Episode Transcript
[00:00:00] Speaker A: Hello, my name is Chris Bolland, and I'm interviewing Professor John Mitchell at the headquarters of the Royal Meteorological Society in Reading on the 14 December 2022. So welcome, John, to this discussion about your life and career. Now, I know you were born in Belfast in 1948, but I don't know anything about your family, so be grateful. Tell me who your parents were and what they did, particularly when you were growing up. And it would also be useful to mention your parents educational background.
[00:00:39] Speaker B: Okay, Chris, first of all, can I thank you for agreeing to do this and interview me?
My father was born in England. He trained as an architect. He went to Liverpool University to do a degree. He didn't finish his degree because unfortunately, his brother died of tb. And I think there was a bit of a hiatus in the family. So although I think he attended most of the courses, he didn't actually graduate. But he went from on there to be a professional architect and town planner. He met my mother in England and they got married. And then shortly, then the war came along, and after that he moved to Northern Ireland because there were younger people ahead of him in Cheshire, and he became the county architect for county down. Initially he was in Belfast, and then he moved to the county town in Downpatrick. My brother was born before they moved from England. He was seven years older than me, and he was somebody I actually looked up to.
My mother was born in Dublin. She was unusual in 1930s that she and one of her sisters actually did science degrees. My mother did a degree in mathematics in Trinity College in Dublin. There wasn't much for women to do as professional scientists at that time, so she moved to England, where she was a teacher, met my father's sister and hence met my father. They got married, settled in England. As I said, my brother was born, and then after the war, they moved over to Belfast and settled in down Patrick, my early interest. We lived out in the country, so I very much spent most of my time out of doors.
I loved exploring the fields, the trees.
There was a river nearby where we fished and often fell in it.
I went sailing. My father had a boat at one stage, and later when I was at school, one of my friends had a yacht which he sailed in the local yacht club, and I crewed for him. My brother later on had a boat as well.
I liked fishing. I liked climbing mountains, particularly later with my brother. We did a tradition on Boxing Day of going down to the mourns when everything was frozen and climbing.
I was in the cubs and the scouts and also the army cadets.
I did my Duke of Edinburgh award, which included hikes, and also a hobby, which I chose. Meteorology. Not particularly because I had an interest in it, but it seemed to be an easy thing to do at the time.
We lived in a big house, so we had, when the weather wasn't good, I had dinky toys, I had electric trains and quite an elaborate system. Since I built on the system, my brother had Meccano.
I was interested in sport, mainly rugby. I'm still interested in rugby, although I don't play it anymore.
[00:03:37] Speaker A: Okay, John, thanks.
So during this time, were there any secondary school teachers or perhaps other people who had a particularly strong influence on you, either in science or other areas?
[00:03:52] Speaker B: I think in terms of early influences, my brother probably had the strongest influence. He was seven years older than me, so two too much older to be competitive, but certainly somebody who I admired and wanted to emulate. He was good at school. He also was interested in science, and he was, probably has been the main influence in my life, really.
My mother was very keen that I said, do well at school and encouraged me to do homework. She would questioned me in homework and really kept my nose to grindstone, if you like. My father, I don't remember so much taking an interest in my schoolwork, but one of his sayings was, if a thing's worth doing, it's worth doing well. So I was encouraged to do things well at school. I don't remember any strong influences apart from headmaster was a very good maths teacher and we had him as maths teacher. And then in the year before my final year, a young graduate from Queens who was doing a diploma in number at Queen's, but was looking for some employment at the same time. And he made mathematics fun.
Not always on the syllabus, but as an example, we'd one person in the class who wasn't very interested in academic work, but he's very keen on betting on horses. So this teacher would explain how the betting system worked and how the odds were lined up so that the bookie would make the money.
And as I say, that made schoolwork fun. And in my final year, fortunately, I had a math teacher who was much more disciplined, kept my nose to the grounds down, so I learned the basics then much more carefully. I'm just thankful that I had them in that order because I had them the other way around. I think it would have been a bit of a disaster, given that background.
[00:05:52] Speaker A: Did you expect from a relatively early age to go to university? And what influenced you in that direction and also when you first went to university what did you expect your education would lead to? Perhaps you can say at what point you decided on the subject of your degree, what this action was and how to remember. How did this evolve in time?
[00:06:20] Speaker B: I suppose because I wanted to do everything my brother did and he was seven years older than me.
That was probably the first thing that made me think I wanted to go to university.
Both my parents had been and my mother has been. So I suppose there was an expectation that that would happen. I didn't really think it through very carefully. My brother went on to do a PhD in chemistry, so I thought, well, that's probably what I'd like to do. But then, following the influence of the teacher had in penultimate year at school, I also had an interest in computing.
So originally I'd actually set up to go to Manchester to do computer science, but then my brother had actually won a foundation scholarship to Queens, and because of that, I think my parents lent on the headmaster of the school to enter me for a foundation scholarship and I actually got a.
I was lucky enough that I was successful in the foundation scholarship, which is quite a lot of money in those days.
And so rather than going off to Manchester to do computer science, I decided I'd rather stay in Belfast and keep the money. And so in a sense, I was expecting to go back and do chemistry.
But when I met the advisor to studies, I had the advantage of going straight into second year and that rather upset my choice of subjects. So the end, I decided to do a four year degree, three years past degree and one year's honoured. And because of that I was offered a couple of alternative courses, statistics, which turned out to be very useful in my career, and second and third year maths, which I was good at, but I wouldn't say I was particularly confident at it. Although I sort of set my mind into chemistry after the first year, I wasn't particularly confident in my chemistry practical.
I wouldn't say I was bad at it, but I thought, at least if you get a chemistry practical round, you've used your materials up and you're sort of stuck. If you make a mistake in your calculation, you can cross it out and start again. So there's sort of odd considerations like that. I ended up doing applied maths, I think the third thing was I realized that the chemistry room was over. My brother stayed on and became a lecturer in chemistry at that stage, but I didn't think the same thing. Opportunities were going to be available to me, and applied mathematics in particular, I could see there'd be a wide range of options, and particularly if you had a computing background at the same time. So my choice to what I did really changed as I went along. So at each stage, I thought, what am I good at? What are the opportunities?
What will bring me a reasonable salary?
[00:09:09] Speaker A: Yeah, well, that's actually quite a normal situation to be in, actually, at that stage of your career. So, given that, were there any undergraduate teachers who made a strong impression on you, did you go on any short term placements in your undergraduate period, or did you have any gaps in your studies where you went somewhere else? How did you enjoy the course and did you complete an undergraduate dissertation on any specific topic as part of your degree?
[00:09:44] Speaker B: Well, it was a straight university course. I didn't have any placements or a dissertation at the end, even for honours, I generally enjoyed the course. There were some parts I find easier than others.
After three years, I was interviewed for the honours course and Professor Bates, who was the head of department at that time, a very well known astrophysicist, I said to him, I'd find some exams very easy in my past degree and some of very difficult. And he just looked up at me and sort of stared over his glasses and said, that's like life, which is true.
So I qualified with a two one honours in applied mathematics, and as I say, that included degree level pure maths and statistics, and also some computer science, all of which were helpful in my career.
[00:10:45] Speaker A: That actually is very nice combination, as it turns out, in retrospect.
So what point did you decide that you wanted to make your career in science itself, or was your interest still at that time more than applied mathematics? I noticed your PhD was in theoretical physics.
Was it just doing the research that attracted you?
And did you, at that stage, want to become more a theorist than an experimentalist or an observational scientist? What was the balance of your interest in those areas?
[00:11:26] Speaker B: I don't remember taking a choice about doing science. I'd always been interested at school, and at each stage, again, I consider had I enjoyed what I'd just done and what it might lead on to in the future. So, as I say, I enjoyed mathematics, I enjoyed applied mathematics, and I was looking to, again, just keeping my options open. And although theoretical physics, which is what I went on to do, sounds perhaps a bit theoretical, for lack of a better word, it did have the application. There was a lot of computing in what I did, so it had a very practical side, having a broader applied master's degree, that put me in a good position when I came to look for work.
[00:12:16] Speaker A: So after that or during that period, as you were contemplating the future, what did your family think about your general direction of the career that you are moving towards?
[00:12:31] Speaker B: I don't really know what my father thought because unfortunately he'd a severe stroke in early 1965 and through a series of strokes in the end, he died in 1969. So I never really had a deep discussion with him. My mother, again, I can't remember her expressing an opinion, except that I did my best to kept my nose to the grindstone.
She would say, your brains if you'd use them.
But my brother, he lectured. By then he was lecturing in the university and he knew that applied maths was a good department. So generally I think he was approving.
[00:13:11] Speaker A: So coming on to the next period at Queen's University, you chose to remain there, in fact, to do a PhD. So why did you choose to remain at Queen's? What was the subject of your PhD thesis?
[00:13:28] Speaker B: Okay. I did look for jobs after graduating, atomic energy authority at ICL and a firm called Tube Investments. But I guess the thing in the background is I wanted to stay in Belfast because my girlfriend Katrina, now my wife, was doing a postgraduate course in social work in Belfast in 1970.
And it's easier to change a job than a wife, I guess, if that's what I was looking for.
And applied maths was more marketable than pure maths and the undergraduate courses, which I did, fitted well with the research in the department. So whereas somebody coming from another university would probably have had to done an MSc first, that wasn't the case for us, because most of the people in the department had lectured on the topics they were working on in research.
Things that we needed to know, we'd already been taught. The PhD was in theoretical physics. It was entitled some aspects of electron atom scattering.
Two parts, one of pseudo states and the other was collisioned with alkali atoms. And in principle, the idea was very simple. You had a ground state of an atom which had a function and you could fire an electron at it and you solve for the radial equation of the electron and that would give you the cross section how much the electron was scattered. And in fact, if you needed a whole set of the ground states of the atom, which mathematically would form a complete set, and so you had an infinite series of the sum of the product of the ground state and then the radial function of the scattered electron. But in fact, Uten had only used two or three ground states of the eigenstates of the atom because that was enough to give you an accurate enough cross section but once you started knocking electrons off, so that was at higher energies, that would be a very inefficient way to do things. So the technique was to invent these pseudo states which would cover the, if you like, most of the space that's covered by the atom. And that was a very efficient way of calculating cross sections. And the reason we're looking at alkali atoms, of course, they have only one electron outside a shell. So again the first ground state of the atom, the first excited state, were probably enough to look at the low, particularly the low energy cross sections. So it was a very, very simple in principle, but involved a lot of computing. My supervisor was Professor Phil Burke, who was both a good supervisor and a very good computational physicist, and later went on to be a fellow of the Royal Society. He was interested as a supervisor, because I worked in a, in a house in Belfast, and the professors had offices upstairs, and if I got stuck I'd go up to talk to Phil and he start talking about subject in electronaus and scattering, but it wouldn't necessarily about the problem I was doing, and I don't know whether he knew what he was doing or not, but by the time I'd finished the conversation I go downstairs and say ah yes, that's what I need to do. But he never told me directly what to do, which I thought was rather, rather clever of him.
And it turns out some of the Mars has parallels in meteorology and in detection and attribution, as I can probably explain later.
[00:16:59] Speaker A: That's actually very interesting, I appreciate that, so bring that out later on.
So did you do all your PhD training at the university or did you spend any periods of time elsewhere during your PhD?
[00:17:17] Speaker B: I did it all at university, although I do remember ongoing some of my colleagues to, I think it was a SERC course in Liverpool for people doing PhDs.
Cynically, I suspect this was for people doing subjects that they thought they wouldn't be able to get jobs when they went out to the wide world, or at least to that subject. So that was quite an interesting experience.
[00:17:42] Speaker A: When did you meet your partner, who of course became your wife, what was her background, education. And has your partner, your wife has a separate career?
[00:17:56] Speaker B: Well I met my wife Katrina at school and I was going to go to Manchester University, she was going to go to Edinburgh, but then as I explained I won an entrance scholarship to Queens, so I went to Queen's and unbeknownst to me her accommodation fell through. So she ended up in Belfaste and doing a degree in English.
We ended up in the same group of halls of residence, and by the end of the year, we started going out again.
She did an honours degree in English and literature, which took four years. She did a diploma in social studies in Queens, then a year working at Royal Victoria Hospital, which was rather an interesting time because the troubles were in full swing then. And then a diploma in social work and another qualification in social work at the University of Edinburgh. We then got married and moved to the south of England. So she worked in Newbury and Bracknell as a social worker, spent a time at the Oxford council for, sorry, the Oxford district.
Let me start adding the Oxford Diocesan Council for Social Work, which involved in fostering adoption of children. And then she had a break to have children. We have three children, all now in their forties, one in data science, one in biochemistry, and one teaching French.
Gradually, as the children got older, she went to work part time in our local church, doing social action and also children's work, which was part time and paid, and she did things like setting up a food bank and a charity shop. So, although the work, I think, was interesting to her, it wasn't as well paid as she might have been if she stayed on. Social work perhaps was interesting, but in a different way.
[00:20:07] Speaker A: Okay, I think now is the point where we should move on to your scientific career.
So tell me, what attracted you to the Met office and how did you start? Did you start as a scientific officer or what grade were you at that time?
[00:20:24] Speaker B: Coming up to finishing my PhD, Katrina and I wanted to get married, so I had thought of staying on and doing postdoctoral work. But by that time, the sort of all the opportunities in science were beginning to close down a bit and people were coming back from the states and so on. And I realized I wasn't the brightest theoretical physicist.
So I looked at a number of places, iclhemen, the Atomic Energy authority and the Met office. And the reason I looked at the Met office, I think two things. First of all, one of the PhD students in the room next to me, a year ahead of me, had gone to the Met office.
Alan McElveen said, well, it seems to be a good place. I find it interesting you should apply. He ended up being the head of the computing branch much later on.
The pay was good and there seemed to be a wide range of opportunities that you could do when you could get there. So if you didn't like the first job that you did in the organization, you might be able to another one, which was better.
So I applied.
I had an interview in London.
In Northamlin Avenue, which I remember, and they took me on, and I actually visited the office in the summer of 1973. So Fred Bushby, I think, was then the head of the forecasting search branch. He tried to get me into the training course for graduates, but I was too late, so I joined in October 1973.
[00:22:11] Speaker A: When you joined the Met office, you first went to Meto 20 for about four years, which I think was called the dynamical Climatology branch, and particularly concerned with developing models, climate models, if you like, of ideas about the general circulation of the atmosphere. Can you tell me about the science that you did there and what you think you accomplished most there before you were posted? The central forecast office in 1977. And please mention any benefits you got from your central forecast office posting, which of course occurred after the first four years. Renowned climatology branch.
What did you think about the metal training course? Was it actually matched which you did during this period?
Was it actually matched to what you needed?
[00:23:09] Speaker B: My first job was, as you say, in the dynamical climatology branch.
I was super numary, so they had to find something for me to do. And at the time, they were developing a model of the tropics for use in the gate experiment. So Garp Atlantic tropical experiment and the metro 20 had developed a model for running across the tropic high resolution model, and they had appeared in the summer of 74, running that model and daily forecasts. And I think they had a meteorological analysis scheme, but they asked me to look up, setting up a variational meteorological analysis scheme after some of the work which a japanese scientist, Sasaki, had set up.
So I took that as a base, and I spent the year programming the variational scheme and instantly the variation scheme for deriving equations. It's the same technique that you use to derive scattering equations. So I had a start from what I'd been doing at university, and over the year, I managed to set up an analysis scheme. It was never really tested in anger. I think I learned a lot both about the programming, the programming system at the office, and quite a lot about meteorology, particularly tropical meteorology.
Then after that, I was sent to the scientific officer course, which I hadn't been able to go the previous year. And I found that useful, I think, for two reasons. One, already having been in the office, seen, been in the central forecasting office, and heard a lot about forecasting in the office. There are questions I wanted to ask. I think it was easier for me than for some of. For example, the PhDs would come in with PhDs in electrodynamics or something high powered like that.
They, I think, find a bit more difficult with some of the ordinary forecasters trying to teach them thermodynamics.
And so I enjoyed it, interestingly, about forecasting, but also useful, because later I was posted to go forecasting, and I wouldn't have been able to do it without that training.
We also spent a month at an out station and we did a small topic at the end, which I did on heavy rainfall in Scotland.
So then I was posted back to Meadow 20. I spent about a year modeling the upper layers of the ocean, if you like, the boundary layer of the ocean, and based on a mixed layer model formulated by Krauss and Turner. Essentially the mixed layer is heated by sunlight.
That means that the top layers, particularly in summer, are light, and then that light water is mixed down and forms a mixed layer as a discontinued the bottom. So it's essentially, if you like, a conservation of energy.
And I developed that and tested it.
I then was posted to the.
No, sorry.
I was asked to do something else. And Chris Gordon took over, developed the mixed layer further for use in the couple model. I think Chris make a very good job of it. But at that time, summer 76, John Mason gave the Simmons lecture, and he wanted to talk about the end of the last ice age and the role of the change in orbital parameters, which led to much stronger heating over high latitudes in the northern hemisphere in the northern summer. And Ken Carpenter and I were asked to do an experiment at very short notice. So we ended up having a global five layer model with interactive cloud and then 1 meter mixed layer, which isn't very satisfactory, but it allowed the model to respond and it was written up in the quarterly journal. I think it was communicated by Andrew Gilchrist in John Mason's memorial lecture.
I felt it was very unsatisfactory, so I redid the experiment with climatological ssts, in other words, fixing the sats at present day, but allowing the land surface to react, because the main mechanism of particularly the monsoon strike thing was an increase in land sea contrast. And you get most of that by letting the land response. I wrote it up as a technical note, and that actually helped me later on, as we will discuss at that stage. I was then posted to the central forecasting office. It was very much the policy of the office to take some of its research graduates, PhDs, and send them forecasting to give them some practical experience, and also, particularly if they're going to enter the forecast research branch. They had some idea of how forecasts were used and they would see the practical side of forecasting and not just the theoretical side. And that I did. I spent most of the time doing the shipping forecast, along with David Parker, who was on at the same time. So we often handed over shift to each other.
It was very interesting. You got quite a kick. If you were, for example, there was a gale down the North Sea in the middle of winter and a yacht would ring up and said, how long are these gales going to last? And you'd be able to tell them, hopefully, what the expectation was. I think the lowest part of my forecasting career was I put up easterly gales for Newcastle Weather Centre and they turned out to be easterly because the low just went a little bit the wrong side of Newcastle.
So I knew my future wasn't in forecasting.
[00:29:22] Speaker A: Well, after your experience in forecast office, you were posted back to meso 20. That was in 1978. So did you ask to go back and what research were you involved with next? And I think this coincided with the time that interest in anthropogenic climate change first developed politically in the UK and the USA was a national academy of Sciences report on climate change, I think in 1979. So can you tell us a bit about that period?
[00:29:58] Speaker B: Indeed, and the political background is important.
In 1970, the Department of Environment was formed from housing, local government, public works and transport. And I guess, having a bigger remit, they tend to take a broader outlook on what they should be looking at. In 1971, there was a study of man's impact on climate. 1972 limits to growth. And in the 1970s, within government, there were discussions on climate change, particularly in the late seventies. And could the UK do something? There's a note by Jonathan Agar in 20th century climate change, which talks more on the political side of that. So I didn't ask to go.
I was told I was coming back and posted back to Metto 20 to help them get up to speed and do something. And I think the Met Office wanted, or I suspect it was forced by government to do something pretty quick on climate change, on the effects of CO2.
So that's how I got in fine in climate change.
There was a group set up in the Met office to work out what we could do. We didn't have a couple model, so I had to use prescribed ssts to get up to speed. I looked up the papers by Manabe and Jim Hansen, Warren Washington Gates and others, the modellers of the time. And it's worth noting that just as I had a conversation going back to aside to my time, in Belfast as an undergraduate, I was a member of the physics and applies math Society on the committee. And we invited David Bates, the head of the department, out for a meal. And he told me that when he started in his field, he could read all the papers in his field.
And he said he couldn't do it now. And I was in that lucky situation in 1978. I could read pretty well all the main paper, climate modelling papers, so I was able to get up to speed quite quickly. That's something I would never be able to do now.
So following that, I did a trip to the USA.
Andrew Gilchrist wanted me to visit JFDL and Suki Minabe, who was pretty well the world leader in this field, and of course, has since become a Nobel Prize winner. And we applied to NATO for a travel grant to visit GFDL.
And Andre Berger, who was interested in paleoclimatology, had seen my take note on the Holocene and the effect of orbital perturbations on climate. So he knew a bit about me and he, as a result, I think that's why I got the place, I got the money to support me, and I spent about three weeks there.
Suki was at first a bit of dubious about whether I was worth talking to or not. I didn't have a name or I didn't have any experience.
And he sort of part me with Ron Stouffer, who by then had begun to be one of his main collaborators, who was a sort of similar stage in his career as I was. And I spent a lot of time talking with Ron, and he persuaded Suki that I was worth talking to. And I've since had a friendship with both of them, particularly with Ron.
I had only two sessions with Suki, but it had a huge influence on my approach to modeling, and it helped that I'd read a lot of papers before I went, so I was up to speed.
He had the annoying habit of, I would say I was doing something, and he leaned over to his drawer and pulled out and said, yes, we did something on this five years ago, but we never actually wrote it up. But the things I learned from Manabi's philosophy and influence really set my philosophy for modeling.
A number of principles ensure the models are based in physical understanding, because unlike forecast models, you can't validate them against a whole series of realizations. The real forecast, we've only got one or two climates that we know anything about.
The second thing was to don't make models more complicated than necessary, particularly for this particular problem. Not every science question needs a high resolution atmosphere biosphere chemistry model. In fact, sometimes it can make things more difficult. Include the processes that you need to address the problem that you're investigating.
Don't make the model processes more complicated than can be justified.
So he was again some of a very complicated problematization to convection when we don't know much more convection at that time, about the fact that transfers heat and moisture into the atmosphere according to the limitations of the stability watch signal to noise, and particularly for things like rainfall, where there's a very low signal to noise ratio in the small changes that we've seen to date, and if necessary, back up with idealized experiments or experiments where you've got a much bigger forcing to help you see what's going on at the more realistical level level of changes and finally understand the model mechanisms leading to change and to extent which they are credible.
[00:35:58] Speaker A: Well, John, that's very interesting, and that was obviously a valuable set of hints for you to help you with what became the most important single part of your career. Youre CO2 experiments. So can you now tell me about your first climate experiments with changing carbon dioxide?
[00:36:25] Speaker B: Because of the urgency of situation and the fact that we didn't have a couple model, I probably broke out most if not all the rules I've just outlined because we didn't have a credible couple model.
I did experiments both with a climatological sea surface temperatures prescribed, so the seasonal cycle of the sea temperatures was prescribed.
We did a long control run. We then did a run where we doubled atmospheric CO2, which didn't give a big response because obviously the sea torpedoes weren't allowed to change from their control values. And then finally a simulation where we increased the sea temperatures by two degrees and double CO2.
And the model had five layers, but it had quite, for then what's quite a high resolution, three degrees.
And that sort of the resolution sites are quite important for precipitation. And the preliminary results were actually quoted in the Charney report, the famous where the famous one to 0.5 to 4.5 degrees just in passing.
But it obviously didn't contribute to that conclusion.
Later I added runs to that. The original runs were appeared in the quarterly journal, which is my first meteorological environmental paper in ten years.
And that changed over the next few years.
And I was helped by the fact computing power in 1982 increased considerably. We started running the Sabre 205 and I repeated the experience we'd done with an eleven layer model and a two and a half by three and three quarter grid and concentrated on precipitation.
I noticed a couple of things. One, the two models had different control climates, so the centers of precipitation were slightly different. Monsoons were slightly different than and so on.
But the pattern of changes was similar in the two models. So, in other words, to first order, precipitation tended to increase where it was wetter and to reduce where it was drier. It was far from as simple as that. And the reason for that, of course, in a warmer atmosphere, you've got a greater moisture content. And so areas where you've got strong convergence, like the monsoon, in depressions, that's going to lead to more rainfall.
I think that was the first thing of interest. The second thing of interest was how CO2 and water vapor affect the atmospheric heat balance. So, in the present day atmosphere, the long wave heating due to precipitation is largely balanced by, sorry, the long way cooling in the atmosphere due to water vapour is largely balanced by latent heat release from precipitation. There's obviously a contribution from solar and sensible heat, but that's the dominant mechanism. And so by looking at the changes in the long wave cooling due to increasing CO2, and also, of course, then the accompanying increase in water vapor, you notice, first of all, CO2, if you increase it instantaneously, you actually heat the troposphere. There's a lot of cooling in the stratosphere. And that explained when I look back at the simulations where I just increased CO2. When you start the experiment, the precipitation reduces because you've reduced it, long wave cooling. But then as the surface starts to heat, then you get more water vapor, you increase the longwave keeling, and that then leads to more precipitation. So precipitation does actually increase. And that work was published in the quarterly journal in 1987, and where I was beginning to start to look at mechanisms and understand some of the mechanisms which might be also important for regional climate change.
[00:40:46] Speaker A: I think many of those insights, if not all of them, in fact, have been shown to be very much true to this day. And then you came onto another fascinating insight.
A key uncertainty in climate change feedbacks are changes in cloud amount and other climate changes properties, of course, as well. And I remember an early experiment you published about 1989, I think it was, that showed that cloud amount feedback was very sensitive to the way cloud feedbacks are parameterized. And this led to one of the most clear memories I have of that period, where in three different experiments, you found surprisingly large variations and the best estimates of surface global warming, for example, carbon dioxide. So, can you tell me more about this and indeed, your later cloud feedback work and how significant progress actually you made since that time on modelling cloud feedbacks.
[00:41:53] Speaker B: So the next stage was really to build a climate model that we could do proper climate experiments with.
And I prompt you to just say a bit about building climate models. It's a very labor intensive task. We had some very good staff in the Met office at the time. In my group at this time, I had people like Trevor Hills, Clive Wilson, Julius Lingo, or Walker, as he probably was then. Now Dame Julia and Martin Cunningham.
But not only the people in my group, but the people from the other groups modeled the development, including Peter Roundtree, who worked on convection, Tony Slingo, his group, and particularly the radiation schemes. The Edward radiation scheme, which he developed was critical.
The ocean group, led by Howard Cattle, and often overlooked model software and diagnostic groups, including people like Mick Carter, and finally observational data to test them. Models led by people like Chris Follans and Dave Parker. And with that help, we managed to produce, led by Clive Wilson, a low resolution atmospheric model with eleven layers coupled to a slab mix layer. And for the first time, I felt we had a model which was fit for studying long term climate change. And it was used for many experiments.
Probably the most significant experiment is, is one we did looking at cloud feedbacks, which were probably the biggest uncertainty in global sensitivity.
We were able to do that because Rob Smith had developed a parameterization of cloud amount and radiative properties for the atmospheric model, and we were able to put the scheme into the mixed layer model. So we ran three CO2 doubling experiments. The first had the standard radiative scheme played based on relative humidity. The second had the cloud water scheme with prescribed radiator properties, so the cloud amounts could change, but the radiator properties couldn't change. And the third, the cloud amounts, like the reflectivity, were dependent on the cloud water amount with the improved scheme. So the original relative humidity cloud scheme gave a warming of 5.2 degrees. When we put in the cloud water scheme, it went down to 2.7 degrees. Castinho, who was running the experiment, spent a lot of time checking it to make sure we hadn't made a mistake, that we hadn't got some energy sink that was leading to this drastic cooling.
But she did a good job checking it. And then we allowed the cloud water changes to influence the radiative properties of the cloud. And that led to a further cooling down to 1.9 degrees, largely because when you allow water cloud to turn into ice cloud, it dissipates more quickly. So in reverse, if you heat the cloud, you tend to get more water cloud than you had ice cloud. And we did a further run, actually, which we improved some of the aspects of the scheme, which led to a final warming of 3.2 degrees that appeared in nature with cast senior and William Ingram as co authors. And it highlighted just the degree to which small changes in parameterization, in this case, clouds, could alter climate sensitivity. So I think it was, if you like, a warning to be careful, to be modest about how much we knew about climate change and led to a lot more work on cloud properties, cloud feedbacks, cloud mechanisms.
[00:45:42] Speaker A: So have we improved our understanding of cloud feedback to be more confident about this topic, which is obviously key to climate?
[00:45:55] Speaker B: I think progress has been slow and of course, there are a lot more models around now, so that if anything, tends to, in one sense, increase uncertainty. But I think in general, probably some of the more unlikely cloud schemes have been eliminated. We've had improvements in observations, particularly things like cloud radar and so on, satellite cloud radar, because the important thing is the vertical profile of the clouds and the properties. So we still have way to go. But I think there has been slow progress.
[00:46:30] Speaker A: Okay, so we come on to really what I would regard as the next phase of your activities. They just follow on naturally. So, from about 1988, you played a key role in planning for an early development of the Met Office, Hadley Centre for Climate Change, as it was called at that time, which started formally in 1990. So can you tell me a bit about your role here, including the development and initial implementation of the Department of the Environment climate change contract as it came to be?
[00:47:15] Speaker B: Well, the events leading up to the Hadley Centre, they're sort of the wider things in terms of the political interest of misses satcher speech to the Royal Society.
At the same time, there was a setting up the IPCC at the end of 1987, and that was funded by the Department of Environment.
And we'd link to the department Environment. Jeff Jenkins had been seconded in there from 1984 to 1948, and he told the chief scientist, Dave Fiske, that climate change would become important. And we already, as a result, had some contracts with the Met Office.
In January 1988, I was asked by Peter White to write a ten year plan for climate change, which I did.
By November 1988, John Horton had written today, Fisk with a national plan for climate change without appendices.
There was then quite a lot of debate. Would it be a national centre? Where would it be a european centre? A new building and finally existed on existing facilities in the Met office with extra staff, computing and a three year rolling plan.
By the end of 1989, DoE were seeking treasury approval and that went through. And the Hadley centre was opened by misses Thatcher in. I think it was the 25 May 1990, which was also the final day of the first IPCC report. And we had delegates bussed in from Windsor. I remember afterwards, misses Thatcher went round the displays with her two local mp's and was briefed by various scientists, including myself. I showed her the warming over North America and she turned to her mp's and said, we should buy land in Canada.
But then I showed her that it also got drier and she said, perhaps not.
And at the end of the opening, all the IPC delegates, including myself, were bused back to Windsor.
[00:49:36] Speaker A: Thanks, John. Now, it's clear that climate modelling was a key part, the key part at that stage, particularly of the setting up of the hamper centres. But I remember that in the early years of the Hadley Centre, it was quite a lot of difficulties in the climate modeling area, which were, of course well resolved later. So why did this happen? How did it affect the development of the science, how and exactly when the problem largely resolved?
[00:50:16] Speaker B: Well, running a contract with the government department was a new experience for us, because up to now, almost our funding, we'd been vote funded.
When we met David Fisk, around the time the opening, he was chief scientist at Department of Environment. He said no prizes for coming second, which set the tone for the centre.
The Hadley centre coordinator managed the implementation of the contract.
First David Bennett, then David Dinkins.
They had regular and frequent meetings with their dov equivalent to develop the contrast. And one of the things they did which was very useful, they involved the appropriate Hadley Centre scientists.
And that meant scientists learned what the issues are, political interest from the DOE, and they can anticipate new science needed to address them. And the DOE staff learnt of the scientific limitations and the opportunities that arose. And then the whole thing was there was a progress was reviewed by Science review committee of UK and international scientists.
For my group, I had two initial priorities. How big is future climate change likely to be and what are the regional changes? And I put that in a view for greet meetings.
So we experienced staff from the existing groups with a lot of new staff, including people like Jonathan Gregory, Simon Tutt, Peter Stoughton, Chris Hewitt and my group.
We had other modeling groups from the old climatology group and we had the old crate building at the end of the Met office car park, which gave us our own identity and we had computing capacity, but we had problems.
At this time, the office had chosen to introduce a new unified forecast climate model.
This is something I supported strongly because it would hopefully reduce maintenance and also give some cross fertilization between our experience forecasting as well as climate change.
And unfortunately, at the same time, CDC pulled support from the expected new T EtA ten computer, I think late 89 90.
So cray provided two cray machines instead. They were known as Ronnie and Reggie.
One was blue for mod and one was green for DoE. And the new model for unified weather and climate. For weather worked very well. But for climate, it didn't work so well. There's quite a lot of coating and it worked okay at low resolution, but it simulation of the North Atlantic was just not suitable for a couple model.
And also in the couple model, the antarctic simulation was not acceptable. Hence we'd require them to do a run with a gradual increase in greenhouse gases and new computers to run it on, but no coupled model to run it out.
So what we did, we kept the old cyber 205. And James Murphy completed a run on a model that had been developed on the old cyber machine.
We used what we call flux adjustments to heat and salt to keep the model temperatures from drifting from present day climate. And those fluxes were then kept the same for anomaly runs.
And an initial experiment with one year half year increase CO2 was completed. A couple of years later, results were published and I went with Dave Bennetts, who presented the results to the John Gummer, who was the department, the minister for the environment at the time. At the same time, we used the new low resolution atmospheric model coupled to a slab ocean for sensitivity studies, continuing to look at mechanisms climate change, including cloud feedbacks, notably by Cass senior, on summer drying and effective orbital variations. So the new group remains scientifically productive. And department environment had a flow of results, even they weren't always the ones that they wanted.
Meanwhile, all who could contribute it to the model, the new model did. And in the end, we solved those problems. One was a problem with gravity drave drag. The sea ice problem was occurred by including infection and movement of sea ice.
And that model provided the basis for a large number of simulations using looking at scientific questions, simulating the instrumental temperature record of the last hundred days, detecting climate change, global mon predictions under IPCC, and also looking at past climates.
[00:55:30] Speaker A: Thank you John. So it would be useful at this stage to give you some more detail about how your research and that of your group developed in the decade after these early years, especially in the era of climate change. And of course, a key background was increasing in the successful development of the new couple models after the period that you mentioned, with better computers, of course, helping, culminating in the remarkable, certainly globally stage of the art model head cm three.
And it's worth mentioning before you say all this that you've been promoted in what today's language is called the science Fellow in 1998, and during the period you're going to discuss this was upgraded in fact in 1998 to a higher level.
And importantly, you developed as head a world leading climate change group which is still peace. And I remember particularly the Seminole work you did in this later period to include the influence of anthropogenic aerosols in model. And this was published in Nature. And I think during this period you awarded the Daliano Norbert Gervais International award particularly for this. And you also won the same award again the very next year. What was that for? And you also published many other papers in nature and science.
So can you please give me some prominence to this period, this decade, to results that were published in high profile gels where this is appropriate?
[00:57:29] Speaker B: Okay, so I'll start with the first simulation with greenhouse gases in aerosols.
And it was fairly obvious that if you just include greenhouse gases that increases over the last hundred years, the warming in the models tended to be larger than observed. So that was one motivation for including aerosols, but it is probably the second largest forcing on climate.
So with the model that we developed first in the contract, we set up a simulation which not only included increases in gases but greenhouse gases represented by their CO2 equivalent. But we also wanted to increase the cooling due to sulfate aerosols. And this we represented in a very simplified way. Following Bob Charlson, Henning Rodd and others, we represented the scattering by the aerosols as an increase in solar albedo at the surface.
They provided a data set of the geographical onion meal sulfate loading, and Peter Rowntree provided a time series of emissions to give an estimate of the annual mean sulfur loading. And that gave us a dataset of aerosols which increased geographically over the period of the last hundred years. So we ran a control.
We ran a control with greenhouse gases only and then a run with greenhouse gases and sulfate. And when we compared with observation, the greenhouse plus sulfate gave better agreements with the observers in the global mean and also over Southeast Asia, where the inclusion of aerosol weakened the increase in summer monsoon seen in most experiments with increases with greenhouse gases.
The original results were published in Nature in 1995. We were just two months ahead of NPI, who find similar results in similar experiments with their model.
We had a bit of an issue with DoE, the Department of Ireland, because they'd insisted on some of the results being put in a brochure, which they wanted for cop two, and that slightly upset nature. But we managed to come to an amicable agreement.
The study was highly significant, both scientifically and for policy, for the comparison with past observation, for the comparison with past observed surface temperatures, and also for the projection with a more plausible set of emission scenarios.
And those simulations featured in the second IPCC assessment. And the paper won the Gorbiat mum prize in 1997, as Chris has said.
So that was the. So the first thing, we went on to do more work in detection attribution, which I'll talk about a little later time.
The next thing that Chris mentioned by in this period, we developed a new ocean model, had CM three, and this was with much higher horizontal resolution, which allowed the use of a smaller viscosity, which allowed for sharper currents and so forth. And the new model under the leadership of Chris Gordon. And it became available, I guess, around 1999, just in time, as it happens, for the third assessment, for work to be done for the third assessment, an update in computing second, t three a enables to run the higher resolution, and there are many atmospheric improvements, including cloud formulations, additions of a fully interactive sulfate cycle, and improved, and its parameterized effect on radiation and cloud. So, not just the direct effect of sulfate, but its effect on clouds. And the basic model simulation at the time was really quite far ahead of other groups around the world, and it was probably the best model at the time and continued to be for quite a while.
So, one of the first studies we looked at in this model was the effect of the increase in green gases on the thermo halide, sorry, the thermoheline circulation in the North Atlantic.
And of course, this is something where if the model had had flux adjustments, it would be very, have very little credibility, because if the model wasn't simulating the proper North Atlantic circulation without help, whereas that circulation was well represented without any artificial fluxes, and most models show a weakening of that circulation. And through the leadership of Richard woods, we looked at the overturning circulation in the Labrador Sea and around the Greenland and Iceland ridge, and we found that indeed, particularly in the Labras or sea, the convection, the ocean convection was greatly reduced.
But over the either side of the Greenland, Iceland, Scotland Ridge, it remains stable.
This is an observable result. You can actually go out in measurement that 30 years, 20 years ago, 20 years ago, and the observations have been made since, but they're not long enough to determine whether this result is right or not. What was has been noticed that although the thermohalate circulation in the laboratory is deep, it's not particularly strong.
So we'll have to wait some time to see whether these results have been validated.
[01:04:05] Speaker A: Yes, and I think at the moment we're seeing a lack of warming, in fact, south of England, which is thought by some to be related to a slowdown that's been published as the global warming coal. Can you just remind me though, John, what your second Norbert gov international reward was called? I think that would have been in 1992.
[01:04:30] Speaker B: Okay, so this goes back to the.
Really to the star, I guess. And we did some, we combined with Ben Santor and a lot of other groups, GFGL, for example, to look at a lot of simulations of the effects of increased CO2 over the last and other factors over the last hundred years. And that appeared in, I think it was, I can't remember, was a nature of science, but the paper was led by Ben Santor and quite a few other people. And essentially it showed that largely through looking at the vertical distributions of temperature change, that there was an observable effect of global warming on climate change.
And that formed the basis of the second IPC statement. The first time that we claimed detection of the effects of CO2 and that received the without everyone received the second Gerbert Norbier mum prize.
[01:05:52] Speaker A: Was this an example of where the improvement in observations, which I think were partly due to those some people in Hadley Centre radio observations, did that help a lot?
[01:06:06] Speaker B: I think generally the observations have been improving with time and course is as we progress, we've got a longer record which helps.
The one specific example where observations made, developed and by the Hadley Centre made a big difference was a radio sound data set, which I think Dave Parker was involved with, and in a separate paper, I think led by Simon Tett, but with others, including David as an author, we showed again agreement in those patterns. The difference with what we'd done in the IPCC report was that we'd added an effective ozone and that makes a bigger difference in, particularly in the southern hemisphere.
[01:06:56] Speaker A: John, you were from about 1988, you were very much involved over a considerable period time with the intergovernmental Panel on Climate Change, especially as a lead or convening or sometimes called coordinating lead author. Now can you tell me how this interaction with other scientists around the world developed in your area through the successive reports you're involved with?
Can you as part of that, give me a little sub prominence to your important involvement in later reports about the key topic of detecting and attributing climate change, particularly the method of optimum detection.
[01:07:43] Speaker B: Over this period, we've formed a collaboration with Miles Allen at University of Oxford and Simon Tett, Peter Stott and myself, and also William Ingram, concentrating statistical techniques for detecting patterns in time and space.
The discernible influence of climate was based on pattern correlations. It doesn't tell you anything about the size of the warming. So we started using optimal detection, which essentially is taking the patterns of the change due to CO2 and other factors and fitting them to the pattern of change over the observed periods. And that will give you an amplitude of each pattern and how significant it is. And that's the basis of optimal detection.
The first study we did with the old model, we included greenhouse gases and aerosol patterns, plus changes due to volcanic eruptions and solar variations, as two natural factors. And to cut a long story short, for the periods of 46 to 96, it excludes the influence of natural factors and attributes most of the warming due to human causes. And that was peered in tete Dalit nature.
And because this gives you the optimum amplitude, you can create that pattern and you can scale it into the future according to emission scenarios. And because models give different results, what this allows you to do is to take the results from one model, find out what your optimal amplitude is in the pattern, fitting for the observed changes, and then use that scaling on the simulated changes in the future. So if it turns out that the model is producing a response to two times CO2, which is say, twice the amount that you see in the observations, then you can halve the amplitude in the future and correct them.
We did this paper in nature with Miles Allen and others, and the prediction was tested later against the decadal change over 2001 to 2010, published in 2013. And you can see good agreement over that. So it does give some confidence. I think there's still more testing to be in that, but it is a useful way forward.
I then became the coordinating lead author for the detection attribution, the third report, along with David Caroly. And there were a couple of things I wanted to do. I wanted to make detection and attribution of climate change more accessible to non specialists.
So, showing diagrams, mean temperature changes observed, and from simulations with and without anthropomorphic factors, whether it be global mean changes, the vertical cross sections or the maps of changes in temperature temperature, as well as including a strong scientific basis.
And secondly, I'd expect to show more confidence with more and better models, better techniques and six years more observations.
So we had a dream. Team David was an expert on simple tests which people can see what's happening. Gabby Hagril on past climate changes. Francis Weirs covered the mathematical basis and pattern testing, and mine, Alan and I, the application to studies and our contributions from the Hadley centre were simulations using Hadzean three, which doesn't have flux adjustments. We did three sets of simulations led by Peter sot. The first was natural forcing only, the second was greenhouse gases only and anaphylactic factors. And the third, which I must admit I was initially against, but I was persuaded by Peter Stott and Jeff Jenkins was with all the anthropogenic and natural factors included.
And that appeared in the policy makers summer and we. After three years of debrilation, we came out with the phrase that a substantial amount of the 50 year warning during the last 50 years was due to greenhouse gases.
The Saudis didn't like this. They said substantial is a very. They didn't use the word dangerous, but a very persuasive word in you, in circles. And they wanted us to change us.
We went back, we talked to our other lead authors while we were in Shanghai at that final meeting, and we convinced ourselves that we could actually say most of the warming was due to the increase in greenhouse gases, which I think rebounded a bit on the Saudis because. Because that, to me, seemed to be a stronger statement. But we were very careful to make sure we could justify it.
[01:13:15] Speaker A: Thank you, John.
Now, the Intergovernmental Panel on Climate Change, indeed, anthropogenic climate change in general, has been very contentious in certain quarters, with very shade, sometimes vehement scepticism in several countries, which clearly persist today in some quarters. So how, during the period you're involved with IPCC, did this affect you and your colleagues? And have you got any interesting stories to tell us about this, especially again, in the context of IPCC?
[01:13:54] Speaker B: I suppose, for me personally, I was the review editor on the paleo chapter in the fourth assessment, and I did get chased under FOI for some of the correspondence between the authors of that chapter. And it was near the beginning of Foi in the UK. We didn't have a lot of experience of dealing within the Met office, and I suspect I and others didn't handle it very well. So we. It probably caused us more pain than it should have.
In the end, it went away because climate had occurred and suddenly Phil Jones was in the target of all the skeptics, which was very unfortunate for Phil, but it is stressful.
I was lucky.
Other people like Phil suffered horribly. People may have seen a recent television program which went through Pill's experience, and earlier, not through so much through FOI, but again by skeptics. Ben Santor had a very rough time after the second assessment and the discernible influence of climate change, which the skeptics hated. He was accused of changing the chapter after it had been approved, but he had actually behaved totally in accordance with the IPCC rules and there was no fiddling the result.
The problem for me is that if scientists are assessing something, I felt I wanted them to be free to have a conversation where I didn't have to be thinking about each thing they said or wrote in the conversation up to the final, if you like, the final contribution to the report, and to be able to use phrases like the phrase that Phil Jones used of the trick of taking that and supplying it, a trick being a technique which helps you get around it, difficulty, rather than a trick which is something to deceive people.
That's the sort of thing I think is unacceptable, but it's hard to avoid.
[01:16:18] Speaker A: Indeed, no. I remember another topic which you're involved with, which is never, I think, an official part of the Met Office department contract. And that's your work. And with many others, I think, outside the Hanley centre of the modeling paleo climate, particularly the Holocene climate maximum, the last ice age.
The Holocene climate maximum was particularly interesting to me because it supported the idea, from paleoclimatic observations, that the Sahel region of West Africa, and even further north, was much wetter during Hollowseat maximum. And this probably relates to the current climate variations, where sea surface temperature variations in the tropical North Atlantic particularly and probably elsewhere, affects our health. Rainfall, that's droughts and wet years and decades, as well as rainfall variations in adjacent regions of the soil. Now, I can see research these topics over many years, and at least one because it wasn't an official part of the Met Office contract, getting colleagues in the USA to rather willingly provide the leading computing effort. So, can you tell me a bit more about your key findings in the many papers that you and your colleagues, of course, have published on these topics over the years?
[01:17:52] Speaker B: So, in 1985, I was invited to a NATO Advanced Studies institute in Grenoble on abrupt climate change. And on the way home, on the plane, I sat beside Lane Street Perrott from the University of Oxford, who's an expert on past lake levels, especially over North Africa, during the summer monsoon during the mid Holocene, 6000 years ago. And there's a lot of evidence from then that the Sahara was much greener and at present implying a wetter or more extensive of monsoon.
We'd already done a couple of experiments at 9000 Bp where the orbital parameter changes are bigger, but there was also an ice sheet, laurentide ice sheet, so that whereas at 6000 bp there was no ice sheet. So with those earlier experiments the african monsoon did strengthen and shifted slightly north, leading to a wetter surface and also a lighter cooling at the surface. However, the north extent of the precipitation was less than suggested by late data.
The removal of the Laurentide ice seat did lead to a slightly bigger response, but it didn't really affect the northward movement of the monsoon.
So Elaine suggested doing an experiment where we used a darker albedo, replacing the bright sand with over the Sahara with a lower albedo, in other words a lesser reflectivity. And it did indeed induce a greater northward shift in the summer rain belt, which suggests that there may be a positive feedback due to vegetation.
That collaboration continued, and about that time Larry Gates started the PMIP, the paleoclimatic model into comparison project led by Carl Taylor and Sylvie Joe Som, similar to Ames of the couple model into comparison but dealing with paleo.
And they chose two periods where the climate drivers were thought to be well known and simple, and also that the signal to rose ratio was relatively large.
And also climate might also help shed light on future climate change. So they came up with 6000 years ago as one period to look at, especially its effect in northern hemisphere vegetation and the precipitation response in high latitudes in North Africa.
And there was already a Holocene data project, so the data for that period had been collected. And similarly the last glacial maximum, which had a good signal to noise ratio in the global mean temperature and reasonable data because of the again there was a data project collecting that had collected data for that period, although neither were ideal and hence a lot of the work on paleoclimate then concentrated in those two periods. We also did extra stuff on 6000 bp.
Moving from a model with present day ssts to one with the fully coupled ocean also led to an increase in northward movement of rain built. So it seemed that both the vegetation response and the ocean response may contribute to that northward response.
We also looked at the last glacial maximum one doing an experiment with a mixed layer model where we tried to estimate the radiative forcings compared to the last glacial acetyl maximum. We found about two thirds of the forcing of the cooling was due to the ice sheet and about a third due to changes in greenhouse gases.
We also wanted to do a fully coupled experiment for that period using the new model, which didn't have flux adjustments. But as Chris has noted, this work wasn't supported by the environment and there's no way we could sort of sneak in the computing to do it. But fortunately, due to a chance conversation with Ron Stouffer, an IPCC meeting, he told me that they had a small t three computer at GFDL, the same as we were using, same brand as we're using at Hadley Centre, which he thought would be available for them to us to run the experiment on. And so with much groundwork by Ron and Tony at GFGL, and help from Paul Valdez, who was Chris Hewitt's supervisor, Chris Hewitt's work, helping with the word, the model was transferred, the experiment was completed, and the changes were largely in line with paleo reconstructions, except for a local warming in the North Atlantic, which seemed to contradict some of the paleo data. And the mean surface cooling was just under 4k.
So we had some evidence from past climate that the model was in the right ballpark.
[01:23:03] Speaker A: Thanks, John.
Now we come on to the later part of your career, and a highlight of your career in many ways, was your promotion to chief scientist of the Met Office in 2002. So can you tell me how this happened?
What were your roles related to this in other respects, in the rather complicated periods at Met Office that followed? We had the move of the Met office to Exeter in 2003, and for a variety of reasons, there were four chief executives in four years around this time.
[01:23:50] Speaker B: Well, with the move to Exeter, I really wanted to stay in reading, perhaps at the Met Office post output there, but it would split the group. And then I was persuaded by the head of the Hadley Centre and Paul Mason, who was my predecessor and chief executive, Peter Ewans, to ply.
And to cut a long story short, when Peter asked me to meet him at the Fox and hounds after the final interviews, I thought it was to tell me no, and I was quite shocked when he said yes, so I had to think very carefully about it. But in the end, I negotiated a day and a week in reading to do research, and it was fairly obvious I'd probably at least spend one more day in the southeast during the week.
And so after much heart searching, I said yes.
As Chris has indicated, it was a difficult time.
The trading fund put pressure on funding and it had just started to bite. We'd had some practice years, but now it was being implemented for in earnest.
There was a move, and again, as Chris has said, we'd had four different chief executives in the four years. And also I didn't have much administrative experience at this level.
My sort of aim as chief scientist was to make sure we recruited good science and good science leaders and get on with it.
And I have to say, Peter was very generous when I first joined. He helped me settle into the board, ensured that the research funding was preserved and not raided to support commercial development.
And he retired, was due to retire then, and David Rogers took over. He was there for a very short time, but he did actually, there are a number of things that were very helpful. He signed off, he set up a proper basis for the periodic renewal of the computer, rather from starting from scratch each time, and that was particularly important.
And he renewed the contract with the NERC for the research plane, National Environmental Research Council with the research plane, which I think otherwise would have gone and not been replaced then had Mark Hutchinson. I actually got on very well with Mark, but he was only there as a stop card and he only wanted to stay for as long as he needed to sort some of the administrators issues out. But unfortunately, he actually wasn't able to get away for two years.
But he did bring a lot more stability and at that time, the support for the organization at what was much stronger, the realization that one of the things that determined our success was to be at the edge of the curve in computing.
The interesting about Mark is that he was not a meteorologist. He asked me to take on the role of UK permanent representative WMO, and this included things like signing agreements with Met services in South Africa and also with China in Beijing.
And then John Hurst brought, I think, a much better approach to commercial work in balance with research.
He was very supportive of research because he'd come from the ICL paints division, where the new developments were, what kept them competitive.
[01:27:46] Speaker A: Thanks, John.
Perhaps we can turn away from administration again and come back to the last part of your Met Office career. So you took on other roles which you perhaps could mention.
But can you complete a discussion of your metaphors career by telling me, including how those roles helped in how it developed over that last decade and what you achieved before you retired in 2018?
[01:28:26] Speaker B: Early on in Mark Hutchinson's tenure, the MoD chief science at the time, Sir Roy Anderson, had proposed setting up a climate scientist post in the office of the same greatest chief scientist. And I was asked, was I interested?
And I said yes. But both Mark and I were dubious that anything would happen. But indeed, about a year later, Roy had got agreement from MoD for the post and I agreed to take it on.
I've been privileged to be appointed chief scientist and I enjoyed, I would say enjoyed most of my time in the row and meeting and working with a lot of interesting and often powerful people. However, administration isn't my favorite occupation and I wanted to have a stronger connection with science.
And so unfortunately the first recruitment reign to replace me failed and a year later Julius Lingo was recruited and again unfortunately Julius husband died in October 2008 and Julia delayed her arrival for six months. So by then I'd completed over six years of chief scientist.
And this time I also started reached the stage where I could start to reduce my working hours considerably.
My role included advising Julia on climate science where needed. For example, when the so called pause in global warming occurred early in the new millennium, I contributed to a couple of Met Office papers which Julia had commissioned. I also helped with running the program for updating the UK Climate Impacts program, Climate change scenarios, which was a huge operation involving setting up and running ensembles of a new, largely intestate climate versions of the global model, regional model and climate models. And Julia also asked me to look after research fellows for which you created the post of principal research fellow. And that was probably the part of the job I enjoyed most. As I say, I ran down my r's over that time and my last official talk was the 30 year anniversary of IPCC in Paris, talking on the achievements of working group one, which I thought was a nice way to end my career.
[01:30:44] Speaker A: Thanks, John. Well, as a result of this outstanding career, in fact up to a very propitious time in meteorological science, you have a number of awards of considerable magnitude. You awarded the OBE in 2001 and became a fellow of the Royal Society in 2004.
You also won the Royal Meteorological Society Symonds gold medal in 2011 and I think the European Union's Oscar medal in 2004. So can you say briefly, for what achievements did you get at these awards?
[01:31:31] Speaker B: The OBE was officially for services to the meteorological office. But I feel in a way, it was almost an award to the Hadley Centre and all the people who had contributed.
Royal Society if you look at the website, it talks about my contribution to the roles of climate, of clouds, of climate, and to detection of human influences and global climate, and my contribution to the IPCC. The Simons Memorial Medal mentioned most of those things over my career. I gave a talk, the Simons Medal talk in the bank of England with the governor present, which was a nice way to receive a reward. The Oshko Medal, I think was to do with my contribution to paleoclimate.
[01:32:21] Speaker A: John, you've served on many national international committees on your list over the decades, but which were the most important to you and which were the most satisfying?
[01:32:35] Speaker B: I think if I have, I was to name just one. I talk about the working group and coupled models which I was involved in in various incarnations over the. I think it was about 18 years.
It started, it went through various names, I won't go through those. But in November 1990, it was set up with represents the major modelling groups at the time, notably Ron Stouffer and GFGL and Jerry Mill at NCAR.
The three of us remained in these various incarnations over those 18 years. It's first chaired by Larry Gates, then Leonard Benkson, then I chaired it, or co chaired it with Jerry Meal. And it was a very, very useful forum for modelers to work together, to share problems and their solutions.
And one of the problems that we had in IPCC was agreeing standards for keeping data and comparing models and evaluating their performance. And each model tended to do its own thing. So Larry had set up a standard experiment for atmospheric models, Amit, which had been very successful.
And he felt that at the first IPCC reportin fact, the first two IPCC reports, if there'd been a similar organization for coupled model experiments, it would made his life a lot easier. So that's essentially what he did. He set up the center for collecting, I think Lawrence Livermore PCMDI the project for climate model, into comparison, was the host for storing the data. And they agreed to keep it and distribute it. The reform is set. The fields were agreed that we'd keep, and generally in the same way as Amit, we were able to do standard evaluations of the couple models without having to people do a lot of work.
It was all agreed and standardized. The first experiment we did was just a simple long control stimulation to evaluate the climate of the couple model. And from there on we started to develop experiments, not just to evaluate the model performance, but to look at how various climate response to various factors, including greenhouse gases and aerosols and so forth. The first in the comparison showed a lot of common areas like Southern Ocean being too warm in a model, and hence not enough antarctic sea ice, lack of subtropical stopic humulus, which led to the subtropical oceans being too warm. And that was a fairly persistent error. But I think progress has been made and it's slow. We also set up a lot of idealised experiments.
The first was CMIP two, looking how I modeled at responding to 1% increase in CO2. And we tried to characterize that in terms of climate sensitivity based in slab experiments with the atmospheric model and the ocean mixing the effective ocean mixing coefficient. And that was to enable us to look at some of the projections in the future and attribute some of the differences in those projections due to either differences in the ocean, the atmosphere. Of course, it never turns out to be as simple as that, but it was histired. And since then there have been many more experiments designed to elucidate and understand some of the responses of coupled models to different things.
CMIP has, I think, has made an enormous difference to the development of climate modeling, not just by standardizing experiments and the formats used, but also storing the hostage of model data, sorry, the storage of model data, and make it available to bonafide modellers around the world. And this meant that climate analysis, the analysis of climate data, is not confined to people in models or associated model centers, but is available to a much wider climate and academic community, enabling a much wider and more thorough analysis of the data. And that's particularly true for the broader climate impacts. So I think if that has made it, I mean, Larry needs to be recognized for that confirmation that's made to the science in my own involvement, just the ability to get modelers to do controlled experiments, so we compare what's going on has made life much more exciting, because you can really work out why the model is different, rather than experiment, where different models have done different things and it's very difficult to understand what happens.
[01:37:25] Speaker A: Yes, in fact, we see the initiatives of being absolutely key to the development of our science, and indeed to IPCC. And for those of us that compare models with observations, the seeing data, especially the more recent data, has been really important. So I think that indeed it must be a very satisfying thing to help pioneer. So, coming on to professional societies, what did you or do you still belong to and have you had any formal positions on them?
[01:38:01] Speaker B: I've done the usual things of giving talks and attending meetings. I have been on a couple of committees, one, the committee on communicating climate science. I've been in the climate science Special Interest group and now the history group. So yes, I haven't had any formal positions on the Medsoc committee.
[01:38:19] Speaker A: So what's about the Royal Society?
[01:38:21] Speaker B: The Royal Society again, I've been involved in a number of reports.
One, knowledge networks, nation's global scientific collaboration, the 21st century, where I think my IPCC experience was useful.
And then in a couple of reports of climate change. Climate change, a summary of the science in 2010, and then with the National Academy of Science in the US, Climate Change Evidence and causes in 2004. And I served on the procroys editorial board.
[01:38:55] Speaker A: Thanks very much, John, for telling me about your outstanding career in the Met office. You made wide ranging contributions at the time for the office of Substantial Transformation and challenge.
[01:39:10] Speaker B: Thank you, Chris. I'd just like to say that I feel I've been very fortunate to be in a career where I have one. Had a really interesting job to do, but also the people that I've worked with.
