A magnetar / SGR / radio pulsar only 3“ from Sgr A*

In case you haven’t been following this, there’s a really interesting story unfolding in a series of ATels about a magnetar/SGR with a radio pulsar counterpart at a very small angular distance (3“) to Sgr A*.

If this were a pulsar in an orbit around the central black hole, this would be one of the holy grails of astronomy, since it could potentially be used to probe the space-time around the black hole with unprecedented precision. I’ve tried to summarize the whole story with its various twists and turns below.

NuStar finds a periodic X-ray signal

It started with ATel #5020 (http://www.astronomerstelegram.org/?read=5020) on Apr 27. This ATel reported NuStar X-ray observations of Sgr A* initiated by a flare detected earlier by Swift. A periodicity analysis of ~17k photon arrival times found a signal with a period of ~3.76 seconds and a three-peaked pulse profile. The ATel concluded by saying: „The bursting, timing, and spectral properties strongly suggest a previously unknown magnetar undergoing an outburst.“

NuStar X-ray observations of Sgr A*
NuStar X-ray observations of Sgr A*, showing the detection of a periodic signal with a period of ~3.76 seconds and a three-peaked pulse profile. Plot from ATel #5020.

Later on Apr 27, there were two radio flux monitoring observations using the NICT Kashima-Koganei VLBI (2 GHz and 8 GHz; http://www.astronomerstelegram.org/?read=5024) and the Jansky VLA (L, S, C, and X band; http://www.astronomerstelegram.org/?read=5025). While they were targeted at monitoring possible emission due to the gas cloud G2, which should be closest to the Galactic Centre about now, they also didn’t see any enhanced radio emission from the supposed magnetar / SGR.

Non-detection in the radio band?

The radio astronomy community chimed in on Apr 29, with ATel #5027 (http://www.astronomerstelegram.org/?read=5027). Astronomers at the Max Planck Institute for Radioastronomy in Bonn had used the Effelsberg telescope to obtain deep 1-hour observations at each 4.85 GHz and 14.6 GHz. They made no radio detection, and could only set upper limits.

At that point one might’ve given up hope on a radio detection, but note that Sgr A* is pretty low in the sky for Effelsberg, making this a not an optimal target for the telescope.

Chandra confirms the X-ray periodicity and improves sky position

Apr 30 brought ATel #5032 with the Chandra confirmation of the SGR flaring activity (also at the same period of 3.76 seconds) initially found by NuStar (http://www.astronomerstelegram.org/?read=5032). The Chandra observation also had an improved sky position for the new source, about 3“ away from Sgr A*. This angular distance corresponds to a distance of 0.12 pc if the source is at the distance of the Galactic Centre at 8.0 kpc. The Chandra observations also showed that Sgr A* itself was not flaring.

Chandra image of the newly-discovered magnetar/SGR.
Chandra image of the newly-discovered magnetar/SGR, only 3“ away from Sgr A*. Left: the new HRC-S observation. Right: archival ACIS-S image (obsid:13017). Note that Sgr A* is not flaring itself in the left image. Plot from ATel #5032.

Yet another radio non-detection…

On May 1, with ATel #5033 (http://www.astronomerstelegram.org/?read=5033) a report of yet another radio non-detection, this time with the 76-meter Lovell Telescope at Jodrell Bank. Could that have been the „coup de grace“ for a possible radio pulsar counterpart?

…followed by a report of a slightly mysterious radio detection just hours later

While ATel #5033 looked like the end of a radio pulsar counterpart for the SGR/magnetar, there was another ATel (#5035, http://www.astronomerstelegram.org/?read=5035) less than six hours later that claimed „detection of radio pulsations from the direction of the Galactic center Soft Gamma-ray Repeater with Parkes and the GBT“. The description of what they found was very interesting, but a bit vague: „For all the observations on-source a strong signal (visible in less than 1 minutes at 1369 MHz, 2000 MHz and 3094 MHz) with a large duty cycle (approaching 100%) was found, showing a rapidly changing period with values close to the 3.76 s detected in the NUSTAR, and Chandra  observations.“

There appeared to be a periodic, transient signal, seen only for a very short time. Apparently the emission is not seen in narrow pulses, but is occupying almost a full period, which is unusual for pulsars. The ATel does not explain where the rapidly changing period comes from. The ATel also gives an estimate of the dispersion measure (DM, integrated column density of free electrons in the interstellar medium) of the radio source of < 50 pc/cc, cautioning that „…an accurate determination of DM is very hard because of the fact that the shape of the pulses is varying in time and possibly also in frequency.“ This very low DM was very disappointing, since if it was true, the radio source would be very close to the Earth and not even close to Sgr A*, just coincidentally along the same line of sight.

Unambigous detection of a radio pulsar by the Effelsberg telescope

The–as of now–last bit of the story came last night with ATel #5040 (http://www.astronomerstelegram.org/?read=5040), again from the Max Planck astronomers at the MPIfR in Bonn. They did another 2.5-hr observation with the Effelsberg telescope, this time in a 500 MHz band centred on 8.35 GHz, using the improved sky position as obtained by Chandra. The MPIfR astronomers detect „(…) a narrow (duty cycle ~ 2%) bright (signal-to-noise ratio ~140) pulse with a spin period matching the NuSTAR discovery period.“ There’s a *lovely* detection plot at http://www3.mpifr-bonn.mpg.de/staff/reatough/public/pdmp-cand.pdf

Detection of the radio pulsar counterpart
Detection of the radio pulsar counterpart with the 100-meter Effelsberg telescope, operated by the MPIfR in Bonn. Plot from ATel #5040.

This narrow pulse profile is in direct contrast to the GBT and Parkes observations at lower frequencies, but is likely due to lower scattering at higher radio frequencies. Most importantly, the Effelsberg observations do not show any change in the spin period over the 2.5 hours of the observation. This excludes that the pulsar is in a tight binary, but a wide orbit is still possible (see below) Further observations are necessary to test this.

Also, the estimate of the dispersion measure for the radio pulsar is 4000 +/- 3100 pc/cc. Although the errors are large, the values are much larger than the incredibly small ones reported from the Parkes/GBT observations. They are also consistent with a source at the distance of the Galactic center, so it might not just be a coincidental source along the same line of sight. Given the 3“ angular distance that would be a a cruel coincidence to boot.

It’s going to stay very interesting

So at this point it’s not completely clear to me whether this actually is a pulsar very close (or in an orbit around) the Galactic center. The expected distance of only 0.12 pc to Sgr A* certainly sounds tempting. From Kepler’s third law, I’d estimate an orbital period of at least ~1800 years. That is, assuming we look at a putative orbit face-on, 0.12 pc orbital radius from a Galactic centre distance of 8.0 kpc, and a central black hole of 4.1 million solar masses.

I expect the other large radio telescopes worldwide (Parkes, GBT, Lovell) to have another look at the bright radio pulsar identified with the Effelsberg observations, if they haven’t done so already. It’ll be illuminating to find out where the big difference between the initial Parkes/GBT and the latest Effelsberg observations comes from.

Let’s see what the next few days will bring, in any case it’s going to be very interesting! I’m keeping my fingers crossed that this thing is in an orbit around Sgr A* and not just one of nature’s cruel coincidences.

13 people like this post.

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert