Learning from the Great East Japan Earthquake. Building a Disaster-Resilient Society with Satellite Data [Space Brothers Collab #23]
“Space exploration” describes a wide range of developments and their objectives. In this series, we will contemplate the present and future of space development with Astronaut Serika Ito, who has been appointed Chief Dream Officer of WarpSpace.
In this 23rd issue, we welcomed Professor Shirasaka Seiko from Keio University’s Graduate School of System Design and Management to discuss the use of satellites for disaster prevention and disaster-resilient social systems.
Space Development in the Post-3.11 Era
Serika: Today, 100 years after the Great Kanto Earthquake, we have Prof. Shirasaka, a specialist in System Design Management and emergency response to natural disasters.
Dr. Shirasaka is a professor at the Graduate School of System Design and Management, Keio University, and was also involved in founding an Earth observation startup, called Synspective. Why did you decide to start an Earth observation business as a startup?
Prof. Shirasaka: In 2015, I became a program manager for the Cabinet Office’s program, the Impulsing Paradigm Change through Disruptive Technologies (commonly known as ImPACT). This program aims to create innovative science and technology innovations that, if realized, will revolutionize industry and society. Our theme for ImPACT was disaster prevention. The program began four years after the Great East Japan Earthquake.
In the rescue of human life in an earthquake disaster, it is important to find and rescue victims as quickly as possible. However, due to timing constraints, Japanese Earth observation satellites were unable to contribute to the initial lifesaving efforts in the aftermath of the Great East Japan Earthquake. The Japanese Advanced Land Observing Satellite (ALOS) “Daichi,” which was equipped with a SAR sensor, passed over the Tohoku region just after the earthquake struck, so it was not possible to make observations immediately after the disaster.
A large amount of taxpayer money has been spent on government-sponsored space development. One of the reasons for this is disaster response. Despite all the talk about the need to develop space technology and launch satellites in preparation for disasters, it was still frustrating that when a major disaster occurred, we were unable to use them to save lives in the early stages of the disaster.
Serika: The ALOS equipped with optical sensors, made an emergency observation in the morning of the day after the earthquake, and helped us assess the situation in the entire disaster area, including areas where airplanes and helicopters were prohibited from flying due to the accident at the Fukushima Daiichi Nuclear Power Plant. Unlike SAR (Synthetic Aggregation Radar) sensors, however, optical sensors cannot observe areas covered by clouds. I heard that it took several days for the clouds to clear in the coastal areas and for us to be able to observe the full extent of the damage caused by the tsunami.
If we deploy a large number of satellites, the cost is simply increasing from manufacturing to launch. To reduce these costs, we need to make satellites small, and lightweight at low cost. In addition, it is necessary to develop SAR satellites because they must be able to observe day and night and in all weather conditions. This is why ImPACT has been working on the development of a small SAR satellite system that can make observations in tens of minutes to hours, even at night or in bad weather.
When we thought about how to implement the technology of this small SAR satellite system in society, the Nankai Trough earthquake, which is predicted to occur with a high probability between 2025 and 2045, came up in our discussion. According to the National Research Institute for Earth Science and Disaster Resilience, the Pacific coastal areas of Tokai, Kinki, Shikoku, and Kyushu will be affected. The government’s only hope to assess the extent of damage in this area is to use satellites, as airplanes and drones are completely inadequate for this task.
Furthermore, it is said that the government will set up a disaster task force at the Prime Minister’s official residence in approximately two hours to make initial decisions. In the case of a wide-area disaster like the Nankai Trough Earthquake, the government must quickly assess the damage and decide all the resources to allocate, such as how many rescue teams to send to which areas. If we really want to deliver information to the prime minister’s office within two hours, satellites that fly only occasionally will not help. We need to launch dozens of satellites and build a system that can deliver information at any time.
In order to realize this system by 2025, speed was a top priority. That is why we founded Synspective, a startup that develops its own small SAR satellites and provides satellite data solutions.
SAR Satellite Imagery Helps Rescue Lives at the Time of Disaster
Serika: Is it difficult to get satellite images of the affected area from other countries when a disaster strikes? Should we still have our own satellites?
Prof. Shirasaka: Of course, there is a framework called the International Disaster Charter, which provides help for other nations to cooperate in the event of a disaster. However, at present, even if we use satellites from all over the world, there are not enough of them, so we will not be able to get there in time for the two-hour period. Recently, overseas start-up companies have been active, but even so, there are not enough. It may become sufficient if there are 50 satellites in the world, but it would not be strange for Japan, which is exposed to high natural disasters risks such as earthquakes and tsunamis, to build a satellite system with a sense of urgency.
Serika: That is certainly true. I hope Japan can lead satellite utilization in disaster response. What kind of information would you be able to obtain if you could use satellites to observe disaster-stricken areas immediately after a disaster occurs?
Prof. Shirasaka: If we can assess the damage caused by the earthquake and tsunami, we can determine which roads and bridges are passable for rescue teams’ vehicles, and thus determine the route to take to the disaster area. In the Great East Japan Earthquake, car manufacturers released information on traffic records using driving performance data, and this was useful.
Serika: I see! But still, if we want to provide information within two hours, wide-area observation by satellite is effective. How accurate would satellite images need to be to provide information on roads and bridges that are passable for vehicles?
Prof. Shirasaka: 1-meter resolution (one pixel in an image corresponds to one meter). That is why ImPACT also aimed to develop a 1-meter resolution SAR satellite. The major difference is that defense applications require detailed observation of specific locations, while disaster response requires detailed observation of a wide area. For this reason, ImPACT was the first in the world to develop a small SAR satellite capable of high-resolution and wide-area observations.
Ordinary Earth observation satellites are evaluated based on two major axes: spatial resolution, which indicates how finely the Earth’s surface can be observed, and temporal resolution, which indicates how often the Earth’s surface can be observed. However, we, who are engaged in disaster response, evaluate satellites not on either of these axes, but on “responsiveness. This is the idea that it is important to be able to shorten the time it takes to acquire and provide information after a disaster occurs.
Serika: With 1-meter resolution SAR satellite imagery, the data volume is large, and it may take some time before it can be received on the ground.
Prof. Shirasaka: As you mentioned, high-resolution SAR satellite images have a large data volume, so ImPACT has developed a high-speed communication device. We were also considering the introduction of “onboard deep learning,” in which data is processed in orbit after the satellite observes the Earth’s surface and only the necessary information is transmitted to the ground.
For example, it is possible to extract information on roads that are passable in the event of a disaster and send it to the ground. However, since we can only obtain pre-set information, we cannot know what the condition of the stairs is at the same location when asked, and we have to observe the images again with a satellite. To solve this essential problem, it will be important to develop a system that can immediately transmit SAR satellite images to the ground, even if the original data volume is large.
Serika: In the case of natural disasters such as floods and landslides caused by typhoons, I think it will be possible to identify areas of high risk in advance based on the path of the typhoon and observe them with a satellite. On the other hand, in the case of natural disasters such as earthquakes, where we do not know when and where they will occur, it seems difficult to shorten the time required to provide information.
Prof. Shirasaka: When a disaster strikes, it is difficult to locate the affected area.
That is why we have considered, for example, a system in which sensors are attached to oceans and rivers, and when water levels change, the satellite can be instructed to observe directly without human intervention, as ImPACT has done. Currently, when a disaster occurs, local government officials go to the disaster site to check on the situation, and it takes time to assess the situation. If a system that can automatically task observation instructions to a satellite is realized, the time required for human intervention can be reduced. However, under the current Japanese legal system, it is difficult to obtain permission for such an initiative.
Thus, to consider responsiveness, we need to design the entire system, which will reveal issues not only in technology but also in the legal system. Through my activities at ImPACT, I realized the possibility of creating a system that can realize its objectives through the use of technology by considering not only local optimization but overall optimization.
Serika: So this is connected to your specialty of system design and management, isn’t it?
System Design Management, the study of interaction
Serika: What do you mean by “systems” in systems design and management?
Prof. Shirasaka: In Japan, the word “system” has a strong sense of IT. The same perception is actually held in many countries around the world, not only in Japan. There are two main types of systems: “engineered systems” created by humans and “natural systems” that originally exist in nature. Both have in common that there are multiple elements and that something is created by their interaction.
If I put it this way, almost everything in the world is a system. We can say that communities are systems because of the interaction between people, and laws are also systems because they are made up of multiple rules. There is a professor at Keio University’s Graduate School of System Design and Management who is researching happiness studies!
Serika: So it is an academic discipline that deals with a wide range of topics, regardless of whether it is in the humanities or the sciences? How did you come to study system design and management?
Prof. Shirasaka: I majored in aerospace engineering at graduate school, and after that, I was involved in satellite development at an electronics manufacturer. Keio University’s Graduate School of System Design and Management was established in 2008 as Japan’s first graduate school specializing in system design and management.
Prior to its establishment, a summer school on system design was offered from 2004 to 2006, where I taught methodology for large-scale complex system development as a part-time teacher, which is how I came to be appointed at the school. Many of the participants in this summer school were involved in space development. The Graduate School of System Design and Management was created based on this summer school.
Serika: Are there any graduate schools abroad that specialize in system design and management?
Prof. Shirasaka: There is one at the Massachusetts Institute of Technology. In fact, they learned about it after the decision was made to open our school. In addition, there are two other schools in Singapore.
Serika: Even looking globally, there are only a few universities that have a Graduate School of System Design and Management, right?
Prof. Shirasaka: That’s right. Systems design and management is a field that does not fit well with academia. Universities are places to pursue expertise. In general, they pursue a theme in depth. I think it is difficult for a complex academic field like systems design and management to expand. On the other hand, since it is close to social implementation, it is easier to collaborate with commercial sectors.
Serika: From a global perspective, where does Japan stand in terms of system design management? Is there anything that Japan can learn from overseas trends?
Prof. Shirasaka: I think the U.S. is leading the way in system design from a systemic perspective, or in other words, in system design that considers overall optimization. In the world of the latest system development, there are many areas where we are still learning from the United States. However, it is not so simple to say that everything originates from the United States. There are times when the American way is better, and there are times when the U.S. adopts the Japanese way. By repeating this back and forth, more advanced social systems have been created.
The next move was in 1994. It was a time when Japan was in a bubble and globally competitive. The U.S. studied the source of Japan’s competitiveness, and it was, after all, the so-called Toyota Motor Corporation’s “kaizen” (efforts to improve efficiency by having front-line employees take the lead in sharing opinions and ideas) that caught their attention. Until then, systems engineering had focused on what was called operations research, which incorporated optimization methods using mathematics into management. However, when it was revised in 1994, a standard process for implementing “kaizen”, rather than operations research, was proposed. The U.S. applied “kaizen” to systems engineering, whereas Japan had adopted it for on-site work processes.
The U.S. is still advanced, but it is evolving while incorporating good things from around the world. Now is the time when Japan is trying to introduce the good things from the U.S.
From Disaster Response to Creating a Disaster-Resilient Society
Serika: From the perspective of system design and management, what do you think is necessary to create a disaster-resilient social system?
Prof. Shirasaka: Essentially, we want a society that is less prone to disasters. Perhaps we are the cause of making natural disasters more likely to occur and are now thinking about how to make it easier to cope with disasters when they occur, but we have yet to address the question of why so many disasters occur in the first place. We need data to track down the causes of disasters, and we need to analyze what is happening. We need to find out what we can do and what we cannot do (planetary boundary) for the global environment. However, disasters will happen until we find out what to do and take countermeasures, so we are taking steps to save lives when disasters occur first.
And now I am concerned about what to do with the lunar version of the planetary boundary. It is only when the global environment has deteriorated so much that we have begun to be aware of the planetary boundary. We are going to destroy the lunar boundary, aren’t we?
There may be no flooding on the Moon, but there is the possibility of unknown disasters that don’t happen on Earth. In the future, when humans start living on the Moon, we may have to take disaster countermeasures. Natural disasters have already occurred frequently on Earth, and if human activities are causing the degradation of the global environment, it is important to consider the lunar planetary boundary in advance so that the same thing will not happen again on the Moon.
Serika: I see. So, what should we do to find the planetary boundary for the Earth and the Moon?
Prof. Shirasaka: There are many things we do not yet understand, including the Earth’s environmental system. I think that many people have been through school education in which they were given questions with known answers and were taught how to solve them efficiently. But this era is over. We should train people to find out what needs to be done. It is also no good if expertise is confined to a narrow field, but this does not mean that specialists are not needed, people with “expertise to unite specialists” are becoming necessary. It is precisely such human resources that our graduate school is fostering. Recently, entrepreneurship education for elementary and junior high school students is beginning to be recommended, isn’t it?
Japan is a country where education for working adults has not yet matured. Although the rate of university enrollment is high by global standards, the rate of master’s degree and doctoral degree attainment is quite low among developed countries. This means that learning stops at the undergraduate level. While some companies may offer training, much of it is for operational efficiency. With such rapid changes in the social environment and the 100-year life period, it is difficult to cover the remaining 80 years of one’s life with the knowledge gained at the undergraduate level. That is why Japan, as a society, needs to create an era in which people can learn while they work. Because everyone in the world is learning.
Also, even though new perspectives lie outside of their expertise, specialists can only think within their own expertise. So, how can we get outside of our own expertise? However, it is not enough to simply have a community of people from different fields. The Graduate School of System Design and Management is researching mechanisms for successful interaction. If we can provide a mechanism for a diverse community to function, it is more likely to generate ideas that have never existed before.
Serika: So, to solve global social issues, it is important to reform the educational environment and create a mechanism that allows diverse people to discuss issues beyond the boundaries of expertise. Thank you very much, Dr. Shirasaka!
This month, which marks the 100th anniversary of the Great Kanto Earthquake, we brought you a conversation between Astronaut Serika and Professor Success Shirasaka of the Graduate School of System Design and Management, Keio University. Please take this opportunity to think about what you can do to create a society that is resilient to disasters.
Writer: Haruka Inoue