Materials

Shaun Lovejoy: "Scaling, fractals and the search for high level laws in weather, macroweather and the climate"

Macroscopic bodies involve huge numbers of molecules, yet for most purposes, the micro-details are irrelevant. Weather models are based on thermodynamics and continuum mechanics and are successful because they retain only the relevant variables: they don’t even acknowledge the existence of atoms. Similarly, the number of atmospheric degrees of the atmosphere - the weather “details” - is ≈ 10²⁷. Starting with Richardson in the 1920’s, this has motivated the development of higher level turbulence laws that ignore irrelevant aspects of the jumble of vortices. These laws are based on the physical principle of scaling; for realism, they have been generalized to anisotropic (especially stratified) multifractal processes. In the last decades, such processes have been identified not only in the atmosphere where they are ubiquitous, but also in numerous geo and other complex systems.

Beyond about 10 days - the deterministic predictability limit, the macroweather regime - standard weather models become stochastic and to tame this complexity, new high level laws are needed. I describe several (high level) macroweather and climate models based on energy balance and scaling. I argue that they already make optimal monthly and seasonal forecasts as well as improved multidecadal climate projections.

Tim Palmer: "Ensemble prediction"

I shall discuss my personal involvement in the development of ensemble prediction techniques - first the monthly prediction system at the UK Met Office in the 1980s - the first operational ensemble prediction system I believe - and then the ECMWF operational ensemble forecast system which became operational in 1992, and finally the DEMETER multi-model ensemble system for seasonal and longer timescales which became operational in the 2000s as the EUROSIP and now COPERNICUS ensemble forecast system. I shall summarise the rationale for ensemble prediction, discuss some of obstacles to progress, and shall look forward to give my perspective on how numerical weather prediction should evolve by 2030.

Hans von Storch: "Climate science as a social process – history, climatic determinism, CUDOS and post-normality"

Since ages, the topic of climate – in the sense of “usual weather” - has attracted attention in the western tradition as a possible explanatory factor. Climate, and its purported impact on society, has become an integrated element in western thinking and perception.

In this lecture, the history of ideas about the climatic impact on humans and society, and the emergence of the ideology of climatic determinism are sketched. This ideology favored the perception of the westerners being superior to the people in the rest of the world, giving the legitimacy of colonialisms.

In modern time, when natural sciences instituted self-critical processes (repeatability, falsification) and norms (CUDOS @Merton), the traditional host for climate issues, namely geography, lost its grip, and physics took over. This led to a more systematic, critical and rigorous approach of building and testing hypotheses and concepts. This gain in methodical rigor, however, went along with the loss of understanding of climate as culturally being falsely perceived as a key explanatory factor for societal differences and developments. Consequently, the large segments of the field tacitly und unknowingly adopted the false concept of climatic determinism. Climate science found itself in a “post-normal” condition, which lead to a frequent dominance of political utility over methodical rigor.

Cecile Penland: "Time and chance happeneth to them all: Stochastic processes affecting life and nature"

It is impossible to separate the scientist I have become from the socio-political events of the 1970’s and ‘80’s, over none of which I had much if any control. And so, stochasticity became my friend. I started off in astrophysics, fascinated by the statistical mechanics of elliptical galaxies, where two-body collisions are not very important and the stars wander around under their collective gravitational force like hungry blind beings drawn to the smell of a gourmet restaurant. Although I passed my oral qualifying exams unanimously, it was against the recommendation of my advisor and I could not continue with that study. Given the choice of dropping out completely and working in underwater acoustics, I chose the latter. By this time, I had discovered stochastic differential equations and was able to convince my employers to finance my physics dissertation on wave propagation through stochastic media. By being exposed to the physics of the ocean, I learned to love geoscience. It was assumed that I would go to work in one of the military labs after graduation, but I defected to dynamical probability theory applied to climate instead.

Climate studies present a wealth of beautiful problems to work on, and we’ll talk about some of those that have shaped me. The concept of probability as a physical, conserved quantity underlies very useful limit theorems that are tragically underutilized in our field so far. But things are changing, and stochastic methods in climate appear to be entering a golden age. After all, it takes thirty years to be an overnight success.

Brian Hoskins: "Potential Vorticity"

The development the concepts of potential vorticity (PV) and quasi-geostrophic PV, and their use by largely separate academic communities will be traced. In my work on frontogenesis in about 1970, I was able to bring them together in the context of simple situations with uniform PV, and this was later generalised in semi-geostrophic theory. With the availability of good quality global analyses from forecast centres, dynamical ideas were immediately applicable to the understanding of weather developments, and our 1985 paper on Isentropic PV Maps was written to advertise this. In the subsequent 35 years, these maps and those for potential temperature on a PV2 tropopause have become a general resource. I am now using them to gain insight into tropical weather and climate.

Klaus Hasselmann: "Klaus Hasselmann's perspectives on climate sciences: an interview"

A conversation with Prog. Klaus Hasselmann, founding director of the Max Planck Institute for Meteorology in Hamburg, previous director of the Deutsches KlimaRechenZentrum (DKRZ), the German centre for high performance computing in climate research.

Prof. Hasselmann performed seminal research on a number of fundamental topics, concerning stochastic and nonlinear dynamics, attribution, optimal fingerprinting, and many others, setting the ground for a significant part of the current research on climate change.

We discussed about his long and outstanding career, the importance of his background in physics, Hamburg's research environment across the years, how to open a new field of research and jump to another when you are tired of it, without ever losing curiosity, fun, and always having an impact on the research community.

Susan Solomon: "Meeting the scientific and policy challenges of the Antarctic Ozone Hole: A global success story"

The discovery of the Antarctic ozone hole shocked the world in 1985 and contributed to remarkable changes in policy as well as in environmental science and public understanding. In this talk, I will review key aspects of the history of ozone science. I will also summarize the roles of science, public engagement, international policy and technology in the international process that has effectively phased out the production of ozone-depleting chemicals. Finally, I will discuss some of the ways in which science continues to advance the understanding of ozone depletion chemistry as the ozone layer begins to heal.

Kerry Emanuel: "History of the scientific understanding of Hurricanes"

Since their identification as rotating vortices, in the early 19th century, hurricanes have been the subject of scientific scrutiny, which waxes and wanes with the occurrence of the destructive storms themselves. In this talk I will review the rocky history of hurricane research including the development of false ideas, some of which persist to this day. I hope to provide a few object lessons to students on the importance and hazards of received wisdom in beginning a career in science.

Eugenia Kalnay: "It's not just Climate Change"

How can we model Sustainability and improve Predictability in Earth System Models, not just for Climate? We need to bidirectionally couple (with feedbacks) Earth System models with Human System models. In the real world, the Human System has become the main driver of change in the Earth System, and these two systems are bidirectionally coupled. Therefore they should also be bidirectionally coupled in our models. Such two-way dynamical coupled models will be more realistic, and their predictions more accurate. They are also needed to estimate the impact of policies on Sustainability.

David Ruelle: "Chaos Theory: the multidisciplinary origins"

Classical deterministic time evolutions exist with apparent random features, as is seen in hydrodynamic turbulence. Such phenomena have been called deterministic chaos, and are associated with sensitive dependence on initial conditions. We discuss chaos theory with emphasis on the multidisciplinary work concerning chaos in natural phenomena during the three decades 1970-2000. Work in that period has involved developments in pure mathematics, new experimental techniques, and the use of digital computers. The problems addressed include hydrodynamical turbulence, meteorology, chemical kinetics, and the astronomy of the solar system. These problems can be handled with precision. More general applications of deterministic chaos theory remain open.

Pascale Braconnot: "Paleoclimate modeling to test climate feedbacks and variability"

The paleoclimate modeling Intercomparison Project (PMIP) was launched in 1991 with the aim to better understand the mechanisms of climate changes, identify key feedbacks operating in the climate system and test the ability of climate models to reproduce climate different from today. During the same period models have evolved from atmosphere models to Earth System models in which the energy, the water and the carbon cycles interact, coupling the atmosphere, the ocean, the land surface and the ice not only through dynamic and thermodynamic processes, but also through biogeochemical processes.

During this seminar I will discuss how the different phases of PMIP have helped to discuss climate sensitivity, different feedbacks affecting monsoon changes in the tropics and the linkage between interannual variability and changes in the climate mean state. I will highlight the new steps that were possible thanks to new model developments and comparisons with paleoclimate reconstructions. I will also discuss uncertainties and the confidence past climate simulations provide on the results of future climate projections.

Berengere Dubrulle: "On the concept of energy cascades in turbulence: from Richardson/Kolmogorv picture to multifractal and beyond"

Turbulent flows are characterized by a self-similar energy spectrum, signature of fluid movements at all scales. This organization has been described for more than 70 years by the phenomenology of "Kolmogorov/Richardson cascade": the energy injected on a large scale by the work of the force that moves the fluid (e.g. a turbine) is transferred to smaller and smaller scales with a constant dissipation rate, up to the Kolmogorov scale, where it is transformed into heat and dissipated by viscosity. Such cascade phenomenology is at the basis of most turbulent models.

I will discuss in this talk how progresses in numerical simulations and laboratory experiments gradually changed such simple vision (starting from Landau objection in the 50's), leading to a new picture where quasi-singularities living beyond Kolmogorov scale play a central role. This has important impact on resolution requirement of numerical simulations and call for new models of turbulence.

Giovanni Jona-Lasinio: "Understanding non-equilibrium via macroscopic fluctuations"

In non-equilibrium there is an enormous variety of phenomena so we cannot hope to formulate a unique theory having a generality comparable to classical thermodynamics. We have to restrict to subclasses of problems. One difficulty is to define suitable thermodynamic functionals in far from equilibrium situations. Large fluctuations offer a way out as large deviation rates provide, via variational principles, genuine thermodynamic functionals. Singularities of these functionals describe new types of phase transitions.

Clara Deser: "New Perspectives on the Role of Internal Variability in Regional Climate Change and Climate Model Evaluation"

Natural climate variability occurs over a wide range of time and space scales as a result of processes intrinsic to the atmosphere, the ocean, and their coupled interactions. Such internally generated climate fluctuations pose significant challenges for the identification of externally forced climate signals such as those driven by volcanic eruptions or anthropogenic increases in greenhouse gases. This challenge is exacerbated for regional climate responses evaluated from short (< 50 years) data records. The limited duration of the observations also places strong constraints on how well the spatial and temporal characteristics of natural climate variability are known, especially on multi-decadal time scales. The observational constraints, in turn, pose challenges for evaluation of climate models, including their representation of internal variability and assessing the accuracy of their responses to natural and anthropogenic radiative forcings. A promising new approach to climate model assessment is the advent of large (10-100 member) “initial-condition” ensembles of climate change simulations with individual models. Such ensembles allow for accurate determination, and straightforward separation, of externally forced climate signals and internal climate variability on regional scales. The range of climate trajectories in a given model ensemble results from the fact that each simulation represents a particular sequence of internal variability superimposed upon a common forced response. This makes clear that nature’s single realization is only one of many that could have unfolded. This perspective leads to a rethinking of approaches to climate model evaluation that incorporate observational uncertainty due to limited sampling of internal variability. Illustrative examples across a range of well-known climate phenomena including ENSO, volcanic eruptions, and anthropogenic climate change will be discussed.

Valerie Trouet: "A paleoclimate perspective on large-scale climate dynamics"

We use paleoclimate proxies, biological and geological archives that record past climatic conditions, to study natural climate variability and to put current and future climatic changes in a long-term context. Climate history of the past ~1,000 years is of particular interest, because it allows us to look at policy-relevant (decadal to centennial) time-scales and to link climate history to the best-documented period of human history. Tree rings are the most widely used climate proxy to study recent climate history and a century of dendrochronological research has resulted in a broad network of tree-ring chronologies that allows us to study the drivers of continental- to hemispheric-scale climate dynamical patterns over multiple centuries.

Here, I will present two studies in which we used tree-ring data to reconstruct long-term variability in the position of the jet stream over (1) the North Atlantic and (2) the North Pacific. I will discuss what our reconstructions tell us about past variability in these climate patterns, how they are linked to ecosystem and human history, including wildfires, and why this information is important for future climate research.

Raymond Pierrehumbert: "A journey from GFD to exoplanets, with Snowballs and global warming along the way"

The richness of behaviour of climate systems is an emergent property of the interaction of a relatively small number of components governed by well-established fundamental physical laws. For atmospheres the core components are Newtonian mechanics, thermodynamics and radiative transfer. Oceans can be largely regarded as a dense form of atmosphere, and indeed for many planetary atmospheres there is no clear distinction between "atmosphere-like" and "ocean-like" behaviour. As one broadens the field of inquiry to include origins and evolution of atmospheres and their associated climates, additional components are engaged, including atmospheric chemistry, geochemistry, crustal processes and the planetary interior processes that govern what is outgassed into the atmosphere. In this lecture I will give some examples of the ways the fundamental building blocks of climate have given rise to diverse behaviour in a selection of the climate physics problems I have engaged with, building outwards from idealized fluid mechanical problems to my current obsession with exoplanets. In the course of this exploration, I hope to provide a small tribute to the many people from whom I have learned over the years. A key emerging theme is that a certain amount of stochastic forcing is good for one's intellectual development.

Ken Golden: "On Thinning Ice: Modeling sea ice in a warming climate"

Polar sea ice is a key component of Earth’s climate system. As a material it exhibits complex composite structure on length scales ranging from tenths of millimeters to tens of kilometers. A principal challenge in modeling sea ice and its role in climate is how to use information on small scale structure to find the effective or homogenized properties on larger scales relevant to coarse-grained climate models. In other words, how do you predict macroscopic behavior from microscopic laws? Similar questions arise in statistical mechanics, materials science, and many other areas of science and engineering. In this talk I will give an overview of recent results, inspired by theories of composite materials and statistical physics, on modeling effective behavior in the sea ice system over a broad range of scales. We consider fluid and electromagnetic transport through the brine and polycrystalline microstructure, advection diffusion processes, ocean wave propagation through the ice pack, and the evolution of melt ponds on the surface of Arctic sea ice. We will also discuss how sea ice physical processes influence microbial communities living in the ice and upper ocean, and vice versa. This work is helping to advance how sea ice is represented in climate models, and to improve projections of the fate of Earth’s sea ice packs and the ecosystems they support.

James Yorke: "Small dynamic models to understand large models"

I will discuss a variety of models emphasizing small models in dynamical systems, and I will relate them to high dimensional models.

Gabriele Hegerl: "The Detection and Attribution of anthropogenic climate change"

This talk discusses the history of detection of climate change, starting with a string of similar papers setting out a methodology for early detection of climate change in noisy data by Bell, Hasselmann and North. I will discuss its first applications by a small group of scientists which lead to the conclusion of a 'discernible' human impact on climate. These results have since strengthened by quantifying the contribution by greenhouse gas and aerosol influences to the historical temperature record, and moving to its detection in precipitation, water vapour, satellite temperature and even attempts at circulation change detection. I will also discuss the role the record of the last millennium has played in this discussion, and the public perception of this detection of climate change. Recently, attribution has been broadened to discussing individual events, which is related to attribution of longterm change yet with important differences.

Denisse Sciamarella: "Topology of Chaos and Climate Dynamics"

Packard et al. (1980) first attempted to identify the key features of a dynamical system from observational or experimental data. They used time series to reconstruct a finite-dimensional phase-space picture of the sampled system's time evolution within an embedding space, and to characterize it geometrically. Topological properties can replace geometrical ones and have the advantage of providing information about the mechanisms that act in phase space to generate the flow. These mechanisms — stretching, squeezing, tearing, and folding — are topological in nature, and they are intimately related to the governing equations. The duality of dynamics and topology opens several doors. In the deterministic realm, it can provide clear-cut definitions of categories such as “coherence” and “regime.” Recent research shows, for instance, how to rely on topological tools to unravel coherent sets from sparse data in fluid flows. These tools can also help validate, emulate or refute models from data, as well as in comparing data sets. Enlightening surprises arise if one takes one step beyond, and extends the topological perspective to random dynamical systems, which provide the appropriate mathematical framework to tackle ocean–atmosphere coupling and climate change. Noise modifies the behavior of a random attractor: at each instant in time, though, the random attractor’s structure is still well represented by a branched manifold, defined as an integer-dimensional subset of phase space that supports the invariant sample measure. Different “stages” in the “life” of a random attractor can be identified by monitoring the abrupt changes of the branched manifold’s topology. These findings hold promise for the understanding of the climate system’s “tipping points” in unprecedented ways. This talk will present joint work with Gisela Charó, Mickaël Chekroun and Michael Ghil.

Michael Mann: "The rise and fall of Atlantic Multidecadal Oscillation"

For several decades the existence of interdecadal and multidecadal internal climate oscillations has been asserted by numerous studies based on analyses of historical observations, paleoclimatic data and climate model simulations. We use a combination of observational data and state-of-the-art forced and control climate model simulations to demonstrate the absence of consistent evidence for decadal or longer-term internal oscillatory signals that are distinguishable from climatic noise. Only variability in the interannual range associated with the El Niño/Southern Oscillation is found to be distinguishable from the noise background. A distinct (40–50 year timescale) spectral peak that appears in global surface temperature observations appears to reflect the response of the climate system to both anthropogenic and natural forcing rather than any intrinsic internal oscillation. These findings have implications both for the validity of previous studies attributing certain long-term climate trends (e.g. in Atlantic hurricane activity) to internal low-frequency climate cycles and for prospects for decadal climate predictability.