A DISTANT, COMPLEX SUPERNOVA REMNANT G344.7-0.1

Wow, hey guys! It has been quite the hiatus and I do apologize. But, hello!π As you can see, I have moved things over to a new web host (WordPress). I lost quite a bit of formatting in many of my old posts so it took me some time to go through and fix it all, though I need to go back (again) to fix how some of the posts appear on mobile devices, so thank you all so much for your patience with me during this time ππ₯°

I thought it was about time to not only come back here to continue my regular blog posts, but to go ahead and conclude the second paper that I started discussing back in September and is available here.
To briefly remind us what we are dealing with:
- We have a known supernova remnant (SNR) located along the Galactic plane with the Galactic coordinates (G) 344.7 (longitude, in degrees), -0.1 (latitude, in degrees). Hence, the identifier “G344.7-0.1”.
- Faint, point-like gamma-ray emission, which is detectable at energies > 10 GeV with the Fermi-LAT, overlaps with the Western edge of the remnant, suggesting some kind of connection.
- Extended very high energy emission above 1 TeV is adjacent to the > 10 GeV emission location, suggesting that whatever is responsible for the > 10 GeV gamma-rays is also responsible for the very high energy gamma-rays above 1 TeV.
- Analyzing old archival X-ray observations from the XMM-Newton X-ray Space Telescope, we discover the SNR is dominated in this energy band by thermal X-rays with no hint of non-thermal emission in or around the SNR.
- Finally, there is a notable IR filament that overlaps well into the confidence region for the gamma-ray emission (i.e. the region where the gamma-ray emitting source is most likely to be located).

There are a lot of cool things happening here. Because my first source turned out to be a shock-cloud interaction (probably), I was definitely wondering if I was somehow looking at another one! The evidence seemed to line up in a strikingly similar way to the Vela SNR. But G344.7-0.1 is a lot farther away from us than Vela, so that introduces some extra challenges to trying to figure this out.
In fact, Vela is only about 1,000 light years away from us. That means it takes 1,000 years for LIGHT to travel from the pulsar to us, which is roughly 5,879,000,000,000,000 miles from Earth or 9,461,000,000,000,000 kilometers. But really, as far away as that might sound, that’s still relatively close. Look at some similar systems plotted by distance from us (the Sun/Solar system) with respect to the larger Galaxy we belong to, the Milky Way π₯°.

The above image is borrowed from an analysis on objects like the Vela SNR in the HESS Galactic Plane Survey released in 2018. The free version of the article can be found here. Look how close Vela-X (a component to the Vela SNR) is to us within our Galaxy!
The black outlined star with the initials “GC” represents the Galactic Center (GC) which is about 8 kiloparseconds away from our Solar system. That’s roughly 26,000 light years away from us. That’s right! That means that the light from the center of our own Galaxy has to travel for 26,000 YEARS before it reaches us! And yes, it’s moving as fast as it can!
Now, this is all relevant to G344.7-0.1 for the following reason: Measuring the distance of a far-away SNR like this one can be tricky, so our estimates leave us with some pretty big margins of error, but we estimate this particular system is at least 3 kiloparseconds away (about 10,000 light years) but could be on the opposite side of the Galaxy anywhere between 9 kiloparseconds (kpc for short) and 14 kiloparseconds (or almost 46,000 light years away from us π§). The uncertainties attached to these huge ranges in distance haunt us throughout our analysis!
It also haunts us in the quality of data that is available for this source. For Vela, we had so much literature and surveys to sift through, which ultimately gave us the “smoking gun” for the shock-cloud interaction, the hydrogen cloud that shared the same shape and location as the higher energy emission we had discovered! But this source, G344.7-0.1, being much farther away, does not have adequate imaging for those same surveys, so we aren’t able to make any firm conclusions about what the SNR could be running into.
This is because the angular resolution of most available surveys for data covering our SNR region was just not gonna cut it to resolve any meaningful connections. This time, we had to really think about the physics involved here to be able to interpret our findings and offer a consistent explanation.
So what did we do? Our most favorite thing! We took all of the data we could measure for this source in light waves — from radio wavelengths to TeV gamma-rays — and we tried to put it on a plot (remember those spectral energy distributions we discussed some time back?). Then, with some software tools like Python, we can apply relevant physics equations to the processes going on and try to predict what we would observe and then compare that to what we actually see. This way, we control the physics and processes that explain the observations so we can make meaningful conclusions on what the possible scenario is going on here.

Recall that we had all of the following wavelength ranges on this source: radio, X-ray, Fermi (MeV-GeV) gamma-rays, and HESS TeV gamma-rays. In the plot above, however, you’ll notice that only the gamma-rays are plotted on our spectral energy distribution (SED) plot. This is because these models assume that the particles radiating the observed emission is all one population with the same characteristics — same radiative mechanisms, same interaction processes, same average energies, etc — and this may not always be a good assumption to make. It is entirely possible that the particles responsible for emitting in radio waves are totally separate from the population responsible for the higher energy emission!
As we plotted all of the radio — gamma-ray data, it became clear that there must be more than one particle population present: one to explain the radio and X-ray data and one to explain the higher energy data. However, our model is limited in this regard. If we need to “add” more particle populations to the model, it gets very complicated, and so we are unable to investigate this further. As a result, we limit ourselves to only trying to characterize the population behind the high-energy emission in the gamma-ray regime. Hence, the SED above only plots the gamma-ray data points.
Now, we are almost there for coming up with a way to understand what we are seeing. We now know that the high energy emission is disconnected in a particular way from the lower-energy (radio and X-ray) emission from our supernova remnant. This is interesting since the radio and X-ray emission is confined to the supernova remnant itself: the radio and X-ray emission fill the SNR, but both the radio and X-ray emission steeply decline just beyond the SNR shell. On the other hand, the gamma-ray emission is located on the Western edge of the SNR, with higher-energy gamma-rays extending to the South-East of the SNR shell. This picture seems consistent with the current model results that these are two separate populations.
Based on the X-ray and radio location and properties, we know the SNR has accelerating (i.e., radiating) particles that are emitting synchrotron radiation largely in radio, but these particles are not the same ones generating the gamma-ray emission. We can also tell by the SED above, that we cannot make any distinction with the data alone to say if it’s more likely to be protons (hadronic scenario) or electrons (leptonic scenario) that is responsible for the observed emission… Did we hit a road block? Is this where our journey ends?
The answer is no. We still have some information to consider. We need to consider now both the morphological properties (aka how the SNR looks in each wavelength) and compare it to our best-fit model and the estimated parameters (properties of the particles) to try and understand the most likely origin for the high-energy emission.
This is a lot to digest though, so you know what time it is! π Next time, we will wrap up with the big picture for this object and I’ll also explain a little more about what measurements we can make from our best-fit model (particularly the right panel shown in the above image).
Quiz! Take it here.
And as always, check out the free version of the article being summarized here.