Topband: Re: US Navy and Solar Cycle 23

Cary Oler oler@mx05.erols.com
Mon, 26 Apr 1999 11:30:50 -0600 (MDT)


KH7M wrote:

> How can these Navy guys be so smart??

A few points...

First, make sure you interpret the chart correctly. I don't have a copy of it
in front of me, but I believe the Navy chart used smoothed values, didn't they?
If so, you must remember that smoothed values lag real-time values by many
months by virtue of the smoothing method used. So the smoothed value you see
listed in the Navy chart corresponds to predicted activity about 6 months into
the FUTURE. The latest estimated smoothed sunspot number is valid for July of
1998 and is computed to be 90.3, which is probably well above the predicted
values given by the Navy chart.

Second, since I believe we are dealing with smoothed sunspot values, you can't
compare current observed sunspot numbers with the smoothed values, since the
current observed values are daily observations which are then used to compute
the smoothed values. The variance of daily values is simply too large to
compare with smoothed values.

I've mentioned this before to a few others and I think I'll state it again for
clarity sake (since apparently this message may end up on a reflector). The
predictions issued by EVERY forecast center in the world are based on empirical
methods that _do not_ take into consideration any serious quantifiable physical
data that is known to be directly linked to the behavioural characteristics of
the solar cycle. A lot of people THINK they understand the basics of how the
solar cycle is driven (there are numerous theories), but no one really knows
for certain what the driving parameters are yet. The process is a slow and
tedious one and I suspect it will be many years before anyone can assemble a
true physical model that can depict future levels of solar cycle activity to
any significant degree of accuracy. It is amazing that the empirical methods
currently in use yield results that are as close as they are, particularly
given the fact that they are almost all purely statistical in nature. The model
which is currently considered as one of the more reliable ones is known as the
geomagnetic precursor method. It, at least, is more closely linked to physical
processes on the Sun than the older methods (still in-use) such as the famed
McNish-Lincoln model (which is responsible for producing the primary extended
forecast issued by the Space Environment Center).

If you take a look at the latest McNish-Lincoln modelled predictions issued by
SEC (http://www.sec.noaa.gov/weekly/prf1231.pdf, last page), you'll see that
the model is still predicting a peak smoothed sunspot number near 160 in the
early part of the year 2000. I personally am beginning to find that as a less
likely scenario (I feel the maximum will be delayed until later in the year
2000 or perhaps even in 2001). One of the quirks of the McNish-Lincoln method
is that it becomes progressively more accurate the closer you get to the true
cycle maximum. This is a side-effect of the modelling method employed.
Nevertheless, some weight must be given to the output of the model since we are
getting closer to the cycle maximum. My own 'gut feeling' is that the cycle
will peak at a level below the level predicted by the McNish-Lincoln model, but
above the lower decile value imposed by the model (see the "lower threshold"
plot line in that prf1231.pdf report).

Periods of lulls in solar activity are common. They occurred in most of the
rising phases of previous solar cycles, to one degree or another.  They often
coincide with periods of large-scale restructuring of the solar activity
centers (known as the "active longitudes"). The last couple of solar rotations
have shown weaker active longitudes while other areas of the Sun have shown
increased strength. I believe it is possible we might be seeing some
restructuring going on at the present time. I fully expect to see activity
begin swinging upward again (perhaps more significantly than many might be
anticipating) within the next several months.

Comparing previous solar cycles with the current, some interesting information
comes to light. If you look at the yearly mean sunspot numbers and how they
behaved within one year of the observed initial peak near the cycle maximum
during the last six solar cycles, you'll find that the difference between the
observed yearly mean value one year before this peak and the actual initial
peak near the cycle maximum varies by an average of about 50. It should be
noted that the difference values varied between 23 and 73, so the variance is
fairly high. Nevertheless, four of the six cycles I examined had differences
very close to the 50 mark. The two other cycles were extremes, one having a
difference of 23 and the other of 73. So the average value of 50 might be
considered as being fairly representative. In any event, if you use this 
difference value of 50 to extrapolate what we might observe during the current
cycle, together with the current observed 365-day mean value of 98 (not the
smoothed value, the mean of the last 12 months), you end up with a peak 365-day
mean sunspot number of near 150, give or take 10.

It is also interesting to note that during almost every one of the last six
solar cycles, the sunspot number progression toward the maximum has stalled for
a period of a few months before resuming it's course toward a cycle maximum.
I believe we are observing one of these 'stalled' periods. The interesting
feature about this solar cycle is that the stalled period is longer and has
occurred at a lower sunspot number value than has been observed in most of
the previous solar cycles. But this should not come as a surprise to anyone,
as the Sun is such an unpredictable source of activity. Each cycle surprises
us in more than one way.

Let me remind everyone that only two and a half months ago we observed the
highest solar flux level so far this solar cycle (in mid February). That isn't
very long, but when you're looking at the values daily and hoping for
increases, that period of time can SEEM much longer and might lead to doubts
about the performance of the cycle.

As I've said before, I think it's still far too early to render any firm
verdicts. The best we can do is make some educated 'gut feeling' guesses. And
really, that's all they are: guesses.

Everyone seems to have an opinion when it comes to the solar cycle. Most
professional solar physicists with an interest in cycle predictions have
opinions fairly closely in-line with the consensus report published by SEC
where the predicted maximum should be near 160 (+/- 30). Others like to go out
on limbs, predicting abnormally high or abnormally low values. Some have even
suggested that a new "Maunder Minimum" might begin soon (a period of time
between roughly 1645 and 1715 associated with what has become known as the
"Little Ice Age"). Don't hold your breath on that one... you'll almost
certainly pass out.  Others are predicting the end of the world at this solar
cycle maximum, sporting some very colorful, unproven, unfounded (and in my
opinion irresponsible) theories concerning interactions between the Earth and
abnormally large coronal mass ejections. This one is undoubtably an artifact
of some poor guy's inability to control his lucid dreaming.

As anyone familiar with statistics knows, you can recognize patterns in almost
anything if you look at the data long enough. Unfortunately, statistical
analyses is all we really have to work with right now, at least until we gain a
better understanding of the internal workings of the Sun -- which is something
we really hope the GONG research group will continue to make progress in.

I hope this helps place current activity into a little better perspective.

        -Cary Oler
         Oler@Holly.CC.Uleth.CA, Oler@Uleth.CA,
	 Oler@Solar.Uleth.CA, or
         COler@Solar.Stanford.Edu


--
FAQ on WWW:               http://www.contesting.com/topband.html
Submissions:              topband@contesting.com
Administrative requests:  topband-REQUEST@contesting.com
Problems:                 owner-topband@contesting.com