A Better Future for the Planet Earth
I. Birth - Infancy - Childhood
I was born in Nagano City in October 1934 and lived there until I was three; but I have no memories of my time there. My father was a bureaucrat at what was then called the Home Ministry, so we transferred around. I remember that my father was strict and taciturn. Immediately before World War II, when I was between 3 and 4, we lived in Kyoto. After my father was drafted into the military, my mother often brought us to visit her parents in Nara. My maternal grandfather was a principal at Nara Women's Higher Normal School and we had plenty of science books at home. I loved reading books about the natural sciences, especially about the earth. My family environment nurtured my interest in natural science and shaped my future path.
When I was small, I did some odd things that my older sister still kids me about. I was enrolled in the elementary school attached to Nara Women's Higher Normal School at first, and I said some strange things during the entrance examination. When I said, "I am the emperor," I was severely scolded since it was during the war, and at that time one had to be careful not to disrespect the Emperor. Then when we were taught the story of how the gods created Japan, I did not believe the myth about how Izanagi (god) and Izanami (goddess) had created the country of Japan. I learned from one of the earth science books how the terrain of the Japanese land had been formed, and I believed only what I could trust and verify. I thought that the most important thing was to think for myself and to confirm facts. I carried forward this way of thinking from my childhood into my career as an adult researcher. My kids showed many of the same qualities. Looking at them now, I can imagine what I must have been like as a child.
After my father was discharged from the army, he was transferred to Ehime, Kagoshima, Saga and many other places. I remember we often had typhoons in Ehime and Saga. There was a large camphor tree in our yard in Saga, and I remember vividly a very strong storm that tore away large branches. I lived in Saga when I was in the 4th grade of Kokumin Gakkou (elementary school operated during the Japan-China War from 1941 to 1947). I watched with excitement different types of planes, such as Douglasses and Martins flying overhead. We moved from Saga to Mito when my father was transferred at the end of February 1945, which was the fiercest time of the war. Ten days before the bombing of Tokyo (Great Tokyo Air Raid), we arrived in Mito after stopping over on the way in Kyoto, and moved in to an official residence for government workers. I was briefly enrolled in the fifth grade of the elementary school attached to the Ibaraki Normal School.
At the beginning of the school year in Mito, smoke from Allied bombing raids over Yokohama covered the sky and spread to Mito one raid after another. The school prepared for the bombings, and I remember that we had long, unscheduled vacations during which we lived through several air raids. I was frightened by the machine guns in a mopping-up operation from carrier-based aircraft that targeted the arsenals near Mito. In Saga, allied aircrafts flew high in the sky so we amused ourselves guessing the name of the aircraft, saying, "Hey, it's a Curtiss!" In Mito, however, I was really terrified. I could hear the whiz of the machine gun bullets.
On August 1, Mito was also heavily damaged by air raids. My father at that time was the Ibaraki Prefecture Police Commissioner. He was in charge of painting the office buildings in black providing camouflage patterns, and storing rice and other food. On the night before the air raid, my father had coincidentally invited his staff to our home party. When numerous illuminating flares fell, everybody at the party feared something awful would occur. On the very next day, there was a big air raid with flares. Mito was attacked in horrific air raids. I spent that night near the Naka River that runs through the middle of Mito and Katsuta. That was one of the most awful events of my life. Everything around our house was burnt, but our house miraculously survived.
When Japan lost the war, the best thing for me was the sudden freedom. Although I was a child, I experienced a sensation of great liberation. Everything was burnt down, even my school was gone; but I remember the fun we had playing around the wrecked machines at the plant destroyed during the air raids. Immediately after the war, we studied under big trees. Soon, though, classes were taught at the barracks, and then moved to classrooms that were built in the gymnasium that had survived the bombings. In this way, our education got its bare beginnings. Under such circumstances, we children enjoyed freedom in our own way. From the 6th grade of elementary school to the 1st and 2nd grade of junior high school, the school schedule was loose. During lunch break, for example, we were so free that we continued playing baseball unconcernedly even after the break time. Teachers just watched us playing. Around this time, teachers were prohibited from assigning physical punishment. That also increased our sense of freedom. In social studies classes, we were allowed to choose our own projects. I was very interested in ancient tombs and remains, and our group was excited whenever we found something. We were constantly hungry at the time because food was short, but it was a wonderful time because we could do whatever we liked to do.
I was a junior high school student when I became interested in meteorology. GHQ named typhoons in alphabetical order, always with female names. The most famous was typhoon Catherine in 1947. The embankment of the Tone River collapsed and water flooded the town. The radio and newspapers reported the flood widely. In the following year, typhoon Ione destroyed the embankment of the Naka River that flowed north of Mito and floodwaters inundated the city. I vividly recollect that the train was suspended because of the flood. Immediately before one typhoon hit, my class had the chance to make an inspection tour to the local weather station. After the visit, I noticed that I had forgotten my pencil and returned there by myself. There, one of the station staff taught me about the weather map. Since it happened to be before a typhoon arrived, I learned that by drawing a weather map, we could understand something about weather.
After the war, weather information was no longer a military secret, and the radio weather forecasts started. While weather information was important for those who worked on ships, such as fishing boats and merchant vessels, it was of no real significance to the general public. Such information was also not utilized for agriculture. Those who were interested in weather information like me seemed unusual. Radio weather forecasts only gave information about the next day, and that information wasn't always accurate. In the summer of 1949, when typhoon Kitty hit, I drew weather maps utilizing data from the radio every day during summer vacation. I showed the maps only to some of my good friends. It was probably from typhoon Jane in 1950 that newspapers started posting weather maps. I continued drawing weather maps at high school. Even my geography teacher asked me to draw weather maps in class. When I was in junior high school, I made a slide rule using pieces of cardboard on which I attached a scale capable of multiplication and division, utilizing a logarithm scale that I found in my sister's textbook. This surprised my teacher.
Starting from the study of meteorology, I advanced to learning science and mathematics systems in earnest, such as physics and mathematics. After the war, the sky was so clear and beautiful, that this motivated me to study astronomy. I was very attracted to the stories of constellations written by a famous astronomy ethnologist Hoei Nojiri. I memorized the names and learned other interesting facts about the stars. By reading books, I increased my knowledge of astronomy. I could point out some star in a constellation and say, for instance, that "Spica belongs to the Virgin, and its temperature is 10,000 degrees Celsius and its color is blue." When I graduated from junior high school, I was honored with a Special Award for mathematics, which was unusual because normally the Special Awards were given to outstanding students in art and music. Students who excelled in mathematics and English usually got Excellence Awards. However, my performance in subjects other than mathematics and science was not that good, so the teachers may have thought to recognize my outstanding performance in mathematics by creating a new special awards in mathematics. Generally, excellent students were excellent in all subjects, but in my case, my grades varied. I was especially fond of meteorology, astronomy, and mathematics independently.
II. The University of Tokyo Faculty of Science - The University of Tokyo Graduate School
Undergraduate Studies
I entered the Faculty of Science at the University of Tokyo in 1953. I have an indelible memory about my entrance examination. It was about my studious friends. During my high school days, some of my friends studied very hard after school, and I asked one of them why he was studying. He said, "I have a test next week." I realized for the first time that we should study before examinations. Because of the situation after the war, I never thought about studying hard before tests. However, I studied the subjects of English and world history a little bit during third grade summer vacation, being conscious of the university entrance examinations.
General education programs in the science fields at the University of Tokyo were divided into Natural Sciences I and II, Natural Sciences I was for students in the Faculties of Science and Engineering. Students (freshmen) lived in the Komaba Dormitory for one and a half years. One of my high school friends joined the rowing club, so I did too. When I became a junior after advancing to the specialized department of science, we had a regatta with other faculty rowing teams. The formal race was among eight-man sculls, while the informal one was with four-man sculls. Because only four students were available I participated in the four-man counter rowing race with a few members from the chemistry course.
When I became a senior, I had to decide my specialized course and only those who earned good grades were allowed to enroll in popular departments. The Geophysics course was not as popular, so I felt easy and enjoyed the boat club. However, when the competition for the course increased, I became a little anxious. When I was accepted, I was relieved. The capacity of the course was only 12 to 13 students. After enrollment, I noticed a shadow of sorrow fall over the students after Japan's defeat. Some students were obliged to quit school to find work. Others quit university to take over their family's businesses. Many students were veterans of Japan's overseas battlefields, who worked hard to enter prestigious universities, and then, were forced to give up their studies to focus on family businesses, unrelated to their chosen studies. Only a few of my classmates entered the university right after high school. Among these older, sadder veterans, I think, we must have appeared childish.
For the first year, I studied mathematics, dynamics, electromagnetics, and quantum mechanics, but geophysics was not included in the first year. In the last year, I studied seismology, oceanography and solid-earth geophysics. Meteorology was not yet a specialized field at that time and the class was given once a week.
Graduate Studies - Numerical Weather Prediction
I took my undergraduate degree in 1957, and was really happy to continue on to graduate school. It was a time when not all students could advance to graduate courses however much they might have wanted to. As I mentioned, some were obliged to work in their family businesses. In fact, I too applied for employment, with the Meteorological Agency and Japan Airlines. Before the interviews were scheduled, I talked to my father. I asked him if he could financially support me if I advanced to the graduate course. With his support, I turned down the interviews. With the money from my father, together with a scholarship and revenue from part-time work, I would be able to afford school.
At graduate school, I joined the meteorological research group under Professor Shigekata Shono. It was when Japan was adopting the numerical weather prediction system made possible by the computer developed by Von Neumann at Princeton University. Members of the university's meteorological research group created the Tokyo Numerical Weather Prediction Research Group with researchers from the Meteorological Agency and the Meteorological Research Institute. We carried out preparatory research using U.S. research results as a reference and engaged in introducing computer systems needed for numerical forecasts.
As a result, the Meteorological Agency decided to bring online a super-jumbo computer, the IBM 704, and to implement numerical weather forecast services from 1959. Japan participated in the project carried out at Princeton University, and the late Dr. Kanzaburo Gambo, who later became a full professor at the University of Tokyo after the Meteorological Agency, moved to Princeton in 1952, and played a leading role in the project. He is probably the first Japanese researcher to use computers in research. By the time he returned to Japan in 1954, all researchers in the field had studied the project and read the relevant papers. A group of researchers formed and started practical activities, recognizing the urgency of the work. We rented a computer paying an annual fee from 100 to 200 million yen. No one was quite sure if the system would improve weather forecasting or not; however, a tremendous amount of money was invested in the system based on the bold decision of Kiyoo Wadachi, Director General of the Meteorological Agency, that the system was important for the future. It was much later, probably in the 1980s, that Japan was finally capable of producing computers equivalent to these American models.
Even after we implemented numerical weather predictions, weather forecasts did not improve. Dr. Gambo moved to the Meteorological Agency and became the director. However, the capability of the conventional computers and observation data were insufficient; and we did not have satellites then. Under these circumstances, Dr. Gambo had a very hard time. They could not even correctly forecast the movement of low pressure. However, in those days, each grid covered large areas, using rough scales like every 380 km at minimum, making the outcomes very approximate. We also couldn't sufficiently deal with the Himalayan ranges in Tibet. They needed to forecast weather for three days within a few hours, which was very difficult because of these rough scales. I heard that at a workshop held in Boulder, Colorado in the 1970s, many participants expressed serious concern that they might have made some mistakes or they might have missed something. Numerical weather prediction wasn't put into practical use until sometime around the 1980s.
Recently, Hiroyuki Yoshikawa, former President of the University of Tokyo, divided technological development into three stages: The Age of Dreams, The Age of Nightmares, and The Age of Dream Realization. In the 1950s and 60s, numerical weather prediction was in the wonderful Age of Dreams when the University of Tokyo and the Meteorological Agency worked together. For the 20 years from 1959 to around the 1980s it seemed to be the Age of Nightmares. Staff in charge of numerical weather prediction provided data to forecasters at the Meteorological Agency for final calculations; however, they did not trust our data at the beginning. They didn't have anything to do with us because the data were not reliable. Nonetheless, I continued the basic scientific research solemnly.
Now of course numerical weather prediction system is firmly established and used as the basis for weather forecasts by local meteorological observatories in every country. I do have some reservations on this however. Current numerical weather prediction started in the United States and Europe and these countries are located in relatively higher latitudes. Forecasting weather is different between areas of high latitude and low latitude. In the Northern hemisphere where latitude is relatively high, by the effect of the earth's rotation, low pressure areas become counterclockwise vortexes, which affect pressure and wind. Based on their specific locations, the United States and Europe established their own basic concepts and theories for numerical weather prediction. However, Japan is located much farther south. This fact has two key elements. One is that the effect of the earth's rotation is not as significant as the U.S. or European areas. The other is that when vapor becomes liquid for instance during heavy rain, energy is generated. Since energy is added to the atmosphere, atmospheric currents are changed.
Recently, I have consistently pointed out that the approach to weather forecasts need to be adjusted according to latitude. The southern tip of Europe is the Strait of Gibraltar at about 36 degrees latitude, approximately the same latitude as Tsukuba, Japan. For areas south of Tsukuba, the model for higher latitude areas cannot apply. However, Europe and other developed countries at that time simply applied their model to all areas. When I thought this over after marshalling all elements, I saw there was much more to be clarified through research. For example, no studies had been made on conditions around the equator. Most research seemed to have been done for those who were interested in northern areas above 36 degrees latitude, and research under the latitude 36 degrees area had not been clarified. I recall an interesting episode that illustrates the problem. When Dr. Gambo visited local meteorological observatories in Sapporo, Sendai, Osaka, Fukuoka and other regions to convince them to use numerical weather prediction data, people in northern regions such as Sapporo and Sendai thought the data were "great," while people in the south from Osaka and Fukuoka complained that the data were "not reliable." Since Japan is located at a lower latitude (between low and higher latitude), it seemed we might be better served by weather forecast research of the tropical side. We needed to adjust the model to match the geographical conditions. We therefore needed something plus alpha, which was totally different from what had been achieved.
Graduate School Studies - What I Learned and My Master’s Thesis
For my master's thesis research, I tried to calculate emissivity in the infrared rays domain of the cloud layer utilizing aerially observed infrared radiation data researched as a part of the International Geophysical Year (IGY) projects. Unfortunately, I couldn't publish the collected data because they were identified as unreliable. The calibration of the imported measuring device was insufficient. However, I managed to extract passable data from my own adjustment. Later, I heard that my data was referred to by Dr. Shukuro Manabe when he explained atmospheric temperature structure based on the emission convection equilibrium theory.
At the end of July 1957, when I was in the master's program, there was a severe rainstorm in Isahaya, Kyushu. I participated in a joint project with members of the Meteorological Research Institute and researched the mechanism of concentrated torrential rainfall collecting data together. In addition to this research, after a time teaching as an Assistant Professor at the University of Tokyo, I participated in a study of the Ise Bay Typhoon in 1959 with the late Dr. Michio Yanai, Professor at the University of California, who became a leading expert in the study of typhoons and tropical meteorology. Another graduate program fellow was the late Dr. Takao Takeda, who later became a professor at Nagoya University, who, by utilizing numerical modeling, clarified for the first time the mechanism of the Mesoscale Convective System that causes long-lasting concentrated torrential rainfall. Research and discussion with these people gave me the chance to start a project developing an innovative atmospheric model that could directly present global tropical convection with a 5-km grid, which later became my starting point for the challenge of utilizing the Earth Simulator system.
After 1959, when numerical weather forecast came to be in practical use, research on this theme slowed down; and young members of the meteorological research group shifted their interest from short to long-term periods of weather forecasting and general circulation of the atmosphere based on physical principles. This shift was partly motivated by the first numerical experiment on the general circulation of the atmosphere carried out by Dr. Phillips, which was published in 1956. These young researchers, however, had already started from their own viewpoint focusing on changes in heat (energy) and water in the atmosphere as a step toward utilizing numerical weather prediction. Among these were Dr. Shukuro Manabe, a GFDL/NOAA research fellow from 1958 and the first recipient of the Blue Planet Prize in 1992, and Dr. Kikuro Miyakoda, his assistant at that time and a GFDL/NOAA research fellow from 1965 after his professorship at the University of Tokyo. Both were actively engaged in the world's first attempt at numerical prediction of rain.
Dr. Manabe and I spent a year and a half together as graduate students in the same study group. We learned a lot from young researchers at the meteorological study group the importance of advancing research on themes based on our own values rather than following advanced ideas or global trends. I appreciated the free and comfortable atmosphere which was similar to what I had experienced immediately after the war. I never felt that there were walls between me and my mentors. I am deeply grateful to Dr. Miyakoda and other researchers. The most important thing I learned from them was to think for myself. What I owed them was not only learning but also the importance of thinking independently. Dr. Gambo, who transferred to the Meteorological Agency from the University of Tokyo, was the leader at that time and many researchers followed his advanced theories and topics. Still, everyone here had the attitude that they were "different from the others," which accorded with my own character and I felt thankful for this.
Assistant at the Laboratory of Meteorology, University of Tokyo
In 1962, although I had not yet obtained my doctorate, I was assigned to the laboratory of meteorology as an assistant. I developed new ideas about numerical integration methods, which would later be called the Matsuno Scheme. This was a basic research to advance the next stage of numerical weather prediction utilizing new computers introduced by the Meteorological Agency. I also conducted numerical experiments on typhoon development with young graduate students.
Doctoral Thesis
While I was working at the laboratory of meteorology as assistant, I continued to seek a research theme for my doctoral dissertation. I finally completed writing it in 1965, and published it in the Journal of the Meteorological Society of Japan at the beginning of 1966. The dissertation was entitled, "Quasi-Geostrophic Motions in the Equatorial Area." In the thesis, I theoretically examined wave motion based on mechanics equations under conditions where wind direction around low pressure is reversed across the equator because Coriolis Force, which is decisive in the mid latitude, weakened in the Equatorial Area. The significant fact I clarified was that, besides the wave motion assumed from the middle latitude, there existed two particular kinds of wave motion One is the Kelvin Wave near the equator (measuring 3000 km from south to north) that moves to the east along the equator, and the other is the mixed Rossby-gravity wave that moves to the west. The Kelvin Wave was named after the 19th century physicist William Thomson (Lord Kelvin). Kelvin studied the wave along the banks of canals, and these were later named Kelvin Waves. There are no banks in the equator, but the Kelvin Waves show a similar motion because the equator plays the role of the banks.
These theoretically forecast wave motions were actually observed shortly after; the mixed Rossby-gravity by Mr. Yanai, assistant professor, and Mr. Maruyama, a graduate student, both from the University of Tokyo in 1966; and the Kelvin wave by professor Wallace, and Mr. Kousky, a graduate student, of the University of Washington. In the stratosphere, a strange phenomenon called quasi-biennial oscillation (QBO), a quasiperiodic oscillation of the equatorial zonal wind between easterlies and westerlies, had been observed. Careful analyses on a small amount of observation data were conducted to clarify this mechanism.
Afterwards, in the 1970s, a Kelvin Wave in the ocean was also found, measuring approximately 400 km wide from north to south. It was also clarified that the Kelvin Wave played an important role in the phenomenon of El Nino, which was then a popular research theme. Because active convective clouds in the tropical troposphere atmosphere cause complicated fluctuations, it is not easy to identify the equatorial wave. However, in 1994, Mr. Takayabu, a researcher at the National Institute for Environmental Studies (currently Professor at the University of Tokyo), brilliantly verified, based on comprehensive analysis of huge data, that the equatorial wave existed behind convective cloud fluctuation. Since then, utilizing more sophisticated methods, analyses on the convective coupled equatorial waves have been carried out.
Meteorological phenomena are complex, making theoretical forecasts difficult. Fortunately, in my case, my forecast has been verified, and this brought me great satisfaction. The Kelvin Wave is unique and its existence had not been predicted, so when Mike Wallace at the University of Washington found a Kelvin Wave in stratospheric data I was relieved to find that after all, I was not mistaken. Basically, it is no wonder that the Kelvin Wave was the phenomenon seen in the ocean. The wave was clearly identified and moving beautifully. During El Nino, the Kelvin Wave moves frequently. Again, I realized with great satisfaction that I had been correct. In fact, it seemed better to have named it the Equatorial Kelvin Wave, which has become the name of the wave at present. It has a commonality with the Kelvin Wave generated in canals; however, I think it is a unique wave in that that it propagates only toward the east.
III. Kyushu University, University of Tokyo, and Hokkaido University (1966-1998)
Kyushu University
After completing the doctoral program, I joined the faculty and continued my research while teaching in the natural course. Professor Sawada at the Kyushu University Faculty of Science invited me to join him as an assistant professor. I was engaged in research on dynamic meteorology in the stratosphere and mesosphere with young researchers, including assistants and graduate students, under Professor Sawada. It was the beginning of the Space Age, which started with the success of the first artificial satellite, Sputnik, launching into an elliptical low earth orbit in 1957. Research on the atmosphere and especially the stratosphere and mesosphere (later jointly called the middle atmosphere), which are at the edges of outer space between the latitudes of 11 and 80 km, was dominated mostly by the United States and caught the world's attention as cutting-edge research.
In 1957, when I graduated from the university, jet aircraft began regular passenger services in the United States, Europe, and Japan. The World Meteorological Organization, the federation of the Meteorological Agencies of member countries, decided to fly a radiosonde, a balloon, to observe the stratosphere. The balloons tended to explode when they climbed above 16 km; however, the organization was determined to obtain data until it could succeed. Until then, the troposphere had been considered important since clouds existed and rain fell in that area. However, after airplanes started flying, the stratosphere became a more important area for research for practical reasons since, even without rainfall, we could collect data about temperature, level of wind force, and occurrence of turbulence. In the stratosphere, a wide range of unknown phenomena and facts were discovered, which made the stratosphere a frontier for world researchers.
Kyushu University had a great research environment. I did not have many chores; and in summer, I went to the seashore to think quietly. It was also rare at that time, but Kyushu University had air conditioning. In a cool environment, we could read, discuss and ponder in the early evenings. On the other hand, we had problems with computer work since only the University of Tokyo had a computer at that time. I had to send punch cards there and to wait for results so that I could proceed with my work. Kyushu University was planning to install a computer, but had to postpone the plan for a long time because a U.S. fighter aircraft crashed in the neighborhood.
At that time, there were two unresolved questions regarding the meteorological phenomena in the stratosphere. In short, since a lot of data began to appear, two strange phenomena, which had not been anticipated by conventional meteorology, emerged. These brought decisive purpose to my research. One was a phenomenon not clearly understood called the quasi-biennial oscillation (QBO); the wind around the equator changes direction to the east and west nearly every other year. The other was sudden warming, and I was interested in both phenomena. I assumed that the Kelvin Wave might have played a significant role as a mechanism of the QBO. Richard Lindzen, researcher at MIT then and later at Harvard University, tried to verify the phenomena using the Kelvin Wave. Although I did my best to clarify the mechanism of this phenomenon at Kyushu University, Lindzen achieved success earlier than me. Sudden warming was a phenomenon that we had known about for a long time. Dr. Miyakoda, my senior in a sense as well as my dissertation supervisor, also worked on this theme. From the first, every researcher showed interest in this phenomenon because of the big question, and I was happy to be the one who verified its mechanism.
Invitation to University of Washington and Princeton University
Mike Wallace, who had first observed the Kelvin Wave and conducted research on the equatorial stratosphere, invited me to the University of Washington in 1968. The University of Washington was one of the best in the field of atmospheric science. Outstanding project leaders invited young researchers from around the world to promote research in the field. This helped to improve atmospheric science research at that time.
I spent one more year at Princeton University as a visiting researcher working with Dr. Miyakoda, Dr. Manabe, and Dr. Kurihara, who belonged to the National Oceanic and Atmospheric Administration Geophysical Fluid Dynamics Laboratory (GFDL/NOAA). During that time, we clarified the mechanism of sudden warming in the stratosphere. High and low pressures, usually seen in the troposphere, do not exist in the stratospheric atmosphere. Instead, the "planetary waves" on the scale of the Northern hemisphere are generated by geographical conditions and land-sea distributions, and reach upper latitudes via winter westerly winds and reshape the vortex of westerly winds circulating around the North Pole on a hemispheric scale. Occasionally, the vortex gets deformed significantly and splits, which weakens the west wind and changes it into an east wind. Along with this, the temperature in the Arctic region increases by 40 degrees C over a period of one week. I analyzed the mechanisms of this series of phenomena and eventually verified them with numerical experiments utilizing a simulation model.
I felt fortunate and happy to have been able to clarify the mechanism development of the sudden warming in the stratosphere, which, despite various attempts, had not been hitherto verified. It was considerably complicated by multi-layered mechanisms that made it hard to clarify. However, by approaching the core of the problem step by step, I finally made the correct judgement. At that time, I partly understood the steps I had taken. My progress was the fruit of discussions with researchers, including the young assistant Dr. Uryu. All of them understood these things very well and what we discussed together was instrumental in my success.
This proved that basic research could be the foundation to clarify changes in the natural environment. I think that my work was the fruit of these efforts. As a way of thinking, placing high value on basics led to practical use. When I was President of the Meteorological Society of Japan, I wrote that "the most fundamental research is the most practical research." What I meant was the importance of not seeking answers in hurry.
The thing that I appreciated the most in the United States was being able to use computers for research. Although my research did not require many calculations, it would have been difficult to research in Japan at the time. I was also pleased to be free from administrative work. I established the foundation of my research strategy through discussions with staff at Kyushu University and completed it in the outstanding research environment in the United States. In other words, I thought over the solution while I was in Kyushu and implemented my ideas in Washington and Princeton. Since Kyushu University provided a wonderful research environment, I don't know whether I would have moved to the United States if we had a good computer.
The University of Tokyo Faculty of Science
In 1971, I moved to the University of Tokyo as an assistant professor, invited by the late professor Gambo, who replaced professor Shono. I taught geophysical fluid dynamics to graduate students, my first attempt to teach in this field. This was the foundation for clarifying the mechanism of the planetary atmosphere and phenomena of the atmosphere and oceans. I also expanded my teaching to mesoscale phenomena targeting tropical convections. At the same time, I continued to research the middle atmosphere dynamics (stratosphere and mesosphere) based on my experience.
Just after I moved to the University of Tokyo at the beginning of 1972, I was invited by the United States Department of Transportation to deliver the keynote speech at the workshop on the ozone layer. The purpose of the workshop was to initiate a large project to investigate the effect of nitrogen oxide (NOx) on the ozone layer. In the United States, the proposed development of a supersonic aircraft, which would fly at an altitude of 20 km in the stratosphere, had been under examination. If the project was realized, it seemed possible that emissions from aircraft would destroy the ozone layer. The workshop was held to investigate this issue, prior to commencing the project.
This was probably the first full-scale study of a global environmental issue, and the first issue I had been involved in about environmental problems. This project connected my specialized study of the stratosphere to socially important environmental problems and the topic was to be a research centerpiece at NASA. For the whole world, research objectives were shifting away from basic science motivated simply by curiosity to nature. As for myself, I thought it would be too challenging to conduct a large-scale project focusing on atmosphere beyond the stratosphere at the small-scale meteorology research group of the University of Tokyo. I decided therefore to limit the research to theoretical issues. Meanwhile, overall, Japan's research on the stratosphere was ahead of other countries, and I continued research activities such as organizing the Middle Atmosphere Dynamics seminar jointly with my United States counterpart in 1982.
After Professor Gambo retired in 1984, I became a professor of meteorology at the University of Tokyo. I managed to establish a research system at the university utilizing numerical climate models, and, as the first step, I looked for an assistant professor who had experience in creating or using numerical climate models. I asked Dr. Akimasa Sumi, who was working at the Meteorological Agency Numerical Prediction Division, to join me. Since then, as a representative of the Japanese meteorological community, I have participated in many research projects, such as international collaboration research and the Earth Observation Satellite Project. Dr. Sumi went on to become a professor at the University of Tokyo. Without his support, I would not have been able to fulfill my role as a representative.
The University of Tokyo Center for Climate System Research (CCSR)
Global warming became an international political issue after the extreme heat wave that hit the United States in the summer of 1988. This led in 1991 to the establishment of the Center for Climate System Research (CCSR) at the University of Tokyo as a base for environmental research. I was its head until 1994. The center was supported by University of Tokyo President Akito Arima and had the aim of addressing measures and policies presented by the Ministry of Education to strengthen scientific research on solutions to global environmental problems, such as destruction of the ozone layer. The CCSR was a small joint-use facility, headquartered at the University of Tokyo, with four professors, four assistant professors, and two assistants. It was independent from any department. As the hub of climate research from universities across Japan, the center implemented climate change research and contributed to the 3rd IPCC report in 2001 and the 4th in 2007, focusing on climate change. I was in charge of the development and improvement of the basic Model for Interdisciplinary Research on Climate (MIROC) in the 4th report as well as the 5th report to be released this year.
After I became a professor at the University of Tokyo, I participated in many activities as a representative of the meteorology researcher community in Japan. The principal reason for my assuming a major role was that I was the oldest researcher remaining in Japan after other senior and major researchers had moved to the United States. Further, I felt a sense of responsibility, recognizing the social importance of meteorology in addressing global environmental problems. Given that my senior researchers have become world leaders in this field, I assumed responsibility as a representative of Japan.
Hokkaido University Faculty of Environmental Earth Science
I taught at the Hokkaido University Graduate School of Environmental Earth Science for four years from 1994 and participated in the creation of the Division of Atmospheric Ocean Science. Prior to that, in 1993, Hokkaido University established the Graduate School of Environmental Science (equivalent to a department), which specialized in researching global environmental problems as well as bringing up environmental specialists. The school set up four courses: Material Environment, Biosphere Science, Earth System Science, and Atmospheric-Ocean and Climate Dynamics. The Atmospheric-Ocean and Climate Dynamics course was established in 1994, and I was assigned as the head of the course. Together with the university members who had been involved in preparation, I was responsible for establishing the new course from the beginning, including the selection of specialized academics.
Looking back on my 32-year career working at universities, I consistently esteemed independence in my students and guided them to think by themselves and to exercise their ingenuity, rather than to chase academic trends. Many students were studying existing or published research results. I guided them to research only what they were curious about and they thought most important. The most wonderful aspect of my seniors was they carried through the principles exactly as I had guided my students to do.
IV. Frontier Research System for Global Change (FRSGC) (1997-2005)
Establishment of the Frontier Research System for Global Change (FRSGC)
For one year from 1995, I chaired the Earth Science Technology Task Force, subcommittee of the Science Technology Council of Aeronautics and Electronics and held discussions with specialists in relevant fields. This new task force was established by the former Science and Technology Agency to address global environmental problems that had become significant in the 1990s. I was the head of the subcommittee and drafted a report entitled, For the Realization of Global Change Forecasts, in 1996. In response to this report, the Agency established, as a new research system, the Frontier Research System for Global Change (FRSGC) with the goal of making global change forecasts. The FRSGC was a joint project of the National Space Development Agency of Japan (NASDA, currently Japan Aerospace Exploration Agency JAXA) and the Japan Marine Science and Technology Center (currently Japan Agency for Marine-Earth Science and Technology JAMSTEC) based on a plan setting 10 years for the 1st term and another 10 years for the 2nd term. This system started in 1997, and I led the team to complete a new research system.
This project was open to researchers both in Japan and overseas encompassing all fields of environmental studies, including climate change, water circulation change, global warming, atmospheric composition change, and ecosystem change. This was a unique research from the perspective of world standards. About 100 worldwide experts who joined through a public application process gathered and collaborated. By harnessing the unique characteristics of each group, this project took the initiative in the development of the earth system model, which included the effects of carbon dioxide emitted by humans circulating in the environment and absorbed by the ocean, as well as the effects of movement of the vegetation zone caused by global warming.
Development of the Earth System model
What is the earth system model? Let me start by explaining the numerical weather forecast model that was developed more than 50 years ago. Numerical weather forecast is the mathematical model that basically formulizes values such as atmospheric temperature, wind, pressure, and vapor in accordance with the laws of physics. Then, based on collected data, we make mathematical calculations. To do this, we divide the earth in grids, from north to south, east to west, up to down, and then express the figures relating to pressure, wind, and temperature in each grid in formulae or charts and calculate them. After observing the current figures, we predict figures in the near future based on the laws of physics. For example, we can forecast pressure worldwide after 5 minutes, then after 10 minutes, etc. In this way, we can forecast the temperature and wind direction, for example, in Tokyo after 24 hours. That is as if we have the atmospheric model in the computer. For example, we can predict changes in weather tomorrow by inputting the data of the current condition into the computer and manipulating the time one day earlier.
Starting with the weather forecast as a model, we can apply the same to the ocean. On the basis of the laws of physics, we can know the currents of the ocean. We learn the changes in speed, rises in temperature caused by the sun, and changes in salt content in ocean currents; salt is concentrated after vaporization, and, conversely, diluted by rainfall. The density of the current changes according to the salt content and causes extremely irregular movements. When wind blows, ocean currents are generated. As for the relationship between salt content and temperature, cool currents with high salt content run deeper while warm currents with low salt content run shallower near the surface. Considering these elements, we can establish a method of calculation in accordance with the laws of physics.
On land, trees influence the climate. Water doesn't evaporate from the ground, but is instead absorbed by trees through their roots and evaporated from their leaves. Accordingly, in modeling, we have to take account of how trees work. In addition to this, forests absorb CO2 that they require for photosynthesis, an extremely important process in CO2 circulation. CO2 is absorbed into seawater as well. This is also based on the laws of physics. Trees absorb CO2 and compose organic substances through photosynthesis. We include all such elements in modeling. The current model has become more and more complex, taking in various factors that have been discovered. This is why we call this the earth system model.
Another important factor is the distribution of broadleaf and evergreen trees. When temperature rises, cold areas become warmer and the population of broadleaf trees increases. Thus, the areas where evergreens grow also change. Until recently, we have not taken such factors into consideration. However, on the basis of a research into how broadleaf and evergreen trees change according to climate change, the earth system model calculates earth environmental changes accompanied by complicated factors of global warming.
When we develop models, we do not strictly follow the laws of nature; instead, we supplement the models to a certain degree. At times, this results in differences and discussions among the team as well as discrepancies with actual observations. This raises questions about the validity of the models, which becomes a serious problem.
It was Dr. Manabe who started calculations using such computer modeling. Up to a point, he originated everything himself, including ocean and ground humidity in rainy weather. Beyond that, the models became extremely questionable since calculations were not based on the laws of physics, and this called their legitimacy into question. Accordingly, Dr. Manabe stepped back from modeling, noting that it was "meaningless to include what we don't understand accurately."
The Frontier Research System for Global Change (FRSGC) started in 1997, but didn't cover all groups in the beginning. We added two groups to the System: the chemical research group in 1998 and the ecosystem research group in 1999. I thought they would become indispensable in the future. The most unclear research fields at that time were ecosystem and carbon circulation, and their models had not matured yet. Almost no research in these fields had been done either in Japan or overseas. As a result, our research proved to be of great importance. Through my experience, I understood that researchers tended to pursue only what they thought significant in their own areas. One reason why overseas researchers were sometimes not productive was that they prioritized their own study, placing less importance on cooperation with others. The FRSGC recruited researchers to develop the earth system model as a part of the Project for Sustainable Coexistence of Humans, Nature and the Earth. It was a unique project because researchers from various organizations gathered for the same purpose, which was one reason why the FRSGC succeeded.
In 1999, I was awarded the Carl-Gustaf Rossby Research Medal, the highest award in the field of atmospheric science, from the American Meteorological Society. Dr. Arakawa, Dr. Miyakoda, and Dr. Manabe, who carried out research in the United States, had received the same award. However, I was the first Japanese recipient whose research was based in Japan. Since the prize was awarded by the American Meteorological Society, most of the winners had been Americans, except for the two Englishmen and myself.
Participation in the Earth Simulator Project
In the 1980s and 1990s, climate changes that caused natural disasters and global environmental changes by human activities became serious problems. Under these circumstances, the former Science and Technology Agency launched measures to promote research on climate change prediction. The agency decided to make the best use of the strength of the large-scale research infrastructure of the FRSGC for oceanic and space observations and to connect their observed data to social needs. These were difficult research fields for universities. The summary of the project was reported in 1996 in a paper entitled For the Realization of Global Change Forecasts.
In order to forecast climate change, an extraordinarily powerful computer was needed. The Earth Simulator Division, subcommittee of the Computational Science and Engineering Council, announced the Earth Simulator Project in 1997, which planned to develop the world's fastest super computer. I participated in this project as chairman of the division and worked with researchers in the climate science field. In the report, we set the goal of developing an atmospheric model that covered the whole earth with a set of grids of 10km at most.
In response to the announcement of the Earth Simulator Division, the former Science and Technology Agency decided to implement the project jointly with three affiliated institutions: the Japan Atomic Energy Research Institute, the National Space Development Agency of Japan, and the Japan Marine Science and Technology Center. Hajime Miyoshi, who developed the world's fastest computer system (Numerical Wind Tunnel) at the National Aerospace Laboratory of Japan, was assigned as development director. He set the goal of developing a computer with a processing power of 5 T Flops (5 trillion calculations per second), by the scheduled time of 2002. The outstanding ability, passion and power of Dr. Miyoshi, the general director, brought the project to success. Dr. Miyoshi was originally an aeronautical engineer, and he used computers to calculate air flow, aerodynamic lift, and resistance around wings. Over time, he became increasingly interested in the computer itself and added knowledge to the field of computer science.
In 2002, they succeeded in developing a computer that exceeded their expectations, achieving a processing power of 36 T Flops, which was 90% of theoretically peak performance. It was five times the world record, and maintained the No.1 position for two and a half years. Dr. Miyoshi was cautious in setting the goal, but, eventually, he achieved a performance level of three times more than he had promised. During the development of the ultrahigh-resolution atmospheric model, which had been expected to test the maximum performance of calculation ability, young researchers were anxious about outcomes. They were concerned that if they couldn't achieve the expected result, they would face difficulty in future employment or in publishing research papers. However, the results were the opposite. They started in 2002, and produced an admirable result by 2004 to 2005. The development period for the Earth Simulator was four years. The cost of production was 40 billion yen, and the annual maintenance fee was 4 billion yen. I deeply regret that Dr. Miyoshi passed away in November 2001 right before completing the supercomputer.
We encountered unexpected problems in using the supercomputer for climate change prediction, the use that climate researchers had been longing for. We faced hardship in overcoming this obstacle. The research group, including researchers in solid earth sciences like earthquakes, who had proposed development of the Earth Simulator, kept asking officials in charge who would use it, and how it would be used. However, no discussions were held on these questions until the Earth Simulator was completed. The facts were as follows. In accordance with tacit consent or habitual mode of behavior of the government office territory, the basic policy was that since the computer had been developed by the former Science and Technology Agency, its use would be prioritized by its affiliated organization, the FRSGC. It would however be available for use by other institutions when possible. However, the FRSGC was a newly formed organization staffed by young researchers who had difficulty with research utilizing large-scale models. It was difficult for them to fully use the Earth Simulator unless the FRSGC worked together with the University of Tokyo CCSR and the Meteorological Research Institute. I continued to insist on the need for cooperation with the aforementioned research institutions, but my suggestion was rejected. I felt responsible for spending 40 billion yen of public money and was pained to consider that the project would not be used to its full potential.
This dilemma was, however, solved in an unexpected way. As a part of governmental reforms, the former Koizumi Administration ordered a large cut in the budget of Government affiliated corporations where retiring bureaucrats were parachuted. At that time the Earth Simulator was owned and the decision had been made for it to be operated by JAMSTEC, but outside users had not yet been decided. However, because of these new government measures, JAMSTEC would not be allocated a budget for operational fees. So, the agency created a mechanism that enabled payment of fees to the JAMSTEC; firstly, it created a new large-scale project, secondly it commissioned other institutions to research climate change forecast, thirdly, it ordered JAMSTEC to conduct mathematical calculations using the Earth Simulator, then, finally the project paid fees for those calculations to JAMSTEC. To implement this mechanism, the agency launched the Project for Sustainable Coexistence of Human, Nature, and the Earth, which recruited partners through public announcement. In the end, we established an ideal Coexistence Project, which might be called the "All Japan System," in which major climate research groups across Japan participated.
The members of the group were the University of Tokyo CCSR, the Central Research Institute of Electric Power Industry, the FRSGC, and the Meteorological Research Institute. They carried out model development and climate change forecast experiments making the best use of their own characteristics and strengths. As a result, they accumulated data which enabled the forecast of climate changes in the future from diverse aspects including global warming, global environmental changes, and extreme climate changes caused by typhoons and heavy rains. They were also able to predict risks of global warming and establish necessary measures. The results were published in a paper, and most of them were introduced in the IPCC 4th Report. I believe only Japan could have implemented forecast experiments in global warming and climate change in such an organized manner.
The major reason why the Japanese group could produce such excellent results was full use of the Earth Simulator. The 20-km grid developed by the meteorological research group was far ahead of any other in the world. In those days, a grid of 100-150kms was applied over the globe. Even application of 50-km grids had been regarded as producing superior resolution. However, Japan used 20-km grids for the entire world utilizing the power of the Earth Simulator. Otherwise, it would not have been possible to predict typhoons and heavy rains. The Earth Simulator made it possible to predict such concentrated rains accurately. In that sense, it exhibited overwhelming power.
Development of a New Non-Hydrostatic Icosahedral Atmospheric Model (NICAM) by the Earth Simulator
In addition to global warming predictions, in order to make the best use of the Earth Simulator, the FRSGC implemented its own project, to develop the super-high resolution New Non-Hydrostatic Icosahedral Atmospheric Model (NICAM) which could directly calculate convective clouds in the tropical zones by dividing the entire globe into 5-km or smaller grids. When the Earth Simulator project started, the advanced atmospheric model used for weather forecasting business and research around the world applied a 100-km grid. However, when we considered actual phenomena in the atmosphere, such as fronts, the 100-km grid model was not sufficient. Therefore, weather forecasting centers around the world tried to make smaller grids to exploit the improvements in the performance of computer calculation.
When we launched the Earth Simulator project to meet our goal of creating a capacity 1000 times faster than conventional computers, we aimed for a 10-km grid model, which was one tenth of the 100-km grid atmospheric model. A 10-km grid model was successfully applied for testing, and a 20-km grid model was used to investigate changes in the power of typhoons and achieved remarkable results.
However, I was skeptical of the 10-km grid model. For high resolution, the 10-km grid, which could express more detailed structure than rough grids, was surely better when targeting a front associated with low pressure in the middle altitude range. However, the 10-km grid model was insufficient since, at cloud clusters, a mass of convective clouds in the tropical zone, upward flows of convection at the core were generated every 10km. In general, a 2- to 3-km grid was considered necessary and at minimum, a 5-km grid was required to express the upward current. When we tried to establish a 2-fold higher resolution grid model, the number of calculation steps increased 10-fold. Further, since expressing convective clouds is based on the original equation added by abandoned effects, the volume of calculation would increase. As a result, employing a higher resolution model would be impossible. However, computers are making steady progress, and someday in the future, we will be able to model the convective clouds in the tropical zone appropriately. We should continue to try things no one has ever done with the coming new era in mind in which we will be able to use the world's fastest computers. We thought that would be the most valuable way to utilize the power of the Earth Simulator.
When I made this proposal, Masaki Sato, a graduate student at the University of Tokyo (currently professor at the University of Tokyo), fortunately showed interest in my proposal and played a leading role in a high-risk project. Soon, we were fortunate in finding several sharp young researchers who were interested in the project. In this way, in 1999, we started developing the world's first and only 5-km grid atmospheric model that covered the entire globe. To do this, we developed a grid model that segmented each triangle of the icosahedron. This is how it came to be named NICAM. We started from nothing but proceeded in the development step by step and achieved a good result from the first aqua-planet experiment, the results of which were published in "Science” in 2005. In our computer model, we recognized the reproduced image of an explosion of great masses of white clouds, a phenomenon which we could see only from satellite. I was deeply impressed and it is no wonder that the two referees evaluated our paper as "landmark." For eight years since then, we have maintained our unique position in this research and achieved various innovative achievements.
V. Japan Agency for Marine-Earth Science and Technology (JAMSTEC) (2004 - Present)
Following the shift of JAMSTEC into an independent administrative institution in 2004, the FRSGC, a research institution under the umbrella of JAMSTEC, was also reformed and renamed the Frontier Research Center for Global Change (FRCGC). I continued in the position of Center Director of the FRCGC until 2005, and after retirement, I worked for the center as an adjunct senior researcher.
VI. Promotion of International Joint Research
The World Climate Research Programme (WCRP) (1986 - 1994)
The World Climate Research Programme (WCRP) is an international joint research program launched in 1979 under the joint sponsorship of the World Meteorological Organization (WMO) and the International Council for Science (ICSU) aiming at forecasting changes in the atmosphere, oceans, and land for longer than one month and to forecast potential climate changes caused by human activities. From 1986 to1994, I promoted the program as one of 20 members of the Joint Scientific Committee, which led the entire program of the committee. As a Japanese representative of the committee, I played the role of a bridge between international research and various Japanese research institutions, such as the National Space Development Agency of Japan (NASDA) and JAMSTEC.
There were two major activities. One was the Tropical Ocean-Global Atmosphere Coupled Ocean Atmosphere Response Experiment (TOGA-COARE) field program held in the waters around the Marshall Islands between 1991 and 1992 to clarify the interaction of atmosphere and ocean in the tropical zone, focusing on the phenomenon of the El Nino. The University of Tokyo professor Akimasa Sumi and many other Japanese researchers participated in the program. The other was the Tropical Rainfall Measuring Mission (TRMM) satellite project, a collaborative venture advanced by the United States and Japan. The satellite was successfully launched later.
International Geosphere-Biosphere Programme (IGBP) (2005 - 2010)
While the World Climate Research Programme (WCRP) focuses on the physical aspects of the global environment, including climate change, the International Geosphere Biosphere Programme (IGBP), launched in 1987, is a collaborative international research program, which aims at addressing biological and chemical changes in the global environment, including changes in ecosystems and ozone layers. I was selected as a member of the Science Committee in 2005, and served in the position for two terms. The IGBP encompasses a wide range of specialized research fields and each specialized field was researched by a specifically organized international team such as the International Global Atmospheric Chemistry (IGCA) Project. Since I specialize in physics, I simply participated in general discussions.
4th Intergovernmental Panel on Climate Change (IPCC) Report (2003 - 2007)
In 2007, the IPCC published the Fourth Assessment Report consisting of contributions from Working Groups I, II, and III as well as the Synthesis Report of the three working groups. I was engaged in the following activities:
- Review editor for Chapter 8 "Model Assessment" in the Working Group 1 report
- Drafting author of the Technical Summary and Summary for Policy Makers in the Working Group 1 report
- Core team member for the writing of the Synthesis Report by Working Groups I, II, and III
I was the only Japanese who played so many roles in this project. One reason for this was that Dr. Susan Solomon, a stratosphere researcher who played the role of Co-chair of WG I, has been my acquaintance for many years. The IPCC was awarded the Nobel Peace Prize in 2007, and I was honored as co-recipient of the prize. I assume we were awarded because the Fourth Assessment was a huge turning point in assessing global climate change.
VII. Epilogue - My Regret
I am very sorry that the working positions of young meteorological researchers remain unstable. The problem of insufficient positions for post-doctoral fellows, which did not exist when I was young, is serious, and I deeply regret that this problem has not been solved yet. Dozens of young researchers are working in temporary positions, an issue to which I do not at this time have a solution. This is a serious problem we should address prior to research. As for research itself, I hope young researchers are able to continue to think for themselves. When I was young, I was afforded the luxury of taking the time to think through the problem in a relaxed way, almost like daydreaming. It may be insensitive of me to mention this, given that younger researchers no longer have that environment. I think this is extremely regrettable and how to change this system is a most difficult issue.
What I would like to do from now is to develop a numerical model for plate tectonics, similar to the atmospheric modeling, which was an issue before the problem of global warming emerged. This field of research has not been explored enough and the basic equations have not yet been established. I would like to simulate plate tectonics such that the plate of the Pacific Ocean sinks under the Eurasian plate. The basic equations have not yet been established for this phenomenon, which prevents us from creating simulations. I became aware of modeling plate tectonics decades ago, and I planned to start working on this theme when I moved to Hokkaido University. However, I had to commit myself to another difficult project that aimed at reducing greenhouse gas emissions by half by 2050, which was adopted at the Heiligendamm Summit in 2007. To achieve this goal, I started working with researchers at the Central Research Institute of Electric Power Industry and created an alternative proposal from the perspective of climate science. Under these circumstances, I was obliged to postpone my plans. Since I completed writing my paper last summer, I would like to resume my simulation plan as soon as possible. The ultimate source of my challenging a new research is, after all, interest in a research theme.