Summaries of Themes
Summaries of the six key themes
Ice sheets
Theme leaders: Julian Dowdeswell and Tony Payne
The interchange of ideas between glaciologists studying contemporary ice masses and those working on the records left by former ice sheets has always been fruitful. Examples of the reciprocal nature of this relationship include the impact of the discovery of Heinrich Layers on studies of the dynamics of ice streams, and the influence of studies on the contemporary ice streams of the Siple Coast on the discovery and understanding of palaeo ice streams. The strengths of palaeo studies remain ease of access to the former ice-sheet bed over wide areas, and the ability to view dynamics over millennial or greater time scales. Progress has, however, often been confounded by the difficulty of dating material and issues associated with viewing a composite, palimpsest record that reflects the time-integrated sum of many processes. Both contemporary and palaeo studies have benefited enormously from the advent of satellite-based observations, while the use of high-resolution marine geophysics has seen major advances in our ability to map the entire bed of former ice sheets. The following overlapping topics served as an initial focus to our discussions.
1. Future evolution of ice sheets
How good are the palaeo analogues for the future evolution of the Antarctic and Greenland ice sheets? What work is required to tightly constrain their sizes during the Eemian? Do former ice sheets, such as the one that covered the Barents Sea, represent useful analogues to the loss of a marine ice sheet?
2. Past deglaciations
How well constrained are the records of ice-sheet retreat from the Last Glacial Maximum? How well understood are the causes of ice loss during this time and the relative roles of atmospheric and oceanic forcing? To what extent do models of global deglaciation require additional physics to explain the complete disappearance of North American and Eurasian ice masses? How well are the geographical source and causal mechanism of Melt Water Pulse 1A known?
3. Natural variability of ice sheets
Recent satellite-based studies of the present-day ice sheets reveal a range of new and exciting processes whose time scales (decades and centuries) lie well short of those traditionally associated with ice sheets. These observations raise issues about the natural (as opposed to anthopogenically induced) variability of ice sheets on these time scales. What opportunities exist for interrogating the palaeo record at these time scales to add a pre-anthropogenic context to these issues?
4. Ice sheet processes: grounding lines
How can the geological records of former grounding lines be used to understand processes occurring at this key location? To what extent can these records differentiate between different types of grounding-line retreat? Do complementary records of potential forcing (primarily oceanic?) exist and to what extent can we differentiate between genuine instability and heavily forced, rapid change?
5. Ice sheet processes: ice streams
How close are studies of former ice streams to generating quantitative information on ice velocity and basal sedimentological and hydrological processes? How is this type of information best integrated into modelling studies? Are the roles of ice streams and internal ice-flow instabilities in Heinrich Events fully understood?
Atmospheric composition & circulation
Theme leaders: Gavin Schmidt, Eric Wolff & Sandy Harrison
Changes in atmospheric circulation play a crucial role in determining how climate change is seen at local and regional scale. Signs of such changes in the past are recorded in multiple terrestrial and glaciological archives. For example, dust concentrations in ice cores, isotopic signals in cave records and pollen records of changing vegetation are indicative of changes in winds, temperature and precipitation over time. Changes in atmospheric composition for chemicals such as methane or nitrous oxide give an integrated picture of a range of changes in sources and sinks (e.g. in the terrestrial or oceanic biosphere). Understanding the reasons and mechanisms behind these changes is crucial for assessing the predictability of similar changes in the future. In this theme, we addressed:
1. Atmospheric composition
What changes in atmospheric composition (non-CO2 gases and aerosols) occurred over glacial-interglacial cycles, do we understand what caused them, and are present-day models inclusive enough to simulate them? Can models provide a self-consistent assessment of the feedbacks involved in Last Glacial Maximum (LGM) changes in composition? What data is necessary for different models to be distinguished?
2. Atmospheric circulation
Observations today show that atmospheric circulation is extremely dynamic. However, there are certain modes of variability - such as the North Atlantic Oscillation (NAO), or the Pacific-North American (PNA) pattern - that can be statistically defined and shown to have significant impacts on temperatures and rainfall. Can long-term shifts in these patterns be discerned in the palaeoclimate record and is there evidence that they are responding to external forcings? Can these results be used to improve regional climate projections?
3. Understanding of past changes
Large shifts in tropical rain bands have been recorded in response to orbital changes and correlated with rapid changes seen in the Greenland ice cores. Do we understand the mechanisms of these changes and are models sufficiently sensitive? Can we use this information to assess changes in rainfall patterns seen in future simulations?
4. Use of palaeoclimate data with models I
What are the best approaches to using palaeoclimate data with models? Is palaeo data assimilation possible? Do we need forward models for all proxy data embedded within the models? How important are multiple proxy compilations? Can we identify crucial datasets that truly distinguish between competing models?
5. Use of palaeoclimate data with models II
How do we move from observing discrepancies between different models and between model outcomes and data for periods such as the LGM to making changes that address these discrepancies? Should palaeo model-data comparisons be used to weight future projections from those models, and if so, which periods or events are most relevant?
Ocean circulation
Theme Leaders: Jean Lynch-Stieglitz & Jochem Marotzke
Changes in the ocean circulation are thought to make an important contribution to climate variability over the past 100,000 years on time scales from decadal to multi-millennial. However, the mechanistic understanding of the processes involved is poor, and there are conflicting interpretations of the palaeoclimate records of past circulation. In this theme we discussed our current understanding of past ocean circulation based on data and model results, and debated the relevance of this for future climate. We addressed the following issues:
1. Last Glacial Maximum
What robust findings concerning ocean circulation at the time of the Last Glacial Maximum can be inferred from the palaeoclimate record? Why do coupled climate model Last Glacial Maximum simulations appear to be inconsistent with the palaeoclimate data (and with each other)?
2. Atlantic meridional overturning circulation
Can coupled models generate rapid changes in the Atlantic meridional overturning circulation? Do we have a good understanding of the rapidity of past changes in the Atlantic meridional overturning circulation based on palaeoclimate data? How well are models able to reproduce past changes? Could palaeoclimate data be used to help identify model weaknesses?
3. Characterising past ocean circulation
What is needed to diagnose the past state of the ocean circulation, in contrast to diagnosing past changes in water properties? What palaeoclimate data would be most appropriate for this? Can improvements be made to methodologies for diagnosing past ocean circulation? What model development is necessary to better represent past ocean states and to incorporate knowledge of the palaeoclimate record?
4. Combined use of models and observations
Can palaeo-observations lead to improvements in the ocean components of climate models? Or will climate models instead lead to better interpretation of palaeoclimate records? Should we even aim for quantitative palaeoclimate reconstructions? Should palaeoceanographers be attempting to discover hitherto unimagined ocean/climate states?
Carbon cycle
Theme leaders: Ralph Keeling, Corinne Le Quéré & Danny Sigman
The ice core record over the last 650 thousand years shows that atmospheric carbon dioxide (CO2) concentration varies in step with climate over glacial cycles, reaching as low as 180 ppm during ice ages and as high as 295 ppm during interglacials. This remarkable finding suggests that CO2 is an amplifier of glacial cycles and may be a necessary ingredient of these cycles. Understanding the effect of CO2 on climate over glacial cycles may clarify the climatic and biotic significance of the anthropogenic CO2 rise. Moreover, the causes of glacial/interglacial CO2 change may coincide with processes influencing the buildup of CO2 in future and reveal the sensitivities of those processes. In this theme, we addressed central questions regarding glacial/interglacial CO2 change:
1. Polar surface ocean
Recently, efforts to explain glacial/interglacial CO2 change have returned their focus to the polar surface ocean and its connection to the deep ocean reservoir of dissolved inorganic carbon: how strong is the convergence of ideas, model simulations, and data? What are the best candidate processes (physical, biological or both) for controlling the flux of carbon through the polar surface ocean, and how can they be distinguished with measurements and models?
2. Terrestrial carbon reservoir
The terrestrial carbon reservoir is thought to be an important dynamic in glacial cycles, with carbon isotope data indicating it contributed CO2 to the ice age ocean and atmosphere. Do we now have a clear consensus in support of this view, or might the paradigm be overturned? What progress has been made in understanding the amplitudes, rates, timings, causes, and impacts? To what degree does this information affect our understanding of the structure of the CO2 ice core record?
3. Ocean alkalinity and nutrient content
Mean ocean changes in alkalinity and nutrient content are no longer considered the dominant cause of glacial CO2 cycles: do measurements and models justify this, and if so, how important was the ocean alkalinity budget as an amplifier of atmospheric CO2 changes driven by other processes? Do we fully understand the biogeochemical processes of importance in glacial cycles and CO2 change (e.g. weathering and calcium carbonate burial, ocean nutrient budgets and biomass stoichiometry)?
4. Glacial/interglacial CO2 change
To what degree are current hypotheses for glacial/interglacial CO2 change consistent with our larger understanding of glacial cycles, including changes in continental ice volume? Does climate perennially act as an intermediary between glacial ice and atmospheric CO2, or are there more direct connections between ice and CO2?
5. Link to modern climate
Beyond providing a broader perspective on the natural variability of CO2, in what ways does the study of glacial CO2 cycles advance our understanding of anthropogenic CO2 change? Do records of past change provide insight into how land/marine ecosystems will respond to carbon-related changes? Do modern conditions help to distinguish among different views of glacial/interglacial CO2 change?
Solar variability
Theme Leaders: Edouard Bard & Joanna Haigh
The Sun is the ultimate driver of the climate system and it therefore seems natural to look to variations in the solar energy reaching the Earth as a source of climate variability. The amount of solar radiative energy input to the Earth’s system is modulated by three different mechanisms: (i) geometric factors related to the Earth’s inclination and orbit around the Sun; (ii) variations in the radiation emitted by the Sun and (iii) variations in the Earth system which influence its reflectivity. There are also variations in the Sun’ s output of energetic particles, and solar modulation of galactic cosmic rays, which, while energetically very small compared with the radiative input, may influence atmospheric composition. In this theme we addressed the following issues:
1. Measurements of solar activity
What do we know about solar activity, and the magnitude and spectral composition of solar radiation incident at Earth, on palaeo timescales, over recent millennia and on timescales down to the 11-year activity cycle? To what extent can recent measurements of solar irradiance inform reconstructions of past activity?
2. Solar influence on climate
Many studies have shown correlations between measures of solar activity and climate over a wide range of timescales, but the statistical robustness of many of these relationships is questionable, and some of the correlations seem to appear and disappear over time. What signals of solar influence have been robustly detected in climate in palaeo records and on timescales down to the 11-year activity cycle? Do geographically similar signals appear at different timescales?
3. Mechanisms for solar modulation of climate
The different influences on the solar radiative energy input listed above act on timescales ranging from seconds to millions of years, so that the physical processes involved may be numerous. Thus the detection of solar signals in climate, and interpretation of the mechanisms behind these signals, requires a range of analytical and modelling tools. What mechanisms, in addition to direct radiative forcing, may be involved in the solar modulation of climate?
Rapid climate change
Theme leaders: Peter Huybers & Gerard Roe
For the purposes of this symposium we proposed that “rapid climate change” be taken to mean climate changes that occur fast enough to be relevant to society and therefore might, in principle, be of concern to policy makers in anticipating and planning for future climate scenarios. The purpose of this theme was to discuss how we can best focus the study of past climates so as to obtain useful information about climate change on societally relevant timescales. This implied the following issues:
1. Current understanding of rapid climate change
What features of past rapid climate changes are current models capable and incapable of representing? What are the clearest examples of where palaeoclimate has (or has not) proven useful for anticipating future climate? How and why? What elements of the problem were the reasons for its usefulness? Do these case studies of success (or otherwise) serve as a guide for how to conduct palaeoclimate studies in the future?
2. Relevance of past rapid climate change for the future
Which features of past rapid climate changes are sufficiently analogous to possible future changes that they are useful? What are the clearest examples of where future research in palaeoclimate can (or cannot) be useful for anticipating future climate with regard to sea ice, ice sheets, sea level, meridional overturning circulation, precipitation, ecosystem dynamics, biosphere feedbacks, and carbon cycle changes? Which are the best analogies to our future, have the maximum consequence for society, and can we anticipate being the most tractable/accessible?
3. Model simulation of rapid climate change
How much of a past rapid climate change must a model simulate before we have confidence in its ability to characterize future changes? Where has past climate served to test the completeness or accuracy of models? Are there certain features of past climate that we would want a model to simulate before we trust its predictions for the future? For example, is it a source of concern that models fail to reproduce Eocene climates? Does it imply we are missing some crucial piece of physics?
4. Use of the palaeoclimate record
How has the palaeoclimate record forced us to rethink the way the climate system works? Specifically in the context of making climate projections for our immediate future, what are the ways in which palaeoclimate observations have helped? In sorting through evidence of various past rapid climate changes, which are merely observations of what might be possible, and which have the capability to quantitatively alter our projections of future climate?
Changes in atmospheric circulation play a crucial role in determining how climate change is seen at local and regional scale. Signs of such changes in the past are recorded in multiple terrestrial and glaciological archives. For example, dust concentrations in ice cores, isotopic signals in cave records and pollen records of changing vegetation are indicative of changes in winds, temperature and precipitation over time. Changes in atmospheric composition for chemicals such as methane or nitrous oxide give an integrated picture of a range of changes in sources and sinks (e.g. in the terrestrial or oceanic biosphere). Understanding the reasons and mechanisms behind these changes is crucial for assessing the predictability of similar changes in the future. In this theme, we addressed:
1. Atmospheric composition
What changes in atmospheric composition (non-CO2 gases and aerosols) occurred over glacial-interglacial cycles, do we understand what caused them, and are present-day models inclusive enough to simulate them? Can models provide a self-consistent assessment of the feedbacks involved in Last Glacial Maximum (LGM) changes in composition? What data is necessary for different models to be distinguished?
2. Atmospheric circulation
Observations today show that atmospheric circulation is extremely dynamic. However, there are certain modes of variability - such as the North Atlantic Oscillation (NAO), or the Pacific-North American (PNA) pattern - that can be statistically defined and shown to have significant impacts on temperatures and rainfall. Can long-term shifts in these patterns be discerned in the palaeoclimate record and is there evidence that they are responding to external forcings? Can these results be used to improve regional climate projections?
3. Understanding of past changes
Large shifts in tropical rain bands have been recorded in response to orbital changes and correlated with rapid changes seen in the Greenland ice cores. Do we understand the mechanisms of these changes and are models sufficiently sensitive? Can we use this information to assess changes in rainfall patterns seen in future simulations?
4. Use of palaeoclimate data with models I
What are the best approaches to using palaeoclimate data with models? Is palaeo data assimilation possible? Do we need forward models for all proxy data embedded within the models? How important are multiple proxy compilations? Can we identify crucial datasets that truly distinguish between competing models?
5. Use of palaeoclimate data with models II
How do we move from observing discrepancies between different models and between model outcomes and data for periods such as the LGM to making changes that address these discrepancies? Should palaeo model-data comparisons be used to weight future projections from those models, and if so, which periods or events are most relevant?
Ocean circulation
Theme Leaders: Jean Lynch-Stieglitz & Jochem Marotzke
Changes in the ocean circulation are thought to make an important contribution to climate variability over the past 100,000 years on time scales from decadal to multi-millennial. However, the mechanistic understanding of the processes involved is poor, and there are conflicting interpretations of the palaeoclimate records of past circulation. In this theme we discussed our current understanding of past ocean circulation based on data and model results, and debated the relevance of this for future climate. We addressed the following issues:
1. Last Glacial Maximum
What robust findings concerning ocean circulation at the time of the Last Glacial Maximum can be inferred from the palaeoclimate record? Why do coupled climate model Last Glacial Maximum simulations appear to be inconsistent with the palaeoclimate data (and with each other)?
2. Atlantic meridional overturning circulation
Can coupled models generate rapid changes in the Atlantic meridional overturning circulation? Do we have a good understanding of the rapidity of past changes in the Atlantic meridional overturning circulation based on palaeoclimate data? How well are models able to reproduce past changes? Could palaeoclimate data be used to help identify model weaknesses?
3. Characterising past ocean circulation
What is needed to diagnose the past state of the ocean circulation, in contrast to diagnosing past changes in water properties? What palaeoclimate data would be most appropriate for this? Can improvements be made to methodologies for diagnosing past ocean circulation? What model development is necessary to better represent past ocean states and to incorporate knowledge of the palaeoclimate record?
4. Combined use of models and observations
Can palaeo-observations lead to improvements in the ocean components of climate models? Or will climate models instead lead to better interpretation of palaeoclimate records? Should we even aim for quantitative palaeoclimate reconstructions? Should palaeoceanographers be attempting to discover hitherto unimagined ocean/climate states?
Carbon cycle
Theme leaders: Ralph Keeling, Corinne Le Quéré & Danny Sigman
The ice core record over the last 650 thousand years shows that atmospheric carbon dioxide (CO2) concentration varies in step with climate over glacial cycles, reaching as low as 180 ppm during ice ages and as high as 295 ppm during interglacials. This remarkable finding suggests that CO2 is an amplifier of glacial cycles and may be a necessary ingredient of these cycles. Understanding the effect of CO2 on climate over glacial cycles may clarify the climatic and biotic significance of the anthropogenic CO2 rise. Moreover, the causes of glacial/interglacial CO2 change may coincide with processes influencing the buildup of CO2 in future and reveal the sensitivities of those processes. In this theme, we addressed central questions regarding glacial/interglacial CO2 change:
1. Polar surface ocean
Recently, efforts to explain glacial/interglacial CO2 change have returned their focus to the polar surface ocean and its connection to the deep ocean reservoir of dissolved inorganic carbon: how strong is the convergence of ideas, model simulations, and data? What are the best candidate processes (physical, biological or both) for controlling the flux of carbon through the polar surface ocean, and how can they be distinguished with measurements and models?
2. Terrestrial carbon reservoir
The terrestrial carbon reservoir is thought to be an important dynamic in glacial cycles, with carbon isotope data indicating it contributed CO2 to the ice age ocean and atmosphere. Do we now have a clear consensus in support of this view, or might the paradigm be overturned? What progress has been made in understanding the amplitudes, rates, timings, causes, and impacts? To what degree does this information affect our understanding of the structure of the CO2 ice core record?
3. Ocean alkalinity and nutrient content
Mean ocean changes in alkalinity and nutrient content are no longer considered the dominant cause of glacial CO2 cycles: do measurements and models justify this, and if so, how important was the ocean alkalinity budget as an amplifier of atmospheric CO2 changes driven by other processes? Do we fully understand the biogeochemical processes of importance in glacial cycles and CO2 change (e.g. weathering and calcium carbonate burial, ocean nutrient budgets and biomass stoichiometry)?
4. Glacial/interglacial CO2 change
To what degree are current hypotheses for glacial/interglacial CO2 change consistent with our larger understanding of glacial cycles, including changes in continental ice volume? Does climate perennially act as an intermediary between glacial ice and atmospheric CO2, or are there more direct connections between ice and CO2?
5. Link to modern climate
Beyond providing a broader perspective on the natural variability of CO2, in what ways does the study of glacial CO2 cycles advance our understanding of anthropogenic CO2 change? Do records of past change provide insight into how land/marine ecosystems will respond to carbon-related changes? Do modern conditions help to distinguish among different views of glacial/interglacial CO2 change?
Solar variability
Theme Leaders: Edouard Bard & Joanna Haigh
The Sun is the ultimate driver of the climate system and it therefore seems natural to look to variations in the solar energy reaching the Earth as a source of climate variability. The amount of solar radiative energy input to the Earth’s system is modulated by three different mechanisms: (i) geometric factors related to the Earth’s inclination and orbit around the Sun; (ii) variations in the radiation emitted by the Sun and (iii) variations in the Earth system which influence its reflectivity. There are also variations in the Sun’ s output of energetic particles, and solar modulation of galactic cosmic rays, which, while energetically very small compared with the radiative input, may influence atmospheric composition. In this theme we addressed the following issues:
1. Measurements of solar activity
What do we know about solar activity, and the magnitude and spectral composition of solar radiation incident at Earth, on palaeo timescales, over recent millennia and on timescales down to the 11-year activity cycle? To what extent can recent measurements of solar irradiance inform reconstructions of past activity?
2. Solar influence on climate
Many studies have shown correlations between measures of solar activity and climate over a wide range of timescales, but the statistical robustness of many of these relationships is questionable, and some of the correlations seem to appear and disappear over time. What signals of solar influence have been robustly detected in climate in palaeo records and on timescales down to the 11-year activity cycle? Do geographically similar signals appear at different timescales?
3. Mechanisms for solar modulation of climate
The different influences on the solar radiative energy input listed above act on timescales ranging from seconds to millions of years, so that the physical processes involved may be numerous. Thus the detection of solar signals in climate, and interpretation of the mechanisms behind these signals, requires a range of analytical and modelling tools. What mechanisms, in addition to direct radiative forcing, may be involved in the solar modulation of climate?
Rapid climate change
Theme leaders: Peter Huybers & Gerard Roe
For the purposes of this symposium we proposed that “rapid climate change” be taken to mean climate changes that occur fast enough to be relevant to society and therefore might, in principle, be of concern to policy makers in anticipating and planning for future climate scenarios. The purpose of this theme was to discuss how we can best focus the study of past climates so as to obtain useful information about climate change on societally relevant timescales. This implied the following issues:
1. Current understanding of rapid climate change
What features of past rapid climate changes are current models capable and incapable of representing? What are the clearest examples of where palaeoclimate has (or has not) proven useful for anticipating future climate? How and why? What elements of the problem were the reasons for its usefulness? Do these case studies of success (or otherwise) serve as a guide for how to conduct palaeoclimate studies in the future?
2. Relevance of past rapid climate change for the future
Which features of past rapid climate changes are sufficiently analogous to possible future changes that they are useful? What are the clearest examples of where future research in palaeoclimate can (or cannot) be useful for anticipating future climate with regard to sea ice, ice sheets, sea level, meridional overturning circulation, precipitation, ecosystem dynamics, biosphere feedbacks, and carbon cycle changes? Which are the best analogies to our future, have the maximum consequence for society, and can we anticipate being the most tractable/accessible?
3. Model simulation of rapid climate change
How much of a past rapid climate change must a model simulate before we have confidence in its ability to characterize future changes? Where has past climate served to test the completeness or accuracy of models? Are there certain features of past climate that we would want a model to simulate before we trust its predictions for the future? For example, is it a source of concern that models fail to reproduce Eocene climates? Does it imply we are missing some crucial piece of physics?
4. Use of the palaeoclimate record
How has the palaeoclimate record forced us to rethink the way the climate system works? Specifically in the context of making climate projections for our immediate future, what are the ways in which palaeoclimate observations have helped? In sorting through evidence of various past rapid climate changes, which are merely observations of what might be possible, and which have the capability to quantitatively alter our projections of future climate?
Page last modified: 17th Mar 2008 - 13:58:09
