Image Credit: C. Smith, S. Points, the MCELS Team and NOAO/AURA/NSF
The picture above is of the Large Magellanic Cloud, or LMC for short. The LMC is one of the closest galaxies to our home galaxy, the Milky Way. (Read here if you need to remind yourself what a galaxy is).
One of the troubles with trying to understand the Milky Way is that we are in it (can't see the forest for the trees), and large portions of the Milky Way are blocked from our view by clouds of dust and gas. We can see the entire LMC, on the other hand, and it is close enough that we can see fairly typical individual stars! So, with the LMC, we can study both the forest and the trees. There is another, smaller galaxy called the Small Magellanic Cloud (SMC) that is roughly the same distance away, in a slightly different direction, which is likewise useful for studying an entire galaxy along with the component stars.
Supernovae are the explosive deaths of some stars. There are two main types of supernovae, Type Ia supernovae, which are thought to be the explosions of certain white dwarfs, and "core-collapse supernova", which are the explosions of massive stars at the end of their lives. Our theories on the lives of stars predict how often supernovae of each type should occur. The rate of core-collapse supernovae should be ties to how fast a galaxy is producing new stars, because massive stars don't live very long lives. But the rate of Type Ia supernovae is a matter of debate, because we aren't sure exactly what produces them.
One way to get a handle on which theory for Type Ia supernovae might be correct is to measure how often they occur in a given galaxy. This is hard, because the typical rate seems to be about one supernova per galaxy every 100 years. This means that, to measure an accurate rate, we'd have to watch a single galaxy for several thousand years. Try getting that observing proposal approved, even if you ask for long-term status.
So, people have tried other techniques, such as looking at huge numbers of galaxies. These searches find new supernovae all the time, but it is difficult to figure out what the rate means. The astronomers have to be careful to lump together only similar types of galaxies, in case the rate depends on the type of galaxy. Several groups have worked on this, but they come to different results depending on exactly how they analyze the data. Are there any other ways?
In the picture of the LMC at the start of this article, you can see several reddish blobs. These are nebulae. Many of the nebulae are nurseries where new stars are being born, but many of them are also the remains of old supernova explosions. In X-rays and radio light, supernova explosions are more straightforward to identify, and many teams of astronomers have used optical, radio, and/or X-ray observations to identify the location of supernova remnants in the LMC, and also what kind of supernova explosion (Type Ia or core-collapse) caused each supernova. While a supernova itself is only visible for a year or two, their remnants can be visible for tens of thousands of years -- just the time baseline we need to get accurate rates of supernovae!
Earlier this week, astronomers Dan Maoz and Carles Badenes, released a paper that analyzes all of the known supernova remnants in the LMC and the SMC. They use this to determine the supernova rate for both Type Ia and core-collapse supernovae. But they also go one step further. They use the ages of the stars around the supernova remnant to estimate how old of a star exploded. How does this work?
A good friend of mine, Jason Harris, calculated how fast each part of both the SMC and LMC is making new stars, and (even cleverer) how fast each part has made stars in the past. I don't have too much room to explain here, but the basic idea is that he and his thesis advisor, University of Arizona astronomer Dennis Zaritsky, made a map of what kinds of stars are present at each point in the galaxies. Some parts of the LMC and LMC have mixes of young stars and old stars, other parts just have old stars, and yet other regions may have had a few intermediate-age stars, too.
Let's say that in one section of the LMC galaxy around a supernova remnant, Harriz and Zaritsky found that half of the stars had ages spread between 1 billion years and 10 billion years old, and that the other half of the stars were all younger than 20 million years old. If the supernova remnant is from a core collapse supernova, then we can be pretty sure that the supernova belonged to the young stars. This is because stars that live to be a billion years old or older end their lives as white dwarfs, not as supernovae. But if the supernova was a Type Ia, it probably came from the older stars, because Type Ia supernovae need a white dwarf to explode, and stars that only live 20 million years or less don't make white dwarfs.
Back to this week's paper. Maoz and Badenes looked at the ages of stars that Harris and Zaritsky calculated around every known supernova remnant in the LMC. From this, they find that the current rate of supernovae in both Magellanic Clouds is around 3 or 4 every millennium, or roughly one every 300 years. More importantly than the total rate, they find that there appear to be two different sources for Type Ia explosions. In places where there are only very old stars, there are some Type Ia supernova remnants. In places where there are stars younger than 330 million years old, there are a lot more Type Ia supernova remnants. This means that a significant number of Type Ia supernovae in the Magellanic clouds come from youthful stars, though these stars are still old enough to be turning into the white dwarfs that we think are needed for Type Ia explosions.
This work strengthens the case for an idea that other astronomers have been suggesting for some time, that maybe there is more than one way of making a Type Ia supernova (see my blog posts on Type Ia supernovae from last August and last month). It has only been in the past few years that this idea has really caught on in the community, and yet it could be important. Type Ia supernovae are the kind of supernovae that other astronomers have used to probe dark energy, and in order to better understand dark energy, we need to fully understand exactly what is exploding.
I also was happy to see this paper make good use of the huge amount of data on stellar ages that Harris and Zaritsky collected. I was in grad school with Jason when he was working on that effort, and I saw how much work went in to getting those stellar ages out. (In particular, I remember one day when Jason was a little embarrassed because he realized that he'd written a computer program that, if he'd let it run to completion, would have taken about 10 billion years to run. Oops.) This is the power of well-run, fully published projects in astronomy. A decade later, someone can come in and use your data wisely for purposes you may not have imagined.