even worse than that, for the claim that the Earth shifted 8 cm on its axis and days are 1.26 milliseconds shorter now is not based on any observational data at all. Rather we read: “The computer model used by Gross and his colleagues to determine the effects of the Chile earthquake effect...” etc (italics mine).

At this point, I have to wonder if the computer model even takes into account the fact that the earth is not a true sphere, but rather bulges at the equator. I certainly hope it takes that into effect. But even if so, there’s absolutely no way it accounts for all of the Earth’s topographical features—the mountains and the valleys—which help determine the distribution of mass on Earth. Nor could it take into effect all the various mine shafts that have taken mass from inside the crust of the Earth and moved it to some of the many millions of tons of building material on the face of the crust. Even if it did, it would have to be able to accurately show the composition of the Earth’s crust across the entire Earth (density of material would affect where the center of mass is, and thus where the figure axis is).

Finally, this came so quickly after the earthquake that it’s impossible for any of this to have been peer-reviewed. Instead, it’s just a NASA press conference. Which is fine as far as it goes, but when people assume you’re presenting a scientific point of view when you’re really just rushing a computer model out, that is a disservice to science.

And some other scientists have noticed. As reported here, several German scientists have questioned NASA’s conclusions. One said:

It is highly doubtful that these calculations are correct. The changes to the Earth’s axis caused by an earthquake would be so tiny that it isn’t measurable and therefore impossible to reliably detect.
Another said: “It is impossible that there could ever be such a severe earthquake which would observably move the Earth’s axis.”

Given all the factors involved, I would have to agree. It is true that we can know quite a bit. For instance, I have no doubt that we could probably get within several kilometers of where the figure axis of Earth is. We could also measure to within fractions of a second the exact length of a day. But “several kilometers” cannot justify “8 cm” and I doubt that we measure the time of day precisely enough to justify a 1.26 millisecond change. In other words, I agree with these German scientists that any effects from the earthquake would be unobservable.

The problem is, this is not what people will remember. What people will remember is that NASA claimed the Earth’s axis moved and days are now shorter. This is what the news ran with, even though it is highly dubious and not based on actual observations at all.

The greater problem is this is how most of science gets reported! From Darwinism to Global Warming to the Hole in the Ozone to the Fire Danger Levels of the Western U.S., possible scientific claims are made on shaky evidence and, if they fit the current paradigm, are given free reign to profligate until it appears that consensus has been formed. This is a disservice to the scientific method, and ultimately a disservice to everyone who relies upon science.