Meet the researcher

Prof. Toshio Yamagata

Principal Research Scientist, Japanese Agency for Marine-Earth Science & Technology JAMSTEC

Professor Toshio Yamagata is one of the worlds leading climate scientists, his research findings include the discovery of the Indian Ocean Dipole. We had the pleasure of talking to Toshio last week. Here is his fascinating story.

Toshio, how did you end up becoming interested in climate science?

My hometown is about 100 kilometers north of Tokyo, and in the winter strong monsoon blows down from the mountainous areas. When I was a child, the dust from the fields was amazing. Thunderclouds come down with lightning from the mountainous areas every summer in the afternoons every day. In the spring, flowers bloomed in the gardens of each family, and in autumn, the foliage of deciduous trees was beautiful. My parents liked gardening, and I naturally became interested in the changing seasons.
At the University of Tokyo, I majored geophysics as a field where research can be done while enjoying nature. My supervisor was Professor Kozo Yoshida, who is famous for his research on coastal upwelling in the field of physical oceanography. I was interested in mathematical physics, so I chose the field of geophysical fluid dynamics (GFD), which treats fluid phenomena that occur on a rotating, stratified planet.

My younger work at that time includes the weak non-linear theory that explains why anticyclonic vortices keep longevity in the ocean and Jupiter, and the mechanism by which planetary waves become unstable to form vortices. In 1976, the GFD program of the Woods Hole Oceanographic Institution approved me to stay there for three months as a GFD fellow. This was an amazing opportunity to meet many giants in the field, such as Drs. Joe Pedlosky, Peter Rhines, Melvin Stern, Jack Whitehead, Geroge Veronis, Andrew Ingersoll, Willem Malkus, Louis Howard etc. They opened my young eyes to the world. In particular, main lecturer was Dr. Dick Lindzen; I was impressed by his simple Albedo-feedback model for climate. I didn't understand it well at the time, but I think it may have shaped the deep psychology of becoming a climate researcher.

After turning the age 30, I gradually wanted to deal with real phenomena that occur in the atmosphere and the ocean to understand the meaning of seasons, I daringly sent a letter with my articles to Dr. Kirk Bryan of GFDL at Princeton University in the United States, asking possibility of staying a couple of years as a visiting researcher.  He immediately accepted me.  There I met Dr. George Philander, and decided to go deeper into the study of tropical climate. This set the course of my life in science; it was in my early 30s. I moved to the US with my wife and my three-year-old son. At that time, I thought that the United States might become a place for my research. In retrospect, I think there are decisive moments in life that determine what follows.

When did you become suspicious that the Indian Ocean had a see-saw condition similar to the El Niño - La Niña phases in the Pacific Oceans?

The summer of 1994 was extremely hot in Japan, and a research team investigating the cause was formed by meteorologists. I had been studying El Niño model since I was in Princeton, so I joined the group. A little later, the AGU Western Pacific Geophysics Meeting was held in Brisbane, and I had the opportunity to exchange ideas with Dr. Gary Meyers of CSIRO. He informed me that there were abnormally dry conditions in Australia, and, according to his XBT section data, the eastern Indian Ocean was also colder than normal. We checked the oceanic semiannual Yoshida-Wyrtki Jet (which is trapped along the equator and flows eastward only during the monsoon break season: spring and fall) using the satellite data (SST and altimetry data) and found there was almost no eastward Yoshida-Wyrtki Jet in fall of 1994. It was amazing!  This is the beginning of the story. It turned out that the unusual westward winds along the equator from early summer to late fall hindered the evolution of the fall Yoshida-Wyrtki Jet. By the way, the Yoshida-Wyrtki Jet (equatorial long Kelvin wave) was predicted theoretically by my supervisor Prof. Kozo Yoshida in the late 1950s and reported by Prof. Klaus Wyrtki in 1973 using the ship drift data. I was excited at finding this unusual event similar to El Nino in the Pacific and started checking all available atmospheric and oceanic data with my young postdocs from India. Those are Saji, Vinay and Behera. I named this Indian Ocean Dipole Mode and introduced the IOD index. There is another interesting story behind the scene just prior to publishing our paper in NATURE, but this will enter the realm of science historians.

Positive IOD - weaker westerly winds, cooler ocean temperatures and less convection reducing the moisture in the atmosphere over nort west australia

So a casual conversation led you to look at linkages with Australian rainfall?

Yes, the role of my old friend Gary’s suggestion at an early stage was very important.

How important was computing power in data analysis when these discoveries were made?

Negative IOD - Intensified westerly winds, warmer waters around australia, increased convection and increased chance of rain

At that time of the discovery, we started the big supercomputer project called the <Earth Simulator> with the Science and Technology Agency and NEC. However, it was not completed. We just adopted a very simple analysis using an ordinary computer.  What is most important is the curiosity to notice strange things based on solid knowledge of GFD.  Dr. David Anderson described our work in Nature in such the way that it was amazing we still can discover things with a simple method.

Farmers are curious as to how the size of the IOD regions and locations are chosen – can you explain in broad terms how this was done?

The IOD is an Indian version of El Nino/La Nina. Since I developed a simple theory of El Nino in the early 1980s in Princeton, I immediately recognized that the IOD is a “basin-wide” “equatorial” phenomenon. One problem which annoyed me was the existence of the monsoonal Somali Jet as the western boundary current. The mechanism of this oceanic current is totally different.  So, we deliberately excluded the region off Somalia.

Do you think an air pressure measurement, like the SOI would help communicate the Indian Ocean?

We prepared the atmospheric component of IOD, but we could not include that part in our Nature paper. We tried to publish it in the journal of Royal Meteorological Society but it was declined because the editor thought that the IOD was a dubious concept. So, I published it with Dr. Behera in Journal of Meteorological Society of Japan with a slightly shocking title to show how the pressure at Darwin is disturbed by the IOD. Some skeptics still thought IOD in the Indian Ocean was just a response to El Nino/Southern Oscillation in the Pacific. Later on, we, and many others, published so many papers on how IOD influences the climate in many places on the globe.

Saji, N.H, Goswami, B.N, Vinayachandran, P.N & Yamagata, T. (1999) A dipole mode in the tropical indian ocean

The first IOD scientific paper, published in 1999 of which you co-authored has been cited an incredible 4,400 times, yet this climate driver was not acknowledged by the Australian Bureau of Met in communications until around 2012. Can you remember some of the debate in the scientific community about the reality of IOD?

There was a tough time.  But almost all my highly cited papers in my career were originally rejected.  I know the originality denies common sense (Please see my address from Dean’s office in 2011 (right after the Tohoku disaster) ). I forgot the exact date. Perhaps, it was the early 2000s when BMRC (at that time) organized a special session and invite me, perhaps to ask me to retreat from the IOD concept. However, I was prepared well with carrying more than 200 OHP sheets (so heavy!) to respond to all questions expected. Almost every week, I had a strategic meeting at FRSGC with my young postdocs and they helped me a lot.  I remember I said at the opening of my talk at the BMRC conference that “this is a boxing game and I am afraid of the home town decision. However, I find one lucky aspect; the judge is my friend Gary Meyers who understands my work.” One of the skeptics Dr. Neville Nicholls was there.  He was very close to the concept of IOD but, unfortunately, he mixed up several phenomena and retreated his idea on the Nichols’ dipole. After the conference, he, with his wife, invited me to a nice restaurant and I remember we had a wonderful evening.  He said, “Toshi, as you know, I cannot support your IOD explicitly but I never criticize it anymore.” I admired his attitude and said,” You have a real knight spirit.  He replied, “ You have a real Samurai spirit.” I was really happy to be a scientist.

How do you think technology can help better predict the IOD, as it’s a fairly tricky driver to nail down from an Australian climate point of view – would more comprehensive ocean monitoring help?

IOD is more difficult to predict than El Nino/Southern Oscillation because of many players of different time and space scales at one stage.  However, our predicting skill is improving day by day thanks to efforts of climate research agencies and institutions. We have just published a paper which discussed why we were successful in predicting the 2019 super IOD; we have realized the El Nino Modoki (another important climate mode in the Pacific) played an important role in triggering as well as strengthening the IOD. Efforts to develop the Ind-Pacific Ocean observing system are very important and we need to support the international partnership fostered by IO GOOS and WCRP.