Charlotte Christensen. Photo: Engage
Web site.
Charlotte Christensen is on a universe-wide hunt for missing molecular hydrogen, and she's using computer simulations in her search.
Christensen, a graduate student working on her Ph.D. in astronomy at the University of Washington, gave a talk Wednesday evening titled, "Galactic Gas: Simulating Stellar Birthplaces." The talk was part of the Engage Science Speaker Series being presented by UW grad students working on improving their communication about science.
Astronomers are in agreement that stars form when molecular hydrogen gas condenses and becomes dense enough to fire up a nuclear reaction. Christensen explained that "the problem with molecular hydrogen is that it is very difficult to see." In fact, what astronomers look for instead is carbon monoxide, which forms in the densest clouds of molecular hydrogen. In most galaxies it works.
The catch, Christensen said, is that in dwarf galaxies and non-spiral galaxies there are stars, but no observed molecular hydrogen or carbon monoxide. She said that one possible explanation for this is that stars form differently in such galaxies. It's plausible, she said, but so far nobody has come up with a good theory for how that would work. Her thesis is different: The molecular hydrogen is there, but we can't see it, and there's no carbon monoxide.
"In small galaxies, there aren't that many supernovae because there aren't that many stars," Christensen explained. "Without supernovae, we don't produce much carbon and oxygen, which means they can create very little carbon monoxide. Furthermore, the gas in these galaxies does not appear to reach as high of densities as it does in the Milky Way. You put that all together and there might be H2 there, but we wouldn't be able to see it."
Christensen is using computer simulations to test her theory.
"The nice thing about simulations in astronomy is that we can actually use them as a laboratory," she said. "Observational astronomers can't poke a star and see what happens. In our simulations we can tweak things and see what the effects are. And if we can manage to use our simulations to build a realistic galaxy, this is very good evidence that all that physics we put into building it is right."
These simulations are incredibly complicated and take immense computing power to pull off. Each particle in a simulated universe is given characteristics which can change over time depending on its interactions with other objects. Christensen showed a video of a simulation created at the UW that crams 14 billion years of galactic evolution into two minutes and 21 seconds.
"We can take it one step further and make mock observations of these galaxies," she said of the simulations. "Looking at the stars, I can figure out how much light each of these bunches of stars would emit, and from that figure out what light would reach your telescope."
Christensen still hasn't found the missing gas.
"I have a lot more dwarf galaxies to simulate," she said. "I will be matching the star formation and where it's located to where the molecular hydrogen is, and comparing these simulations to observations to see if I'm on the right track, and if the physics I'm putting in creates something that's realistic."
Comments