It’s one thing to design, assemble, and install a next-generation telescope instrument. It’s another to bring it to life.
The Dark Energy Spectroscopic Instrument (DESI), installed on the Mayall Telescope at Kitt Peak National Observatory near Tucson, Arizona, contains more than half a million parts in one of its central components, known as a focal plane. Making sure that all of DESI’s parts work together as designed – and that it is functioning in tandem with DESI’s other components – involves a lot of testing and troubleshooting.
Satya Gontcho A Gontcho, a research associate at the University of Rochester, nearly a year ago was appointed as a lead observing scientist for DESI. She is part of the team that has helped to prepare the instrument for the start of its five-year observing run, which is slated to begin later this year. The Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) is the lead institution in the DESI project, which is supported by a large international collaboration.
Gontcho A Gontcho previously participated in an imaging survey in Chile that was used to help select galaxy and quasar targets for DESI. Quasars, among the brightest objects in the universe, are powered by supermassive black holes at the center of galaxies. She also worked on a predecessor experiment to DESI called BOSS, the Baryon Oscillation Spectroscopic Survey.
DESI will collect the light from about 35 million galaxies and quasars to help learn about dark energy, which is driving the accelerating expansion of the universe. DESI could also supply new insights about the life cycle of galaxies and how the cosmic web – the large-scale structure of the universe – took shape. It will collect the galaxies’ and quasars’ light using an array of 5,000 robotic positioners that each carry a fiber-optic cable.
Every positioner can point a fiber-optic cable at a galaxy or quasar, so DESI can capture the light of 5,000 objects at a time. The light is split up into different colors and measured by a bank of 10 instruments called spectrographs, and this data will be used to gauge galaxy distances and the rate at which they are moving away from us.
In this Q&A, Gontcho A Gontcho shares her experiences at the Kitt Peak site, including evening observing stints to run through detailed checklists and probe how DESI’s components are working.
Q: What is it like to work on DESI directly at the Kitt Peak site?
A: As soon as you set foot there, it is kind of like a black hole – you get completely absorbed by the tasks at hand and time passes really quickly. At the moment, we are testing the instrument and making sure it is ready for the five-year survey. While overseeing an observing run, not every task is challenging on its own; however, there are so many things to monitor at once that you must always be alert. As a result, you need to quickly develop an understanding of what is critical and what is not.
Most of the people who I have been working with are instrument experts in their specific areas. I have been learning a lot of things from them on the fly – everyone is incredibly helpful and collaborative and pleasant to work with.
On the busiest days, I have seen about 15 to 17 people on-site. Most notably, there are about 30 people who are not on the mountain but are always available online and ready to help. Seeing such a large group of people working so well together when the stakes are this high – and considering the work virtually never stops – it is truly inspiring and worth taking a moment to appreciate.
Q: What is it like to participate in an evening of observations with DESI on the mountaintop?
A: My shifts at the Mayall Telescope are for seven nights. I wake up around 2 p.m., grab breakfast and go up to the telescope around 3:30 p.m. Then I get a rundown from the day crew who are working on instrument tuning and upgrades. At 4 p.m. I will usually welcome the new observers beginning their four-day shifts, and at 4:30 p.m. we have a teleconference with other DESI collaborators to reassess our priorities and design an observing plan for the night. Around 5 p.m. we have dinner, and by 6 p.m. we are back in the control room and ready to proceed.
The control room is located on the middle floor of the telescope, which is essentially a huge concrete building. The fact that you don’t see much sunlight for a week can be pretty alienating.
Most of the electronic noise in the control room comes from the warning sounds of the DESI operating system. As things progress, we’re expecting to hear those less and less! On occasion, the wind can blow very, very strongly at Kitt Peak – I’ve experienced sustained winds of 60 mph, with 90 mph gushes – and in those instances you can feel the dome vibrating all the way down to the control room several floors below.
During a night of observing you have about four people in the control room, including a telescope operator who is a member of the OIR Lab (National Optical-Infrared Astronomy Research Laboratory) staff, and a lead observer – that’s the position I occupy. The telescope operator is in charge of the safety of the telescope, and the lead observer is in charge of the safety of the DESI instrument and running the observations.
To kickoff our night, we start by turning on the spectrographs. We need to turn them on a minimum of 30 minutes before observations. We have cameras on all of the critical sites in the building that we need to see from the control room.
Then we start the calibration. We point the telescope to a white screen inside the telescope dome and take test exposures. We check that everything is working as expected from the spectrographs’ point of view.
Then we go on sky (start observing) and calibrate the guiding and focusing camera system to ensure the quality and stability of the images we will collect. When the telescope moves it is incredibly smooth –you do not feel anything or hear it.
Collecting enough information from a particular part of the sky for our tests usually takes about two hours. Since the fall, we have been going through a list of critical things to optimize and solve.
Q: What have you enjoyed about the experience of helping to bring DESI online?
A: Coming here every few weeks, I have been able to see how much things have progressed from one month to another: There is a lot of hard work on the part of many, many people in order to make this survey a success, and it always shows. To contribute to this effort, while having a front-row seat to what it takes to get quality data, it has been tremendously enriching for me. It is enriching as a scientist to experience the process and to get a grasp of all of the various steps involved. But also it is enriching as a human being to experience people working together, collaborating with each other in the best way possible.
I am being trained by literally the handful of people who have this specific expertise on Earth. It is a big privilege – a once-in-a-lifetime chance to learn from them – and for me that is tremendously exciting.
Q: How did you become involved with the DESI project and how has your role changed since you joined?
A: I was a postdoc in London when I joined DESI a couple of years ago. I observed for the DESI imaging survey in Chile. Before that I had been involved in the BOSS survey, specifically using quasar spectra for cosmology purposes.
When I joined BOSS, all of the data was flowing in. At that time it seemed that the bulk of the work started there, with analyzing the data. But that was a rookie assessment. Being involved earlier in the process of getting a cosmological survey on its feet has allowed me to develop a deep appreciation for all of the work that goes on behind the scenes.
I have been observing with DESI for one week every month since we saw first light in October. We have 10 lead observers on rotation, and half of the lead observers are also instrument experts who need to be working during the day shifts on their issues.
The next step for our research is to collect a representative subsample of the dataset we are expecting to get from the five-year survey. The priority at this stage is to test the methods that have been developed for the past few years on real DESI data to determine how far away quasars and galaxies are.
The previous phase, called commissioning, was successful and upon restart of activities DESI will enter the survey validation phase. This represents the last stretch before operations normalize and we reach the point where everything runs as it should during the next five years.
For more information, visit energy.gov/science/office-science
Tell Us What You Think!