Artwork by Akshita Arora
From conversation on:
Nov 01, 2020
Listen to full episode
"What has caught the curiosity of a kid, if not the rolling of thunder, of how the winds change and where all those clouds travel to!" Helping us figure out the ideas and efforts it takes for mapping out the farthest skies and the deepest oceans in all painstakingly fascinating details is our guest – Dr. GovindanKutty Mohankumar. Through the scientifically involved conversation, Dr. Kutty made sure that he leads us on dusty detours of everyday ideas about atmospheric science, climate change and meteorology— he spoke about his first-hand experiences on living in a tornado alley through one of the stints of his career (close to one ‘Flying Cow Cafe’ named very literally), and faring the seas to gather data through another stint, among the many varied parts of his journey. Quipping how it’s common to underestimate the depth, and complexity involved in the process of predicting accurate forecasts (dealing with a very direct manifestation of the chaotic principles of nature and the numerous effects that go into the research) — he cleared out multiple misconceptions about the popular concepts and the discussion brought out the technicalities with examples of daily life understandings and observations, and that’s been a great joy in itself!
If God appears before you and gives you a weather model which is perfect in all sense, you will still not be able to give an accurate forecast! This is because of the fundamentally chaotic nature of the atmosphere… that we cannot change.
ABOUT THE GUEST
Dr. GovindanKutty M. Associate Professor, Department of Earth and Space Sciences, Indian Institute of Space Science and Technology, Thiruvananthapuram.
Dr. GovindanKutty Mohankumar is a scientist who works in the field of atmospheric sciences and meteorology. He pursued his Masters in Meteorology from the Cochin University of Science and Technology, then moved to conduct his doctoral work in atmospheric science at the Indian Institute of Technology, Kharagpur. His further research work was carried out at the University of Oklahoma, following which he joined the current position at the Indian Institute of Space Science and Technology, in 2014, and is currently an associate professor there. He has worked on elaborate ideas in meteorology, with research primarily focusing on the study and impact of data assimilation on the performance of weather models, atmospheric dynamics and predictability, climatology, tracking forecast of tropical cyclones and Indian monsoons, to name a few broad directions.
Transcript
Naman Jain (Host 1) :
Welcome to another episode of Zeroing In. I am Naman Jain and hosting this episode with me today is KVNG Vikram who recently graduated from IIST with a masters in earth system sciences. Our guest for today is a scientist who works in the field of atmospheric sciences and meteorology. He pursued his masters in meteorology from the Cochin University of Science and Technology and then moved to conduct his doctoral work in atmospheric science at the Indian Institute of Technology in Kharagpur. His further research work was carried out at the University of Oklahoma following which he joined the current position at the Indian Institute of Space Science and Technology in 2014 and is currently an associate professor there. He has worked on elaborate ideas in meteorology with research primarily focusing on the study and impact of data assimilation on the performance of weather models, atmospheric dynamics and predictability and climatology, tracking forecast of tropical cyclones and Indian monsoons to name a few broad directions. While these fields can be understood to be quintessentially non-trivial at the onset, the entailing work encapsulates a plethora of details that makes the understanding of the field and the takeaway from it extremely riveting at its very core while also being singularly relevant to the everyday lives of the whole living community at large.
In our conversation with him, we discovered about his journey and explored the diverse field of atmospheric sciences climatology and meteorology and all that it entails from a scientific perspective with an intriguing touch to the everyday ideas which it imparts that seems to be the pressing points of our generation's concerns for the future. A very warm welcome to Dr. Govindan Kutty Mohan Kumar.
Dr. Govindan Kutty :
A very good morning and thank you very much for inviting me to the session.
Naman :
We're really glad to have you sir.
Naman :
We usually begin the discussion with the idea about the beginnings and how your formative years have been as you remember them. Are there any peculiar ideas or stories that you recall from your early years?
Dr. Govindan Kutty :
Yeah, so after completing my schooling, I was actually clueless what to do after that. I was from a village background and there was no proper guidance. My parents always encouraged me to study a lot. But, I didn't know which direction I should move. Some of my school teachers had advised me to go for physics. I completed three years of my physics but some portions of physics were scary for me. I still remember my professor writing down the Schrodinger equation on the blackboard and it was so scary for me. So I thought this is not my track. I thought that physics would not work with me but I don't know what to do next. So I was in fact looking out for different options. I also discussed it with my professors in my university. So we have a project which has to be done in the final semester of my bachelor of science. One of my professors suggested that I go to IMD. Since IMD is close by, I can go there to do some atmospheric science projects and also collect some data. So that was a revolution to me. This was something different. So, I just went over there, collected some data, and did some data analysis. It was some trivial work but I just got excited with this and then I looked for options to study more on the subject. It was then that I found out that Cochin University is actually offering courses on atmospheric science for people who are having a physics background. But you have to qualify certain entrance exams. So I wrote the entrance exam and I qualified that. Once I joined, I realised that physics is not leaving me, it's actually running behind me. So this subject is not far from physics but this physics actually comforts me. I'm okay with classical mechanics but if you ask me to have abstract thinking and all this I'm a bit scared. So this physics, which is mostly dealing with the classical mechanics regime, was okay with me.
Naman :
But then when was it that you decided to move into research or did you have a hint that probably research was a field for you?
Dr. Govindan Kutty :
When I was doing the final years of my project in MSc, I did some work on Indian ocean weather and how it affects Indian climate. So I got some significant results with that which actually excited me. So I thought that I'm not bad at research. After that, I went for some cruises which actually collected data from the Arabian Sea. So it was like a several day cruise. We have been given certain coordinates when the ship reaches this particular coordinate it may be at midnight at 12 o'clock or 1 o'clock. So you have to set an alarm. You have to wake up, put the instrument in the sea and collect the data. So I was actually assisting as a research scholar. So all these things, my project, and the cruises got me excited. So I was sure that I'm going to pursue research in this. But I decided that I will do research only if I get admitted to either some of the foreign universities or some premier institutes in India. I searched for the options and found out that there’s CSAR which is a MEHRD exam, which if you qualify, you will get five years of uninterrupted grant. I qualified that and at that time getting the CSAR is a very rare thing for people, and once you get CSAR it is very easy to get into a premier institute. I joined the physics department at IIT Kharagpur. Once I joined there, I realised that I have to work on numerical modelling of the atmosphere. So we have all these dynamic equations which govern the atmosphere. So you have to write or you have to solve these equations numerically to predict the future of the atmosphere. So the problem which is given to me is that these numerical models are very sensitive to their initial conditions. So this is based on the theory of chaos which involves Lorenz butterfly effect which you might have heard of. For example, a flap of a butterfly wing in Brazil would generate a tornado in Texas. So this means that non-linear dynamical systems are very sensitive to initial conditions. The numerical models which predict that the atmosphere cannot be accurate just because of the sensitivity to its initial conditions. So the job which is assigned to me is to generate some mathematical models to improve these initial conditions. That method is what we call data assimilation. So we use all the satellite observations and develop some mathematical techniques which improve the initial conditions which further improve the forecast. This can be then used by operational weather centres. I worked on this data assimilation system from almost like 2006 to 2010 and I evaluated this with respect to the forecast of tropical cyclones which are forming over the Bay of Bengal, monsoon depressions, etc. so as to know how well my method works.
KVNG Vikram (Host 2) :
Okay then how does that translate further to your later works? Did you wait and explore other ideas or how was your transition to further research positions and the experiences there?
Dr. Govindan Kutty :
Interestingly, the day I had my defence was the same day I got the offer in University of Oklahoma. Actually it was not in the University of Oklahoma, it is the National Weather Centre which is a federal building where there is a Centre for Analysis and Prediction of Storm which is a very famous centre. They make a lot of models and I got an offer letter from the caps. So I left for the US. I did my Post-Doc for almost like three and a half years. The National Weather Centre is located in Norman. Once I reached there, I understood that this is a tornado alley. You will see like 50 to 60 tornadoes every year so that was in fact another scary experience for me apart from Schrodinger equations. The problem with tornadoes is that tropical cyclones or hurricanes are forming over the oceans and days before we can see which direction the hurricane is proceeding but tornadoes are not like that. So this place is the hot spot of tornadoes. You can always predict the probability of the formation of tornadoes tomorrow, but where exactly it is going to happen and when exactly it is going to happen is very difficult. That's why tornadoes are very unpredictable and very dangerous. I have seen that it can, in fact, cause a lot of destruction to the buildings and a lot of deaths in that region during April-May, which is the season in which usually the tornadoes were formed. That was one of the scariest experiences for me. I've also seen the students and teachers sometimes chasing these storms. You might have seen the movie Twister where these people actually follow the storms. They will put the sensors and the track of the tornadoes and if the tornadoes get to that place they will get a lot of valuable information and they can publish papers with that. So those kinds of, you know, adventurous things were also happening. I was so scared that I'll never go for this adventurous sort of thing. I'm not fit for that. These are all very new experiences for me. You know people who are following tornadoes and all those things, and before the formation of tornadoes you will see all these different types of clouds like Matus clouds hanging from us. So when you see that, you know it's an alarming situation and you have to be very careful. We have seen this type of cloud so there’s a possibility of forming a tornado over there. So then we will rush to the shelters. Those are very interesting experiences I had in Norman. In the National Weather Center we had a cafeteria and the name of the cafeteria is very interesting. It is called Flying Cow cafe. So, I asked why this name? It's a weird name right. Flying cow cafe! And they have a cow which flies in with the tornadoes. Then they said that there is a movie called Twister. They have a shot in which a tornado takes away a cow, so the cafeteria is named after this. So I mean in every location you'll see some imprints of tornadoes because they are so much into the tornadoes. They are seeing tornadoes every year and so that was one different experience for me and there is a big group actually working on the predictions of tornadoes in that caps. But I was not actually involved in tornado predictions. I was rather involved in weather phenomenons in the global scales. My project was a NASA sponsored project. So, NASA has launched a satellite called AQUA which has the AIRS instrument. So they have to see how much this data can improve the forecast. I worked with the global models not with the tornadoes but I had to chat with all these people who are working with tornadoes and all these things. So it was a really nice experience for me.
Naman :
So probably I would like to interject here at this point and and jump back a little bit probably when you talk about how it's like a continuing pattern and the things that you talk about and we just probably noticed that you look at the things in a very scientific manner. Like you say that there were tornadoes. But then, there were tornadoes for everyone perhaps right but then you look at them as someone who's probably so scientifically involved that you say that these were the kind of tornadoes that were happening and there were all these clouds. I'm sorry but I don't know the exact term but then basically you identify the kind of clouds that were there and you remember those details from such a scientific perspective that it flows in some sense from a scientific imagination or scientific background that you have. So where do you think this took roots for you in some sense like where do you remember yourself making these kinds of detailed investigations in the kind of everyday life that we all see but then we don't make these kinds of investigations in life so how do you remember yourself?
Dr. Govindan Kutty :
That’s because I have been trained as an atmospheric scientist from my MSc onwards right. So I have been learning all these phenomenons. So I read and I study that in my books and once I see those phenomenons which are appearing in the real world it's really exciting for me. I think that is one reason why I don't like abstract things which we cannot see. So these things we learn in the books, we can see in nature also. That's something which I hope is because I learned a lot of courses. When you do PhD, you will think a lot about the systems and all these things so it naturally gets ingrained to your thought process. When you see certain things, you will think in a particular manner but we’ll not realise it. It is very subtle when you say that you are thinking in a different manner. I realise okay I'm thinking in a different manner but it is already ingrained in me because I have been working on this area for a long period of time.
Naman :
So we would like to jump a little bit here and talk about your current works. We know that you're pursuing research on the effects or the impacts of data assimilation and with your work on ensemble Kalman filters as well. Could you explain how this works exactly? I mean how do you internalise this work in a sense for you? How do you look at it and how would you explain it if you can for a beginning undergraduate or so in science?
Dr. Govindan Kutty :
I have never been exposed to ensemble Kalman filter-like methods. There are two different ways of approaching a data assimilation problem. I've been working with what is called as minimization problems like variational data simulation when I was doing my PhD. But when I got into the National Weather Centre, that time they were working with a hybrid method which actually combines the variational approach. That's a minimization technique with an ensemble Kalman filter which got some new good results. So I got experience with both the strips with variational as well as ensemble Kalman filter methods and how to combine them optimally. So ensemble Kalman filters can be used to improve the initial conditions but there is another use of ensemble Kalman filters i.e. to understand the predictability of the atmosphere. So as I said, the model is very sensitive to initial conditions. So if you give a model forecast with one initial condition and if you add perturbation to the same initial condition and give another model forecast, the two will evolve in two different directions. So where do we get these perturbations? The perturbation should be consistent with the flow of the atmosphere. So because we are dealing with a non-linear dynamical system, we cannot just take some random perturbation and add it. So I know that using the ensemble Kalman filter system you can generate these perturbations, which are consistent with the flow of the atmosphere. So that was something new for me. I thought that this can be used for understanding predictability. So then I used the ensemble Kalman filter system to obtain those perturbations and these perturbations will be added to the initial condition. So let's say I'm studying one particular tropical cyclone case which is forming our Bay of Bengal. What I will do is, I'll make 80 or 90 initial conditions by adding perturbations which are generated from the ensemble Kalman method. Now I will see how it evolves with respect to forecast in the model forecast so if it is spreading too much then that's an indication that at that particular time the model is very sensitive to that initial conditions. So on certain another day if you add the same type of perturbations the model will not evolve that much. So the spread in the ensemble or the standard deviation of the ensemble members will not be much larger. So then that predictability of the atmosphere is more at that particular time. So in that way, we can discuss the predictability of the atmosphere. So an ensemble Kalman filter can be used for improving the model initial condition as well as it can be used for purposes like understanding the predictable system. For example I will just give you more details. Let's say I'm trying to forecast a cycle which starts from Bay of Bengal and one model forecast says that the cyclone is going to land in say Andhra Pradesh and other model forecast says that the cyclone is going to land in Chennai and yet another model forecast is that it is going to land in West Bengal. So there is a huge difference in the ensemble spread so there is an unpredictability. The predictability is limited so then I will just take the model forecast and compare what has happened to the one which has gone to Bengal, what are the dynamical changes which have happened which took the cyclone to Bengal and what are the dynamical changes which took the cycle to Chennai. So then I can really understand what is the underlying dynamics which actually drives the cyclone in different directions. So in this way I am trying to get a more in-depth understanding of the phenomenons which are happening in the given region. So I use an ensemble Kalman filter system to generate perturbations or to update perturbations and using that I add this particular set of perturbations to the initial conditions and I'll evolve the initial conditions to grow and I'll see how the errors are growing. And from this we'll gain an understanding of the dynamics of the atmosphere. That is exactly what I’m working on right now.
Naman :
It’s quite fascinating to hear this because these are the ideas that we never are able to understand because we always like to hush away the details or put it under the carpet since there are so many variables that we cannot account for them. And here you are actually literally accounting for them in mathematical ways so this is something that's extremely fascinating that we wanted to talk about as well and I think we'll have further questions on that later as well.
Dr. Govindan Kutty :
Yeah actually you're right in the sense that the dimensions of adverse remarks are of the order of ten to the power of seven or so. It is very difficult to twist every model, twist every variable and so that's the reason why we are adding perturbations and just doing this. It's a different way of representing the probability density of that weather event.
Vikram :
Yeah sir. Since you mentioned that you are working with Kalman filter but for a general undergraduate who is not into atmospheric science, Kalman filter is being used in a lot of different fields like from automation to all sorts of things. So how is the Kalman filter applications in atmospheric science? Is there going to be any contrast or anything unique to Kalman filter usage in atmospheric science?
Dr. Govindan Kutty :
No. Actually, there is nothing unique in it. It is the same thing. Only thing is that usually in the electrical engineering field and other areas, I have seen the Kalman filter which is mostly a linearized model which is being used. But applications may be different for different areas. But it is the same Kalman filter that we are using. In atmospheric science, the problem is multi-dimensionality. So we are dealing with high dimensional systems. So we cannot use the Kalman filter as such. So we need an ensemble Kalman filter. We cannot just represent the probability distribution of the entire system. So instead, we are generating random pits or random troughs from the population and we make ensembles of them. Because of this high dimensionality, we use the ensemble Kalman system. Otherwise I think the Kalman filter system which is used in electrical engineering is mostly the same thing but the purpose or the objectives of using it may be different in different areas. But I don't think atmospheric science is going to settle with the ensemble Kalman filter system because the Kalman filter system has a lot of issues. One is its linearity and the second one is that it assumes the distribution to be a Gaussian distribution. Those assumptions are actually not true for atmospheric systems. There will be non-gaussian distributions as well. So that needs to be incorporated so the estimations of the initial conditions may not be accurate because of underlying assumptions like linearity and like assuming that the distribution is gaussian. Now the atmospheric scientists are slowly moving on to this particle filter which in fact accounts for the non-gaussianity of a system as well but it is very difficult to work with high dimensional systems using particle filters. I am sure that there are a lot of people who are working to find a way out of this and hopefully in a few years this particle filter will get implemented.
Naman :
It's quite interesting to know. I would like to again go a little bit wider in perspective and I would like to ask you about some of the rapid changes in the world. There are newer environmental challenges coming up in some sense, some of which were probably unforeseen a few decades ago. This has perhaps started to even affect the climate patterns and in turn various forces, factors, and variables that one might need to account for while making these elaborate atmospheric models that you've talked about. For instance, there are so many phenomena that probably did not happen so often in the past and then they have started to get more and more frequent in the recent years. So how do you see it in a broader perspective, the long-term effects on the predictability of the natural phenomenon and climate patterns in general? How do you see your work or in general the field of atmospheric science is addressing that? Do you have any future or clearer ideas about this?
Dr. Govindan Kutty :
Yeah I think that's a very good question. So a few years before, people would make fun of you, say in 2000, if you say I am working with the weather and climate. But now I think the attitude itself of people is changing. They have started seeing the signals of climate change. Previously there were only predictions in terms of the number. There was always uncertainty. We are not sure whether in warming weather the number of cyclones is going to increase. But one thing which is for sure is that if there is global warming, and if it exists, this is going to increase the intensity of the cyclone for sure. So in the coming years I think people will start seeing more signals about climate change. I am sure that then people will depend more on scientific research and scientific conclusions. They will take it more seriously because that's something which they really experience. Once people experience this and then you say that this is the science behind it, then they will really understand it. As far as weather is concerned, you can see in India a lot of weather companies are coming up very recently because our computational power has actually increased and the weather forecast has become more and more accurate these days. So whenever you say “the weather forecast is not accurate” e.g. when you say rainfall and there is no rain, that is becoming an old story. Now we have all these big computers, we have new artificial intelligence, machine learning methods. With all these combined improvements, the forecasts are getting more and more accurate. There are a lot of companies that are actually using the weather for their own purpose. For example, the power sector companies require accurate forecasting of weather to decide what amount of power they want to generate for the next day or the day after. So that's going to save a lot of money for them. Once the money is involved, naturally there will be a lot of companies which come up and try to give accurate forecasts. And the cyclones are increasing and more weather phenomena are coming up so that's producing a lot of destruction. So if you can get an accurate track or accurate intensity of a cyclone, then the destruction can be reduced a lot. There is also the money involved. That's the fact in which the impact of the significance of the weather forecast is actually improving in recent decades. I'm sure that in the coming years what is being predicted in the climate sense is that, in the future, due to global warming the extreme weather is going to be higher. So in India, we have monsoon from June to September. So there will be rainfall from June to September. So due to climate change, the total rainfall is not going to change much. But, instead of getting distributed from June to September, it will precipitate in pockets like it will have extreme rainfall at some point in time and other times there will be droughts. So these kinds of situations are something which we are going to see in the future. So when extreme weather events increase, the significance of giving accurate forecasts is also going to be higher. With the rising computational power on the other side and with new techniques coming up, I'm sure that the atmospheric science and weather forecasting has a real future because people have started realising the significance of that subject.
Vikram :
Sir, for a general listener, can you explain what climate change looks like and how is it caused? Because everyone comes across the term in popular articles but they often do not paint a clear scientific picture. Can you explain it from a more scientifically informed perspective?
Dr. Govindan Kutty :
Actually, there are a lot of misconceptions about climate. So when you ask any student, say a plus two student, what is actually causing climate change or global warming, all of a sudden the immediate response you get from this is that- “We are burning fossil fuels and this is actually causing global warming”. So that answer you can expect for sure. I've given talks in many schools as an outreach program and I’ve got the same answer. I was expecting this answer and that same answer I am getting. One thing you need to understand is that climate change is not just because of human involvement. It is still not very sure how much human involvement has caused climate change. One thing we can say for sure is that of course anthropogenic involvement or influence is affecting climate, but to what extent the anthropogenic influence is changing the climate is something we are not sure of. That is a real scientific fact. Climate change or global warming can also happen due to changes in natural phenomena. That has happened in the past. So in such past climatological studies, what they do is that they will break open the ice in Greenland or Antarctic regions. So these ice layers will tell you the stories of different centuries. So air is trapped between the ice layers and from that air and from the compositions you can gain a lot of understanding about the atmosphere which was prevailing at that point in time. Okay so when I say past climatological studies, I'm talking about the order of one million years ago. So those studies show that basically the climate of the earth was never constant. It just kept on changing. There were warming events and there were cooling events and all these things kept on happening with respect to the natural changes or natural phenomena that were happening. But that's on the scale of several thousand years. Now the confusion here is about the changes that we are observing now. Is it because of some naturally induced thing or is it only because of anthropogenic processes? So recent modelling studies have shown that anthropogenic sources such as the burning of fossil fuels have some influence on the climate. But we cannot say for sure that it is only because of humans. So it is basically a combination of both. One thing I can tell you for sure is that whatever amount of radiation is coming into the earth’s atmospheric system from the sun can affect the climate. So if that radiation is trapped or blocked due to some reason, the climate is changed. That's a one-liner for that. The one thing which drives climate change is the balance or imbalance in the radiation which is coming from the sun. So that's something which is actually changing the climate.
Naman :
In this sense do you think that there are also some long-term effects or something of that sort? So basically what people go at or what we’re again made to believe in this aspect is that over the last 30 years we've been seeing more and more extreme climates in some sense. And then there are always newer and newer records being set in the climates or weathers each season and so on. So I think that is one of the main ideas or ideologies that's given to support the fact that we are causing it and I believe there is, of course, some aspect of it which we contribute to. But how much do you think is the whole natural system sensitive to our perturbations? Do you think that there is something that is actually going to change that?
Dr. Govindan Kutty :
I probably cannot give a scientific answer for that because it is not still clear how much humans are contributing. But let me reiterate that I have not said that humans are not causing climate change. I'm not Trump to say that humans have no role in it. So I am saying that how much humans are contributing, we are not sure. But, looking at the time scales with which the changes have happened, it is in a few decades we are seeing this. Such a change I am sure that humans are actually pretty much involved in the change and in the current phenomenons occurring over here. So even if you reduce fossil fuel emission, this will not immediately settle down. It will take a lot of time for the climate system to bring it back to normal. Let me give you an example. So, when Mount Pinatubo, which is in the philippians, has erupted i.e. volcanic eruption has happened, it sent out a lot of dust or sulphur ashes to the air in the atmosphere. This blocked the incoming solar radiation and has actually affected the climate system of the earth. We know that the monsoon, which happens every year, is driven by the land-sea temperature contrast. So when Mount Pinatubo erupted, it blocked the sunlight coming in which actually reduced the temperature contrast. This has actually weakened the monsoon for a few years. Once the ashes settled down, the monsoon returned back to normal. It took several years to get it back to normal. So I'm saying that just an eruption can in fact set an imbalance to the radiations which can in fact affect the climate system. So even small perturbations can in fact produce a lot of changes there. So that's natural. That's not a human-induced thing but it is reversible so such things can in fact influence the climate system. So I cannot just say for sure how much humans have influenced but looking at the time scales humans have a significant influence on climate change.
Vikram :
So we come across many misconceptions about the field of atmospheric science and the details that go into this kind of complex research that is perceived. Would you like to talk about any of these from your experience?
Dr. Govindan Kutty :
Okay, so one of the misconceptions which I face every time I talk about the weather is- “Why can't you get the weather forecast very accurate? So in the U.S. and Europe, people will simply look at the weather forecast for the day and they'll carry an umbrella only if rain is forecast. If snow is forecast, snow will be there. So why is it not happening here? Is it because we are lacking something?” So in my experience, people ask such questions because they don't listen to the answer. They already have this understanding or preconceived notion in their mind that in India we have less people who are actually working on it and that they don't have that much experience. But actually that is not true. That's a common misconception. The reason we are not getting our forecast right is because we are over the tropical regions and the US and Europe are over the middle latitude region. Suppose we are in the era of Pangea when continents were floating and suppose we have been pushed further north of the latitude. If we have the same technologies that we have now and the same models, we can get a very accurate forecast. So it is not about the technologies, it's not about the computation, it is about the weather systems in the middle latitude. It is different from the weather system in tropical regions. In tropical regions, we have a lot of convection. Convections are something which is a gray area, as the physics of convection is still to be revealed and very active research is going on in convections. So in tropical regions, the weather systems are mostly driven by convection, which as I said is not well represented in the models. So that's the reason why over tropical regions, wherever it is, be it in India or in African regions, predicting such phenomenons are very difficult. Even in the US or other high latitude countries, predicting the summer weather is more difficult than winter. In this scenario also, there is a role of convection. Convection is embedded in large-scale systems and models can easily predict that. But small-scale convection is very difficult to predict and that's the reason why we are struggling. It's still improving but it won't be as accurate as what we are having in the US because of the difference in the dynamics of the weather systems. That's one common misconception and purposefully I wanted you to ask this question because this is something which I faced a lot even when I started. Even as a PhD student when I mention that I’m working on atmospheric science, people ask why is the weather forecast not very accurate here and what is lacking in India? I kept getting those sorts of questions.
Naman :
That's so true. I think there is this one question that I want to follow up with. You're working on atmospheric modelling and all these ideas that you mentioned to us it's been extremely interesting to hear about them. I just wanted to know in the long term, probably it's a very naive outtake from someone who's from outside the field, so is it something that you're trying to converge to a weather model that can correctly predict at all times? Do you want to converge to one model which encapsulates everything in a proper manner or is it something that is always going to have multiple dimensions that one can control? So basically there are multiple models that one can use. Is it something of that sort?
Dr. Govindan Kutty :
Actually it’s not just about the accuracy of the model. Suppose if God appears before you and gives you a weather model which is perfect in our sense, you will still not be able to give an accurate forecast because of the chaotic nature of the atmosphere and that’s something we cannot change. We have no control over it. So even if you have a perfect model, you will start getting false alarms every time. That's because the very nature of the atmosphere is chaotic. Or else the god has to change the atmosphere and say make it more predictable just like the movement of planets which is very easy to forecast right. So if you know the physics it's very easy to forecast but this is a non-linear dynamical system that is actually very difficult to capture. That's the beauty of it. The uncertainty and trying to predict it is actually where the beauty lies. The forecast of the model improves with respect to time. But will you get a perfect model in your future? I am very doubtful about that. So if you ask me about my future work I will be working on this to continue working on this and understanding the dynamics of the system. At the same time, I am just trying to map the experience which I've gained from here into the planets like Mars and all, where predictability problems have a lot of scope for it. So now we started getting the atmospheric models for Mars as well. There are a lot of agencies from the west as well as in the east which are releasing the models for Mars and all those things. So even they have the data simulation systems and all. I have not really started working on it but in the future, I am thinking of contributing a bit towards the planetary atmospheres as well.
Vikram :
So when you said you have models for other planets, one thing I don't understand is why should there be a different model? The governing equations are the same on both planets. So why is it a big change?
Dr. Govindan Kutty :
That's a very good question. The governing equations are the same. It's true. But the model is not just of governing equations. The model has a lot of approximations. It needs to include a lot of parameterization schemes as well. So the parameterization schemes are not absolute and are not universal. That needs to be changed with respect to different planets depending upon the situations over there. And you have to, of course, make some changes for the Martian atmosphere as compared to the earth’s atmosphere. Plus say if you talk particularly about the Martian atmosphere, we don't have to include so much of moist processes. Since the moisture is less,moisture-related aspects can be reduced or can be taken away from the model or the contribution of such effects can be reduced from the model. So you have to make serious changes. I mean you cannot simply plug in the atmospheric model to Mars. So you have to understand those phenomena which are relevant over there and then you have to make the changes accordingly. So to explain the relevance of studying mars particularly, let me tell you that Mars is usually governed by all these dust storms. These dust storms will actually block the solar radiation. So if say ISRO is sending a lander in the future, it will have solar panels with it. So if there are dust storms it will actually reduce the amount of radiation that the solar panels are receiving. So if you can just use these models to give an accurate forecast about when and where the dust storms can happen, they can adjust the systems accordingly. This is something which I am looking forward to. I mean this kind of knowledge of the predictability of the Martian atmosphere can be used for future space programs. So that's one context I’ve started thinking about. I slowly started moving towards it but I am still a student in that. I’ve started reading and learning and all these things are still happening.
Naman :
That's a really interesting bit. On a concluding note if you would like to add something that you had and probably did not discuss, you can answer that.
Dr. Govindan Kutty :
Okay, so let me say some final words about climate change. So ancient Mayans actually believed that there is a saying that storms are sent by gods if they did something wrong. But I can see a correlation between what ancient Mayans believed to the current situation. In this world of scientific knowledge, I would say that it is not the god, this is basically the nature that punishes us for all our deeds. So without understanding nature, without caring for nature we are polluting nature. And the fury of nature which we are seeing is because nature is punishing us for our bad deeds. So we have not traveled far from the Mayan era to this current era. So Mayans believed that it is god but now we know that it is not God, it is nature who is punishing us for our deeds. So we have to care about our nature for our own well-being.
Naman :
This was Zeroing In with Dr. Govindan Kutty. It has been brought to you by the sounding rocket in collaboration with the IIST alumni association from the Indian Institute of Space Science and Technology. We extend our sincere gratitude to Dr. Govindan Kutty for sharing instances from his wonderful journey and ideas on various aspects of research in the field of atmospheric sciences, the understanding of climate and weather patterns, and how a complex field is dealt with in its entirety, on behalf of the Zeroing In team, which included Fenil Shah, Shreya Mishra, Manish Chauhan, Prajwal Patnaik, KVNG Vikram and I am Naman Jain. Thanks a lot for listening to this episode. If you have any suggestions you can write to us on zeroingin@outlook.in or contact us on our Instagram handle @zeroinginpodcast or The Sounding Rocket page on Facebook.