This data comes from the GOES 12 weather satellite. Each GOES satellite carries a number of Space Weather instruments. Their data is available from NOAA’S SEC in Boulder, Colorado.

The solar spectral irradiance constantly changes at all wavelengths and over different periods of time. All wavelengths change over the 11 years of a sunspot cycle. Flares cause the most rapid changes, increasing the X-ray irradiance in a few seconds. Fluctuations in TSI are associated with the five-minute oscillations of the Sun, although the effects of a flare have been measured by the SORCE satellite. Watching many wavelengths over many years is how we can understand how these variations are created. Watching from space is necesary to see all of the wavelengths (some are absorbed by our atmosphere).

The absorption of solar irradiance by the Earth is linked to many important processes. However, this absorption makes it difficult to accurately detect the total incident energy; an instrument on the ground can only detect the total energy minus the energy that has been absorbed!

Many Space Weather effects are caused by irradiance variations in the ultraviolet part of the spectrum. A solar flare can increase the X-ray irradiance 10 to 100 times for several minutes. The effects at Earth are dependent on wavelength and timescale. Here is an example of how a large flare changes the solar irradiance.

This data comes from the SEE instrument on the TIMED satellite. The data is available from the SEE website at LASP in Boulder, Colorado.

During October and November 2003 the Sun had many spectacular flares. A lot of them came from active regions 484 and 486. Here is a record of the X-ray irradiance just before and during the largest of those flares at 1929 UT on November 4, 2003. This may have been the brightest flare ever recorded! Active region 486 then rotates out of view only to return two weeks after for more fireworks.

The X-ray irradiance increased almost 10,000 times during the flare! The only flare thought to be larger was seen in 1853 and caused problems with the telegraph lines in the United States and Europe. A direct comparison is difficult as we no longer have an extensive telegraph network. Heck, we don't even have telegrams any more!

We also see changes in the solar irradiance over longer periods of time. The SEE instrument was launched on TIMED and has returned spectral irradiance data since early 2002.

During 2003 and 2005 the Sun was in the declining phase of Solar Cycle 23. The SEE instrument measured the spectral irradiance over the period, which included the Halloween Superstorm. Here we show the spectral irradiance at 5 different wavelengths during the two year period. The bottom two curves are far more variable (move up and down more) than the top three curves. Because these data are averaged over a day the flares seen above don't appear. Although the X-ray and EUV irradiances change by large amounts over a solar cycle the TSI varies by only a small amount.

Total Solar Irradiance and the Solar Cycle - Data from this figure is from composite_d41_61_0604.dat, provided by the PMOD/WRC, Davos, Switzerland, using unpublished data from the VIRGO Experiment on the cooperative ESA/NASA Mission SoHO mission. The data is described in C. Fröohlich and J. Lean, 1998, The Suns Total Irradiance: Cycles, Trends and Related Climate Change Uncertainties since 1978, Geophys. Res. Let., 25, 4377-4380, 1998.

Solar cycle variations in TSI can be clearly seen in the upper plot showing a composite of more than 25 years of satellite measurements. The red points are daily values while the green line is the annual average. Note how the changes over a day easily exceed the typical amount of variation over the solar cycle. The daily points vary by 0.3% while the annual points vary by only 0.07% (shown by the vertical lines at the left.) The annual sunspot is shown in the lower plot. Comparing the plots shows that TSI increases when solar activity increases.

Accurate measurements of the solar irradiance must be made from satellites. These have been available for the past few solar cycles. Before that, we have only the sunspot number observations to use as a proxy of solar irradiance. We have records of sunspots since Galileo started sketching the surface of the Sun in the early 1600's. Some observers did better than others, but the 400 year record of sunspots is a rare long-term dataset in science. These observations show that the sunspot cycle has persisted over the past few hundred years and that more dramatic variations were seen in the past. We even find a period when sunspots were few and the Earth was cold—the Maunder Minimum (1645-1710). This is not due to lack of looking. Scientists tried to see sunspots during the Maunder Minimum and did not. Noticing that TSI is higher at solar maximum, perhaps the lack of spots tells us what caused the cooler temperatures.

Sunspot Number Over the Centuries - The sunspot number has been measured since 1610. Here we show the annual values of the sunspot number (using the yearly.plt file from the NGDC. The blue is the 25-year running average to show some of the trends in the data.

The period of low sunspot number in the mid to late 1600's is called the Maunder Minimum. We can extend the record further back in time by measuring isotopes of carbon and beryllium that are produced by cosmic rays. During solar maximum it is more difficult for cosmic rays to reach the Earth than at solar minimum, causing a solar cycle to appear in the isotope data.

Sunspot Number Over the Centuries

These cosmogenic isotopes, which are found in tree rings and ice cores, are solar activity proxies that suggest long-term fluctuations in solar activity may exceed the range of contemporary cycles. In this figure minimum and maximum refer to surface temperature. The Maunder Minimum is labelled, along with several other temperature extremes. There is a tendency for cooler temperatures during times when the Sun is less active.

The changes in isotopes are tracked fairly well by the temperature measurements from other records?

Looking further back in time, we find that the Sun has increased in brightness about 25% since its birth 4.5 billion years ago. How did the Earth's atmosphere and the changing solar brightness interact to create a livable climate over the past billion years?

Some important points:

• At every wavelength the solar spectral irradiance changes over the 11-year sunspot cycle.
• The sunspot cycle also varies in time, and there is evidence of significantly different behavior in the past
• All types of energy input to Earth exhibit greater changes on shorter timescales (flares, CMEs)