We are currently in the middle of the 11 year solar cycle, with the Earth experiencing the effects through changes to radio communications, power distribution, orbiting spacecraft and even the weather. So predicting what the Sun will do next is in all our interests.

by Leslie Mullen

Most people think of the sun as a featureless, unchanging ball of light. But the Sun actually has seasons, or cycles of activity and relative inactivity. Right now, we are in the middle of the maximum activity phase of the current 11 year solar cycle. The Sun is daily exhibiting many sunspots and flares. We feel the effects of an active Sun here on Earth - radio communications, power distribution, orbiting spacecraft and even the weather are all affected.

Sunspots are relatively cool areas on the Sun that appear as dark blotches. Scientists count the number of sunspots to measure the magnitude of a solar cycle, and to determine how long the cycle lasts. If scientists were able to predict sunspot activity, not only would we know ahead of time what the Sun will do, but we might gain a better understanding of how the Sun operates.

Dr. David Hathaway, along with Robert Wilson and Ed Reichmann, looked at many different ways scientists predict sunspot activity. They tested each statistical method to see which worked best, and then combined the top two methods to develop an even better prediction method of their own.

"There are many different ways of predicting the sunspot cycle," says Hathaway. "but until now there has never been a systemic study to determine whether one method works better than another. After examining various methods, we found that some of the techniques currently used and touted are basically useless."

By looking at more than 15 methods, the scientists found that 8 or 9 were better than average at predicting solar maxima - when the sun is at its most active. The two best methods essentially used the same information - disturbances in the Earth's magnetic field.

"Explosions from the sun travel through space and hit the Earth, causing the magnetic field to wobble and shake," says Hathaway.

Joan Feynman from NASA's Jet Propulsion Laboratory developed one of the top two methods, the Australian astronomer Richard Thompson developed the other. Although each scientist took a different approach to the data and reported different results, they both looked at how the Earth's magnetic field shook during the previous solar cycle to predict the sizeof the next one.

Solar Activity affects the Earth's Magnetic Field.

Scientists don't know why previous solar activity is connected to the next active period, or why the Earth's reaction to that activity helps in solar cycle prediction. But the connection allows scientists to estimate what the next solar season will bring.

The statistical model developed by Hathaway's team uses both Feynman's and Thompson's methods and integrates them with a curve-fitting technique. The "precursor" methods used by Feynman and Thompson try to determine the total number of sunspots that will appear before the season actually begins.

The curve-fitting method finds the best curve to fit recent solar activity. Based on years of observations, solar scientists have developed a library of curves that follow the average of solar cycles. By using their prediction method, Hathaway's team can pick a curve from this library before the solar cycle even begins, and than make adjustments as the cycle progresses.

For the current solar cycle, Hathaway's team predicted an average sunspot maximum of 154, with an uncertainty of plus or minus 20. This prediction has a narrower range of error than a previous, widely accepted prediction, which placed the sunspot maximum at 160 with an uncertainty of 30.

Graph of predicted sunspot activity

"We are in a period where the sun is very active," says Hathaway. "Until mid 2001, we'll see daily sunspot numbers between 100 and 300, with an average around 154."

After that, solar activity will begin a slow descent into solar minimum. So far, the Sun seems to be following the curve picked by the scientists.

"Monthly values are actually jumping all over the place," says Hathaway. "You have to remember the curve is only an average of what is really going on."

On one day for example, the Sun had a sunspot number above 300 - far more than the 154 average. But for the previous 5 months, there were fewer sunspots than expected. The average number of sunspots met in the middle to follow the curve picked by Hathaway's team.

As good as this method is,"physical models to predict sunspot activity several years in advance are not available," says Hathaway. "We don't understand well enough why the sun does this to be able to predict like a meteorologist does."

A meteorologist can input weather factors like temperature and barometric pressure into a computer model to get a weekly forecast. Solar predictors don't have a physical model, however, because they still don't know how all the factors of the Sun's activity work together.

The Suns Four Regions

A simple model of the Sun shows the solar interior separated into four regions. Energy is produced in the core, and this energy radiates outward through the radiative zone in the form of gamma-rays and x-rays. In the convective zone, fluid flows in a boiling motion. These fluid motions are visible as granules and supergranules on the surface of the Sun. A thin layer where the Sun's magnetic field is thought to be generated lies between the convective and radiative zones.

Hathaway, along with most other solar astronomers, believes the Sun's magnetic field is the key to understanding the solar cycle. Sunspots are formed when magnetic field lines just below the Sun's surface become twisted and poke through the solar photosphere. The photosphere - or "ball of light" - is the familiar, visible surface of the Sun.

The Sun is actually a ball of gas, so it does not rotate rigidly like solid planets and moons do. Instead, the Sun's equatorial regions rotate faster than the polar regions. Because of this "jet stream" near the equator, the magnetic fields become wrapped around the Sun.

"The magnetic field is a lot like a rubber band," says Hathaway. "Fluid flows within the Sun called 'dynamos' stretch, twist and fold the band, wrapping it around the sun many times over 11 years. When the magnetic field loops into the Sun's convective zone, it rapidly rises to the surface. As it rises, it twists a little bit. This provides a change in field direction that helps to reverse the poles."

The Sun's magnetic poles reverse at solar maxima. Starting at the equator, a slow flow at the surface drags the magnetic field toward the poles. Conversely, sunspots first appear in the mid-latitudes and then congregate toward the equator later in the solar cycle.

The extra ultraviolet (UV) and X-ray radiation created by the magnetic field around sunspots causes the Earth's atmosphere to heat up and expand. This creates added drag in the area where satellites and the Space Shuttle orbit. This drag could slowly pull such spacecraft out of orbit earlier than expected.

The extra UV produced by sun spot activity also increases the amount of ozone in the Earth's upper atmosphere.

Although sunspots are cooler areas on the solar surface, the Sun is actually hotter when sunspots appear and cooler when they are absent. Scientists believe that a long period of solar inactivity may correspond with colder temperatures on Earth. From 1645 to 1715, astronomers observed very little solar activity. This time period coincides with an era known as the Little Ice Age, when rivers and lakes throughout Europe (and perhaps the world) froze.

Although there are good records of solar activity since the invention of the telescope in 1610, scientists have to look toward other sources to determine if there were even earlier periods of low solar activity. Because it is believed that sun spot activity correlates to the amount of Carbon 14 and Beryllium 10 in the environment, scientists can use ice core samples on Earth to determine solar activity levels.

"We can go back in time, before telescopes, by looking at ice core samples," says Hathaway. "Based on these samples, there appears to have been other, earlier sunspot minima."

In 1843, the amateur astronomer Heinrich Schwabe found that sunspots come and go in a predictable 11-year cycle. Ever since that announcement, many have tried to correlate the Sun's cycle with all sorts of events on Earth - some have even believed the Sun influences the stock market! Although there is no evidence that solar activity affects economic trends, by predicting what the Sun will do in the future we can better prepare for the many other impacts solar activity has for life on Earth.