Interview of Prof Chris Folland by Prof John Mitchell

[00:00:00] Speaker A: My name is John Mitchell, and I am interviewing Professor Chris Folland at the headquarters of the Royal Meteorological Society on the 30 November 2023. So welcome, Chris, to this discussion about your life and career. I know that you were born on the 3 June 1945 in Port Cole, South Wales. Can you tell me something about your early life? And can you include something about your parents education and also their occupation at the time you were in secondary school? [00:00:33] Speaker B: Yes. So I lived for the first five years of my life not far from port, call it a little village called Ogmore by sea. And this has probably had the earliest influence, actually, on my eventual career choice, because we lived about 150 foot above the sea, very close to the coast, very exposed to the winds from the south to southwest. And I remember the storms that occurred at that time and also the wonderful sunset, especially in the winter, to the southwest. And from that moment, I think I always had an interest in the, in the daily weather. And I remember particularly, particularly in 1950, when we moved to Swanach in Dorset. When my father moved his profession, a great snowfall occurred in Swanage. Well, in the records, it occurred on the 10 December 1950, and it was certainly over the tops of my five year old boots. And I remember the mountains of snow at the school entrance, which would have been quite several feet high that time. Of course, that's very unusual for a place like swanage. And then a little bit later on, my interest in storms and weather was confirmed by the fact that my parents used to let me hear the news each evening on the BBC at 07:00 and I remember a great storm in 1952 when there was a ship called the Flying Enterprise that was lost. It took several days of attempts by various salvage people to try and save it. I think most of the people were rescued, but that left a deep impression on my mind, and so did the floods in late winter 1953. I remember following that very closely. So I was following weather related stories from early age. Now, my parents, my mother was born in Aba Kenfig in the valleys of south Wales, and she went, of course, to our primary and secondary school. And I think her major certification in education was the national certificate in the early 1930s, where you had to pass at least six subjects at the age of 16. And then she became a secretary eventually at Bridgend Grammar School. Now, my father had a very different background. He was a scholar, and he actually got a scholarship to Cambridge from a school in Birmingham and attained just. He just missed getting a double first in classics and philosophy at Cambridge in the early 1930s. So that was a big influence on my life later on, I was quite keen, actually, on Latin. I suppose that was influenced by my father, even though I didn't actually go in that direction, though I was actually quite good at it. And I think I got quite a good grade at O level as it was in those days. [00:03:51] Speaker A: So you said a little bit about weather events that sparked your interest. Any people who have an interest in science in general and meteorology in particular? [00:04:02] Speaker B: Not to begin with, I think my interest was very much influenced by events. And I remember one particular event was September 1959, which is one of the finest September's after a fairly fine summer in 1959, in the 20th century. And I remember immediately buying a sixes maximum minimum thermometer and started to take readings on the back of my father's shed, though I was always amazed that we had record temperatures, until I realized when he opened the shed door, the hot air came out, that readings were wrong. So soon afterwards, I built myself my own thermometer screen, something like a steamson screen, and took observations for a number of years and they in fact compare quite well with the observations around that area. And at that time we were living in Poole at a village called Broadstone. And I remember that in 19 62 63. My interest is further piqued by the extreme snowfall that we had for such a south coast location. At the same time, though, I did become aware, and I can't remember how this happened, of the work of Hubert Lamb and the reduction in the westerlies. In other words, especially in the winter, when there was a tendency towards an increase in blocking in british winters, which of course fitted in very nicely with the winter of 62 63. [00:05:38] Speaker A: I wonder who, of all these things and people you think had the greatest influence on you while you were growing up. [00:05:44] Speaker B: I think because I've been interested in the weather to start with, right from a very early age. And then Lamb's work got me interested in climate. I was really interested for myself in almost all time scales of meteorology, though I think as time went on, my interest in the longer time scales for potential career overtook those on the short time scales. In other words, weather forecasting. [00:06:16] Speaker A: Did you expect from an early age to go to university? [00:06:19] Speaker B: I think the answer was yes. I went to a village school when we lived in Poole, called Litchipminster school, and it was not a very well funded school. And I remember 25 of us taking the Lebanon plus, and only five passed. And I actually had the highest mark of the five when I went to Poole Grammar school. And then when I was at Poole Grammar school. I was an all rounder at the various subjects that I took and I was generally top of the class, in fact, in the a stream. So I suppose it became a natural thought that I would go to university, but it took a while for me to decide what it was I wanted to do. But by the age of 14, after the September 1959, it's fine September and summer, that I decided to go into the science stream, though I was as much interested perhaps in geography at that stage as I was in anything to do with meteorology. You have to hedge your bets at that stage. You're not really clear what it is you want to do. But it generally fitted and it became clear to me that my interest really was in meteorology. I remember giving a talk, it was the fifth or 6th form on some meteorological subject. And I think by that time I decided that physics was the only way forward because I'd heard that geography was not a very good background for going into meteorology, though there are some exceptions to that rule, of course. So that encouraged me to consider taking a physics degree. [00:07:59] Speaker A: Very interesting. When you first went, what did you expect your education would lead to? [00:08:06] Speaker B: I went to reading university because I'd heard that although it didn't have any meteorology at the time, that Professor shepherd was going to set up a department of meteorology about 1966, about the time that I would finish my degree. And in fact I actually applied for doing initially a masters at reading at that stage, before I took my degree, of course, and I was accepted to do that, so that by that time I was pretty set on meteorology, but events were to actually change that in the short term. [00:08:49] Speaker A: Oh, tell me more. [00:08:51] Speaker B: When I inquired into what was happening in meteorology at the Met office, because that's really where I wanted to go. I'd heard that there weren't very many research places going to be available at that time. And also I'd heard that there was a potential at the Institute of Oceanography, which was then at Wormley in Surrey, to do work in air sea interaction. So I actually applied to Wormley and got accepted and I was going to work under Henry Charnock on air sea interaction. Unfortunately, he left the institute three weeks before I was due to join, to go as a professor to Southampton University. So I didn't actually do what I wanted to do at the institute. I actually went to deep ocean currents, which was quite interesting and without a lot of help, perhaps from people at the institute, did research on that particular topic. [00:09:53] Speaker A: Okay, so you decided you were going to physics at university you decided at that stage you weren't going to the Met office. What interested you? Was it teaching or doing research? Did you want to be an experimentalist, make observations? [00:10:07] Speaker B: So I was doing research, of course, at the Institute of Oceanography, particularly on deep ocean currents. I remember one the main thing I did, actually, was to look at the outflow of the mediterranean water into the Atlantic. That's a salty, dense water that sinks in the Atlantic well to the west of the Straits of Gibraltar. And actually I was quite fortunate in some respects, because new technology was available at the time through temperature salinity, depth recorders, which recorded automatically. So we were able to get many more observations than was possible in the past. And I was able to study some of the mesoscale events that occurred in the ocean and noted, in fact, that what happened at depth at about 1000 meters was actually seen to some extent near the surface, which was, I think, one of the first times it has actually been noticed. But I still wasn't convinced that I should be in oceanography because I wasn't really doing anything to do with the atmosphere. [00:11:13] Speaker A: What was your family's view about your choice of career? [00:11:17] Speaker B: My father was very supportive because he saw meteorology as an international subject that was doing a lot of good to humanity. He wasn't a scientist, of course, and he actually was quite critical of many aspects of science. But despite the fact that physics and maths actually weren't my best subjects at school, they were geography and Latin, probably, and chemistry. I won the 6th form chemistry prize. Nevertheless, he supported the fact that I might have a little bit of a struggle, perhaps compared with some people whose best subjects were physics and maths. And that actually worked out well in the end. [00:12:04] Speaker A: You also had another significant meeting in your life. [00:12:08] Speaker B: Yes, when I was in the Institute of Oceanography, I went on two cruises on discovery. The third, which is a ship in the direct line from Scott's discovery in the Antarctic in the early part of the 20th century. And on the first cruise we went to the Faroe, Shetland's channel, where I did quite a lot of work using Nansen bottles to measure temperature and salinity at depth of the ocean. That was quite interesting because you had to stand on the end of a gangplank and help lower the bottles over the side so they didn't hit the sides of the ship. And I read in the library of discovery that Mister Beeching was going to, probably going to close the West Highland Railway and being quite fond of railways, because back in my youth, on my mother's side, my relations, one of my relations had been an engine driver, another one a signalman at seven tunnel junction. I decided I had to travel on this railway and this fitted because I was due to go to south west Ireland to meet some friends, to go on a holiday in Ireland with a Courson caravan. So with a friend, because I was only on the first half of the cruise, we got off at Stornoway, went on the ferry to Mallaig and got the West Highland Railway from Mallaig to Fort William where we stopped for breakfast for an hour. That's what the railway did in those days. And when I got back there was somebody else sitting in my seat. It was in fact my future wife. And because the train was late and took about 8 hours to go from Fort William to Glasgow, we had plenty of time to chat and the rest is history. [00:13:56] Speaker A: What was her background? And did she have a separate career? [00:14:00] Speaker B: Yes, in fact she had a very interesting career. She joined the civil service back in 1963 and became a secretary in the treasury. And in fact she did pretty well. And she got moved into a senior personal secretary post about a year or two before we met Sir Daniel Thompson, who was the treasury medical advisor. And I particularly remember one aspect of her career which was fascinating. She worked on the devaluation of the pound in 1967 when Mister Wilson said, the pound in your pocket, it hasn't really changed. Of course it was devalued against international currencies and she was thanked for a lot of work she did very late on that rather then Chancellor James Callaghan. So that was her background. And that turned out at a later stage of my career to be extremely useful. [00:14:58] Speaker A: Okay, so you've had a couple of years or so doing oceanography, particularly looking at sea surface temperatures, which presumably came in very useful later in your career. In September 1968 you did then join the Met office and went on the senior scientific officer course. How did you find that? [00:15:19] Speaker B: I actually enjoyed it. It was a very nice mix of theoretical things. I think it was. Mister Heasty was in charge of the more theoretical things. He was the senior instructor actually at the college at Stanmore, and Mister Deeks was the more the experimental type. And we were 13 of us on the course. That's where I met my colleague later on, David Parker, who we worked with very intensively on climate issues. And so it was thoroughly enjoyable. And after the course I went to Gloucester for three weeks at Gloucester airport. And what I remember about that was the weather forecasters there had a very interesting idea about drawing charts they would draw far more fronts on the chart than the central forecast office would. And it was the typical rivalry, if you like, between the out station and the central office. [00:16:21] Speaker A: In April 1969, you were posted to the operational instrument branch in Beaufort park and Bracknell, where you stayed until 1974. Was this your choice, or were you sent there? And can you tell me something about your time there? [00:16:37] Speaker B: I didn't really have a very strong opinion at that stage about what I should do in the Met office, though my marginal choice was actually to work at Malvern under Keith Browning. But that didn't happen. So I was actually posted, although I didn't object in any way to the operational instrument branch, which at that stage wasn't located with the main building in Bracknell. It was down East Hampstead park, several miles to the south. [00:17:10] Speaker A: In this period, you were approached by reading university to do a PhD with the Met office. How did that happen? [00:17:17] Speaker B: Okay, so I was moved to the surface instruments part of the operational instruments branch, and I got very involved with humidity instrumentation, and did some work actually there, which changed the Handbook of met instruments in, in several ways. And reading university got to hear about this after I'd been there for about two or three years, and thought that I'd already done half a PhD or half at that time, and would I like to apply for to do a formal PhD, because I'd already done some of the work. And I thought about this, but then, as I was thinking about it, I was summoned by my director. At the time I was in the services section in the Mister Bannon, and he pointed out that in those days, career development was quite a strong thing, actually, for scientists, despite you might think it's more of a recent idea. And if I did this PhD, it would put back my promotion to a principal scientific officer, because I was by then a senior scientific officer by about two years. So I had to turn it down. In other words, I had to spend less time in the operational instruments branch than it would take to finish the PhD. [00:18:42] Speaker A: Okay, in April 1974, you were posted to the anoptic climatology branch to work on long range forecasting climate change. During this period, Yuko authored research published in a short paper in Nature on climate change, particularly in the southe. Can you tell me how this important work started? [00:19:03] Speaker B: Yes, there was a paper by a gentleman called Winstanley, who claimed that the Sahel drought which was developing at that time was caused by a southward movement of the climatic zones in the atlantic region, and we didn't actually think this was likely to be true. So I was asked by my supervisor at the time, Kevin Miles, to look into this. And that was quite interesting, because when you joined the synoptic climatology branch you had to do your computing in a different language to any other branch in the Met office. Having learned Fortran in the instruments branch to do, to do some of my own calculations, I developed a number of models in the instruments branch about instruments, I had to learn metal codes. I did all the coding for the paper that we wrote in nature, which proved that Winstanley was wrong. Really. His claims were based on observation by lamb that the Iceland low was moving further southwards over the last hundred years or so. But actually over the period of the Sahel drought, it was actually moving northwards again. And we also looked at the latitude of the westerly wind maximum in the Atlantic and also the subtropical high, and they were clearly moving in the wrong direction. So whatever the Sahel drought was caused by, it wasn't caused by the things that Winstanley suggested. This was sufficiently definitive a result that nature published it in 1974. [00:20:51] Speaker A: Very good. In May 1976 you were on the move again to the Agricultural and hydrology Meteorology Branch. During this period you were promoted to principal scientific officer, now known as uninformatively as grade seven. Your work there also contributed to the ward later in 1985, of the Royal Meteorological Society. Hubert Robert Mill Medal in prize in 1985. Can you tell me about your work there and why it was so important? [00:21:23] Speaker B: Yes. The most important job I was given to do was to take over from a predecessor to Met Office contribution to a major activity which was to develop an automatic system for designing storm sewers in Britain. It was called the working Group on the hydraulic design of storm sewers. And it was partly funded by Dear E. And it was collaboratively in the Met Office, the Institute of Hydrology and the Hydraulics Institute, both at Wallingford. And in the process of doing this I had to do two things that were particularly important. One was to understand and explain the flood studies report, which had been completed in 1975, which described how extreme rainfall varied over the country and how it was calculated. Because urban drainage design usually requires models to be produced which produce floods of. Of a fairly high return period, but not as high as in some areas of design. So we're talking about five to about 20 years. The other thing I had to do was to explain this to many senior and famous engineers in the country. And this was actually the biggest challenge. And it seemed to work. I have to admit I was helped a great deal during this because moving into hydrometerology was a bit of a shock, and I made great friends with a colleague who was head of another group, Brian Wales Smith, who was one of the most helpful people I had in the Met office. And we became great friends. And the working group led to the program of predicting storm flow through sewers, urban sewers, which today is still recognizable form. So it was a big effort, but well worthwhile. So the other thing that I did, which was probably the bigger contribution, was to study the problem of the loss of rainfall from rain gauges, which had been well known for a long time. That was suggested, actually, by my assistant director. And I used my physics knowledge and some experiments that have been done by hydrologists on the wind flow of rain gauges to develop a model of the loss of rainfall from a cylindrical rain gauge and also from a conical shaped rain gauge. And this led to a model which agreed very well with observations, particularly observations which had been taken in Denmark, where they showed the loss of rainfall function of wind speed. And I was able to reproduce that the losses turn out very simply to be a function of the upward motion on the windward side of the rain gauge, where the small drops are lost, and the acceleration of flow of the rain gauge, which gives divergence of the raindrops, which means some of them are lost behind the rain gauge. And that is the basis of the model. I designed a new rain gauge as a result of that, which is called the Follon Severock Strangeways rain gauge, which was manufactured, and a similar rain gauge is still manufactured today, though by a different company under a different name. [00:24:50] Speaker A: I know you had some experience within the Met office. How did you find the style of management? [00:24:56] Speaker B: I think I was lucky, because in branch very collegiate directors, and my. My boss was equally collegiate. The same thing happened, actually, in the next branch, which, of course, was agriculture and hydrometerology. And I remember at the time, and I can't remember how this happened, that there was a dispute between branches. I think it might have been metal eight and meto three. And I was asked to intervene and bring the two sides together, and I managed to do that, which probably helped with my reputation amongst assistant directors. And even after that, I never really had any problems, despite the fact that at that time, this is before the middle 1980s, there was one problem related to management, and that was discouragement of writing papers. And I remember that very clearly, and that there was also some discouragement from collaborating with people abroad, and that was to have some consequences later when I challenged this successfully. [00:26:14] Speaker A: Good. Okay, how much freedom did you have to develop your own research interests at this stage? [00:26:21] Speaker B: I think a lot. I mean, obviously there were problems that I had to solve, especially in the early parts of my career. But as my career developed, especially later on, when I went to the synoptic climatology branch and started to work in climate change and monthly weather forecast, I had almost complete freedom. [00:26:47] Speaker A: And who outside the Met office at this time most influenced the direction of your career? [00:26:54] Speaker B: If anybody, it was still Hubert Lamb. And I used to correspond with Hubert Lamb. And in fact, when Hubert died in the 1990s, because of my fairly close relationship with him, I was asked to go to his memorial service and gave a eulogy. So probably Hubert, despite the fact he was very agnostic at the best about anthropogenic climate change, was probably the biggest single influence. [00:27:26] Speaker A: Okay, how was your work funded then and how did this change later in your career? [00:27:32] Speaker B: Much of the funding was mod funding to the Met office until I joined the Hadley Centre. Then it became the Hadley Centre contract with the Department of the Environment. But after I partially retired in 2008, still attached to the Hadley centre, some of my funding came from the Swedish Research Council, some of it came from the science and services project that became under the new fund that was linked to China. [00:28:06] Speaker A: In November 1980, you were posted back to the synoptic Climatology Branch working on long range forecasting, UK monthly and tropical seasonal forecasts, recent climate change and global dataset development. By now you had a growing international reputation and we're building a strong research group. Looking back, what you feel were the key developments and achievements in that period from, say about 1980 to 1988? [00:28:37] Speaker B: The answer to that is a little complicated, because the main purpose of the synoptic climatology branch, as defined by the directorate, was the development of monthly weather forecasting, which of course seems to be developed temporarily from about the time that the Hadley Centre was set up. However, there was a lot of effort put into that, and when I took over the responsibility for what? Quasi operational forecasts. The operational forecast had been stopped in 1980 and the branch had been reduced in numbers. So I took over more than PSO post. I looked at what had been done and most of the techniques that were used, which of course was statistical, I rejected. And I was left basically with only one major technique, which turned out to be very useful later on in different contexts called linear discriminant analysis. And the reason why it was so good, it was. It used Bayes theorem. In other words, say you were trying to forecast for the United Kingdom what the atmospheric circulation would look like what we did was to divide the characteristics of the circulation into a set of clusters. You could work out from past data what these clusters look like, perhaps with equal probability. And given the prior probability that was equal, what was the past year of probability? In other words, the forecast probability, given various predictors, which turned out to be eigenvectors of surface pressure thickness, and also some areas of the ocean where sea surface temperature from previous work was thought to be particularly influential. And this system actually, although it had very limited skill, did have some skill. And in order to prove that we had skill in long range forecasting, what we did was to divide the forecast into two halves. The first half was for the first 15 days and the second half was for the second 15 days. Now, the first half was dominated to some extent by short range forecasts and medium range forecasts from east scum WF, which we used for most of our forecasts from 19, 1981 onwards. The statistical methods, the linear discriminant analysis, and there was another one which is less important, contributed to both the first and the second halves. I don't think the dynamical system did. And when we assessed the forecast and the paper that was published in the Metmag in 1986, you could see that we actually had significant skill in rainfall in the second half month. Actually, we had more still in rainfall than we did in temperature. And this was rather fortunate because I was sitting at my desk one day having my lunch, which I did in those days at my desk, and I had a phone call from the central telephone exchange in the Met Office, saying I was about to get a phone call from the government, from the Cabinet Office about what I thought was going to happen, about the ongoing drought which had developed in 1984 and it had become quite serious. Drought alters had been issued in July and did I think the drought was going to continue. So this phone call took place, and I'm still to this day not sure whether it was a minister or, or a senior civil servant, but luckily I got it right. Now, luckily in August 1984, there have been some early indications that the drought might not be quite so severe. And in the second half of the month our forecast clearly showed it would become considerably wetter. And because we knew we had some skill in those days, in the summer in rainfall, because we looked at the seasonal variation, skill in a cautious way, suggested that the drought might be alleviated fairly soon. And this turned out to be true. And in fact it was a bit like the 1976 drought. It ended the end of August and the following autumn was wet very good. [00:33:11] Speaker A: Okay. The other thing, I guess, that you did cloud a lot of work on was on sea surface temperature data. [00:33:18] Speaker B: Yes. Sea surface temperature development had arisen from a project which had been set up by the department of the Environment. Actually, before I left the synoptic climatology branch, before, in 1976, I actually started the planning for this project. And when I came back at the end of 1980, I actually had the fruits of what had been done in the previous four years. And very good work had been done in many ways on analyzing sea surface temperature, particularly the way the quality control was done. But unfortunately, it wasn't completely automated, and this wasn't suitable, of course, for product which could be widely used. So a lot of effort went into that. That. But what we also noticed was that the sea surface temperatures behave rather peculiarly. Now, the way we noticed this was that I was worried. How could we know whether the sea surface temperatures were right, especially for studying climate change? How did we know the low frequency variability was trustworthy? So we decided to develop a different dataset called the nightmare in air temperature data set. So we were the originators of this, because these are now used very widely and have been developed in other places as well. And the nightmare in air temperature data set clearly showed that the sea surface temperatures behave differently, especially towards the end of the 19th century and up to the second world War. And we discovered, through some work on observational practices, why this might have been. We also had to make small corrections, though, to the nightmare marine air temperature, because those ships get bigger, the screens get higher above the water, and so the temperatures go down. So this has to be corrected for. So when we did this, we managed to get the sea surface temperatures to look much more like the night marine air temperatures. And this global dataset was the first of its type that had been produced. And we published a paper in Nature in 1984, which also was the first paper to detect interhemispheric variations in the sea surface temperatures. And there was a parallel paper produced that very much supported our paper by a senior climatologist in the United States. And this was the beginning of a very long period of development of sea surface temperature data sets. And the use of them by IPCC, in a very major way, was that. [00:36:07] Speaker A: With MIT having worked out that the. [00:36:12] Speaker B: Sea surface temperatures were biased, we needed to find out more about the details of this and whether a uniform correction for the world, which we made at 0.3 degrees centigrade, based on the nightmare temperatures before second world War, was correct. And using my physics knowledge, I'd started to develop a theory of why this happened. And the reason why we were able to make a start on this is we noticed that if you took extra tropical sea surface temperatures and produced anomalies of them, there was a spectral peak, which is far too big at the annual cycle. There shouldn't really be a significant peak with anomalies. And that gave me the, shall we say, the clue as to what was happening. And it was quite clear there was something to do with the way the sea surface temperatures have been measured, particularly by buckets, which were covered in canvas, which worked a bit like wet bulbs. They lost heat when you took them out of the ocean. And we also discovered that there were other types of measurements made by other buckets, which were wooden, which we discovered lost a lot less heat, but still lost some. So, in the early stages of developing this model, MIT got very interested in it. And I was invited there for three weeks to work with one of their scientists on developing a more extensive computer model of how this might work. And I went to MIT, I think it was in 1986, and we developed an initial computer model of how this worked and compared it with nightmare temperature data and discovered that it really was on the right lines. And after that, the model was further developed. And it was the basis of a major atlas that we produced in 1990 that was originally going to be funded by the Americans. MIT had tried to get funding for America, but it didn't actually happen, but did here. We got very interested in this because of IPCC in 1990, and diverted 30,000 pounds of IPCC money to the publication of this atlas of temperatures over the last hundred years or more, and night marine air temperatures to compare with. And I think there are about 300 maps in this publication and a fairly full description of the way that sea surface temperatures were analyzed, what the errors were, and how all these corrections for sea temperature buckets had actually been done. [00:39:04] Speaker A: There's also, I think, in this period that you had started a crucial collaboration with UEA. [00:39:11] Speaker B: We decided that we really needed to collaborate with UEA to produce a global temperature data set. And the UAA, of course, had developed land surface temperature data sets. And, in fact, it started looking at global data sets. Now, we were the first to publish a global sea surface temperature data set. And so I approached my assistant director at the time that we should collaborate formally with CRU. There was some hesitation, because there was still the feeling that we shouldn't be collaborating too much with people outside. But Andrew Gilchrist took a very different view and became very enthusiastic about such a collaboration and in fact a formal agreement was drawn up between the climatic research unit, which is the University of East Anglia, and the Met office. And this has been extremely successful ever since. And this led to global datasets, which are now known as the had cruet datasets. And they have played a leading role in all the IPCC reports from the first report to the 6th report, though of course, there are many more global data sets now, so they have more weight than ours did. But in the early days our dataset had very considerable weight and in fact they were the main evidence for observed climate change at the surface in the early reports. [00:40:47] Speaker A: Also during this period, you did some more work on the effect of relations between tropical sea surface temperatures and Sahel drought. [00:40:55] Speaker B: This came about in a rather peculiar way. Having published the Nature paper in 1984, we started looking at the sea surface temperatures in every season of the year, right back to the 1850s, when the data started. And it was David Parker and I were looking through these probably a thousand printouts, and we noticed, because somehow, drought I was very well aware of, as you can see, that there appeared to be a relationship between the Sahel drought and variations of sea surface temperature between the hemispheres, particularly in the north Atlantic, but also elsewhere. And we analyzed these and found that there was a very strong correlation between annual Sahel rainfall, which is mainly in July to September, and sea surface temperatures on this global scale. And it worked that the northern hemisphere, North Atlantic and to some extent the North Pacific, were cold, and the southern hemisphere, particularly in the South Atlantic, and also the whole of the Indian Ocean, including the North Indian Ocean, was warm when the drought occurred, and vice versa in wet periods earlier in the century. Having produced this relationship and done some analysis of it, Tim Palmer, who was in the branch at the time, got very interested and ran the eleven level model. I think it was in perpetual July mode for 180 days and got a similar relationship when it was forced with this pattern of sea surface temperature. And so this encouraged us to think, well, we've really made a breakthrough here, because the prevailing paradigm, which is due to Jules Charney, I think, was that changes in albedo were the main cause. Other people thought it might be more to do with vegetation changes because of the drought. What we showed at that time was, we didn't say that those things were not at all important, but sea surface temperature seemed to be the dominant influence. And in fact that was a new paradigm that became accepted. It probably took about a decade for it to be fully accepted. There were still some people who thought land surface processes were dominant. What we did show, though, that land surface processes were nothing unimportant, especially soil moisture feedback. They in fact did increase the variability of the rainfall in the models that we looked at. And of course we looked at this in the then eleven level model, which had fixed cloud actually as a matter of interest. And I think that was fortunate because later models didn't do quite so well, at least for a decade or so. And we managed to reproduce rainfalls in wet seasons and dry seasons really rather well. [00:44:08] Speaker A: And this original work was published in Nature, I understand. [00:44:11] Speaker B: So the original work was published in Nature, then there were several other publications after that. [00:44:16] Speaker A: In 1985 you had another offer of a change in job and location. Do you want to say very briefly about that? [00:44:23] Speaker B: Yes. John Horton, who had of course been heavily involved with Oxford, Oxford University in his previous career, was keen to increase the amount of work done on climate at Oxford University. And the head of geography, who is a reader, was just retiring. And it was suggested to me rather strongly that I might apply to take over this role, but in a very different way, to do really the work that I'd started to do on climate variability and the influence of sea surface temperature, and that I would be able to have a link directly to the Met Office modeling capabilities. Now, this would take an awful lot of organization, obviously, and the problems of staffing, where you get the staff from, some obviously have to come from the Met office, some possibly elsewhere. It happened, unfortunately, when we discovered the problem, but we hadn't completely done the research and we hadn't written the paper in nature. And I was very worried that possibly two years of effort, which it could have been, to set up a new department, which is basically what it would be, would destroy what I thought probably would be one of the most important things I would ever do. And so I did reluctantly turn it down. [00:45:47] Speaker A: Chris, in May 1988, you started a year at the Climate Analysis Centre in Washington, DC. Can you tell me a little bit about that? I think you co authored another nature paper there and your links to IPCC started. [00:46:01] Speaker B: So when I went to the climate Analysis centre, the first thing I did was to write a confidential report to the director, at his request, on the future directions of research and applications in seasonal forecasting, because there have been some controversy as to whether statistical methods should be used more or less on their own for the foreseeable future, or whether modelling should be included. And I, of course, suggested that atmospheric modelling certainly should be developed as quickly as possible for seasonal forecasting, keeping the statistical methods and in the long term coupled modeling. When I was there, I did a lot of conference meetings on the sea surface temperature. I was invited to many different places to show what we had developed. But the most important thing that happened in early February 1989, we had a phone call from the Met Office saying that I would be chosen to be convening lead authorization the observed climate variability and change chapter of the Intergovernmental Panel on Climate Change, working group one. And that was a complete change of direction, basically. And I was asked to set up a workshop, which we did. We developed a workshop plan which came to fruition in the summer of 1989. The difficulty in those days was communicating large amounts of information to potential participants. And my wife was actually, as an experienced secretary, was taken on board by the director to actually carry out this activity. And that worked out very well. And I think the University of Amherst was the only time I actually worked directly with James Hanson. And that was one of the first IPC workshops. And that probably was the most important thing that I did at the climate analysis centre, except there had been a controversy over satellite data measuring sea surface temperature. A paper had appeared in nature saying there had been a 0.6 degrees centigrade rise over the 1980s and sea surface temperature. This was unfortunately caused by the eruption of Al Qi Chong volcano in the early part of the record, which great the aerosols and that which greatly cooled the apparent sea surface temperature from the satellite records. And we were able to show that this was wrong. We even had an idea about the reasons, but we showed it completely disagreed with in situ measurements. And that paper was published in Nature at about the time I left the climate Analysis center in the summer of 1980. [00:48:50] Speaker A: In 1989, you returned to the Met office and were promoted to personal merit. Grade six and head of climate variability research. This coincided with the development of the Hadley Centre and the associated contract with DoE and the increasingly strong development of the intergovernmental Panel on Climate Change. Also at this time, the Met Office wanted to enhance the observational component of the proposed Hadley Centre. Can you say something about why this is the case? [00:49:21] Speaker B: Because misses Thatcher was very keen on climate change. She thought we had a repairing lease on the planet. I think that was her phrase when she gave her famous lecture to the Royal Society in 1988. Money had been found for the modelling side of climate change, but not very much for the observational side. There was a little bit. So my then director of research, Keith Browning, was very keen to rebalance the contract, especially to add the development of satellite datasets, climate change. And so I was asked to produce a proposal, which I did, which went to Dewey and the Department of Transport, but unfortunately it wasn't funded. And until much later, after I left the data area of the Hadley Centre, we did not develop ourselves any satellite datasets. We used other people's and discovered their problems. [00:50:20] Speaker A: So the years around 1989 to 1990, there was a lot happening in climate change. And before we go into detail about your work during this period, can you tell me a little bit about your approach to research? How did you learn about new developments in your field? [00:50:36] Speaker B: There wasn't any one way of doing it. I suppose some of it was through conferences. When I started the project, I would do something of a literature research, but actually one of the ways of finding out what's been happening recently in your subject is to read a key paper in that area and look at the references. This accelerates the process no end. In fact, it's become easier and easier with time. Computerization really didn't exist very much in 1990, so I had to visit the Met office library quite a lot. Nowadays everything is pretty well matched online and incredibly easy to do. So being in the Met Office gives you a head start in this source of research into what's been done. [00:51:26] Speaker A: How did you decide which journals to read? [00:51:29] Speaker B: Obviously looked at journals which looked at climate, but not all the papers appear in such journals. So as I say, most of it was probably done in an iterative way by finding a key paper, unless you've been to a conference, obviously, which did help, and then finding other papers. [00:51:52] Speaker A: And what were the controversies of the day in your field and if so, how did you deal with them? [00:51:58] Speaker B: I remember talking to had an interview about the causes of climate change and how we knew about climate change. And I pointed out that at that stage in the early 1990s, theory, in other words, the theory of the greenhouse effect, was really the dominant factor that set up IPCC. And the observations at that stage weren't all that supportive of anthropogenic climate change. There had been some climatic warming in the first part of the century and it was starting again. This caused a lot of reaction from climate scale. Who thought that? I didn't think observations were important at all. And in fact I've had feedback on this problem for the last 30 years. Only recently did I have to answer that question yet again. But we wrote up what we really thought at that stage in a letter to the Institute of Physics journal. That was done with Bruce Callender, who was then leading organizer of IPDEM in the Met office, and what we said was exactly that. At that stage in 1993, which is after my interview, theory still dominated. But we look forward to the time that observations have become crucial, especially when climate change detection attribution was developed, which of course at that stage was being thought about quite seriously. And it wasn't many years after that that of course that actually happened and the observations took on a new importance. [00:53:39] Speaker A: Okay, so you now have become head of climate variability research and you initially put highest priority on the development of sea surface temperature datasets, but also encourage others to develop other global datasets and new datasets like the daily central linguist. Temperature seems to be not only a major contribution to the climate science itself, but also underpins much of your other work. Indeed, the wider observation of climate change and validation of climate change. Do you want to tell me something about that work? [00:54:10] Speaker B: I just mentioned central England temperature. First, this was developed between the late 1980s and early 1990s. The idea was to develop a daily data set of central England temperature back from the 1770s, which we did and has been kept up to date to this day. And because the Met Office, the Hadley Centre, had some problems with climate models at that early time, I remember this was seen as one of the more important developments in the first two or three years in the Hadley centre. But in the longer term, what's more important was sea surface temperature and various data sets had been developed. We'd already developed the Met Office historical sea surface temperature data set, which was published in the atlas, which I mentioned. And then because of the need to force models with sea surface temperature, existing dataset wasn't adequate because it wasn't globally complete. So what you need for a climate model is a globally complete sea surface temperature data set that includes sea ice. And this is where the global sea ice and sea surface temperature data set, or gist, came in. An early form of this didn't have any major reconstruction anomalies over data spot areas. They were estimated in particular ways, but this wasn't very satisfactory. But that data set was used. That was gist one with some success. And then we developed more complex systems using eigenvectors, which made better estimates of the sea surface temperature, because the covariance matrix was calculated over a very long distance and the gist gradually improved. And for the more recent period from 1982, we used satellite sea surface temperature data from America, the AVHR data. However, that proved to be a problem because that data isn't very homogeneous and it led to the Americans, I think it was. Hurrell and Trenworth discovered that compared with some of their data sets in the southern Ocean. Our gist data sets had very poor persistence from one month to the next. And we discovered this was largely due to the fact that we had very few sea surface temperature data sets, sea surface temperature data measured in situ. There we relied on the satellites, and we hadn't used the satellite data in the best possible way. It needed correcting. Various other things had to be done. And we developed what's called headist, which used a better analysis scheme called eigenvector optimum interpolation, or reduced based optimum interpolation, which also solves these problems with the satellite data. And the success of that data set for forcing models is shown by the fact that we believe the paper that published that, and a better nightmare in air temperature data set at the same time, which is not globally complete, has been referenced at least 8000 times in web of science, over 10,000 in Google Scholar. And I think it's probably the most highly referenced paper in Met Office history, as far as I know. [00:57:32] Speaker A: And I believe as part of testing the SST data sets. You went to New Zealand twice. [00:57:38] Speaker B: Yes. So one of the things we were worried about was the Southern Ocean, which has the least number of observations and as you can see, to a potentially quite large problems. So I applied, with some encouragement from a New Zealand colleague of mine, to become a fellow of what was then the Waitangi foundation of New Zealand, became the link foundation of New Zealand in 1993. And I was appointed the second meteorological fellow that they actually appointed. And they gave me a grant to go to New Zealand to work with the National Institute of Water and Atmospheric Research, which was the climate organization in New Zealand at the time, on New Zealand temperatures, with the main developer of that dataset, Doctor Jim Salinger. And this proved to be highly successful. And we found very close relationships between sea surface temperature and New Zealand temperatures. And I actually chose New Zealand for one very special reason. It had the best historical air temperature data sets, because at the end of the 1860s, a series of stations have been set up by Sir James Hector, who was the leading administrator at the time, using Stevenson screens. So the data are probably some of the most homogeneous air temperature data in the world world stretching back 150 years. And the data so good, we're able to show that the bucket, the corrections we'd made to sea surface temperature, let alone the general agreement, were very accurate around New Zealand, not only in the annual May, but in all seasons. [00:59:25] Speaker A: Very good. Also in this time, you were looking at rainfall forecasts over northeast Brazil and west Africa. And these seem to be particularly successful. Indeed. I believe you also won an international prize with your colleagues for this. [00:59:40] Speaker B: Having discovered the relationship between Sahara rainfall and global sea surface temperature, we looked at this in some more detail to see whether we could create a forecasting method. And we discovered that patterns of sea surface temperature in the South Atlantic, and to some extent have nothing north Atlantic, were particularly important in addition to the global pattern, especially on inter annual time scales. So we developed a statistical forecasting method, which was quite successful, though we noticed that changes of sea surface temperature in the Atlantic in the period before Sahel rainfall were significant. And you got a better result if you made the forecast just before the rainfall season began, compared with, say, two or three months in advance. Northeast Brazil was better and we found it was dominated by changes in the intertropical convergence zone over the Atlantic, which were dominated by interhemispheric variations in sea surface temperature, and also to a lesser extent, though in major El Ninos, equally by the El Nino la nuna cycle, with El Ninos having a particularly large effect. And so you could summarise it by saying that the correlation between Sahel rainfall and sea surface temperature on the seasonal timescale. If you knew the sea surface temperature was in the region of 0.6 to 0.7 in northeast Brazil, it was over at least 0.75 and you could explain 55% of the variance of sea surface temperature. And this we published together with our model experiments in a paper in 1993 and we were awarded the Norbert Guerda Mum prize. It's an international WMO prize in 1996. [01:01:39] Speaker A: How and when did you change your prior to change to success? Extent to understanding predicting extra tropical seasonal forecasting, especially northern Europe and then the UK. When did these forecasts become operational? [01:01:52] Speaker B: Okay. Discussions in the Hadley Centre suggested with David Carlson at the time that we should start to think about UK seasonal forecasting. This, of course had been done back in the 1960s and 1970s by purely empirical techniques, but the idea now was to try and gain some understanding of whether seasonal forecasting was possible. Now, there's a widespread view, which actually continued for quite a long time, that seasonal forecasting for Europe was basically impossible. The internal variability of the atmosphere was too great. But we did some work using a climate model forced with observed sea surface temperatures, having discovered that there seems to be a link between sea surface temperatures and the Atlantic and the North Atlantic oscillation, and particularly what's called the sea surface temperature tripole and the North Atlantic oscillation. This is a variation in temperature between warmth in the subpolar gyre, coolness to the southeast, warmth in the tropical north Atlantic, or the reverse pattern. And we did some experiments using this pattern, which could be shown to exist using a form of analysis of the ocean and the atmosphere together that showed they were linked. And we took this pattern and applied it to the model. This pattern been produced in the paper elsewhere. We took a strong version of the pattern and we indeed found a very convincing reproduction of something that looked quite like the North Atlantic oscillation in the model. And later experiments over much longer periods of time confirmed these results. And as a result of that, we didn't actually use the climate model as such in forecasting mode. But in a later paper, we looked at this empirically and made a surprising discovery. This pattern actually was best linked to the North Atlantic oscillation in May or in the early part of the winter. And obviously from the model results, we got the early part and perhaps mid winter, because we used simultaneous ssts. So we'd found a forecasting method, and this is Mark Rodwell and myself. It's called the Rodwell follons forecasting method, and this started to be used in operational seasonal forecasting in the early two thousands. I can't remember exactly when we began, but I do remember a couple of us published papers in 2006 to look at the success of the 2005 six winter forecast. And it turned out the fallen painting method had actually been the dominant cause of skill. And we were lucky because this pattern was particularly strong. It was in the mode that favoured a blocked winter, and that winter was the first cold winter for several winters. And although it didn't get quite the magnitude right, we got the sign clearly right. [01:05:09] Speaker A: Is that partly because of the inertia with the winter temperatures, which are over a very deep mixed layer, whereas in the summer you've a much shallower layer? [01:05:20] Speaker B: Exactly. So the seasonal thermocline is a big problem. So the seasonal thermocline develops during June and July and unfortunately persists, at least residually, right up to November. So if you look at the relationship between this pattern and the North Atlantic oscillation, it's strong, probably a bit stronger in December, but that's not the forecasting method than it is in May. But there's not a lot of difference. [01:05:47] Speaker A: The observational data sets which you put together have been the basis of many of your studies of climate variability and oscillations on a wide number of timescales, from seasonal to multi decadal. Perhaps you could outline one or two of your more significant developments in this area. Over the recent years, one of the. [01:06:07] Speaker B: Things that came out of my visits to New Zealand was controversy had developed over whether the pattern of ocean variability we'd found in the Pacific, called the interdecadal Pacific Oscillation, which bears a strange strong relationship to the Pacific decadal oscillation, which is confined to the North Pacific, which had been found by the Americans, really had any real effects on anything other than Australia. Now, found that this pattern was discovered, in fact, in Australia, had a strong effect in Australia, apparently, on the relationship between the southern oscillation and rainfall over Australia. But was that just a lone thing, or was it something more to the interdicadal Pacific oscillation? So we did some work to look at the joint effects of the southern oscillation and the interdicatal Pacific Oscillation on the largest climatic feature in the South Pacific, the South Pacific Convergence zone. And using observed data to define where this zone actually was, we found indeed that there was a strong influence of the intergal Pacific oscillation on the South Pacific convergence zone as well. Of course, there's a strong influence of the southern oscillation, and the combination is needed to be known to actually work out what the behavior actually is. And this backed up, if you like, the fact that there really had been an apparent relationship with australian climate as well. [01:07:50] Speaker A: You also set up the climate of the 20th century project. [01:07:53] Speaker B: Yes. [01:07:54] Speaker A: Can you enlarge on that? [01:07:56] Speaker B: Yes. So this was very much an informal project to begin with that leading scientists were quite keen on for model evaluation. In particular, the amipheen experiments have been set up, but they were limited only to ten years at the time. [01:08:15] Speaker A: So these were the atmospheric, these were. [01:08:17] Speaker B: The atmospheric model intercomparison projects. And what we were trying to do was to look on much longer time scales. So we set up a collaboration with lots of leading modellers, including people like Kiku Miyakoda. And we actually had a major workshop in 19 98 94 which developed some diagnostics which were going to be used in the second assessment reports of the IPCC for the model evaluation chapter. That was the immediate aim. And in fact, they were to give you a couple of examples. Models, some of which have been used in AMIP or were being used in AMIP, were used to reproduce northeast Brazil rainfall and the Southern Oscillation index using a form of gist that we had at the time, the globally complete sea surface temperature data. And, of course, the results look quite good, especially northeast Brazil was quite amazing, really, which just confirmed, of course, because these were independent confirmations of what we'd already done, and you will find those in the IPCC report. After that, it went into a forbade for a while, and then at Schuylkiller, got very interested in resuscitating the project, and we developed a design which was published in Cliver exchanges in 2002 and became a Clive project in 2003, to look much more widely at climate variability, mainly particularly decadal climate variability. And to give you just one example of the thing I think we put most emphasis on at the beginning was what caused the droughts in southwest America in the 1930s, etcetera. Was it due to sea surface temperature? And one particular american author, Schubert, ran his model, I think, nine times, and was able to show that a pattern that looked quite like the into decadal Pacific oscillation was in fact not dominant, but very important to the rainfall variations, though there are other causes, including internal variability. And there's just one example of many papers that were published as a result of that project. However, it became clear that there's probably a limit to the value of doing this now. The coupled models were becoming very much so. In 2013, myself and my colleague, who was the co chair of the climate of this project, Jim Kinter, decided it should become a climate change detection and attribution project. And the idea is that you run models a, with complete sea surface temperatures and natural forcings, and bhdem with anthropogenic forcings on top with the sea surface temperatures adjusted to get rid of the anthropogenic component, which is done using coupled models. They'd be very careful how you do that. But that is the philosophy. And actually, although I've been long retired, I'm actually involved in one of these actual projects at the moment. [01:11:35] Speaker A: So one of the things you initiated, if that's the right word, was the. The real time global temperature a year ahead, and also work looking at the components which contributed to that forecast. [01:11:46] Speaker B: Yes, that was started in 1999, and it became quite clear that global temperatures were being influenced by some factors which had been identified in the literature quite clearly, like ENSO, which has a big effect on global temperature from one year to the next. We decided to make a forecast from one year to the next, which includes several factors, but ENSO is the biggest solar includes ENSO, the Intergal Pacific oscillation is there assumed. So these forecasts started for 2000 and in fact are still going, and they had quite good skill. But initially we discovered that they appeared to be bias warm. And the reason was the verifying data which we were using, which were Hadley Centre data didn't include the Arctic, and the Arctic, of course, is warming faster than the rest of the globe. So this leads to underestimate and reanalyses of this period show that this warming era disappears. In fact, the statistical models we developed have got better. We did a revision in the early part of the think the last correlation I looked at up to 2018 was 0.92 between the global temperature a year ahead and the observations. But in 2008, we decided to introduce climate models and we use a decadal forecasting model depressis. This, of course, has developed over time and it's got better and better. So there will probably come a time when it actually takes over completely. Its correlation with reality is comparable to the observations, though there is some evidence that recently it's got a bit better because the skill is so high. People do take notice of, of these forecasts, and they have been used in the past to coincide with conferences of the party meetings. And probably we will be issuing one again soon to correspond to the conference of the party. 28. [01:14:05] Speaker A: Thank you. And in view of your wide knowledge, particularly in climate observations, you were heavily involved in the first few ports of the intergovernmental Panel on Climate Change. Indeed, you were coordinating lead author three times and the lead order once. What do you see as your main contribution to these reports? [01:14:23] Speaker B: I think the main scientific contribution has been on the temperature side, obviously. Also, although I only had limited involvement with this, the behavior of the troposphere and stratosphere, mainly through David Parker, that's had strong prominence in the IPCC reports. The other thing that I was very much involved with at the end of the 1990s, to the early two thousands, is extremes. And we started work looking at extremes in 1998. I actually held a workshop, a WMO workshop, on this. We, we wrote a plan and the group took a leading role in this and published a paper in, I think it was 2002, which actually got used in the, in the taR, which showed the first examples of things like the reduction in the frequency of frosty nights. And then this was greatly extended for the fourth assessment report, led by Lisa Alexander. And she was able to show things like warm nights for warming, particularly fast, warm days less so, and various other changes of extremes. Now, this was a difficult thing to do, because national meteorological services don't like letting out their daily data, which is necessary for this sort of thing. And I'd given a lecture to WMO, their request actually, on climate data in 1999, and pointed out the great need for a greater exchange of daily data and the way that we solved this problem, to a large extent, was to hold workshops where people brought their daily data. Scientists had developed the software to create the indices of these extremes, and they could then take their daily data back with them. And this worked very well. So a whole series of regional workshops were held. I was involved in one of them, and others of my staff were involved in many others. So by the time we got to the fourth assessment report, we had quite a good range of results from much of the northern hemisphere and a reasonable part of the southern hemisphere as well. And this would have been impossible without the workshops. [01:16:52] Speaker A: Did you have any involvement with the controversies over the surface temperature record? [01:16:58] Speaker B: Yes. In fact, there was a big controversy over the difference between surface warming and warming of the troposphere, which caused skeptics. A lot of headroom models suggested that the troposphere was warming faster than the surface, especially in the tropics. And the microwave sounding unit and our radiosons were not actually showing this. And as I say, this was a global thing, but mostly, mostly in the tropics. And we had a major meeting which was held in the United States. I actually had to be security cleared to go to this meeting. And we had a week long workshop, I think, near Chicago airport. And all these things were thoroughly reviewed. But I don't think the problem was fully solved. Though we were lucky that between the time that this workshop was set up and the time we had it, there were some improvements to the surface data sets, which actually gave a little bit more warming. And my main contribution as a lead author was to the last chapter of how the observational data sets should be improved. And I came up with seven recommendations. And it's interesting that one of them, which might not be obvious, was that in the long term, to get a really holistic view of what was happening, we needed to do it through reanalyses. And so one of the seven recommendations was the development in different places of reanalysis. And I think we can see now, although we haven't quite got there yet, era five is a very different beast from the reanalyses that we started with in the 1990s. [01:18:52] Speaker A: On your partial retirement in 2008, climate variability and forecasting split into two, and each part enhanced. So one was climate change detection and attribution, plus climate data and aspects of observed availability. And the other part was the monthly to decadal prediction and climate variability, though this includes aspects of climate change. June 2008 to March 2017, you continued as a meth office research fellow in monthly to decadal forecasting and climate variability. Also continuing your work on climate variability and change, which has extended into full retirement. What significant activities in this period have, if any, we not already mentioned. [01:19:38] Speaker B: The first thing that I completed in this period was thorough analysis of something that we basically discovered in the form that we presented it in. Anyhow, the summer North's Atlantic oscillation. This is the summer version of the winter North Atlantic oscillation, which of course is very well known. And we were able to show how this behaved in the past back to 1706. And this is done through collaboration with the University of Gothenburg, who had experts on tree ring analysis. And we were able to show that the southern orthostatic oscillation strongly influenced Scandinavia as well as the UK. The tree rings came from particularly Scandinavia and the UK. And we were able to show that this reconstruction was really very good, because with a few exceptions, its correlation with central England temperature was very similar to what we see in the observations. Also, we looked in models. I think one of them has had Gem, the other was had CM three, how the summer north atlantic oscillation might change in the future. And basically we discovered that it increases its anticyclonic drought mode in the summer, which agreed with other analyses which have been done, which haven't actually identified the summer oscillation as a specific mode. [01:21:10] Speaker A: You mentioned University of Gothenburg, from which you have a professorship. You also have professorships at two other universities. [01:21:17] Speaker B: Yes. One is the climatic research unit, and this is really to reflect the excellent results of the collaboration that we've had with CRU, particularly through Phil Jones and of course his colleagues, ever since 1986. And so I got that in 2003. And then the University of Southern Queensland was very interested in developing links with the Met Office, which recently they funded posts in the Met office. On the modelling side, they're very interested in applied climatology and decadal forecasting. And in fact, I was asked fairly recently to go to Australia, having won an award from them, to present some lectures on these topics on climate change, which have been filmed and used in undergraduate courses. [01:22:10] Speaker A: Two of those professorships are abroad. Have you ever been tempted to emigrate to another country? [01:22:16] Speaker B: This is a very interesting question. And in fact, the opportunity arose while I was at the climate Analysis centre, and people were very interested in the sea surface temperature work and the modelling work that went with it on climate variability. And it was Maurice Blackman in the United States that asked me whether I'd be interested in moving to the United States and forming a new group at the David Skaggs Institute in Boulder, Colorado, with some existing scientists involved in climate variability and climate change, but also with a direct link, the modelling system at the Geophysical Fluid Dynamics Laboratory at Princeton. And I thought very long and hard about this, and it wasn't until the Hadley Centre was formally opened by misses Thatcher and the talk she gave the importance of climate change to the government and to society. And I thought, gosh, if somebody on that wing of the political spectrum should be so enthusiastic, then the Hadley Centre, which had just been opened, must have a very good future. And ten days later I came to the somewhat reluctant conclusion that I was better off in the Hadley centre than the uncertainties of the United States. And history has probably proved me to be correct. [01:23:49] Speaker A: Well, I'm very glad you stayed, because speaking as a model in the Hadley centre, I've always felt the strength. One of the strengths of the Hadley centre has been having models and observations together. [01:24:00] Speaker B: That in fact is a big factor, because David Skaggs institute the models would have been distal linked, as you say, in the Met office. As we showed, as long as you can find the computer time, you can always run a model. [01:24:15] Speaker A: So over the years you've had many honours and awards. These have included, among others, fellowships of the Institute of Physics in 1996, the American Meteorological Society in 2003 and the American Geophysical Union 2011, and several Met Office awards and a recent significant award in retirement. You also had a significant award from the south korean Met service. Do you want to say a bit about that? [01:24:43] Speaker B: The South Korean Met Service wanted some lectures from me, having given me the money and award. It was called the Grey Expert Award and they wanted lectures on climate variability, seasonal and decadal forecasting, basically everything that I had been involved with in sea service, temperature. And I gave a course of nine lectures in the south korean met service in Seoul. I also attended the workshop, which was partly organized by the Met service, and I was also as an extra, because one of the lectures was of particular interest to the University of Seoul, which is of course the premier university in South Korea. It took about a month and it was a very enjoyable visit. And as part of this my wife came with me and we went up to the din military zone between South Korea and North Korea and saw for ourselves what it was like there and heard the propaganda from North Korea. [01:25:45] Speaker A: I know you've been involved with the Royal Meteorological Society. What have been your main contributions? And have you held posts in other learning societies or professional bodies? [01:25:56] Speaker B: My main contribution to the Royal Metsoc, I was on council for three years, 1979 to 82. I was on the education committee for six years and remember I organized a big exhibition in London on meteorological instruments. I think that was about 19. 78, 72. More recently, I've been a member of the history group of the Royal Meteorological Society and I'm their designated writer of pen portraits. Past presidents of the Royal Meteorological Society. So, with others, including John Mitchell here, I wrote written papers on Sir John Mason and Raymond Hyde. And the most recent thing I've been involved with is the 150th anniversary. The quarterly journal and a major website was set out by the society. And we contributed to the society and many, many people that helped me in the Met office and outside to a list of papers. And so they've been added to those suggested by past editors of the court. Hawkeye Journal. And the other thing I've been involved with so far, I think I'm on insane. I organised the first hybrid meeting of the Royal Society, met Society on the History of climate change and research and its applications. That was done in 2022, after two years delay because of COVID You've also. [01:27:26] Speaker A: Been involved in various world meteorological organization groups. From which one did you or ones did you derive most satisfaction? [01:27:34] Speaker B: I think the most satisfaction was definitely the working group on climate change detection, or climate change detection and indices, as it came to be known. Of course, indices include everything from sea surface temperature, global mean sea surface temperature. But the main emphasis was on extremes. And it was through my membership of that working group and other related groups, and links with the developers of the cooperative atmosphere ocean data set in America, that I was able to accelerate considerably the development of the analysis of climate extremes, which is a complete. Which was a completely new area. And it's now very important, of course. So I think that gives you the most satisfaction, because I think I had problems, had the most impact of one of these WMO groups through that mechanism. I was deputy chair, of course. [01:28:31] Speaker A: Have you had involvement with government or example, for example, advisory groups additional to IPCC, or have you had involved in briefing ministers or MP's? [01:28:42] Speaker B: Yes. So after 1990, when the Hadley Centre was set up, besides briefly misses Thatcher, when she opened the Hadley Centre, which, by the way, was the most unusual talk that I've ever given, because misses Thatcher had been briefed about my talk beforehand and actually spoke back to me what my talk was going to be. And I remember there was a program on television at the time called yes, Prime Minister. And I remember saying, yes, prime minister full time. She was absolutely right. And after that stage, various ministers and MP's came along and I was briefing them on observed climate change, especially over the UK, but globally, as well, and I also remember briefing the Lords Committee on science and technology, I think it was in 2004, that was headed by Lord Winston, I think, at the time. [01:29:39] Speaker A: To what extent have you been involved in giving public talks or writing books or blogs? Popularising science? [01:29:45] Speaker B: Okay. I've not written any books, but mostly in retirement and probably just before, I have given a series of talks near where I used to live in Devon, on Devon. Climate change in the global context. That wasn't actually the title, but that was the whole variety of organizations, from the Women's Institute to the u three A and I think I've given five talks and each has been developed with time. And I've been very grateful to my old group, to the national climate information centre, which I used to be in charge of at one stage for providing me with information about climate data in Devon, which, of course, I use quite a lot of. Put that in a global context. [01:30:36] Speaker A: You've led a very full research career. What main interests do you have outside that? [01:30:42] Speaker B: So my biggest interest is in Philateli. So I'm a member of quite a few philatelic societies, including the Royal Philatelic Society in London, which is the premier philatelic society in the world, I think. But I'm also very interested in the National Trust. And in the last couple of years I've been a room guide, first of all in Davening Castle Drogo, and currently nearby where I live, in stone underworld at a place called Snow's Hill Manor. And this involves knowing about the history and content of all the rooms. You're not a room guarding one room, you're a room guide in them all. [01:31:20] Speaker A: Finally, have I left anything out that you would like to mention? [01:31:25] Speaker B: The only thing I would mention in retirement is I've also been involved with mentoring students at Reading University. These are students whose courses involve climate change, but not just climate change. And one of them was a postgraduate. The rest have been undergraduates. And one actually has led to a project which I'm in currently involved with, which involves the climbs of the 20th century project. It's actually called the climbs of the 20th century plus project now, simply because it includes the 21st century. And also it's a good time to rename it because it now is detection attribution and we're looking at detection attribution of regional climate over the United Kingdom using this project. [01:32:15] Speaker A: And you've also been involved in the supervision of PhD students? [01:32:20] Speaker B: Yes, in the Met office, I think I was involved in the supervision of four PhD students, some in the Met office, and one at Oxford University, outside the Met office. [01:32:35] Speaker A: So, Chris, can you say a little bit more about what you've been doing since you took partial retirement? [01:32:41] Speaker B: Yes. Another thing that we were very keen to do was to look in the around about 20, 1011 at potential predictability in the north european region, mainly using observations. We did quite a detailed observation analysis that showed that the SST Tripole, which I mentioned, the Roswell fallen pattern, enso and global warming and occasion of the Voltanic index, actually gave a result that showed that the potential correlation, if you knew those values correctly, between the observed and forecast North Atlantic oscillation was about 0.4. And in fact, most of those things can be forecast quite well before the winter. I mean, Enso is easily forecast before winter. We know the SST track, global warming is well known and you usually know about volcanoes. What's interesting is that no sooner had that paper been published, when suddenly the models were revolutionized. And I must say this is not my work, but this is very important to the story. The reason why the SSTs in the North Atlantic had not been picked up correctly, the climatology of the models was wrong, because it was far too cold in the subtropical gyre. And it was when the model resolution, the ocean resolution, was increased to a quarter of a degree, by a quarter of a degree, that suddenly the Gulf stream operated properly and this problem disappeared. And the subtropical gyre is a crucial part of the SST tripole. And suddenly, using hind casts, using the model that had this improvement, went right up from what? Not 0.4, because many people thought it would be zero to 0.6. And that 0.6 remains to this day. And that includes the real time forecasts that have been issued. That was really very exciting. And it showed that several of the factors that I mentioned were actually important when we co authored the paper that actually produced this result. And we show that some of the factors that in the model, which the model reproduces very well, were related to some of the factors that I naturally mentioned. Another thing I was asked to do, there had been a major drought between 2010 and 2012, which actually came to a very sudden end in the summer of 2012. And I was asked by the then director of research, Julius Lingo, to investigate whether we could find out any of the climatological reasons for droughts in the lowlands of England. This is where many people live and where droughts are particularly serious. So I worked with the Centre for Ecology and Hydrology to look not only at meteorological droughts, but river flow and groundwater. And we looked at a whole series of factors, and we could see that there were many factors that had a small effect, which need to be taken into account. Any factor that's big has to affect the surface pressure over the region, because that's well correlated with rainfall. And it turns out that the La Nina has the biggest effect. However, it's not guaranteed there are no silver bullets here. But El Nino certainly increases the probability of droughts in the winter. And so we actually looked at the winter half year, not the summer, in this project, between October and March, because that's when the recharge occurs from the rainfall into the groundwater and the rivers. We found some other factors had marginal effects, like El Nino and the Cube, the crust around the oscillation and the atmosphere. But unfortunately, southeast England is in an area that's not very sensitive to the North Atlantic oscillation, where all the predictability really is. So it's only under exceptional circumstances that factors that affect the North Atlantic oscillation and other atmospheric patterns as well, actually have a strong effect on southeast England. If we were looking at northwest Scotland, that would be different. Finally, I was interested, as a result of the work I've done on forecasting global temperature, is whether I could add anything to the explanations of why global temperature changed, right down to the seasonal timescale. And this particularly motivated by the pause in global temperature that occurred of the so called pause between about, about 2001 and about 2012. And I wrote a paper in science advances which showed, for instance, this is a minor result, but relevant in the winter. Optic oscillation, actually, if it's strong, has a significant effect on annual temperature in the globe. It's about five hundredths of a degree, but that's enough to change your forecast a little bit. But the main cause of the pause, according to my model, which agreed very well in one respect with what had already been published, was the interdecadal Pacific Oscillation in its cold phase. But also an almost equally strong factor, was an unusually strong solar cycle, with an unusually strong decline in solar output over that pause period. We also highlighted the fact that we couldn't reproduce the sea surface temperatures very well in the second World War, and these are known to be too warm. And also we couldn't reproduce very well the coal temperatures in the early part of the 20th century, which are known to be a minimum. And we also compared the results with CMIP, five models, I think faulty models, and they have the same problem. Later work has shown that the sea surface temperatures in the Second World War are indeed suspect, and people have made suggestions about how they could be corrected. I think the early 20th century problem has not yet been solved. So that was interesting because it confirmed, if you like, what we had already in our global surface temperature forecasting model, the rod row fallen method, which had been used very successfully. Seasonal forecasting had been replaced by now because, as we have said, winter forecasting became much more skilful. This was because Adam Scaife had increased the resolution of the ocean part of the coupled model that we used and that implicitly replaced and extended everything that the Rodwell follen method could say about the North Atlantic sea surface temperature patterns. [01:39:46] Speaker A: Thank you, Chris, for telling us about your long and distinguished career in the Met Office in which you have made significant contributions in a wide range of topics including climate observations, the physical understanding of climate variability and prediction, and in hydrology and hydrometerology. And this has been recognized in the last few weeks by the award of an honorary fellowship of the Royal Meteorological Society.