![リガクル[RIGAKU-RU] Exploring Science](/ja/rigakuru/images/top/title_RIGAKURU.png)
Kiso Observatory is all in the mountains
Kiso Observatory sits on top of a beautiful hill bordering the towns of Kiso, Agematsu, and Mitake in Nagano prefecture, surrounded by Mounts Ontake, Kisokoma, and Norikura.
“I remember the awe I felt when I saw it for the first time 17 years ago in 2008. It was the famous Kiso Observatory I had read about in books.”
Associate Professor Shigeyuki Sako, who oversaw the major renovation that made the University of Tokyo's Kiso Observatory the most advanced in the world, reminisces.
The birth of Kiso Observatory dates back to 1974. Established as a large-scale facility of the University of Tokyo’s Tokyo Observatory, Kiso Observatory equipped with the 1.05-meter Schmidt telescope, which boasts an exceptionally wide field of view, has contributed greatly to astronomical research both inside and outside of Japan.
After an organizational reshuffle, the observatory got a fresh start in 1988 as a facility of the Institute of Astronomy affiliated with the School of Science at the University of Tokyo. The observatory has evolved over the years, incorporating the most advanced technologies in each period, making observations using photographic plate, CCD sensors, and digitalization. In other words, Kiso Observatory has promoted and watched over the development of astronomy in Japan, earning its legendary status.
Now, the focal plane of the Schmidt Telescope is equipped with a 190-megapixel CMOS camera named Tomo-e Gozen, which covers 20 square degrees of sky, the equivalent of 84 full moons. It records videos rather than still images, generating 30 terabytes of data in a single night. These unique observational data obtained by the world's widest field video camera are sent to an artificial intelligence (AI) integrated large-scale data processing system and shared in real time with research institutions worldwide.
“It has gone from being an isolated telescope in the mountains to an observatory connected to the rest of the world through a high-speed network, generating and transmitting more data than any other observatory. It has become the flagship of big data astronomy,” Sako proudly says.
Time rolls by slowly in a high school astronomy club
Where the Kiso River flowing southward from the foot of Mount Ontake meets the Nobi Plain is Inuyama City (Aichi prefecture), Sako's hometown.
“Perhaps because the city lights did not drown out the stars, I could often gaze at the night sky as a child with a picture book of space in my hand. When I was in junior high school, I took my camera outside and, alone in our backyard, took pictures of celestial objects.”
The high school he entered in Nagoya had an astronomy club, a rarity in the prefecture. Sako had been drawn to the glitz of the high school athlete lifestyle, but after much deliberation, he decided to join the much quieter and less conspicuous astronomy club.
“I spent my time in a relaxed atmosphere free from instructions, chatting with people and occasionally doing crafts or research. I now realize that those three years were a precious time for me. If I had not joined the astronomy club, I would have lost my connection with space and the universe, and I would not have developed the ability to create things from scratch, either.”
He then followed his vague interest in space and the universe, majoring in physics as an undergraduate student at Nagoya University and astronomy as a graduate student at the University of Tokyo. During his master's program, he traveled back and forth between Japan and the Subaru Telescope of the National Astronomical Observatory of Japan (NAOJ) in Hawaii, where his advisor was stationed.
“I spent my days working from morning to night with senior researchers and technicians at the site of the recently-completed Subaru Telescope. I probably did more chores than research. We would buy lunch in the mornings in the nearby town of Hilo and drive to Mauna Kea in the observatory car. On other days, we would stay in the laboratory and find things to do. I hated reading papers but was excited to be surrounded by cutting-edge equipment. After about two years of living in such an environment, I finally realized that astronomy was fascinating enough to devote my life to it. For my doctoral program, I completely moved over to Hawaii.”
Sako was assigned to a team developing an infrared instrument to be mounted on the Subaru Telescope and oversaw a part related to sensors.
“Taking pictures of the sky as a child, I was deeply moved by the feeling that my camera was seeing space and the universe for me when it captured images of things invisible to the naked eye. The same feeling came rushing back when the infrared instrument on the Subaru Telescope detected signals from astronomical objects for the first time. The infrared camera was looking at the universe for me. It was truly fascinating. I wanted to see more of the universe unknown to humanity. To be able to do so, I wanted to bring into existence the world's most sensitive instrument with my own hands. And this feeling grew stronger and stronger day by day.”
Sako recalls that at the Subaru Telescope troubles typical of starting up a large-scale project occurred daily. Many people came and went, and the days were filled with a whirlwind of passionate energy, like the lively atmosphere of a festival.
Launching the Tomo-e Gozen project
After returning to Japan, he took a position at the Institute of Astronomy. In 2008, at 32, he was assigned to Kiso Observatory operated by the institute. He was tasked with developing a wide-field CCD camera for the visible wavelength range for the Schmidt telescope to enhance the observatory's research output.
“Even though I had majored in infrared astronomy, I was asked to lead a project developing an instrument for the visible wavelength range. I had no idea what to do either from an astronomical or engineering point of view. So, I made rounds with the engineering staff and researchers specializing in visible wavelength astronomy. At first, I did not understand why I was put in a position to develop an instrument outside my expertise. However, my vision became clearer as I spoke more with people. I began to feel that I wanted to leverage my experience and create something interesting together with the team. KWFC, one of the world's widest field CCD cameras, was completed in 2012. I suddenly realized Kiso Observatory was full of talented people. It was like a free and energetic family.”
So, a dream was born of an integrated observation system where all the facilities, including the Schmidt telescope, would work together comprehensively, created by the diverse talent at Kiso Observatory. He also began enjoying managing the project alongside leading the astronomical and technological development. Under these circumstances, at 38, he launched the “Tomo-e Gozen Project” named after Tomo-e Gozen, a local female warrior who served Yoshinaka Kiso in the Tale of the Heike.
Searching for faint changes, not faint light
“One night, I thought I would try making observations with a CMOS video sensor mounted on the focal plane of the Schmidt telescope. As I did so, many shooting stars and artificial objects came into view, making me feel uncomfortable. I had been developing instruments and collecting data on space for a long time, but I had no memory of seeing shooting stars fly across the sky. When I thought about it, I realized that it could not have been any other way as the sensors up until that time had only been capturing still images of slowly unfolding astronomical phenomena. I remember my heart was pounding as the realization struck me. I wondered if we were missing out on many astronomical phenomena that happen quickly or flicker only for an instant.”
Almost all of the world's observatories for visible light used to use CCD sensors for their incredibly high sensitivity to capture even the faintest of light, which somehow had become the conventional wisdom of modern astronomy. However, not prioritizing sensitivity, Sako and his team decided to replace the CCD sensor with a CMOS sensor capable of fast readout and video recording. Sako says they “decided to abandon the desire to capture faint light and focus on capturing faint movement instead.” The Schmidt telescope at Kiso Observatory has an extremely wide field of view. By lining up several CMOS sensors on its vast focal plane, the researchers hoped to encounter dynamic astronomical phenomena that had previously gone unnoticed. This attempt was the first ever of its kind in the world.
“I was an expert in sensor technology. So, I did not feel that the large number of CMOS sensors was an obstacle and enjoyed working with the graduate students on the development. The sheer volume of observational data generated nightly posed the biggest challenge. It is impossible for a human being to look at each frame. Not knowing what to do and searching for answers, I turned to computer scientists and statistical mathematicians. To survey the sky widely and efficiently, not just the telescope but the entire observatory had to be highly digitalized so that the system could operate autonomously in an integrated manner. So, we had to renovate the facilities at the Observatory that were from the previous era... which required research funding. Looking back, we were all absorbed in the project and the number of our collaborators kept increasing. Everyone gave their best effort, getting things done and solving problems to achieve a single goal.”
Completed in 2019, “Tomo-e Gozen” is an integrated observation system consisting of the world's first wide field astronomical video camera equipped with 84 highly sensitive CMOS sensors and a powerful computer that carries out observations independently and analyzes data autonomously.
“The combined effort of experts from various fields at Kiso Observatory and the enthusiasm of the young students have made the experience much richer than I could have ever imagined. I was truly surprised by the evolution of this project,” says Sako.
Tomo-e Gozen and time-domain astronomy
The observation and study of temporal changes in explosive phenomena like supernovae and stellar flares and in moving celestial bodies like asteroids and meteors is called time-domain astronomy. It is a new field of astronomy made possible by recent technological developments such as larger sensors, improved computing power, and faster networks. Tomo-e Gozen has been optimized for time-based astronomy. The combination of a wide field camera and a video sensor enables Tomo-e Gozen to capture phenomena lasting only seconds over a wide area of the sky.
“Tomo-e Gozen has already shown us various new faces of the universe. For example, powerful red dwarf flares have been difficult to detect until now. It lasts only a brief moment: the brightness of the star doubles in under ten seconds. No one had ever imagined that such a phenomenon could occur. Now we know that they occur frequently all over the sky.”
Capturing rapid changes just after a supernova explosion is another area in which Tomo-e Gozen excels. Recently, combined with the X-ray telescope mounted on the International Space Station, researchers have successfully made high-speed simultaneous observations of a star falling into a black hole. Tomo-e Gozen has also contributed to the early detections of asteroids that could potentially collide with Earth.
“We have already detected about 50 events in which a chunk of rock the size of a bus passed by Earth at about the same distance as the Moon. We have also detected trails of satellites that are no longer in use, called space debris. I hope that observations made by Tomo-e Gozen will help resolve this problem as well.”
The autonomous and remote observations continue night after night. The vast amount of video data is transmitted to and stored on the cloud platform mdx at the University of Tokyo's Kashiwa campus using the high-speed academic network SINET. At the same time, AI software checks the data for interesting phenomena and notifies the researchers if it finds potential candidates. The overwhelming amount of data obtained from observations cannot be managed by human power alone.
“I mentioned that we were having trouble handling our data and asked for advice from computer scientists. They were intrigued by the problems and gladly offered their support, giving me a list of papers on cutting-edge AI technology. In the name of supporting science, the local government in Kiso also graciously provided us with fiber-optic infrastructure. I believe that Tomo-e Gozen was built by Kiso Observatory's open attitude toward science and its deep relationship with the local community cultivated over 50 years since its establishment,” says Sako.
Tomo-e Gozen is now a critical facility in the global network of time-domain astronomy. Sako calls this “passing the baton.” If Tomo-e Gozen at Kiso Observatory observes a phenomenon, it immediately transmits the discovery to the rest of the world through the network. Telescopes around the world immediately follow up with further observations. Tomo-e Gozen passes the baton, starting a worldwide relay. Recently, observations of non-electromagnetic waves such as gravitational waves and neutrinos sometimes serve as the starting point for time-domain astronomy. This is called multi-messenger astronomy, and it is garnering attention as a new field. In these cases, it is Tomo-e Gozen that receives information about gravitational waves and neutrinos, immediately following up with autonomous observations. Exploiting a global network is becoming crucial in modern astronomy.
Sako says he hopes that Kiso Observatory and his research group will continue to be a place where young people can test their ideas freely and contribute to the future development of time-domain astronomy.
Unknown unknowns — wanting to find something not even known to exist
When asked about his dream, Sako replies so.
“I am a dreamer. I want Tomo-e Gozen to capture an “unknown unknown” phenomenon. That is a phenomenon that we do not even realize we should know about or do not even imagine that it exists. It is interesting to detect events that physical models predict, but I would rather try to find something that has not even been predicted and start from an observation. Much has already been discovered during the long history of astronomy. However, many short time scale phenomena must have been overlooked. I believe our Tomo-e Gozen will find something "unknown unknown," something incomprehensible, in the large amount of data it generates. That is our raison d'être and our ultimate goal.”
And he addresses young people like so.
“These days, getting information on what you should or is safe to do is easy using a search engine or generative AI. I am not saying it is wrong to choose the safest options. However, I believe walking off the trodden path is a person's raison d'être, and doing so is simply fun. I encourage students to look for “unknown unknowns” in student life.”
He also says that “looking back, I feel like the path from my childhood of star gazing alone in my garden, through my high school astronomy club, and my life in Hawaii, leading to Tomo-e Gozen at Kiso Observatory, was guided by the people who supported me at each stage and by my steady and defiant attitude toward science.”
Looking up at the dome of the Schmidt telescope, which rises above the stone wall-like base, Sako whispers.
“From the outside, it is hard to imagine that a vast amount of cutting-edge observational data is being generated inside this dome. I like how low-key it is.”
※Year of interview:2024
Interview/Text: OTA Minoru
Photography: KAIZUKA Junichi
Related articles
