The Milky Way and the Clouds of Earth

By Sallie Baliunas, Ph.D., and Willie Soon, Ph.D.
Harvard-Smithsonian Center for Astrophysics

The search will continue. Not until the empirical resources are exhausted, need we pass on to the dreamy realms of speculation.

—Edwin Hubble, 1936,
Realm of the Nebulae

Using the largest telescope in the world, the mighty 100-inch reflector at Mount Wilson Observatory, Edwin Hubble forced us to take the last step in the Copernican revolution. The earth, displaced by Copernicus as the center of the universe, now sits in a galaxy with hundreds of millions of stars, in a universe with hundreds of millions of galaxies, which are expanding outward from a cosmic beginning some 15 billion years ago. Hubble’s empirical work with the faint and distant nebulae led to one of the most radical discoveries in science—that there was a physical beginning of time, rather than an eternally existing universe.

One lesson from Hubble’s work is that new knowledge is sometimes best gotten at with an empirical start. For example, we study the idea of a sun–climate link because there are so many observed examples of weather and atmospheric parameters varying with (or against) the 11-year sunspot cycle. Hot research considers the mechanisms by which the sun’s output, in the form of either light or particles, varying with the 11-year sunspot cycle, could explain the changes in climate properties. New results are strengthening this idea, and gradually the sun’s influence, as it is understood, is put into general circulation models (GCMs) of climate change.

If you thought changes in the sun were "far out" in terms of a climate influence, then we know what you’ll say when we bring up the possibility of a galactic influence: "The galaxy? Are they nuts?"

But just hear us out. The Milky Way Galaxy is home to our solar system and is about 100,000 light years across, with many of the stars concentrated in a thin disk about 2,000 light years thick. The disk is structured, too: Spiral arms, carrying most of the gas and dust from which the newest stars are made, emanate from the central bulge. The sun is located in a spiral arm about two-thirds of the way from the center. The arms are turning around the center, like a pinwheel; it takes about 250 million years to complete a full turn. On a clear night from a dark site at the right time of year, you can see the breathtaking sweep of the faint bulge of old stars that is the center of the galaxy— toward the constellation Sagittarius.

So, what does this have to do with the earth’s climate?

In the crowded disk of the galaxy, stars of a range of sizes are formed. Sometimes stars more massive than the sun die not quietly but catastrophically. The supernova explosion creates a long-lasting (tens of millions of years) background of energetic and highly charged particles that are flung outward. The galactic disk contains the skeletal remains of many such events, including a bubble of supernova debris in which the solar system currently sits.

The wake of the supernova has cosmic rays—consisting mainly of fast-moving protons and helium nuclei—that pelt the solar system for millions of years.

The effects of the cosmic-ray shower have been measured near the surface of the earth since the 1930s. Cosmic rays hit molecules at the top of the earth’s atmosphere and make, among other things, subatomic particles like neutrons and antimatter particles like positrons (i.e., antimatter electrons).

The neutron records show the clear influence of the sun’s 11-year cycle (Figure 1). Why? The sunspot cycle is an 11-year change in the strength and coverage of magnetism on the sun. But the magnetic fields also flow outward from the sun’s surface in a wind, so the wind also shows the 11-year cycle. When the wind and its magnetic fields are strong, the cosmic rays from space are deflected and fewer cosmic rays reach the earth. At those times, fewer neutrons will be made in the atmosphere and recorded.

Figure 1. The 11-year cycle of the sun's surface magnetism, and the neutron counts, 1953-1997.

The opposite is also true: Times of weaker solar magnetism mean more neutrons recorded as a result of an increased flow, or flux, of cosmic rays. Because the sun also has a 22-year cycle of magnetism, the neutron record shows that, too. There are also longer periods in the sun’s magnetic record, but the neutron record is too short to study them.

Now comes the galactic connection to the earth’s climate. The cosmic rays are made of charged particles, some of which travel well into the atmosphere. Recently Turco and colleagues have suggested that the galaxy’s cosmic rays make sulfate aerosols in the troposphere that can act as cloud seeds.

Several satellites have been monitoring cloud cover over the earth. Svensmark and Friis-Christensen saw a remarkable relation between changes in cloud cover over the ocean and the neutron flux. (The relationship is better-correlated outside the tropics, presumably because the tropical zone is more shielded from cosmic rays by the earth’s magnetic field.) The cloud cover at mid-latitudes changes about 4 percent from peak to trough in step with the neutron flux—that is, opposite to the solar magnetic cycle. More recent data confirm and extend the relation between cloud cover and the galaxy’s cosmic rays, as modulated by the sunspot cycle (Figure 2).

Figure 2. Percent change in cloud cover (measured from two different satellites) and the percent change in neutron flux. Changes in cloud cover appear to follow closely the changes in the flux of cosmic rays, which varies as the sun's magnetism changes.

The Kuang research team goes a step farther. They find that the properties of clouds are changing along with the cloud coverage and cosmic ray flux during the sunspot cycle, suggesting a physical connection between them. At times of high cosmic ray flux (sunspot or solar magnetic minimum), more thin clouds and fewer thick clouds form. When the cosmic ray flux decreases, the opposite happens: Fewer thin clouds and more thick clouds form.

One cloudy aspect of the analysis is, as Kuang notes, that the satellite records of clouds are still a little short. For example, the sunspot minimum of 1986 occurred near the time of a weak El Niņo. Is El Niņo, and not cosmic ray change, driving the correlation? Longer records may resolve the confusion in finding the cause of the correlation between cloudiness and cosmic ray flux.

Is there another way to look for the cosmic ray influence on clouds? Perhaps we need only look toward the edge of the solar system, at the cloud-covered planet Neptune. Since 1972, Lockwood and Thompson have been measuring changes in the brightness of light at visible wavelengths reflected from Neptune. Because it is so far from the sun, Neptune is a very cold planet. Unlike earth’s clouds, which are made of water, Neptune’s clouds consist mostly of frozen methane, so the planet is quite shiny and reflects much of the feeble sunshine reaching it. But Neptune’s clouds do absorb a bit of sunlight. And like the other planets, Neptune sits in the shower of cosmic rays, some of which its clouds absorb.

Neptune’s reflected sunlight changes on several time scales. Surprisingly, one strong variation, 5 percent in blue and yellow light, has a period of 11 years. Neptune is brightest at sunspot minimum, which is at cosmic ray maximum. Thus, Neptune’s cloud properties are linked to the cosmic ray flux, similar to the situation for terrestrial clouds.

Back on earth, satellite studies are helping to pin down one of the mechanisms for changes to clouds—cosmic rays from deep space in our galaxy. As this cosmic mystery unfolds, our climate models will continue to expand.

References:

Kuang, Z., et al., 1998, Cloud optical thickness variations during 1983–1991: solar cycle or ENSO? Geophysical Research Letters. In press.

Lockwood, G.W., and D.T. Thompson, 1991, Solar cycle relationship clouded by Neptune’s sustained brightness maximum. Nature, 349, 593–594.

Menzel, W.P., et al., 1997, Seven years of global cirrus cloud statistics using HIRS. IRS 96: Current Problems in Atmospheric Radiation, W. L. Smith and K. Stamnes, Eds., pp. 719–725.

Svensmark, H., and E. Friis-Christensen, 1997, Variation of cosmic ray flux and global cloud coverage—a missing link in solar-climate relationships. Journal of Atmospheric and Solar-Terrestrial Physics, 59, 1225–1232.