Léo RamosOn May 22, 2016, professor emeritus Sylvio Ferraz Mello, a São Paulo native and former director of the Institute of Astronomy, Geophysics and Atmospheric Sciences of the University of São Paulo (IAG-USP), was in Nashville. The purpose of his visit to the country music mecca was not to watch shows. Ferraz Mello went to the United States to participate in the annual meeting of the Division on Dynamical Astronomy of the American Astronomical Society, which awarded him the Brouwer Award, a prize granted to researchers who have made important contributions to the field.
Dynamical astronomy studies the movements of celestial bodies, such as satellites, planets and asteroids, governed principally by the gravitational interactions between these objects. This is not an area that tends to generate headlines in non-specialized publications. But its theories, equations and models are the basis for explaining why the solar system and, more recently, the sets of exoplanets are in their current configurations.
A physicist by training, Ferraz Mello is member number one of the Brazilian Astronomical Society (SAB). He completed his doctorate at the University of Paris (Sorbonne) in 1967. From 1987 to 1994 he was adjunct coordinator of the FAPESP Science and Engineering area. During his career, he has studied the orbits of satellites, asteroids and, most recently, exoplanets. His models help explain such questions as why there are so many asteroids in the solar system with an orbital period of eight years and so few or almost none with a period of four or six years.
One of the phenomena that Ferraz Mello has focused on most is resonance, a type of gravitational influence that a celestial body has over another when their orbital periods are commensurable. Or in other words, when the period of one is a rational proportion of another, such as in the case of asteroids that take six years to orbit the Sun, half the time it takes Jupiter to complete the same task. In this interview, in addition to discussing research, Ferraz Mello recalls some incidents from his past, including his time as an astronomer at the Technological Institute of Aeronautics (ITA), the National Observatory (ON), his long career at USP and his work on the task force that chose the site where the Pico dos Dias Observatory of the National Astrophysics Laboratory (LNA), would be installed in the 1970s. “We found the place almost by chance,” he recalls, still active, in his office at IAG, on the eve of his 80th birthday in October 2016.
| Age |
| 80 |
| Specialty |
| Dynamics of our solar system and of other planetary systems |
| Education |
| Undergraduate degree in physics from the University of São Paulo (1959); doctoral degree in astronomy from the University of Paris/Sorbonne (1967) |
| Institution |
| IAG-USP |
| Academic production |
| 137 articles, 40 book chapters, 19 books as author or organizer. Advised 17 master’s students and 23 doctoral students (one in process). |
Did you receive the Brouwer Award for a specific project or for lifetime achievement?
It was basically for my work on asteroids. Two other issues also probably influenced the decision. One was the importance of the dynamical astronomy community here in Latin America today. Three years ago, the Division on Dynamical Astronomy of the American Astronomical Society held its annual meeting in Paraty and was shocked. It was their best-attended meeting. There are few meetings here. So, when one is held, everyone attends. I think that my contributions to the community’s journal, Celestial Mechanics and Dynamical Astronomy, were also important. I have been editor-in-chief since 2001, a position I will leave in 2017.
In 1983, an asteroid was named after you, the 5201 Ferraz-Mello. Why were you honored in this way?
Practically all of us in the dynamical astronomy community have an asteroid. It is not that big of a deal. The discoverer of that one, Ted Bowell, is a colleague with whom I have always been on good terms and he honored me in this way. It is a unique asteroid. I was very happy and began to study it. I discovered that it had notable dynamic characteristics. It has a period of about six years, is resonant, and has a very elongated orbit. It was observed by SOAR [Southern Astrophysical Research Telescope, in Chile] by colleagues in Rio de Janeiro and we discovered that it could be an extinct comet or one that never caught fire. In two years, it will come very close to Earth and we will be able to observe it well. We will be able to check if there is a gas halo around it. When it approaches the Sun, it could heat up and form a coma [a fuzzy halo around the nucleus of a comet]. This would indicate that it has some of the properties of a comet.
Why did you decide to study the phenomenon of orbital resonance in asteroids?
Before, I worked with the dynamics of satellites, a relatively thankless area, extremely important but small. I knew almost everyone. There must be fewer than 100 senior researchers in this field in the world. It is a specialty that requires very precise knowledge in order to support space missions. At one point, I saw that there was some similarity between a specific satellite problem and asteroids. I decided to take advantage of the connection. At that time, there were about 40 or 50 known satellites, while over 1000 asteroids had been identified. It was, therefore, a richer topic, with a much larger community. I began to work with asteroids in the 1980s.
But what exactly attracted you to this field?
I saw that there were unresolved problems related to asteroids that deserve systematic study. There was also a favorable circumstance. I had a very good group of graduate students at IAG. Nearly all had FAPESP scholarships and two of them received several awards after earning their doctoral degrees. With this group, I set up a thematic project that allowed me to buy a top-notch workstation with high processing capacity. I then had computational power greater than many colleagues abroad. The HP Risc processors had just been launched and one of the first machines came here. We were able to study several asteroid problems and solve some.
What problems?
I always use round numbers when speaking, to make things easier. Jupiter takes 12 years to orbit the Sun. The periods of the asteroids range from two to 12 years. Some are in orbital resonance with the planet. They have periods commensurable with that of Jupiter. So, they always cross paths with Jupiter at the same point in its orbit. Asteroids that are not in resonance pass by Jupiter in a different point in its orbit each time. Resonance significantly changes the movement of the asteroid, amplifying its orbit. This saga began with studies by other researchers on asteroids with a period of four years, one third that of Jupiter. They wanted to know why there are no asteroids with four-year orbits in the solar system. This is called the Kirkwood gap.
Personal archive
Ferraz Mello in 1997 with former students Daniela Lazzaro, Tatiana Michtchenko, Fernando Roig, Cristian Beaugé, Alessandro Simula, Roberto Vieira Martins, Tadashi Yokoyama and Rodney GomesPersonal archiveCould you explain this further?
When we looked at the distribution of asteroids, we found bodies with longer or shorter periods, but none with four-year periods. Why were there no asteroids with an orbit one third Jupiter’s period? This question was answered in the 1980s before I began to study asteroids. The resonance provokes chaotic reactions in the orbit of a body. The asteroid orbits Jupiter but, since it receives a gravitational impulse from the gas planet in the same location each time, its orbit begins to deform. It becomes extremely elongated. The asteroid begins to travel through an area much larger than our solar system. This could lead it to a near collision with Mars. When they collide, it is no longer an asteroid with a four-year period. It still exists, but with a different orbit. In our first project studying this issue, we saw that the situation was much more drastic than our colleagues had found.
Drastic in what sense?
In this situation, the asteroid might not only cross the orbit of Mars, but also that of Earth. It can pass less than one million kilometers (km) from Earth, which is very close. Nothing would prevent one of these objects from colliding with Earth one day. This does not necessarily mean a physical collision, but simply passing very close by. If this occurs, the asteroid would be diverted from its trajectory and into a new orbit. This type of instability is common in many asteroids. Then, I tried to use the same method to study asteroids with a period of six years, but the approach did not work. The orbit of these asteroids did not grow enough to lead to a collision with other planets.
What did you do then?
I began to study the family of asteroids with an eight-year period, equivalent to two-thirds that of Jupiter. This case is a little different. Instead of a gap, there is a group of asteroids with this period. Many people tried to understand why this was so. Then, as I said before, we bought a good computer and improved the model for this situation. The more complicated the model, the more complex the equations. In order to study the absence of asteroids with four-year periods, we had adopted a simple, three-body model. We put the Sun, Jupiter and an asteroid in orbit. But, thanks to the new equipment, we developed a more complex model with an additional body, for asteroids with a period of eight years. It had the Sun, the asteroid, Jupiter and Saturn. When we introduced this extra planet into the model, everything worked well. We were even able to explain why there are no large asteroids with a period of six years. There are no old asteroids with this resonance, just small rocks that entered orbit recently.
Do you mean that the model both explained the absence of large asteroids with a period of six years and the excess of asteroids with a period of eight years?
When we wrote the equations for the six- and eight-year asteroids, we saw that they were, strictly speaking, the same. So why the lack of asteroids of one type and the excess of the the other? Why were some diverted to a region near a planet and the others not? To answer this question, we had to do our longest project up until then. We concluded that the two cases were identical, but the time scale associated with each situation was different. It was a question of numbers. The region in which the eight-year asteroids are located will also empty, but this has not yet happened. We calculated that another 10 to 20 billion years will have to pass for this to take place in the region containing the eight-year asteroids. But the solar system is only 5 billion years old. In the region containing the six-year asteroids, the time frame for them to become diverted into other areas is from 100 million to 1 billion years. This time has already passed. So, we uncovered the answer.
Was your answer accepted by the scientific community quickly?
The first time I presented this research was in Belgirate, Italy, in 1993. We were at an event, on the third floor of a building, and they came close to throwing me out the window when I explained my results. The even forced me to delete several passages in the article that I published in the event proceedings.
What was the main criticism of the research?
I think it was that I found something that the others had not and had shown the orbits in which the asteroids make these big gravitational jumps. To arrive at this calculation, that the region that the asteroids with a six-year orbit had emptied in a maximum of 1 billion years, I used an empirical formula that had been deduced by researchers at Harvard based on numerical experiments. We know that the formula is a practical result, based on statistics, but there is no mathematical theory behind it. Shortly afterwards, I had an excellent student from the Czech Republic, David Nesvorny, who was tasked with studying this question using a more robust methodology that would be more easily accepted by my European colleagues with a mathematical background. He did a great job. He spent years programming and calculating and showed clearly and irrefutably that one type of asteroid fell out of resonance in under one billion years while the other needed at least 10 or 20 billion years to do so. There is a humorous element to this story. One of the people at the meeting in Belgirate who asked me several questions — in a friendly way, since we were all friends — was a recent PhD from Milan, Alessandro Morbidelli. Later, he surrendered to our arguments and even used a figure we had shown to illustrate the cover of a book he wrote on the dynamics of the solar system.
Personal archive
In front of the Palais de L’Institut, in Paris, around 1970: doctorate at the SorbonnePersonal archiveHow did you begin to study the orbits of the exoplanets?
There was a funny situation. I was at the National Observatory in Rio de Janeiro for two years, from 1999 to 2001, and then I returned to USP. When I got back, Professor Tatiana Michtchenko, who was a member of my team of students and now works in the office next to mine, was studying the movements in resonant systems of exoplanets, with another former team member, Argentinian researcher Cristian Beaugé. I was also starting to research this question, but with a different approach, studying the evolution of resonant systems under the effect of tides [a secondary effect of gravity that causes the nearer regions of a body to be more attracted than the more distant regions]. One day I showed them a figure and they showed me another, identical to mine. We had arrived at the same results. I think that we were the first to study systems of exoplanets whose orbits were resonant among themselves. One of the systems studied was Gliese 876. The planet closest to its star takes 200 days to orbit it, while the other planet takes 400 days, exactly double.
Was it the improvement in spectrographs that led to the discovery of extrasolar planets?
Yes, exactly. Today a spectrograph measures radial velocities with a precision of 1 meter per second. Usain Bolt runs 10 meters per second. The equipment measures the Doppler effect of Usain Bolt. The precision is really fantastic.
How would you assess Brazilian astronomy today?
It is of good quality. We have several colleagues who are very respected in their areas. The problem, like with all Brazilian science, is that we have one researcher who is an international reference in each field. There is one in specialty A, another in B, another in C. But there is never a critical mass to form teams. We suffer from the problems of the university. When there is an opening, the university has to fill it with the best candidate. The hiring process for that position must be broad. But sometimes, the person chosen, who is the best, is not interested in anything that the other researchers at that university are doing. For the university, there is no problem. Diversity is important to it. But, from the point of view of the scientific community, this does not contribute to maintaining teams. They form, last for a certain time, but there is no continuity. Even the institutes connected with the CNPq [National Council for Scientific and Technological Development], that could have a different approach, end up doing the same thing the universities do.
How did you become interested in astronomy? Was someone in your family a researcher?
An uncle, Ary Ferraz Mello, was a chemist and he had a strong influence on me. He worked as a teacher in technical courses and was also a university professor. He wrote a book on analytical chemistry that was used at several universities and he was even invited to transfer to USP, but he was already on a different path and did not accept. At that time, in the 1960s, working at USP was for rich people. You could go three, four months without receiving your pay until the hiring process was complete. But my decision to study astronomy was sort of by chance. Even today I find it hard to consider myself an astronomer. I am a physicist, not an astronomer. My physics has a lot of math. In France, where I did my PhD, fundamental astronomy was part of mathematics. I do not know how it is today. I think today it would be a doctorate in astronomy, not in physics or mathematics.
You were a technician at USP before becoming a professor. How did that happen?
In 1955, I began working on my physics degree at USP, which was in a building on Rua Maria Antônia, in downtown São Paulo. I stayed there for five years. One day we went to have coffee at the corner bar with the mechanics professor, Abrahão Moraes. He had a notebook in his hand and placed it on the table. There was a stamp on the notebook, from the São Paulo Observatory. I didn’t even know that there was an observatory in São Paulo. So, I began to ask him questions about astronomy. A few months later, he invited me to become a technician at the observatory, which was located at the State Park, in the Água Funda neighborhood [IAG was located there until the beginning of this century]. I was in my second year as an undergraduate and I was an intern in the Van de Graaff accelerator laboratory, with Prof. Oscar Sala. At IAG, I continued working as a technician for several years, doing calculations for the institute’s publications. The story is more complex than this, with a long period teaching at Bandeirantes High School, then teaching physics at USP, but, later, I ended up at IAG and I am still here today.
Personal archive
With Luiz Muniz Barreto, Abrahão de Moraes and the Belgian Sylvain Arend: meeting of astronomers in Hamburg in 1964Personal archiveThere was an interesting study on the first Soviet artificial satellite, Sputnik.
Luiz de Queiroz Orsini and Antônio Hélio Guerra Vieira of the USP Polytechnic School, developed an antenna to measure the radiation passing through the ionosphere and arriving here when the center of the galaxy passed through the meridian. It was a network of wires, not a parabolic antenna. They wanted to take the measurement every day and see how the ionosphere varied. When they launched Sputnik in 1957, the passage of the satellite appeared in the recorder. We followed Sputnik 1 and 2.
When you were a technician, were you already thinking about doing research?
I majored in physics in order to become a researcher. That’s what I had in mind.
What was Brazilian astronomy like at that time?
It didn’t exist. I was one of the first to leave Brazil to study and then return. I worked with Professor Abrahão de Moraes [physicist and astronomer, he directed the Astronomical and Geophysical Institute, the previous name of IAG-USP, from 1955 to 1970] from 1956 until I went to France in 1962. I was with him the entire time. In the beginning, we even shared a desk.
After your doctorate abroad, what was it like returning to Brazil under the dictatorship?
It was a dark time. When I finished my doctorate, there were no jobs at USP. I was ready to return to Brazil, but I did not know what I would do. Years later I found out that there was gossip saying I was a communist and I even had a DOPS record. I had actually been a member of the Socialist Party before leaving Brazil. This could have influenced things here, but certainly not Abrahão. One day, while still in Paris, I met some professors from ITA that were doing some work there and I told them about my problems. They suggested I go to ITA. I laughed and thought “what would I do with all those military types?” Then they told me that ITA had just installed a telescope and the dean, Professor Francisco Antônio Lacaz Netto wanted to develop their astronomy expertise. I wrote him a letter and asked to teach in their math department. He responded that there were no openings in mathematics, but that if I came to work as an astronomer, he would hire me. I went to ITA, where I stayed for eight years. There, I established the first graduate program in astronomy in Brazil, in 1968. Shortly after, Pierre Kaufmann established a program at Mackenzie. I returned to IAG when they created a graduate program here in 1973, in the beginning only part-time, to teach. I continued at ITA, too, until 1975, when I moved to USP definitively.
Is it true that you are member No. 1 of the Brazilian Astronomical Society, founded in 1974?
Yes, I am. But this is because I drew up the list of initial members. If you had done so, you would have been No. 1. There were about 60 people in the beginning. We filled up the room. We were trying to organize astronomy in Brazil in some way. We had a meeting at the old USP administrative building, where IAG had some offices. We had a meeting there and decided to form a society. At that time, there was a joint project by USP, the National Observatory, ITA, the Federal University of Minas Gerais, and Mackenzie to obtain an observatory for Brazil. We were searching for a site to build the National Astrophysics Laboratory (LNA). When I was still at ITA, I coordinated the final years of this search.
Was it easy to select Pico dos Dias in Brazópolis, Minas Gerais State?
No. We found the area by chance. There had been a problem: we were prospecting in the best place to install the observatory in Brazil, but we did not have a good map of the area where the peak was located. We had to guess where things were located. At the time, the best maps published were aeronautical maps. They had good relief maps, showing the mountain peaks and ranges. The state of Minas Gerais occupied several enormous sheets of the map, with many details. But, unluckily, Brazópolis was in the exclusion area of the Guaratinguetá aeronautical school. In that area, the map contained no information about peaks. When we were studying a nearby peak, in Maria da Fé, Germano Rodrigo Quast, who recently retired from LNA, said, holding binoculars: “What peak is that over there?” That is how Brazilian astronomy discovered the existence of Pico dos Dias, in Brazópolis.
