A Primer On Gamma-Ray Pulsars
In which I explain aspects of my research to showcase my communication skills
If you read my previous post, you know that I’ve been listening to a lot of data science podcasts and picking up tips for building my personal portfolio. One of those is to just write and showcase your communication skills to demonstrate that you can explain complex topics in an easy to understand manner.
When it comes to writing, common advice seems to be “write what you know”. So, that’s just what I’m going to do, discuss one area of my research. I’ll gloss over some of the details in favor of providing a framework for readers to understand the aims of my research.
(Author’s Note: I’ve previously discussed this topic during two NASA “Ask Me Anything” events on reddit and on a panel at the Future Con portion of the 2018 Awesome Con in Washington, D.C. Given that, you would think that writing this would be no problem, but the nerves come on anyway. How much to say? Does that phrasing really make things simpler? Will it be received properly? So, if you’re in a similar situation, putting yourself out there, I sympathize, and I guess the trick is to just keep writing anyway.)
How I Got Started With Pulsars
I’ve been working on gamma-ray pulsars since graduate school, using data from NASA’s Fermi Gamma-ray Space Telescope (Fermi) as part of the Large Area Telescope collaboration. At the time, the mission was still called the Gamma-ray Large Area Space Telescope and was a couple of years from launch.
Initially, I got my feet wet playing with data taken during a beam test at CERN. As launch approached and it became clear that I was going to stick with the collaboration, it was time for me to think about trying out a science topic.
For Fermi, in a broad sense, that really came down to narrowing down what type of astronomical objects I wanted to focus on. Talking with my advisor at the time, there were multiple high-profile studies I could get involved with, many of which were aiming to publish early, while the data was still proprietary. The study which attracted me most was a project on the Vela pulsar. There were many people involved in the project, notably Dr. Roger Romani from Stanford University who was the driving force behind the paper, and working with all of them got me hooked on pulsars.
The resulting paper (A. A. Abdo et al. 2009, ApJ, 696, 1084) was one of the first to truly showcase what Fermi could do for pulsar astrophysics, and it was just the tip of the iceberg. I’m really getting ahead of myself, though. First, we need to answer the question, “What is a pulsar, anyway?”
Pulsars
Pulsars are thought to be rapidly rotating, highly-magnetized neutron stars. Theory and observations suggest that these objects are 1–2 times as massive as our own Sun and about 15 miles across. They also rotate anywhere from a few times a second to a few hundred times per second.
To put things in context, imagine squishing the sun down to a ball which could fit snugly inside the Washington, D.C. capital beltway and then spinning it around as fast as a blender. Cool, right?
When this extremely-dense and rapidly-rotating star also happens to have a strong magnetic field, you get particle acceleration and, in turn, emission of light. Though pulsars were first discovered using radio waves by Jocelyn Bell and Antony Hewish (Hewish, A., Bell, S., Pilkington, J. et al. 1968, Nature, 217, 709), they have been observed all across the electromagnetic spectrum (we’ll return to this point later in the post).
To simplify the picture, let’s imagine a giant bar magnet stuck through the center of the pulsar (see image above). Suppose that the bar magnet is tilted so that, as the star spins, it points towards us once per rotation. If light was coming off the end of the bar, you would see a blip of light, a pulse, once per rotation.
The problem, of course, is that the pulsar is spinning way too fast for you to see that the light is pulsing and not constant. In order for you to discern the pulse, let’s imagine that you have special eyes which can enter slow-time mode. These special eyes allow you to watch the pulsar as it spins and see that there is only a pulse of light when the bar is pointed toward you.
As you watch for longer, you see that the pulse you thought was a quick blip actually has structure. For instance, in each rotation the pulse might start building up over time, reach a maximum brightness, and then abruptly disappear. Alternatively, you might see that there are actually multiple little blips and one bright one. As you watch the pulsar spin all the way around several times, you discover that the pattern tends to repeat.
This observation sparks your interest, so you record the pulsar in your slow-time mode (again, using your special eyes), slice the recording up to get videos which are just one rotation, and then average all the videos to get a better picture of what this pulse looks like. Congratulations, you’ve made an average pulse profile.
What we just describes is a simplified version of what we do in pulsar astronomy. “Why?”, you might ask. Well, because when we look at this average profile, how it rises and drops, how many pulses there are, etc.; we can learn a lot about the shape of the magnetic field, how particles are accelerated, and even about the star itself.
You might follow up with a question along the lines of “Why look at a star to learn all this instead of designing an experiment on Earth?” Well, in science, it is a good idea to study theories at the extremes, to really push the boundaries of what we think we know. Pulsars represent extremes of magnetic field strength and density which we cannot currently replicate on Earth, making them extremely interesting objects to study…so we do.
A Multi-wavelength Picture
As I mentioned earlier, pulsars shine all across the electromagnetic spectrum. As a gamma-ray astronomer, I tend to think of light based on energy (not wavelength), which can be a tricky concept.
If you’ve sat watching a campfire, you might have noticed how the hottest part of the flame is a different color from the slightly less hot edges of the flames. This gives us a natural way to talk about the energy of light. The figure below categorizes ‘types’ of light based on ‘temperature’, with a higher temperature ‘type’ of light having more energy than lower temperature ‘types’.
In addition to slow-time mode, imagine that your special eyes can see light from the lowest energy all the way up to the highest. You might wonder, “Does the average pulse profile look the same in different ‘types’ of light?”
To answer this question, you decide to watch a pulsar spin around several times in radio mode, then switch to X-ray mode, to gamma-ray mode, and so on. In this experiment, you use when the end of the bar is directly pointed at you as a reference point and compare when the light is most intense and how the structure is different with your eyes in different modes.
It turns out, as you build pulse profiles at different energies, they’re not all the same! Thinking back to our campfire analogy, this is similar to how different parts of the fire are hotter. In the case of pulsars, the ‘hotter’ parts of the profile give us information about where the most energetic particles, which emit the gamma rays, are. The main takeaway is that, to fully understand pulsars, we need to study them using all the energies of light we can.
Gamma-ray Pulsars
Prior to the launch of Fermi, only 7 pulsars were known to shine in gamma rays. Now, thanks to data from the Large Area Telescope and the hard work of many talented scientists, we know of nearly 300. You can imagine that having more than a handful of sources has enabled a lot of really cool research in this area. Observations using gamma-ray telescopes have not only allowed us to learn a lot more about pulsars, but they have also enabled the discovery of many new pulsars, some of which we don’t detect when we look with radio telescopes.
A big hope for pulsar science with Fermi centered around answering the question of where the gamma rays were produced. Analysis of data from the Energetic Gamma Ray Experiment Telescope (one of the instruments aboard the Compton Gamma Ray Observatory, one of NASA’s four great observatories, the other three being the Hubble Space Telescope, the Spitzer Space Telescope, and the Chandra X-ray Observatory) couldn’t quite provide a definitive answer.
The improvement in sensitivity and performance of Fermi compared to its predecessor had scientists convince that this question could be answered, and answered quickly. The paper I became involved with just before launch was designed to use observations of the Vela pulsar to do just that.
Returning to our toy pulsar model with a bar magnet through the middle, the question at the time was “Are the gamma rays coming from close to the pulsar, right where the bar is poking out, or are they coming from regions higher above the pulsar, possibly some distance from the bar?”
How did we propose to answer the question? Let’s return to our campfire/temperature analogy. In our pulsar-campfire, the hottest parts don’t just emit light of one temperature, there’s a bit of a distribution, with a lot of low-medium temperatures and fewer of the hottest temperatures.
For reasons having to do with physics I’m not going to delve into, if the gamma rays come from just above the pulsar, where the bar pokes out, the ‘temperature distribution’ would have very few of the hottest temperatures, dropping off quickly. If the gamma rays come from higher up, then the ‘temperature distribution’ shouldn’t drop off as fast, with more of the hotter temperatures.
Our analogy uses ‘temperature’ as a proxy for the energy of the gamma ray, so the plan was to record the energy of every gamma ray detected from the pulsar and build up this ‘temperature distribution’ (in this case, a spectral energy distribution) and see how quickly the distribution dropped off at the highest energies. Of course, we weren’t just going to look at it and eyeball how quickly it dropped off. No, we would use a likelihood ratio test to determine at what confidence level we could say that the distribution was more like what was expected for emission near the pulsar or more like what was expected for emission high above the pulsar.
Of course, the next logical question you would ask is “Well, where you able to tell where the gamma rays were coming from?” and the answer is “Yes, but it’s complicated.” We could, with high confidence, say that the emission was coming for higher above the pulsar, but the distribution didn’t drop off as quickly as we would expect for those models either.
Following this initial study, observations of many more pulsars have confirmed that the gamma rays are predominantly produced higher above the pulsar and led to new understandings of the magnetic field shape and how particles are produced. Detailing all of the great pulsar science enabled by Fermi data is beyond the scope of this post (and would make it incredibly long), but you can find some highlights here.
Looking Ahead
If you have questions or comments, leave a note on the post or connect with me on LinkedIn.
My next posts will deal with more coding and data science related topics as I focus on building a ‘data science portfolio’ (and figuring out just what that should look like for me).
