Sunday, January 27, 2013

Prediction is the Essence of Science

Almost 20 years ago, I worked with a guy who was a very senior engineer and one of our go-to guys for the most difficult problems.  The kind of guy who could use a four-function calculator to solve anything - or get you close enough.  When we would work in a lab together it was always a crisis and a problem that had to be solved yesterday.  He'd have me create an experiment to try and determine which subsystem was the problem, and when the result was nowhere near anything he expected, he would use the phrase I've taken for the title of this post.  It took me a few repetitions to realize it was said in total frustration and irony. 

My version has come to be "there's nothing like a clear, useful result, and that was nothing like a clear, useful result".

NASA's office that predicts solar activity must feel like that sometimes.  The current cycle, 24 - the 24th since astronomers started keeping track of them - has stymied experts all along.  The period of solar inactivity between the previous cycle and this one was the longest and lowest minimum in a century (1911-1913) and the start date of cycle 24 was pushed out, while the predicted peak was lowered, every time a prediction was made (sample 2007 prediction here).  The solar wind was the weakest it had ever been measured in the 50s years of satellites that can measure it.  The current predictions are for a peak of this cycle to be in the coming fall, and to be the weakest peak in a century.  So far, this cycle is not living up to even that. Look where the blue smoothed sunspot number curve is next to the red prediction curve. 
NOAA predictions
The prolonged minimum between cycle 23 and 24 was the second longest since the Dalton minimum of the early 1800s.  The thing is, it may not be ending there.  In the '90s, astronomers Dr.s William Livingston and Matthew Penn of the Kitt Peak observatory in Arizona noted that their readings on the intensity of sunspots were trending downward.  They checked data many times and plotted trend lines.  Once they saw the lines, they published a paper showing that by 2015 there may be no sunspots left at all.  The paper was not well received, but about a decade later, they gathered another ten years' worth of data, reanalyzed everything and concluded they were right the first time.  (summary pdf here)  It's important to say they may well be wrong, but completely independent predictions from NASA and others are saying the next cycle, 25, is going to be even weaker than this one, and may approach quiet sun levels at its peak.  Which is to say the cycle simply may not happen.  Should we believe the predictions for cycle 25 when they've been so wrong predicting this one?

What does this mean to us?  Several things.  First the good news: those dire predictions you read about a killer solar flare taking out everything are much less likely than before - and it wasn't very likely to start with.  The other side is much less positive.  Prolonged solar minima have happened before, and they are associated with mini-ice ages.  I use the tentative language because humans simply haven't been able to measure sunspots for most of history.  The growing seasons become shorter and weather changes to become less friendly to crops.  Widespread food shortages are a real possibility.  There's a phrase that comes up that you really don't want to see.  A Real ice age. 

Of course, it's politically incorrect today to assert that the Sun, the ultimate source of every erg of energy on this planet, could effect climate.  That's just talking crazy.

No comments: