A radio image produced by the SWAS team showing water vapor emission in the Orion molecular cloud. The SWAS measurements show water vapor at its maximum in the lightest area, lower right, which is close to the direction of the Orion nebula. The image shows a cloud some 15 light years across. Photo: Cornell University

After more than two years in space, NASA's Submillimeter Wave Astronomy Satellite (SWAS) has provided radio astronomers with one definite conclusion about the clouds of gas and dust that make up the bulk of the mass in our galaxy, the Milky Way. "There's not much to drink there, and it's hard to breathe," says Cornell University astronomer Paul Goldsmith, one of the 12 members of the satellite science team.

When SWAS was launched Dec. 5, 1998, its primary goal was to be a complete radio astronomy observatory confirming the conventional wisdom that the two most abundant molecular carriers of oxygen in interstellar clouds, after carbon monoxide, are water vapor (H2O) and molecular oxygen (O2). Since these two molecules are relatively abundant in Earth's atmosphere, their presence in distant space is hidden from Earthbound telescopes, and it is not known how abundant (or rare) these molecules are.

Now SWAS has dried up that wisdom: Water, it turns out, seems to be 10,000 times less abundant in these so-called molecular clouds than previously thought and, indeed, so scarce that it is found in the ratio of only one part in a hundred million compared with hydrogen molecules, which are the most common component. And as for molecular oxygen: It is at least 100 times less abundant than predicted and, says Goldsmith, perhaps a thousand times more scarce.

Because the interstellar clouds are considered to be nurseries for star formation, the SWAS science team is attempting to understand the chemical reactions that affect star birth. To date, about 100 different molecules have been found in such clouds. But two big holes in understanding the chemistry are how much water and oxygen are present in molecular form.

Also on the SWAS science team are Cornell astronomer Martin Harwit and principal investigator Gary Melnick of the Harvard-Smithsonian Center for Astrophysics, who is a Cornell graduate. SWAS was planned as a three-year mission. But the science team is hoping that the unexpected results recorded by the observatory will persuade NASA to extend the mission to four years or beyond. "We now have between 10 and 20 good interstellar cloud sources, and we think it is well worth having 40 or 50," says Goldsmith.

Paramount among the recorded sources is the giant Orion molecular cloud, 1,500 light years from Earth. The mass of this vast cloud -- some 15 light years across -- is far greater than that of the myriad stars that make up the constellation. In optical images, young stars can be seen lighting up the entire nebula, indicating a prime area to observe star birth.

Both water and molecular oxygen have been considered to be essential constituents of the molecular clouds from which stars form. How then to explain their dearth? It is now realized, says Goldsmith, that the molecular clouds are as cold as only 30 degrees above absolute zero (or a frigid minus 240 degrees Celsius). That means, he says, that the water might be frozen on the dust grains in the clouds and thus not detectable by radio astronomy. There have been some hints of this previously, but few hard numbers.

The almost complete absence of molecular oxygen, Goldsmith says, is harder to explain because the molecule doesn't freeze in the same way as water and, thus, would not be in frozen form on dust grains.

An artist's concept of SWAS in orbit. Photo: NASA

"Since oxygen is the third most abundant atom in the universe, after hydrogen and helium, we ought to be able to understand where the oxygen atoms are and in what form. And they certainly don't seem to be in molecular oxygen in the gas phase as predictions would have it," says Goldsmith. "It really is a dilemma for interstellar chemistry, but it might be a hint that something is wrong with our picture of the structure of these giant clouds."

SWAS does have a limit on its detection abilities, but other experiments are being planned that are more sensitive. However, says Goldsmith, there does seem to be a big gap between existing theories and SWAS observations so far.

The Cornell astronomer is confident that, ultimately, molecular oxygen will be detected in molecular clouds at some level. He admits to being disappointed at the SWAS results at first. But, he says, "in some sense it's much more interesting when your observations contradict prevalent wisdom."