Rama Govindarajan is a renowned scientist whose work lies in the area of fluid dynamics. The recipient of several awards, in 2007 she received the Shanti Swarup Bhatnagar Prize for her “original contributions to the understanding of instabilities in shear and non-parallel flows, flow entrainment, turbulent transition and small-scale hydraulic jumps.” She remains the only woman to have won the prestigious national award for work on fluid dynamics.
Although she originally studied chemical engineering at IIT Delhi and Drexel University in Philadelphia, USA, she made an unusual career choice to begin working in aerospace engineering at the National Aerospace Laboratories (NAL) in Bangalore, where she worked for a decade while also working on her PhD in Aerospace Engineering at the Indian Institute of Science (IISc).
Fifty-four-year-old Govindarajan, who is currently a member of faculty at the International Centre for Theoretical Sciences in Bangalore, has previously worked at the TIFR Centre for Interdisciplinary Sciences in Hyderabad, and the Jawaharlal Nehru Centre for Advanced Scientific Research in Bangalore. Connect interviewed her ahead of the Aerospace Department’s 75th year anniversary about her interest in the field, her memories being at IISc, her work, and more. Here are edited excerpts from the conversation:
How did your interest in aerospace engineering begin, and how did that switch from chemical engineering come about?
I came to Bangalore in 1986 because my husband [Sriram Ramaswamy, Professor at IISc’s Department of Physics], got a job here at IISc. At that stage I did not actually want to do a PhD, I wanted to work for industry and make something useful – that had been my dream all along. I looked for jobs and went to 37 interviews of various kinds. This direction [aerospace engineering] was the only one that worked out.
I joined the Aeronautical Development Agency (ADA) in 1987, but in a few months I decided this was not what I wanted to do, I didn’t see my talent being used in a rapid way. That’s when I contacted Narasimha [Roddam Narasimha, former Professor of Aerospace Engineering at IISc and former Director of NAL], and I moved from ADA to NAL in 1988. I was particular about doing my PhD while working because then I would not have to look for a job at the end of it. NAL gave me the time and computational resources to do my PhD at IISc, which I did as an external registrant from 1991 to 1994.
But it turned out that I did move completely into academics after that.
Was working in aerospace a complete departure from what you’d done previously?
I liked fluid dynamics. That was my connection between chemical engineering and aerospace, although the way fluid dynamics is taught and learned and the areas of emphasis are different. Fluid dynamics is a very interdisciplinary field and it wasn’t so difficult for me to make that jump.
You heard us talking earlier in the lab about singularities – those are questions which appear all the time in aerospace. But those are questions that can actually be asked in chemical engineering, as we found. Later when I became a faculty member, I was able to use both my interests to find new ways to look at problems which originate in chemical engineering, in the petroleum industry, for example, or geophysical flows, which I am looking at right now. A lot of my work now consists of looking at oceans, the atmosphere, clouds, and I wouldn’t say it was that big a jump from aircraft boundary layers to this either.
Was there an aspect of working on your PhD that thrilled you in particular? Or that defined your time at IISc?
I would say it was the recognition of singularities and seeing how far you can take them.
There is the concept of a boundary layer. If you have an aircraft flowing through wind, you can reverse the problem and say the aircraft is standing still and it’s the air that is moving past. The wind some distance away is moving at a constant velocity, which is the negative of the velocity of the airplane. But the wind close to the aircraft is stuck to the aircraft. So it’s moving at zero velocity relative to the aircraft. This very thin layer of air – about a centimetre thick – the boundary layer, determines how much drag there is on the aircraft. If you can zoom it down to a point, that’s what you would call a singularity. If you understand this mathematics well, as well as the math relevant to far thinner layers within the boundary layer, you can write down very simple equations.
I enjoyed that very much. Instead of solving boundary layer equations for stability (in those days they used to run it on a supercomputer, and it used to take two or three weeks to get one answer), because we understood this structure very well, we could minimise those equations, simplify them and I could get an answer on my little desktop in those days within a second, or five seconds. I could get the same answer, or one very close to, the one it would normally take three weeks to find out. This idea was bought from us by Boeing for a small amount of project money.
We could minimise those equations, simplify them and I could get an answer on my little desktop in those days within a second, or five seconds. I could get the same answer, or one very close to, the one it would normally take three weeks to find out
You come from a prestigious line of scholars in the forefront of boundary layer theory – you had the opportunity to learn from Roddam Narasimha, who worked with Satish Dhawan, and like him, worked on his PhD at Caltech. Did that influence your decision to do your postdoc at Caltech? Did being at the famed aerospace department at Caltech (GALCIT), which had been home to other renowned scientists like Hans Liepmann and Anatol Roshko, influence you in any way?
All of us knew about the reputation of Hans Liepmann. He had been a towering figure for us, and it did influence my decision to go there. By the time I reached there, Liepmann would still come in to GALCIT to attend seminars and it was awesome to sit in the same room as him. I worked directly with Professor Anthony Leonard, who is also a very old friend of Professor Narasimha. Professor Leonard’s group and Anatol Roshko’s group would meet every week and we would discuss things, and Roshko has also been a mentor to me. Definitely, being there influenced my work in a very significant way, because for the first time I could spend time in a department where a lot of people were interested in exactly the kind of problems that I was. I learned about vortex dynamics and nonlinear dynamics during my stay there and that expanded my worldview beyond boundary layers. A lot of my future work depended on what I learned there.
One of my biggest interests now is vortex dynamics, and I look at vortices in clouds, how vortices affect mixing, what vortices do to particles – these are all new dimensions, but the fundamentals were inculcated during that period. That and things like chaotic movement of particles in fluid – that was a short project I worked on with Tony Leonard, and that too actually is something I use till this day.
What does your work today involve – is it computational, experimental, or theoretical? What are you working on right now?
My work involves all three. My group does three types of work today – one involves particles in turbulence, or particles in vortical flows. The main thing driving us in this line of work is clouds and how raindrops grow out of them. Clouds are turbulent, they have water vapour, they have aerosol particles, tiny little water droplets suspended in them. Sitting in a super-saturated environment, raindrops could grow due to diffusion. That process should take hours and hours, whereas in a real cloud, big raindrops take between 5 and 15 minutes to form. This is called the ‘droplet growth bottleneck’ – people don’t understand how raindrops can form this fast. There are lots of people who think turbulence is the answer to this, so we didn’t come up with it, but we ask specifically how vortex dynamics can influence this process. In particular, we have shown in a simple model flow that caustics, or regions where raindrops can collide against each other rather than merely coming closer to each other, can contribute to the creation of big drops in a very short time. This work we’ve done analytically. We take a single vortex, write down simple equations for small water droplets in the vicinity of that vortex, see where they go, see how much they collide into each other to make bigger droplets – so those things can be actually written down analytically. In the next step we have done computations in two dimensions where we have a large number of vortices and a large number of raindrops. We’re trying to extend this work to three dimensions, to get numerical confirmation of what we’re talking about in a situation closer to that in a real cloud.
In a real cloud, big raindrops take between 5 and 15 minutes to form. This is called the ‘droplet growth bottleneck’ – people don’t understand how raindrops can form this fast
We do another class of work on fluid structure interactions, where we ask about flexible solids and how they interact with fluid interfaces. We ask how bending, surface tension and hydrodynamics compete with each other to produce some particular dynamics of flexible solids. That is mainly experimental, with a smaller component of theory. We have a very good experimental collaborator, Professor Narayan Menon at the University of Amherst, and the work is carried out in both Amherst and in the KS Krishnan lab at ICTS.
The third kind of question we ask is on flow instability. A lot of instabilities result in flow going from laminar [orderly] to turbulent [disorderly], so we ask questions about the laminar-turbulent transition process in situations where fluid properties such as density and viscosity are a function of space and time. This is directly related to my PhD work on boundary layers, but I now look at these transitions in other scenarios like in oceans, or in industrial flows.
In all these projects, I gratefully acknowledge the huge contributions of fantastic students, postdocs and collaborators.
Since you began working on boundary layer theory in the 1980s, have there been exciting new areas into which the field has branched out today?
The concept of a boundary is not even restricted to fluid dynamics. They can happen wherever there are large gradients of something or other, in a vast variety of systems, so it need not just be flow next to a wall. In that sense, they can be represented as belonging to a class of singular perturbation problems. Many mathematical features of these singularities have been since worked out, and they have been shown to occur in a range of fluid dynamical problems. In our work, we’ve found such singular behaviour in pipe flow or channel flow, inside some critical layers. These behaviours actually have their origins in the fact that there is a wall nearby and shear flow close to it.
The atmospheric boundary layer is another ballgame. It is studied by a vast number of people now. And for all climate scientists the atmospheric boundary layer is very important.
Even the understanding of the boundary layer over the aircraft wing, which I worked on a long time ago, has undergone a transformation.
What was IISc like when you studied there, and what was the Aerospace department like?
Different! I would say this is not just for the Aerospace Department but for the whole of IISc, it was a quieter place then. Not every lab was ambitious as they all are today, as I gathered from discussions with my fellow students in different groups and departments. In those days we used to have pockets of excellence, with some labs doing really really well. Now it’s more like every lab in every department is doing well. There is a sea change in the level of activity and global connectedness, which makes a huge difference. When I started my PhD, I don’t think we had email yet, if I remember right. International travel was harder – so was travel within India. So many faculty and students were far less well-known, especially abroad.
Did you ever get to interact with Satish Dhawan during your studies at IISc? Was access to him easy?
He had already retired by then, but he used to come for seminars and such events, and anyone could go up to him and talk. He was a very nice person. In that sense everybody had access.
I never had an elaborate conversation with Satish Dhawan, and that’s one of my big regrets. Because I was too shy, basically, to approach him. He’d come for a couple of my talks which I felt very happy about and he would ask questions and we would discuss them. But I wanted to learn a lot more from him which I never did, mainly because I was shy in those days.
I never had an elaborate conversation with Satish Dhawan, and that’s one of my big regrets.
Were there any other women in the department at the time?
There was one doing her PhD at the same time as me, and one Master’s student.
Even today few women take up engineering of this sort. In electrical engineering or computer science there are a lot more of them. But even in those days, I met several young women who were very crazy about anything to do with aircraft, or spacecraft. They were completely enamoured by this area but unfortunately most of them maybe fell by the wayside…
In what way?
They went into other careers or into no career. It’s probably a little easier now but it’s still not that easy for women, for a range of reasons.
It was only in 2010 that IISc’s Aerospace Department hired its first (and so far, only) female member of faculty. Does it feel strange to hear that, even after all these years?
That’s probably also been true of Mechanical Engineering until very recently, or, say, Civil Engineering. It’s not just Aerospace. We guys in science have gotten used to it – we shouldn’t be used to it, we shouldn’t be happy with it. It is true that there is this leaky pipeline effect. Certainly, a fraction of women who are PhD students don’t make it to junior faculty and so on. That’s changing a little bit, but not fast enough. Not at all fast enough.