Introduction to the Symposium

Terry Root, Linda Mearns, Jean-Pascal van Ypersele, Ben Santer

Morning Session Moderator: Kevin Trenberth — National Center for Atmospheric Research

Session 1: Aerosol effects on climate

Phil Rasch — Pacific Northwest National Laboratory

Steve Schneider started writing about the role of aerosols in climate during his post-doc at NASA GISS in 1971, with his first work considering the cooling produced by `dust' (scattering aerosols) on the planet. He subsequently worked on absorbing aerosols (nuclear winter), and in 1994 was considering the complexities of aerosol cloud interactions, aerosol feedbacks (the CLAW hypothesis), uncertainties in aerosol emissions, and the difficulty of teasing out the role of aerosols in the presence of natural variability, climate change and changing greenhouse gas forcing (the fingerprinting of aerosol forcing), Steve was also interested in and thinking about aerosols and society, contributing to discussions within the scientific community and in public and policy venues ideas about geoengineering and risk management. All of these scientific and policy questions remain relevant and pressing today. I will cover some of the progress over the last decade resulting in improvements in our understanding of aerosols and climate, issues associated with geoengineering by aerosols, and identify some remaining gaps in our understanding of aerosols and climate.

Alan Robock — Rutgers University

In the early 1980s, Steve Schneider was a pioneer in nuclear winter research, working with Curt Covey and Starley Thompson to conduct general circulation model simulations of the climate response to massive smoke injections. He described his results as "nuclear fall," in contrast to the description of "nuclear winter" of the results of Richard Turco, Brian Toon, Tom Ackerman, Jim Pollack, and Carl Sagan (TTAPS). Even though a nuclear fall would be a disaster, the public interpreted this result to mean that we need not worry about the climatic consequences of nuclear war, and the danger has further receded from the public mind with the end of the Cold War and continuing reduction of global nuclear arsenals. Steve felt justified in publicizing his results, as they came from a three-dimensional climate model that allowed the investigation of processes not considered in the original one-dimensional model results of TTAPS. But their model had no ocean, stratosphere, or mesosphere, and was limited by the speed of the Cray-1. Recently, using modern computers and climate models, colleagues and I have re-done the nuclear winter simulations, and found fundamentally new results. The smoke would be lofted into the upper stratosphere and climate effects would last for more than a decade, much longer than previously thought. For the same 1980s scenarios, there would indeed be nuclear winter and not nuclear fall. Furthermore, there would be massive ozone depletion with enhanced ultraviolet radiation reaching the surface. A nuclear war between Russia and the United States, using the reduced arsenals of 4000 total nuclear weapons that will result by 2017 in response to the New START treaty, could still produce nuclear winter. A nuclear war between India and Pakistan, with each country using 50 Hiroshima-sized atom bombs as airbursts on urban areas, could produce climate change unprecedented in recorded human history and global-scale ozone depletion. This scenario, using less than 0.03% of the explosive power of the current global nuclear arsenal, would produce so much smoke from the resulting fires, that it would plunge the Earth to temperatures colder than those of the Little Ice Age of the 16th to 19th centuries, shortening the growing season around the world and threatening the global food supply. Nuclear proliferation continues, with nine nuclear states now, and more working to develop or acquire nuclear weapons. The continued environmental threat of the use of even a small number of nuclear weapons must be considered in nuclear policy deliberations in Russia, the U.S., and the rest of the world.

Session 2: Cloud effects on climate

Jeff Kiehl — National Center for Atmospheric Research

Clouds are ubiquitous to Earth's atmosphere and play an important role in determining the amount of radiative energy available to Earth's climate system. Clouds reflect shortwave radiation, thus contributing to a cooling of the Earth's surface. Clouds also limit the escape of longwave radiation to space, which is a fundamental process by which Earth regulates surface temperature. Equally important is how clouds may change as Earth's climate warms due to increased greenhouse gases, in which cloud feedback processes may either amplify or dampen the greenhouse forced warming. In this presentation, I look back at how our understanding of the role of clouds in Earth's climate has evolved over the past century. A number of interesting early studies identified clouds as a critical component to Earth's climate system and provided a heuristic framework to look at cloud climate interactions. I then consider the important contributions that Steve Schneider made to our understanding of the effects of clouds on climate. His methodology became the canonical framework for studying cloud climate problems. I discuss how Steve's contributions motivated and directed further cloud climate research up to the present. I conclude with some personal reflections on the challenges that cloud climate research presents to the research community.

I intend to compare the global cooling effect of Clouds as derived from NASA's Earth Radiation Budget Experiment (in 1989) with what Steve Schneider calculated in 1972. It was within 10% of what I and my NASA colleagues obtained with a $250 million experiment. Steve's model cost a lot less money, I am sure. Then with respect to his work on aerosols... I plan to use that study (along with his nuclear winter soot) to go into Brown clouds in Asia, their effect on the Himalayan Glaciers and a Vatican report I co-chaired. I will conclude with Project Surya (cook stove-black carbon experiment in India see http://www.projectsurya.org/) to show how the two of us, while starting at the same place (radiation forcing; climate feedbacks; etc) took two different paths from science to climate mitigation... his was global scale influence of policy; mine was locally focused to translate knowledge into action.

Session 3: The role of the ocean in climate change

Warren Washington — National Center for Atmospheric Research

Some of the history of ocean model development and its role in climate change research will be presented. The early development of ocean models that would eventually be coupled to atmospheric and sea ice models can be greatly credited to the pioneering efforts of K. Bryan and his colleagues at GFDL. The approach taken towards developing coupled climate models by the various modeling groups was mostly incremental, partly due to the limitations of computer time as well as the lack of understanding of ocean processes. There was also a feeling that we should not make models overly complex but refined enough to capture the first order climate effects. The first coupled general circulation model approach was to simulate the ocean surface heating and evaporation in an annually averaged manner so that the ocean feedback would be a moisture source for the atmosphere. The community then moved to seasonal upper ocean heat storage all the while carrying out simple climate change simulations with increased concentrations of carbon dioxide and other forcings. Fully coupled climate models with fully three dimensional atmosphere, ocean, and sea ice components did not occur until the 1970s and early 1980s with very coarse resolution. In addition to presenting the early history of ocean modeling, a few early climate change findings will be briefly discussed and compared with present day models.

The rate of climate change in response to radiative forcing from increasing greenhouse gases is strongly dependent on the efficiency of the processes transferring heat to the deep ocean. Much of the literature on this topic has been based on the vertical advection-diffusion paradigm, in which turbulent mixing processes transfers heat downward into the deep ocean, while high-latitude formation of cold deep waters and broad-scale upwelling serve to cool the deep ocean. It is now apparent that on global average this physical paradigm is incorrect. The global average heat balance of the deep ocean is dominated by processes in the Southern Ocean, where both vertical diffusion and the mean overturning circulation act to warm the deep ocean, while mesoscale eddies, i.e., the weather systems of the ocean, act to cool the deep waters. In this presentation we will illustrate this revised paradigm, and examine how it affects the uptake of heat during transient climate change. Results from a suite of standard resolution simulations, typical of IPCC AR5 class models, in which the mesoscale eddies are parameterized, will be contrasted with very high resolution experiments in which the mesoscale eddies are explicitly resolved.

It is extremely challenging and difficult to talk about a person who stands out as a titan in a society where genuine leaders are becoming a rarity. Stephen Schneider was a person about whom one could write volumes and yet not do justice. In some sense looking at Steve's life one is reminded of what Gandhi said: "my life is my message". It would not be inaccurate to state that Steve's life was also his message, and a message that has to be celebrated. All of us miss him greatly, but he has left us with so much that in some sense his life's work will not be finished for a very long time, and he has left it to all of us to continue with it. His life's work, his spirit, his inspiration – all of these are with us and will only grow as the relevance of his life and work assume even greater importance. It is for this reason I chose the title of my talk to focus on how Steve's life and work will inspire the IPCC in the future. I see the IPCC's future resting on four sets of attributes:

1. Excellence in science and every field of knowledge related to climate change.
2. Understanding the policy relevance of IPCC's work.
3. Creating communications skills and capacity.
4. Dedication to the IPCC and the values and practices it stands for.

These are requirements that Steve set benchmarks in fulfilling to a degree that far exceeds the record set by anyone else that I know of. Steve Schneider was a person who displayed throughout his life and career a consistency of commitment, courage of character and integrity of intellect. These are qualities that everyone working in the IPCC should strive to uphold, for the future of this organization would be secure as long as we treat these qualities of Steve as our beacon-light.

Session 4: Understanding uncertainties in estimates of future climate change

Afternoon Session Moderator: Jerry Meehl — National Center for Atmospheric Research

Linda Mearns — National Center for Atmospheric Research

There are three main uncertainties regarding the climate system in the future: 1) uncertainties regarding the future emissions (and concentrations) of greenhouse gases and aerosols in the atmosphere 2) the way the climate system will respond to the future forcing resulting from the changes in concentrations, on various spatial and temporal scales, and 3) the uncertainty based on the internal variability of the climate system. However, there are other uncertainties, which are harder to quantify, that include important climate processes that are not well modeled in climate models, and/or are poorly understood, as well as 'unknown processes' whose importance is not yet known. From the point of view of climate modelers and climate simulations, the uncertainties with which they are concerned are primarily the second and third as well the incomplete representation of some processes. An important debate regarding uncertainty is whether and by how much the various uncertainties can be reduced, assuming that reduction of uncertainty about physical systems is an important goal of physical science. Certainly, increased knowledge about future climate continues to accrue, and one can view this as a type of reduction of uncertainty. But that is quite different from reduction of uncertainty regarding, for example, by how much annual temperature and precipitation will change in the upper Colorado River Basin by 2050 compared to the current period. It is this latter type of uncertainty that is of particular interest to the many stakeholders who will be making decisions on a regional or finer spatial scale about how to manage future climate change. Furthermore there are important questions regarding by when different types of uncertainty can be reduced. I will discuss some of these issues, (e.g., the role of reducing uncertainty, quantification of poorly understood uncertainties) based on the most recent climate science and make suggestions on where we should go from here.

There is great interest in finding metrics of climate model performance that help determine model trustworthiness for projections of future climate change. Here a method for identifying and employing such informative metrics of performance is presented. The method is based on physical understanding of the climate system. When successful, the method may lead to reductions in the spread of future climate change projections, climate model improvement, and more strategic observation of the current climate system. Under certain circumstances, the method may also be used to reduce uncertainty surrounding future climate change.

Session 5: Impacts of climate change on ecosystems

Terry Root — Stanford

Plants and animals around the globe are showing consistent patterns of detecting regional warming. The biological changes observed include: 1) shifts in ranges, either poleward or upward in elevation, 2) changes in the timing of events (phenology), such as when trees bloom or migrants arrive, 3) change of gene frequencies, 4) morphological changes such as longer wing length or larger egg sizes, 5) behavioral changes such as relocation of nests, and 6) extirpation or extinction. Changes in the phenologies of wild species can be used to attribute changes in regional temperatures to humans. This is accomplished by comparing associations between phenological shifts and actual temperature trends at particular study locations with associations between species shifts and modeled temperature trends. Had CM3 GCM was used to model the regional temperatures with natural, anthropogenic and combined forcings at the locations of the numerous sites where species changes were found. Even at a one-grid-cell scale, the associations between phenological shifts with three different modeled temperatures show that humans are likely to be contributing to the regional warming species detect at various study sites throughout the northern hemisphere.

All across the planet, from the shallow waters of coral reefs to the high mountain forests, species are already responding to climate change. Species' responses to changes and how these responses and interactions scale to ecological communities and ecosystems will play an important role in the impacts of climate change on biodiversity and ecosystem feedbacks to climate change. I will present two example systems, western North America's mountain forests and California's flora, that explore the diversity and dynamics of these responses to changes in climate. Forests store approximately 45% of the carbon found in terrestrial ecosystems, but they are sensitive to drought and climate-induced dieback. Widespread and rapid forest die-off constitutes a major uncertainty in climate change impacts on terrestrial ecosystems and carbon cycle feedbacks. Current understanding of the physiological mechanisms mediating climate- induced forest mortality limits the ability to model or project these threshold events under climate scenarios. I will discuss the individual to regional scale direct tests of two broad physiological hypotheses underlying a recent and rapid trembling aspen (Populus tremuloides) forest mortality. We have combined observational tests of hydraulic conductance and carbohydrate reserves with experimental drought manipulations on potted and mature trees to examine the roles of carbon starvation and hydraulic failure due to water stress in this forest die-off. Second, I will present results from an ongoing project that examines the diversity and drivers of changes in California's flora over the past century.

Session 6: Integrated assessment modeling

John Weyant — Stanford

Integrated assessment models (IAMs) combine concepts and information from many scientific disciplines into systems of mathematical equations designed to facilitate the development of scientific conclusions and policy relevant insights that could not be obtained using methods from a single discipline. Over the last thirty plus years there have been a number of major efforts to construct IAMs to study climate change and climate change policies. Although these models are frequently used for policy analysis they generally have deep scientific roots often stretching the state of the art in each scientific discipline to its limits, and at the same time integrating interdisciplinary knowledge in ways that has never been done before. Stephen H. Schneider was a master assessor of the state of the art of these climate change oriented IAMs. This talk tries to provide a modest update of the work Schneider started with his classic 1997 paper which put forth a set of signposts for measuring the state of IA modeling with respect to what might ultimately be achievable and/or desirable. In true Schneider fashion we start with the basic building blocks of the underlying disciplines and their integration, and a set of design criteria to strive for. We then throw in a large dose of painful, but unavoidable, pragmatism in the form of simplifying assumptions and approximations that are required to implement the theory given our current state of information and about how our world actually works. Life is put into these abstract notions by through examples of results from IAMs over the years to show the types of problems that can be addressed. We end with some thoughts about what the future might bring in terms of better theory and practice in integrated assessment, again using the sign posts from the Schneider IAM global positioning system.

This paper offers a broad overview of how carbon emissions and policy have developed in Sweden over the past couple of decades. It then addresses how integrated assessment models may, or may, not help in developing appropriate abatement policies. Results from our GET model, the Global Energy Transition model, are presented concerning both fuel use in the global transport sector and the role of bioenergy with carbon capture and storage (BECCS) under stringent climate constraints. Interesting key questions, that should be posed by and to all researchers presenting results from integrated assessment models, include: (i) How robust are the results with respect to changes in parameter values and various structural assumptions? (ii) What did we learn from developing and running the model? What did we learn that we could not have learnt without running the model? (iii) Can the new insights be understood without the model? (iv) how should the modeling results be interpreted (are they descriptive, predictive or prescriptive)? The concluding section addresses how the results from the GET model may be used in the context of science-policy interactions, using Sweden as an example. Particular focus is given the extent to which integrated assessment models are used to provide first best policy advice (the cost-effective solution) and whether, perhaps, more effort should be directed to analyze second-best options that although perhaps more costly, stand a better chance of getting wider political acceptance, i.e., the real-world trade-off that policy makers face. A version of the GET model may be run interactively on www.chalmers.se/ee/getonline (this version "only" models the global energy system and the carbon cycle, the full version, not available online also include other greenhouse gases and aerosols, and a three box temperature response model).