Imaging a gamma ray burst

Gamma ray bursts (GRBs), mysterious surges of gamma-ray radiation coming from outer space, have been observed for several years now by spacecraft. (Gamma rays don't make it through the Earth's atmosphere, so can't be observed from the ground.) Nobody's at all sure what causes gamma ray bursts. For many years nobody even knew how far away they were. Were they something in the solar system, or something in the galaxy, or something even farther away than that? If they really were in other galaxies their source would have to be almost unimaginably powerful to be so easily observable at that distance.

For a year and a half or so it's been known that some GRB's also emit visible light, but only if you catch them quickly. The so-called ``optical afterglow'' fades away quickly. Furthermore, it appears that the gamma ray burst intensity and the optical intensity don't correlate very well. Some rather dim gamma ray bursts have resulted in bright (``bright'' used rather loosely here, you understand.. these things are dim!) optical counterparts, and some big burst events have been actively searched for with nothing at all found.

The difficulty is that when these gamma ray burst events happen, it takes a while to get an error box mapping where in the sky the gamma rays came from. The best we can do right now is wait for several interplanetary probes in solar orbit (in particular, Ulysses, the Konus experiment on WIND, and NEAR) to report back when they saw the event, and then triangulate. This takes a while... 24 hours or so after the event.

Very few professionals can mobilize this fast, plus for professional scopes the error boxes are often rather inconveniently large, so they have been actively seeking amateur help in imaging GRB afterglows. (The brighter ones should be within reach of larger amateur instruments attached to modern CCD cameras.) If somebody can manage to image the burst optically, then a precise location can be found, and professional scopes (with great light-gathering power but a very narrow field of view) can then study it in detail as it dims. That is not possible, however, unless the source is located with great accuracy... something that is not yet possible from spacecraft gamma-ray observations alone.

To help amateurs find them, an alert is sent out on the AAVSO (American Association of Variable Star Observers) Gamma ray burst mailing list. They also post a digital sky survey field with superimposed error box showing you just where to go look. Only two amateur groups have been successful at this quest so far... especially after a delay of 24 hours, these things are REALLY DIM (like 19th magnitude or dimmer).

Well, late Tuesday evening (US time) September 26, 2000, there was a burst. By Wednesday afternoon the burst had been triangulated (near the head of Draco), an alert went out on the mailing list, and an error box chart was posted. The Fort Bend Astronomy Club Gamma Ray Burst team had made five previous attempts to image a burst, but without success. I had tagged along twice previously, but as one of the more inexperienced members of the team I had mostly just watched while others ran the show. Our telescope has to be guided to a field by hand (by ``star hopping''), and that takes some skill to accomplish. Several members of our team are quite accomplished at it, but I'm not one of them.

That evening the weather in Houston was uncharacteristically clear, and the burst was ideally placed for observation from our latitude. Unfortunately, I was the only one of the team who could make it. I figured what the heck --- I could use the practice and it was worth a try. Maybe I would be able to find the field by myself.

There are established procedures for how this process is supposed to go. Here's how it actually went. Somehow these things never seem to go exactly according to plan.

I went out to the George Observatory in Brazos Bend State Park. Setting up was a snap --- I've learned how to do that part at least now. The mosquitoes were bad at the start (I smashed at least 5), but as the temperature sank they stopped bothering me much. Focusing I'm still not very good at. I tweaked and twiddled and tried but the diffraction spikes showing up on the bright stars in the CCD images still looked a little doubled, a sign of misfocusing.

Pointing the scope in that far-North part of the sky presented special challenges. I had to untape the CCD power transformer and move it to a new location, because the cord between it and the camera wasn't long enough. After much frustration, knowing from looking through the Telrad that I must be REALLY REALLY CLOSE, but not being able to match anything I could see in the CCD field on my charts, and with my brain unable to make sense of the rotated view I could see through the finder scope, which in any case was WAY UP IN THE AIR at a very awkward angle to look through, I was on the verge of giving up. Then I realized that one of the bright stars in the head of Draco was at nearly the same declination as where I needed to go. All I needed to do was get that star showing in the CCD image, then guide the scope a little bit South, and then slew the scope by hand only in RA. That reduced the problem to a ``line search'', and eventually I was able to recognize the field on the 1-degree search field I had brought with me. (Experienced team members could have, and have since, found this same field within 5-10 minutes!)

About this time as I was zeroing in on the exact field the ``East'' button on the guider jammed, and then as I attempted to get it unjammed (as the scope kept going East, and East, and East) it just went in farther and farther... and then finally it went in all the way leaving a hole. Well, at least that had the effect of stopping the scope from slewing East any more. Fortunately I discovered I could accomplish that motion (at an even faster speed) by simply unplugging the power to the scope's tracking. So I was still able to guide in to the exact field.

I shot 11 GRB fields of 4 minutes each; for only 7 was the tracking good enough that they were usable. At one point I started to hear a funny sound coming from the telescope. It was the sound of the duct tape holding the CCD power supply onto the telescope fork slowly coming loose (not surprising given its unusual near-vertical angle), getting ready to drop the heavy power supply onto the floor. Disaster averted! I quickly looked for more duct tape but couldn't find it. I found some electrical tape in a drawer and used that instead. Except, in the darkness I failed to notice that the electrical tape had welded itself on to a razor blade which was ideally positioned to slash my fingers as I flailed for it in the dark. Fortunately that didn't happen. I resecured the power supply during an exposure. Despite all the bumping it was one of the least trailed images of the set.

I didn't see anything unusual on the original images. I sent them to Bill Dillon to stack them together and went to bed. Little did I realize things were about to get interesting.

> From: Bill Dillon
> Sent: Friday, September 29, 2000 1:43 AM
> To: grb list
> Subject: [AAVSO-GRB] GRB000926: we may have got it
> I processed Joe Dellinger's images from last night.  There may be an
> object near the center which is not on my printed version of the DSS,
> nor on the AAVSO web site chart, but I can't be sure.  I'd appreciate it
> if someone who has a better image can comment on it.

Bill's original stack of the full-frame image can be found here, along with technical details on the processing and a comparison image of the same field taken with the same equipment a few days later.
Here is a compact JPEG version of just the center of our image, with the GRB afterglow pointed out, courtesy of Dan Kaiser:

GRB 000926 image

Several professionals were also able to image this one. An observatory in the Canary islands had earlier nightfall and had already imaged the error box a few hours before we did, although we didn't know that yet. By the middle of Thursday we knew this had been a ``bright'' one, but we still weren't quite sure where to look on our images. We thought we had a good candidate, but wanted professional confirmation.

Turned out that most of the people interested in getting amateur help on these observations were offline, attending a gamma ray burst conference! So we didn't get back any reply from the list for more than 24 hours. Finally, late Friday afternoon we heard back. Guess what? We did indeed get it! Our suspicious blob really was it. It's, uh, a bit ``subtle'' in our image, but, it is there! (If I had only realized I would have stayed out taking images to stack longer!) When you consider that this object (whatever it is) is so far away from us that it is receding at 80 percent of the speed of light, even magnitude 19 seems pretty bright.

I had not realized that we would be only the third (or fourth, depending on how you count it) group of amateurs ever to image a gamma ray burst. We got a congratulations from the NASA-Marshall Space Flight Center, and Sky and Telescope magazine wanted to run our image and picture in an upcoming article on gamma ray bursts. (It has since appeared, in the January 2001 issue.) An amateur group in Finland also imaged the same GRB the next day; their image is much cleaner than ours, in fact!

I had really thought I was just practicing for after the launch of HETE-II! It's a new probe becoming operational very soon, that should give much quicker response with smaller error boxes. Amateurs should then be able to image these things within hours of their going off, instead of days. They should be a LOT brighter then, and so much easier to detect. Still, in any future publication written about ``gamma ray burst 000926'' our club will be able to look at the magnitude data points, point to one of them, and know ``We took that!''.

Return to my home page