As my book has now been published, it’s time to provide a little more detail. My goal in writing the book was to explain what climate models do, how they are built, and what they tell us. It’s intended to be very readable for a general audience, in the popular-science genre. The title is “Computing the Climate: How we know what we know about climate change”.
You can order it here.
The first half of the book focusses on the history of climate models.
In Chapter 1, Introduction, I begin with a key moment in climate modeling: the 1979 Charney report, commissioned by President Jimmy Carter, which developed a new framework to evaluate climate models. The key idea was a benchmark experiment that could be run in each climate model, to study what they agree on and how they differ and offer insight into where the uncertainties lie.
Chapter 2, The First Climate Model, tells the story of Svante Arrhenius’s climate model, developed in Stockholm in the 1890s. I explain in some detail how Arrhenius’s model worked, where he obtained the data, and how well his results stand up. Along the way, I explain the greenhouse effect, how it was first discovered, and why adding more greenhouse gases warms the planet.
Chapter 3, The Forecast Factory, tells the story of the first numerical weather forecasting program, which ran on the first electronic programmable computer, ENIAC, in 1949, and traces the history of the ideas on which it was based. I then explore how this led to the development of “global circulation models”, which simulate the dynamics of the atmosphere.
Chapter 4, Taming Chaos, describes how experiments with weather forecast models led to the discovery of chaos theory, with big implications for predictability of weather and climate. Weather is a chaotic process, so inaccuracies in the initial measurements grow exponentially as a weather model runs, making it hard to predict weather beyond a week or two. Climate prediction doesn’t suffer this problem because it focuses instead on how overall weather patterns change.
The second half of the book describes my visits to a number of different climate modeling labs, and focusses on the work of the scientists I met at these labs.
Chapter 5, The Heart of the Machine, explains the key design ideas at the core of a modern climate model, examining the design choices and computational limitations that shape it, with the UK Met Office’s Unified Model as a case study.
Chapter 6, The Well-Equipped Physics Lab, explores the experiments that climate scientists do with their models, and how these have changed over their history. It features the Community Earth System Model, developed at NCAR in Boulder, Colorado as a case study.
Chapter 7, Plug and Play, explores why it is hard to couple together models of different parts of the Earth’s system – oceans, atmosphere, ice sheets, vegetation, etc. I describe how, with the right architecture, a climate model supports a new kind of cross-disciplinary collaboration in the earth sciences through shared models, and I use the Earth System Model developed at the Institut Pierre-Simon Laplace in Paris as a case study.
Chapter 8, Sound Science, explores how modern climate models are tested, and how modelers maintain software quality. I also explore how well they can simulate recent climate change, as well as climates of the distant past, and discuss how we know what the models get right, and where they still have problems. The Earth System Model developed at the Max Planck Institute in Hamburg features as a case study.
The last chapter, Choosing a Future, concludes the book with a summary of what climate models tell us about likely future climate change, where the remaining uncertainties are, and what pathways we might choose to avert catastrophic climate change. I also explore why action on climate change has been so slow, what we can still do, and why there are reasons for hope.