In an earlier companion post (Sharper Focus, Part 1), I introduced the concept of adaptive optics and talked about some of the ways astronomers and engineers have used those techniques to sharpen the images we capture. Now I’d like to show you some of the specific adaptive optics solutions that will enhance TMT’s vision. As you’ll see, we need to use all of the adaptive optics approaches mentioned in Part 1, and to do that, we draw on the ingenuity, hard work, and dedication of TMT’s contributing teams from all around the world.
But first, let’s come back to why we're spending all the energy and resources needed to design, build, and maintain these systems. In Part 1, I showed you images of Neptune and Jupiter taken after applying adaptive optics techniques. What we'll be able to achieve once TMT is operational promises to be even more impressive. Here’s a simulated comparison of Jupiter’s moon Io, seen through the adaptive optics system currently in use at the W.M. Keck Observatory (left panel), and as it will appear through TMT's adaptive optics system (right panel).
You can see just how much more clearly and in what greater detail we’ll be able to study celestial objects through TMT, compared to the current state-of-the-art ground-based astronomy—and you can imagine what having all this additional information will mean for scientists who want to discover secrets about the origins and destiny of the Universe and everything in it. That’s why we’re putting so much hard work into getting TMT’s adaptive optics exactly right.
Here’s a picture showing what the adaptive optics system will look like on TMT, when the initial systems and science instrument are all installed:
As you can see, the adaptive optics systems for TMT are elaborate. Let’s go through these systems one by one, so you can see how they all work together.
TMT will use both natural and laser guide stars. But what sets TMT apart from existing observatories is that it will have several laser guide stars instead of just one. Remember, this helps neutralize the cone effect, which would otherwise be especially pronounced on TMT because of its record-breaking diameter. When the observatory first opens to start its science program (what we call “first light”), there will be six lasers. But TMT is designed to grow with technology and scientific need. So, as we add more science cameras, we'll also add more lasers.
Currently, our adaptive optics team is looking at how to incorporate this laser system into TMT:
Here’s a photo giving a sense of the complicated electronics that will be needed to power, control, and monitor each laser:
Our adaptive optics team recently performed what we call an interface study, looking closely at how well we can make sure this particular system will work for TMT. Specifically, we examined whether each laser and its electronics will fit into the space available for it, whether we can mechanically support the weight of each laser and its heavy electronics rack, whether the available power supply will meet the laser's needs, and whether we can provide sufficient cooling for each laser, especially with the limited ventilation available in the thin air at the top of a mountain. The good news is that it looks as if this particular system will be ideal for TMT.
As you’ll recall from Part 1, we need to get the laser light from TMT to the correct place in the sky without interfering in any way with the delicate electronics or optics of the observatory, and doing this right takes careful planning and design. For TMT, we’ll build a set of beam transfer optics, as shown schematically in the picture at the beginning of this post. The beam transfer optics will get us from the output of the lasers up to the top (which is the back side) of TMT’s secondary mirror.
From there, the lasers will be projected simultaneously using the laser launch telescope. Our original design of the laser launch telescope used off-axis mirrors to project the laser guide stars to the required places on the sky. You may recall from an earlier blog entry on stressed mirror polishing that fabricating off-axis mirrors can be difficult. So, we’ve now replaced the off-axis mirrors with lenses. Here’s an image of what that subsystem will look like:
You can see that there are multiple components all located together. Asterism, for example, here refers to the artificial constellation of guide stars that we’ll be projecting onto the sky. We’ll leave a more detailed description of the laser guide star system to a future blog entry.
Safety is always the most important consideration for everything we do on TMT, and of course lasers require special care. On TMT, we have a dedicated Laser Safety Officer, and we’ve established a sequence of policies and procedures to ensure the safe operation of the lasers and in particular to make sure that no-one gets hurt.
No matter how many guide stars we have, they can’t help us without a way to gather, process, and use the information from them. These things are not easy to do, and to make them work for TMT we need an instrument hosting a number of complex subsystems: namely, sensors to measure the image distortion; deformable mirrors to correct for the distortion; and a very powerful computer to process the signals from the sensors in real time and send corrections to the actuators attached to each deformable mirror.
This instrument we’ll use is the Narrow-Field Infrared Adaptive Optics System (NFIRAOS; pronounced "nefarious"). Here’s an image of the layout of NFIRAOS, which gives you a sense of the complexity and size of the instrument.
NFIRAOS is so multifaceted and sophisticated that I’ll be writing a future blog entry just about it. But for the moment, it may be interesting to learn that the entire inside of NFIRAOS will be cooled to -30 degrees Celsius (-22 degrees Fahrenheit) to reduce the background infrared glare seen by its camera system—possibly too cold for our Mammoth friend, but just the right temperature for what NFIRAOS has to do.
For reasons described in the previous post, deformable mirrors are really hard to make, and the ones inside NFIRAOS are no exception. As with any deformable mirrors, the surfaces have to be paper-thin and flexible yet resilient, and able to be shaped and reshaped hundreds of times every second by the complex arrays of actuators attached to them. For TMT's deformable mirrors, we’re funding two competitive studies with industry to develop the final designs and fabricate prototypes. Regrettably, I can’t include images here showing how the mirrors might look, because the designs are proprietary.
Once NFIRAOS has done its job of correcting the images, it will feed the corrected light into a science instrument: that is, an instrument for actually making observations. At TMT’s first light, that instrument will be our InfraRed Imaging Spectrograph (IRIS), which will have both a camera to photograph astronomical targets and a spectrograph to identify the chemical composition of those targets.
TMT will have to do all of the adaptive optics work I’ve described here almost as fast as the atmospheric winds shift. Specifically, according to TMT’s requirements, it will have to “perform the complete target acquisition sequence in less than 5 minutes.” This is really hard to do for something the size of TMT, but is needed because, on some occasions, something new and wonderful will flare up in the night sky, and TMT won’t have long to turn to observe that object. The brightness of that glorious object of our affection may well decrease so rapidly that we have to lock onto it and gather what data we can in the few precious moments before—alas—it is lost to our sight forever. An example of this is a so-called long-duration gamma ray burst, where “long-duration” here means up to only a few minutes.
As you can see, creating all of the complex subsystems comprising TMT’s adaptive optics and making them work together presents a daunting challenge. But as with virtually every aspect of TMT, we're meeting that challenge by working together, with contributions from team members who collaborate across multiple continents.
For example, the laser guide star interface study I talked about is for the German/Canadian Toptica/MPBC laser system. Our colleagues at the Chengdu, China, Institute of Optics and Electronics are designing and will build the system that sends the laser beam to the top of the observatory and from there up into the sky. The NFIRAOS processing and mirror-control system is being designed and built by the Herzberg Astronomy and Astrophysics Research Centre, in British Columbia, part of the National Research Council Canada, where our colleagues are currently working hard to develop the final detailed design. A sizeable number of the subsystems for NFIRAOS are being designed by Canadian industry—a direct result of the Canadian Government’s desire to involve and promote its private sector through TMT. To develop the deformable mirrors, TMT is working with AOA Xinetics in Massachusetts (USA) and CILAS in Orleans (France). And the team building the IRIS camera and spectrograph has participation from our partner institutes in many countries across the world—Canada, China, Japan, and the U.S. When it comes to the adaptive optics for TMT, we're all in the project together.
Given the current political environment in many parts of the world right now, I’ve been thinking a lot about something Lawrence Krauss has written. In looking back at the development of the Standard Model of particle physics, he says: “Like any human drama, it also included its share of envy, stubbornness, and vanity, but more important, it involved a unique community built completely independent of ethnicity, language, religion, or gender. It is a story that carries with it all the drama of the best epic tales and reflects the best of what science can offer to modern civilization.”
Just like the scientific challenge Krauss highlights, a project like TMT is a kind of beacon—a guide star, if you will—to show our potential to work together and the amazing things we can accomplish when we do. Krauss’ comments certainly resonate with me in sharing with you the scientific and technological adventure that TMT offers the world.
C. Boyer; B. Ellerbroek, “Adaptive Optics Program Update at TMT”, Proc. SPIE 9909, Adaptive Optics Systems V, 990908 (26 July 2016); Published in SPIE Proceedings Vol. 9909: Adaptive Optics Systems V, Enrico Marchetti; Laird M. Close; Jean-Pierre Véran, Editor(s), doi: 10.1117/12.2232945
Lawrence M. Krauss, “The Greatest Story Ever Told – So Far”, p.273. Atria (2017)
My thanks to Corinne Boyer (TMT Adaptive Optics Instruments Group Leader), Melissa Trubey (TMT Senior Instrumentation Opto-Mechanical Engineer) and Matthias Schöck (TMT System Scientist) for assistance with this blog entry.