At the Helm

Wish you were at the control panel sending commands to the James Webb Space Telescope? Or would you rather dive deeply into its data—which was held in a top-secret location for six months? This year was filled with many deeply rewarding challenges for our engineers and scientists. Below, three members of Webb’s flight operations team share what they experienced during commissioning and how normal operations compare. Plus, colleagues who archive data from more than 20 missions in the Barbara A. Mikulski Archive for Space Telescopes (MAST) explain how they overcame a number of challenges posed by Webb’s data.

Commanding Webb in Its First Year

What is it like to send commands to a telescope that orbits the Sun 1.5 million kilometers (1 million miles) from Earth? Here, three members of the Flight Operations Team—Branch Manager Benee Coleman, Operations Controller Phillip Johnson, and Ground Systems Controller Evan Adams—run through the experience, conveying clear camaraderie and deep trust.

Benee Coleman: I lead the flight control branch, the group of ground systems engineers, operations controllers, and command controllers who execute all of the special activities and send commands to Webb to accomplish mission objectives. There are four crews, each made up of three team members. I am responsible for making sure they know all the processes and have everything they need to be successful.

Phillip Johnson: I fill two roles. I’m an operations controller and a command controller. I ensure the health and safety of the observatory, and work in close concert with the ground systems engineers, like Evan. On each shift, a three-member team is in constant communication, working collaboratively at every step. Everyone is considerate and supportive, and always brings their A game.

Evan Adams: As a ground system engineer on console, I’m responsible for keeping the institute connected to NASA’s Deep Space Network. I make sure data are coming in and that our commands are going out. Practically every second throughout Webb’s six months of commissioning, our team would be sending commands. On the ground system side, we had 24-hour coverage on the Deep Space Network, often with backup antennas. There was never an idle moment. It was very stressful, but extremely exciting. 

Johnson: Our team is full of expertise. They come from all over. Most of us have worked on other satellite missions, including NASA’s Earth Observing System, Sirius XM, and NASA’s Lunar Reconnaissance Orbiter. One person even has experience running underwater remote-sensing vehicles for NOAA. We are always all in, and always learning from one another. And we spend so much time with each other. The supportive feeling that I got from the team really dampened the stress. During commissioning, each milestone we hit successfully gave us more confidence. I felt better every time I came into work.

Adams: Yeah, I was thrilled to come into work. Every day, we did something no one has ever done before, and half the world saw us do it. Then we came in the next day and did the same thing. It was like that for months on end, and it was amazing.

Coleman: I had some really great mentors in my previous roles. Something they stressed was being part of a launch. When I arrived a few years ago, it was everything they said it would be. It was a truly an amazing experience and I’m so proud to have worked with the team through commissioning and as we entered normal operations. Each day, I’m continuing to see their skills sets expand.

Adams: We always come together as a team, especially when there are challenges. Those are moments when everyone truly understands the stakes—and the room goes quiet. This happened a few days after Webb’s launch. 

During the deployment of the sunshield, two membrane covers were slowly released to form the telescope’s kite shape. All 107 membrane release devices had to work for the sunshield to deploy successfully. Each time we fired a pin, it would flip a switch to confirm it unrolled. We went through this methodically, popping one pin at a time, waiting for it to roll and hit the next switch. Near the base, we fired a pin and didn’t get confirmation from the switch. Did the pin fail to fire and release the membrane? Or did the membrane roll free but fail to trip the switch? 

A large team made up of scientists, engineers, and subsystem experts immediately met to discuss the next step. The first breakthrough was from the attitude control engineer, who was watching the rotation rates. When we commanded the pin to fire, there was a little kick, which slightly jolted the spacecraft. That meant the pin did move. We tried commanding the pin to fire again but didn’t see a second kick. At that point, we still didn’t know if the membrane would roll free. 

The next breakthrough was from the thermal control system team. They looked at readings from the temperature sensors, one facing away and one facing toward the Sun. If it unrolled over the edge of the pallet, it would block the Sun, putting the top sensor in shadow and reflecting extra sunlight onto the bottom sensor. 

At the time we fired it, the sensor on the bottom started to get hotter and the one on the top started to get colder. Based on this, the team recommended that we go ahead and pull the membrane out. We did, with all of us watching the data very carefully—there was no camera to show us what was happening. Instead, we watched the level of resistance to the motor that pushes it out. If the resistance started to rise higher than expected, then it was catching on something and we needed to command the motor to stop immediately. 

I was watching the ground station very carefully to ensure our communications weren’t interrupted. If they were, we needed to switch over immediately to another antenna so we could remain ready to send the motor stop command. Fortunately, the team’s analysis was right—it was just a bad switch. 

[Editor’s note: Read the official mission posts about these events, which were reported during and after the successful completion of Webb’s mid-booms.]

Johnson: Seeing the first calibrated image was another moment to pause and take it all in. Webb was outperforming every expectation mission planners had for the telescope. It’s so rewarding to be part of this mission. Many people don’t realize staff send commands to Webb from the Mission Operations Center in Baltimore. They think it’s automatic.

Coleman: I want to highlight the amazingness of these teams and the work they do. The ground systems controllers, including Evan, connect us to the Deep Space Network to make sure we can communicate with Webb at any moment. It is an intricate process to communicate with an observatory a million miles away. The ground system engineer is responsible for making those connections. The operations controllers, like Phil, are responsible for collaborating with the ground systems controllers, the command controller, and subsystem engineers when needed to confirm we are ready to execute mission objectives for the contact. 

They’re looking to one other, “Are we ready?” And when they hear, “We’re go for command,” the command controller sends the next instructions, whether it is firing thrusters or completing calibrations for an instrument. It is amazing to see the screens light up and all of the sudden you are viewing exactly what’s happening on the Webb Space Telescope. These teams make that happen. We have a lot of experts behind us, but they are the ones who actually execute the activities.

Johnson: In July, we transitioned to normal operations. We were all hardened by commissioning—we’ve seen a lot—but we still spot idiosyncrasies that we can remedy. Our work is still all about constant communication. We’re still getting to know the teams at the three ground stations that support the mission, and learning what they expect of us and sharing what we need from them. But everything we encounter is still a “little thing.” Webb is healthy and operating around the clock. It’s super cool. I have to take a step back from time to time to recognize, to realize that we’re in a very unique situation. 

Securing and Releasing Webb’s First Images and Data

Baking a cake can sometimes be a messy process. Experienced bakers know to gather up the ingredients methodically and follow a recipe card precisely to produce a perfect cake—with barely a speck of flour out of place. For staff who help ingest data from more than 20 telescopes into the Mikulski Archive for Space Telescopes (MAST), the process is also orderly and methodical. Longtime archivist Sara Anderson explains what it was like to manage Webb’s data though its six-month commissioning process, and her colleagues David Rodriguez and Larry Bradley share how they supported those activities. Every bit of their work was done in support of the worldwide astronomical community, who raced to download Webb’s images and data as soon as they were publicly released July 12.

Sara Anderson: For over 25 years, I’ve helped process, prepare, and archive telescope data before they are loaded into MAST. It’s a lot like baking. We get ingredients, which includes the telescope data, and recorded science and engineering files. Another key component we need is the observation status file, and that is like our recipe. It tells us how to put together the ingredients. Once we have that, we can bake our “cake,” processing the data. Then we transfer the data to the MAST team, who add metadata, or keywords, to make it very easy to locate. In a lot of ways, what our teams do should be invisible. Researchers should simply run queries in MAST, and download and analyze the data. 

With Webb, we had to process the data differently for the first six months. As Webb was being commissioned, it took test images and data, which were confidential. What went on to become its first full-color images and data could not be uploaded to MAST. Even if data are uploaded to MAST without metadata, there is still a possibility that researchers might stumble on them. As a result, we had to use our in-house processes to ensure that only the people who should be able to access them could. 

Managing Webb’s early data was like being a master baker holding onto secret recipes, handing off the data only when the internal teams requested it. I was constantly checking all the locations where they could possibly be and making sure that nobody else could see them.

David Rodriguez: I remember being asked if I could see if a proposal was discoverable. I had no idea what that proposal was and it was nowhere in my system. I was confused at the time, but it was one of these files. It was so important the data weren’t in MAST, because even if we hid it, either a shrewd user or servers in Europe or Canada, which constantly harvest data from MAST, could have pulled and made the data public. When we were ready to prepare the files for public release via MAST, there was no single button to click to immediately make them public.

Anderson: Exactly. We had to reprocess all of the data from the images used in the press releases, and hand it to David’s team to add the metadata and formally upload over a few days. Some of the individual observations took 24 hours to process. It was a very busy time. It continues to be busy. Now, we’re regularly managing data from Webb. And we continually need to reprocess and replace existing data because the calibration files and software have been updated. We’re also finding out a lot about our data volume throughput and updates to the mission software. It’s all happening in rapid succession.

Rodriguez: Webb is also sending a lot of data! Its data known as spectra, for example, have challenged us. Some observations extract a spectrum for every star in a field, which leads to thousands of extractions in a single image. Suddenly we have folders with 100,000 files. These have taken more time to process and present to users. 

[Editor’s note: See an example of spectra pulled out from Webb’s First Deep Field.]

Anderson: We’re ingesting between 15 and 17 terabytes of data a month for Webb, and it goes fairly smoothly now. For context, a recent Hubble ingestion was about three terabytes in one month. Teams are constantly updating the software and providing new solutions to speed up these processes. Whenever we run into an issue, we send it off to the appropriate team, and each time somebody comes up with a better solution. As we become more familiar with Webb’s data, our work is becoming more routine.

Larry Bradley: My group writes data analysis software that allows astronomers to analyze and visualize the data. Our open-source software is typically downloaded and installed by astronomers on their computers, but one of our tools, Jdaviz, is also available for quick-look visualization and analysis of Webb data within MAST. We also work directly with researchers, answering questions, working through their issues, and addressing bug reports. We’ve gotten to the point where users have higher level questions, which is great. We also always want to know which new features they’re hoping to see. Overall, they are stress testing the data analysis software and we're getting a lot of good feedback. The spike in requests is good, because it means people are using the tools we’ve released.