Light show. Auroral displays reveal the interaction between the solar wind particles and nitrogen molecules in Earth's upper atmosphere.


New Experiments Rattle Space Weather Research

Scientists trying to confirm a long-standing model of atmospheric physics have inadvertently shaken one of the foundations of the field. New tests show that the model, used to interpret energy emissions in Earth's upper atmosphere, is seriously flawed. The findings should help researchers build a better picture of how our planet interacts with solar radiation and the particle stream called the solar wind, and they may give a similar boost to studies of the atmosphere on Saturn's big moon Titan.

For decades, scientists have been studying what happens when Earth's magnetic field grabs high-energy particles from the sun and steers them toward the North and South poles. En route, some of them strike nitrogen molecules in the upper atmosphere, freeing electrons and generating both ultraviolet and visible light. Understanding this process is necessary to unravel what causes auroras (the Northern and Southern Lights), how much solar energy hits Earth's upper atmosphere at any given time, and how the solar wind creates electromagnetic interference with telecommunications.

Researchers can't measure directly how much solar energy is hitting the atmosphere. Instead, they calculate it by measuring how much light the excited nitrogen molecules give off—a bit like determining the horsepower of a car engine by measuring the gases spewing from its exhaust. (Earth's upper atmosphere is about 80% nitrogen.) For 25 years, their estimates have relied heavily on a set of data generated partly by electron-collision experiments and mostly by mathematical models. The data are based on particular wavelengths of ultraviolet light, called the Lyman-Birge-Hopfield (LBH) band, generated when freed electrons strike atmospheric nitrogen. The numbers are critical for building models of Earth's upper atmosphere. But there was a problem: The baseline LBH data in the model covered less than 20% of the observable LBH-band emissions.

So 2 years ago, a team at NASA's Jet Propulsion Laboratory (JPL) in Pasadena, California, set out to expand the data set. The researchers began by duplicating the original experiments. They fired a beam of electrons at nitrogen gas in a chamber—essentially creating a miniature aurora in the lab—and then analyzed the resulting LBH emissions. The results startled them. The team reported online yesterday in the Journal of Physics B: Atomic, Molecular and Optical Physics that the new data differ from the original by almost a factor of 2. Roughly speaking, it means that estimates of incoming solar energy based on the 1985 model have been way off.

Why the discrepancy? The researchers think it stems from the different methods they used. In performing the new tests, they fired the electron beam for equal lengths of time at each measured electron-energy level. By contrast, the scientists who made the original measurements counted a predetermined number of electrons but let the times vary. That approach makes it hard to isolate background noise and can distort the results, the JPL team says.

The team never set out to disprove the 1985 data, physicist and co-author Paul Johnson says: "We sort of stumbled on it." But now that the new results are in, he says, other researchers are going to have to reexamine their models of the upper atmosphere. The new data should also help scientists construct better models of the nitrogen-rich atmosphere of Saturn's moon Titan, in which NASA's Cassini spacecraft has detected LBH emissions.

It's a "very significant" paper, says physicist William McConkey of the University of Windsor in Canada. Not only does it correct the earlier data on LBH emissions, he says, but it also improves measurement of the energies present in the electrons striking atmospheric nitrogen. And both factors are important for gauging the effects of the solar wind on Earth, Titan, and other solar system bodies that contain an atmosphere of nitrogen.