NASA's Hathaway issues new solar cycle prediction

Geoff Sharp writes in comments:

NASA releases their new estimate for SC24. The new number is 64 but not allowing for any 13 month smoothing is noted.

The text is quite comical, they have no idea. By Xmas I predict they will be inline with my prediction made in 2008.

Perhaps. This blink comparator that I made (see below), tells the story pretty well.

Back in October of 2007, the SC24 smoothed SSN prediction was for 150. Now it is 64. But, let us not be too critical of Dr. Hathaway, unlike some scientists we know, he has the integrity and courage to admit when his forecasts and models don’t work, and to revise them in the face of reality. Speaking from experience, Nature can be a bitch to forecast.

Here’s the current prediction below:

Solar Cycle Prediction

(Updated 2010/10/05)

Current prediction for the next sunspot cycle maximum gives a smoothed sunspot number maximum of about 64 in July of 2013.

Predicting the behavior of a sunspot cycle is fairly reliable once the cycle is well underway (about 3 years after the minimum in sunspot number occurs [see Hathaway, Wilson, and Reichmann Solar Physics; 151, 177 (1994)]). Prior to that time the predictions are less reliable but nonetheless equally as important. Planning for satellite orbits and space missions often require knowledge of solar activity levels years in advance.

A number of techniques are used to predict the amplitude of a cycle during the time near and before sunspot minimum. Relationships have been found between the size of the next cycle maximum and the length of the previous cycle, the level of activity at sunspot minimum, and the size of the previous cycle. Among the most reliable techniques are those that use the measurements of changes in the Earth’s magnetic field at, and before, sunspot minimum. These changes in the Earth’s magnetic field are known to be caused by solar storms but the precise connections between them and future solar activity levels is still uncertain.

Of these “geomagnetic precursor” techniques three stand out. The earliest is from Ohl and Ohl [Solar-Terrestrial Predictions Proceedings, Vol. II. 258 (1979)] They found that the value of the geomagnetic aa index at its minimum was related to the sunspot number during the ensuing maximum. The primary disadvantage of this technique is that the minimum in the geomagnetic aa index often occurs slightly after sunspot minimum so the prediction isn’t available until the sunspot cycle has started.

An alternative method is due to a process suggested by Joan Feynman. She separates the geomagnetic aa index into two components: one in phase with and proportional to the sunspot number, the other component is then the remaining signal. This remaining signal has, in the past, given good estimates of the sunspot numbers several years in advance. The maximum in this signal occurs near sunspot minimum and is proportional to the sunspot number during the following maximum. This method does allow for a prediction of the next sunspot maximum at the time of sunspot minimum.

A third method is due to Richard Thompson [Solar Physics 148, 383 (1993)]. He found a relationship between the number of days during a sunspot cycle in which the geomagnetic field was “disturbed” and the amplitude of the next sunspot maximum. His method has the advantage of giving a prediction for the size of the next sunspot maximum well before sunspot minimum.

We have suggested using the average of the predictions given by the Feynman-based method and by Thompson’s method. [See Hathaway, Wilson, and Reichmann J. Geophys. Res. 104, 22,375 (1999)] However, both of these methods were impacted by the “Halloween Events” of October/November 2003 which were not reflected in the sunspot numbers. Both methods give larger than average amplitude to Cycle 24 while its delayed start and low minimum strongly suggest a much smaller cycle. The smoothed aa index reached its minimum (a record low) of 8.4 in September of 2009. Using Ohl’s method now indicates a maximum sunspot number of 70 ± 18 for cycle 24. We then use the shape of the sunspot cycle as described by Hathaway, Wilson, and Reichmann [Solar Physics 151, 177 (1994)] and determine a starting time for the cycle by fitting the data to produce a prediction of the monthly sunspot numbers through the next cycle. We find a starting time of August 2008 with minimum occurring in November or December 2008 and maximum of about 66 in June of 2013. The predicted numbers are available in a text file, as a GIF image, and as a pdf-file. As the cycle progresses, the prediction process switches over to giving more weight to the fitting of the monthly values to the cycle shape function. At this phase of cycle 24 we now give 22% weight to the curve-fitting technique of Hathaway, Wilson, and Reichmann Solar Physics 151, 177 (1994). That technique currently gives highly uncertain (but smaller) values.

Note: These predictions are for “smoothed” International Sunspot Numbers. The smoothing is usually over time periods of about a year or more so both the daily and the monthly values for the International Sunspot Number should fluctuate about our predicted numbers. The dotted lines on the prediction plots indicate the expected range of the monthly sunspot numbers. Also note that the “Boulder” numbers reported daily at www.spaceweather.com are typically about 35% higher than the International sunspot number.

Another indicator of the level of solar activity is the flux of radio emission from the Sun at a wavelength of 10.7 cm (2.8 GHz frequency). This flux has been measured daily since 1947. It is an important indicator of solar activity because it tends to follow the changes in the solar ultraviolet that influence the Earth’s upper atmosphere and ionosphere. Many models of the upper atmosphere use the 10.7 cm flux (F10.7) as input to determine atmospheric densities and satellite drag. F10.7 has been shown to follow the sunspot number quite closely and similar prediction techniques can be used. Our predictions for F10.7 are available in a text file, as a GIF image, and as a pdf-file. Current values for F10.7 can be found at: http://www.spaceweather.ca/sx-4-eng.php.

Here’s my blink comparator: