“You can generally trust the data, but not necessarily the interpretation”
Professor Iain Colin Prentice from Imperial College London is the Science Strategy Lead for LEMONTREE and lead PI for the ERC project REALM (Reinventing Earth And Land-surface Models). He is also the Director of the Leverhulme Centre for Wildfires, Environment and Society. He previously held the AXA Chair in Biosphere and Climate Impacts.
With a PhD in Botany from Cambridge University, he developed the standard model for pollen source area; popularized now widely used techniques to analyse species composition along environmental gradients; and led the international development of successive generations of large-scale ecosystem models – from equilibrium biogeography (BIOME) to coupled biogeochemistry and vegetation dynamics (LPJ). His research now focuses on applying eco-evolutionary optimality theory and multiple data sources to develop and test new, robust and reliable land surface models.
His contribution to science has been recognised with a Nobel Prize certificate for his role in the IPCC Third Assessment Report, being listed as one of the World’s Most Influential Scientific Minds, and his election as a Fellow of the Royal Society in 2018.
We met with Colin to ask him 10 questions. So, let’s get to know more about Colin in his own words…
1. How do your three large projects – LEMONTREE, REALM and the Leverhulme Centre for Wildfires, Environment and Society – all link together and complement each other?
The connection between REALM and LEMONTREE is fairly straightforward, because LEMONTREE is a bigger, more ambitious project building directly on REALM. From the development of model components in REALM, LEMONTREE aims to go all the way to developing a new land surface model (LSM) that can be coupled with a climate model.
The fire connection is interesting because although there are some complex, process-based models (I was partly responsible for one of the first of them…) what we’ve discovered, the hard way, is that we frankly don’t know enough about the processes to build a realistic model. We still don’t properly understand either the relationships between fire and the environment, or much about the nature of the pervasive human interventions in fire regimes. Fire is one of the least understood aspects of the natural world. It has fallen between the cracks of all the disciplines involved in Earth System modelling.
Many people still seem to believe that fire is mainly a human invention. It isn’t. Fire has been around as long as there have been plants on land. Today, in many parts of the world, the greatest number of fires are started by people. Yet even if people did not start fires, there would still be fires – especially the large ones. But as wildfires increasingly impinge on people, it’s hard for them to deny that recent increases in frequency and intensity are due to climate change. It’s not rocket science. But there is a growing understanding that the problem is not just a problem of climate; it’s also a problem of how the land is managed, and that is very complex subject.
The funders of LEMONTREE (VESRI, the Virtual Earth System Research Institute, an umbrella organization for several projects funded by Eric and Wendy Schmidt) were keen for us to factor fire into our land surface model development. So that’s quite an exciting development. And of course, as Director of the Leverhulme Centre for Wildfires, Environment and Society, it’s natural for me to collaborate and bring in fire experts into the project. I hope we’ll be able to develop ways of modelling fire that are better than those that have been tried so far, linking fire ecology with the physical and human dimensions of fire and its impacts.
2. What do you hope to achieve with all of these projects?
Ultimately my interest in all of these projects is to get a better understanding of how the natural world works. And of course, by the natural world, I don’t just mean pristine vegetation and nature reserves, rather the natural world is all around us and we rely on it, and it reacts to things that we do. The world has experienced a great deal of climate change in the past, sometimes quite large and rapid, which is something that people (even ecologists!) don’t always appreciate.
But now, of course, we have a situation unfolding where there is rapid, anthropogenic climate change, and we need to know more about it. Models are models; they are means to an end. I hope to develop better models, but in the end, my motivation is to use models as a means to improved understanding.
3. You said that LEMONTREE is building on the work carried out by REALM. But really LEMONTREE started with the NGDev group. Tell me more about this.
Yes, that’s right, although I must admit I don’t remember exactly how the NGDev (Next Generation Development of models) group started! Sandy Harrison and I had long-standing collaborations in China, and Wang Han was one of my PhD students (and later a postdoc) at Macquarie University. Trevor Keenan became a research fellow at Macquarie. Nick Smith was a postdoc with Trevor after his move to Berkeley. Beni Stocker was a postdoc with me at Imperial. We had this little global community and so for several years, we had been having monthly calls – often at rather unsocial hours! – with an informal group we pulled together through our research. We had people in California and China and places in between.
We were discussing ideas for new ways of modelling based on new theory and making good use of many kinds of data. And so when VESRI put out a call we thought, right, OK, let’s put together a proposal and here we are, those initial members are now some of the PIs and the team has grown from there, with of course, many members of my REALM team involved.
4. Who’s been the person that’s most inspired your research?
When I was in my first year of biology at Cambridge, much of what we had to learn was boring. For example, we spent the best part of a term learning about the anatomy of the different phyla of invertebrates, one by one. But one day we had a different lecturer, the late Professor J.A. Ramsay. His first lecture stuck in my mind, as it was utterly different from anything else I had ever heard about. He was talking about fisheries – an ecosystem with direct human involvement. He started by writing differential equations on the board, and solving them. He ended with a conclusion about the optimal mesh size for a sustainable fishery.
When I think about this now, I realize that this lecture contained the key elements that I still strive for in my research: working with analytical mathematical models simple enough to actually solve and understand, yet realistic enough to reach practically useful conclusions. With hindsight, I think it made me aware of what was possible.
5. You have celebrated a lot of success in your career with many accolades. What do you feel has been your greatest accomplishment?
I think I have three main achievements, in rather separate topics. The first significant breakthrough was my research on how pollen assemblages in sediments – a mainstay of palaeoecology, which I was studying then, and still dabble in – relate to the surrounding vegetation. To make any sense of this subject you had to do some basic atmospheric transport modelling. It was simple stuff, but I came up with the rudiments of a theory to explain what we see. My early papers on this are still cited today as being foundational work.
The second breakthrough was the outcome of a collaboration with a statistician, Cajo ter Braak in the Netherlands, who I’m still in contact with. Cajo provided the mathematical wizardry, and I provided the big picture and some applications. Together we revolutionised the way people analysed vegetation composition, making sense of what was previously a very messy situation. The techniques Cajo developed are now standard and described in textbooks.
Somewhat more recently, I was happy to be selected as one of the leaders of a team at IIASA, overseen by Al Solomon, where our mission was to develop a global vegetation model. At that point, there was no such thing. In fact, whether people were coming from biology or physics or meteorology, they were generally agreed that you couldn’t possibly do this. So, it was a pleasure – with Al’s encouragement – to demonstrate that you could. We spent two summers brainstorming and investigating possible routes. Afterwards, I led – with my then-PhD student, Stephen Sitch, who is now a professor at Exeter – on developing what became known as the Lund-Potsdam-Jena (LPJ) model.
Together with its offspring, LPX-Bern and LPJ-GUESS, LPJ is probably still the most widely used offline global vegetation model. I think the reason it works so well is that it makes a few essential assumptions that are basically optimality hypotheses. Back then we didn’t have the data to test them, and we didn’t make much of them in the papers we wrote at the time. Now we realize that these hypotheses can be formalized and tested, and that they work very well indeed! Hence my pivot to eco-evolutionary optimality theory, and using data to demonstrate when it works and when it doesn’t – the foundations of LEMONTREE, in fact.
6. Aside from a lot of hard work, what’s your key to success?
Well, to some extent, being obstinate, and refusing to take no for an answer. I think I realized quite early on what my strengths are. I have the ability to see where I can use my (limited) mathematical skills, in the context of developing new theory. Those three achievements I talked about are all on very different topics, but they all have in common the use of mathematical analysis to develop new theory in areas where theory was lacking. That’s what I am good at.
7. You have held research positions all over the world: Director of the Max Planck Institute for Biogeochemistry in Germany, Chair of Plant Ecology at Lund University in Sweden, Chair in Ecology and Evolution at Macquarie University in Sydney. Where have you most enjoyed living or where would you love to visit again?
I think from the point of view of doing science, you really can’t beat the UK. It’s amazing how good we are at science despite all the challenges it has faced (and still does today, with Brexit-related idiocy providing the current top issue). There is a very positive attitude about what we can do in science. Scientists here have a habit of tackling big questions, whether they are about fundamental physics, or developing vaccines, or anything in between. The previous President of the Royal Society, Sir Paul Nurse, noted that no one knows why we are so good at science!
I do very much like living in Devon, which is my main home today. But Sweden is lovely. It is a much quieter and less populated place than England, and stunningly beautiful. Sandy and I lived out there for about 15 years, mostly in the countryside. All of our three children were born in Sweden. We visited Dalarna in the summer of 2019, and hoped to make a visit to Scandinavia an annual event – until the pandemic and its after-effects made travel challenging. Maybe next year!
8. I know that you are a rather good pianist. Who is your favourite composer and why?
It’s really a period from the late 19th century through to the mid-20th century. This is the music that excites me most, including “very late romantic” composers like Rachmaninoff and Fauré and impressionists such as Debussy, Ravel, and not to forget Lili Boulanger (who didn’t write enough, and deserves to be much better known). Also British composers of the same era, especially Ralph Vaughan Williams, Frederick Delius (if you count him as British) and John Ireland, all of whom are not as widely appreciated as they should be, especially outside the UK. With some trepidation, I’m about to take on Granados…
9. You are a busy man, but if you had the time, what new skill would you like to learn?
I don’t really have a list of new skills I’d like to learn, but I would like to become better at both bell ringing and piano playing. So, with bell ringing now, we practice every Monday at a local church, usually either in our village or a nearby village where the bells have been newly renovated. I’m gradually improving and of course it’s a team, we have some with more experience that help us stay together and keep in time.
I’d like to improve my piano technique as well. I hadn’t had lessons for years and years, but now I have a teacher again: Caroline Diffley, a brilliant teacher in Exeter who actually taught my daughter Lucy before we moved to Australia in 2010 (Lucy stayed). It’s challenging to find time to practice, but I do. Most of the time, anyway.
10. What advice would you give to someone at the start of their career in Earth System modelling or environmental sciences?
Don’t believe everything you read! Generally, you can generally trust the data, but not necessarily its interpretation.
One more thing: I remember a long time ago, when I was actually a young scientist, reading a book by the late Sir Peter Medawar called Advice to a Young Scientist. I suspect his advice is just as relevant now as it was then. But one of the things that stuck in my mind was very simple. He said: ‘if you want to discover something important, work on an important problem’. He didn’t say how you know a problem is important, but I think it’s good advice anyway.