Fun With Geomagnetism
Posted by The Diatribe Guy on January 6, 2009
One of the things I have been curious about is the impact of the earth’s magnetic shield on climate. According to some charts I have seen (Jose, Landscheidt) there appears to be some correlation with a lower intensity in the magnetic field around earth (geomagnetism) and the occurrence of ice ages in reaction to solar minimums. I admit to not having thought too much more about it than a couple of charts I’ve seen, so I’m not intending here to suggest that I understand that correlation, or even whether or not I fully understood those charts. As time allows, I’ll get back into that.
But that discussion is not really the point of this post. And I’ll state right away that there is no particular conclusion that I’m trying to draw here. But I messed around with some data, so why not post about it?
For some time, I had been intending to look more into the decay rate of the earth’s geomagnetic field, if for no other reason than I’ve heard about these potential reversals. While I heard a lot about the fact that the earth’s field has been decaying for at least 150 years now, and that the last reversal was about 750,000 years ago, I had no idea of what the actual numbers were. Is it a constant decay? Is it accelerating? What’s up with the whole magnetism thing, anyway?
At this point, I won’t go into the things I’ve read about the potential effects of geomagnetic reversals. Nor will I go into the historical record in the lava flows that track these reversals. Nor will I go into the elements in soil deposits that coincide with extinction events that coincide with such reversals. I’m just going to play with data.
In digging around for data, I came across this resource which allows you to pull annual averages from myriad geomagnetic observatories. It makes sense that there is no global average, since, the magnetism intensity is much stronger at the poles. So, in order to do a major study, I’d have to waste some serious time and need a lot of rationalization with my wife to embark on such a thing. So, once I realized that a really large, scientific, study was out the window for the moment, I decided to be far less statistically correct, and just look at what’s been measured in Boulder, Colorado. As we all know, Boulder is the summation of all goodness and light, so it seems a reasonable proxy for the whole world. (Yes, I’m joking.)
The Boulder data starts in 1965 and continues through to 2006. The 1965 average overall intensity was 56,057. The 2006 average intensity was 53,330.
As tempting as it may be to straight-line this, it’s really no fun at all to take such a simple approach. Plus, it isn’t accurate. In playing with the numbers, it was apparent that the drop from year to year has generally accelerated. Here’s a chart of the geomagnetic intensity in Boulder:
We all probably know and understand that 42 years worth of data on something like this shouldn’t be used to project the end of the world as we know it. But since Tamino can make those kinds of generalizations and get an audience, I figured I’d try to do it, too.
So, if I fit the following formula to the data, I actually get a remarkably good fit: 0.99963 x .99996 ^ time. 1965 equals time zero. The problem with this formula is that it doesn’t ever go to zero. I do, however, see the curve reaching 25,000 at t=191 , 10,000 at t=283, and 1,000 at t=439. However, the shape of the curve doesn’t lend itself to the idea that it cannot reach zero, since it isn’t asymptotic.
So, the more appealing method, in my opinion, was to take a look at the incremental changes in the intensity from time to time. Lo and behold, the first few years were all under 30, then it was between 30 and 60, then more, then more, until the last eight readings, which all decreased by 99 or more. No other previous years declined by that much. This seemed to indicate to me that there is an acceleration in the decline that needs to be better considered than a simple power function. I plotted the incremental decreases in intensity:
So, I fit different trend curves to this data, and the best fit – other than some whacky polynomial trend curves – was a power trend. But the end point stretch on the power curve was well below the actual values. The goodness of fit was coming from the front end of the data. The linear fit seemed to fit it best overall, and the r-squared was only slightly less than the power curve. It is almost a value of 0.9.
What this chart shows is how much the previous intensity has declined from the previous year. The slope is 2.0386. This says “add 2.0386 to the previous amount of decline, subtract that from the previous intensity, and produce the projected intensity.
This exercise tells me that the geomagnetism in Boulder, Colorado will hit zero at t=221.
Take it to the bank. And warn your great-great-great-grandchildren.
Now, what I don’t know is if there is some “trigger-point” where the field gets so weak, that it just says “screw it” and falt-out collapses. I also have no idea if this value needs to reach zero before a reversal could ever occur. Also, despite the math, I’m pretty sure that this doesn’t have to go all the way to zero before the intensity could start to strengthen. In otehr words, I don’t really know much about all this, except thos charts are neat, and we’re all going to die in 180 years. In fact, I will go so far as to predict that none of us alive today will survive the chaos of the next 180 years.
Seriously, though. I can’t help but think that this degradation of our magnetic field is our fault. First temperature, now this.