Volume Five, Number One
Summer 2002

Click here for the PDF (Adobe Acrobat) version of this newsletter

Overview of Hilbertian Questions
Analytical Questions
Operational Questions
Normative Questions
Strategic Questions
Upcoming Confrences of Interest
International Conference On Earth System Modelling

The Hilbertian Questions

At the December, 2000 GAIM Task Force meeting in San Francisco, it was decided that an email conference should be convened to develop a set of updated questions for GAIM to address in the coming years, and to be used as a contribution to steering the direction of the next phase of IGBP research in general. The discussion produced questions at various levels and of contrasting type. However, some common issues emerged. Firstly, the questions should address the Earth system as a unit rather than being limited to individual components such as the atmosphere or ocean. Secondly, the questions should be formulated such that they lead to a set of well-posed, testable hypotheses to be explored by GAIM, the Core Projects, and the global change research community at large. Thirdly, the questions can be grouped by their nature and the scope of the issues they address.

The fundamental GAIM questions should, in fact, anticipate the advent of a unified Earth System Science and therefore encompass the natural and socioeconomic dimensions in a balanced way. This accounts for "horizontal integration" across the disciplines, but "vertical integration" across the layers of the problem-solving process is no less important. Consequently, we use the categories

The total number of questions in these groups are 23. This is the number of challenges that Hilbert listed for 20th century mathematics, and so this could be viewed as a "Hilbertian" approach to the Earth System.

Analytical Questions:
1. What are the vital organs of the ecosphere in view of operation and evolution?
2. What are the major dynamical patterns, teleconnections and feedback loops in the planetary machinery?
3. What are the critical elements (thresholds, bottlenecks, switches) in the Earth System?
4. What are the characteristic regimes and time-scales of natural planetary variability?
5. What are the anthropogenic disturbance regimes and teleperturbations that matter at the Earth-System level?
6. Which are the vital ecosphere organs and critical planetary elements that can actually be transformed by human action?
7. Which are the most vulnerable regions under global change?
8. How are abrupt and extreme events processed through nature-society interactions?

Operational Questions:
9. What are the principles for constructing "macroscopes", i.e., representations of the Earth System that aggregate away the details while retaining all systems-order items?
10. What levels of complexity and resolution have to be achieved in Earth System modelling?
11. Is it possible to describe the Earth System as a composition of weakly coupled organs and regions, and to reconstruct the planetary machinery from these parts?
12. What might be the most effective global strategy for generating, processing and integrating relevant Earth System data sets?
13. What are the best techniques for analyzing and possibly predicting irregular events?
14. What are the most appropriate methodologies for integrating natural-science and social-science knowledge?

Normative Questions:
15. What are the general criteria and principles for distinguishing non-sustainable and sustainable futures?
16. What is the carrying capacity of the Earth?
17. What are the accessible but intolerable domains in the co-evolution space of nature and humanity?
18. What kind of nature do modern societies want?
19. What are the equity principles that should govern global environmental management?

Strategic Questions:
20. What is the optimal mix of adaptation and mitigation measures to respond to global change?
21. What is the optimal decomposition of the planetary surface into nature reserves and managed areas?
22. What are the options and caveats for technological fixes like geoengineering and genetic modification?
23. What is the structure of an effective and efficient system of global environment & development institutions?


Analytical Questions

1. What are the vital organs of the ecosphere in view of operation and evolution?
The ecosphere (the Earth system minus human components) can be likened to an organism in that it exhibits self-regulating and other emergent global properties. Organisms possess organs - parts that are specialized to perform one or a number of particular functions - and some are vital in that failure of the organ results in collapse of the system. It is important to understand how the ecosphere is differentiated in terms of functions, and which of the organs so defined are vital to the life-support capacity of the whole system (operation) and its capacity for further development (evolution).

Some key processes for the ecosphere's life-support capacity have been identified, such as complementary biochemical transformations (e.g. oxygenic photosynthesis and aerobic respiration). The question of which 'organs' are vital to enable further evolution has received relatively little attention. Biodiversity 'hotspots' that have been the sites of recent evolutionary radiation are being catalogued, and the future of evolution following the current human-induced mass extinction is only beginning to be considered.

A number of approaches can be applied in the exploration of vital organs of the ecosphere. These include, among others, a) Study of catastrophic events in Earth history and their aftermath; b) Numerical modelling to highlight key properties of the ecosphere, and to simulate the removal of various components from Earth system; c) Reconstruction of the ecosphere on a smaller scale; and d) Real-time observation of the response of the ecosphere to the removal of various components by human activity. The first three of these approaches may be less disruptive to the system than the fourth. Once the vital organs are identified, they can be better represented in future generations of Earth system models.

2. What are the major dynamical patterns, teleconnections and feedback loops in the planetary machinery?
The existence of dynamical patterns in the Earth System is dramatically shown by polar ice-core records that document temporal connections over a period of more than 400 ka between variations of numerous measurable properties, such as isotopic indicators of past changes in polar climate, and the atmospheric content of greenhouse gases (CO2, CH4, N2O) and other atmospheric constituents trapped in air bubbles. This time scale of variation is characterized by glacial-interglacial cycles with an approximate periodicity of 100 ka, shorter periodicities directly associated with the Milankovitch obliquity (40 ka) and precession (23 ka, 19 ka) cycles, and millennial-scale (ca 1.5 ka) oscillations of unknown origin. Dynamical patterns are also in evidence in features of interannual to decadal variability in the present climate (the best-known being the El Nino-Southern Oscillation, ENSO, and the northern annular mode that gives rise to the North Atlantic Oscillation, NAO). These phenomena have observable impacts on biotic distributions and atmospheric CO2 content as well as on human affairs, as mediated by effects on the occurrence of natural hazards such as floods, droughts and fires. There are links across time scales as well; for example, there is evidence for changes in the properties of both ENSO and NAO over the Holocene (the past 11.5 ka), and there is some indication that changes in these climatic variability patterns may be as important as changes in the mean climatic state in accounting for changes in biotic indicators and for events in the archaeological record.
The existence of teleconnections is a general property of the above dynamical phenomena. Proxy indicators of past climates show both synchronous and out-of-phase relationships among regions, depending on the variable and time scale of interest. Climate changes affecting only one region essentially do not occur. On interannual to centennial time scales, although the variability modes originate in particular regional features of the atmospheric and oceanic circulations, they typically have "far-field" effects, as seen for example in such ways as the impact of ENSO on the quality of the Spanish wine harvest.
Comparison of Dust, CO2, and Deuterium (temperature) in Vostok Ice Core, Antartica (Modified after Petit et al. 1999)

The existence of feedback loops in the Earth System has never been in doubt, but recent work has greatly complicated the picture by revealing the importance of biotic processes in regulating the climate, at all time scales from billions of years (evolution of oxygenic photosynthesis, control of atmospheric CO2 content by higher-plant weathering and marine calcification...) through Milankovitch time scales (role of terrestrial dust in promoting marine export production, control of atmospheric oxidizing capacity by biogenic sources and sinks of reactive trace gases...) to the subdaily regulation of the hydrological cycle over the continents by the feedbacks between stomatal behavior and convective boundary-layer dynamics.
Recognizing the significant role now played by human activities in perturbing the planetary machinery adds urgency to the already outstanding scientific problem of understanding the dynamics of the Earth System. It also adds complexity to an already highly complex problem; for example, human involvement adds further dynamics, teleconnections and feedbacks associated with the workings of global markets, the economic implications of adaptation and mitigation responses to climate change, and the evolution of international laws and institutions.

Several fully coupled climate-carbon cycle models have been constructed, and a critical mass of empirical knowledge concerning previously "missing" processes (such as the impacts of atmospheric iron input on marine phytoplankton communities, and the controls of trace gas emissions by terrestrial ecosystems) has been assembled to allow a first generation of more complex Earth System Models to be built. To move forward, it will also be necessary to begin to couple "natural system" models to economic models. This development will build on already-started initiatives to include agriculture, forestry and urbanization in the same modelling framework as natural ecosystem dynamics. This approach is being actively promoted by GAIM.

3. What are the critical elements (thresholds, bottlenecks, switches) in the Earth System?
The planetary machinery is an externally forced, dissipative system operating quasi-resiliently far from thermodynamic equilibrium. A linear and homogenous system could never exhibit such behavior. Therefore, a number of nonlinear processes and non-uniform structures must prevail, i.e. major
Critical switch and choke points within the Eatrh System as Discussed in Question 3 (from H.J. Schellnhuber)
thresholds, bottlenecks and switch elements exist. As a consequence, the Earth System is susceptible to critical perturbations which may either trigger repetitions of paleo-systemic behavior or even novel modes of operation.

Over the last decades, the constitutive disciplines of Earth System science have discovered more and more bifurcation potentials in the planetary machinery, mainly from paleodata. Prominent examples are the instabilities of NADW, continental monsoons, methane clathrates and large-scale ecosystems, as well as the salinity valves represented by the Bering Strait, the strait of Gibraltar and the Skagerrak.

It may be useful to establish a more comprehensive "singularity catalogue" with collaboration between GAIM and PAGES. In parallel to the empirical approach, a number of simulation exercises have to be carried out for the exploration of Earth System. Earth-system Models of Intermediate Complexity (EMICs) could take the lead and set the stage for full-complexity models which should corroborate the existence of singularities. This strategy must be supported by state-of-the-art advice from the nonlinear dynamics community. Finally, if empirical and theoretical hints accumulate to strong evidence for critical elements in the planetary machinery, monitoring programs should be devised which could observe precursor signals of incipient phase transitions.

4. What are the characteristic regimes and time-scales of natural planetary variability?
From the physics of fluids we know that processes taking place on different spatial and tempo-ral scales are governed by different mechanisms and thus, have to be described by different types of models. With respect to Earth system dynamics, we can assume that long-term variability such as glacial cycles or the transition from the "hot house" of the Cretaceous to the Quaternary ice ages is governed by differ-ent mechanisms than climate change over the last century. This does not mean that long-term and short term Earth system dynamics are completely independent nor that assessment of recent climate change can be accomplished in ignorance of past climate change. It does, however, highlight the fact that it is critical to carefully consider "scaling" issues in Earth system dynamics. Therefore the dominant spatial and temporal scales relevant to Earth system processes must be carefully identified and explored.

Schematic spectra of global mean temperature variance indicate a number of peaks that can, by and large, be identified with externally driven variability or internal variability of the climate system. For example, peaks around 40,000 and 23,000 years are associated with direct, or linear, orbital forcing of the system, while the peak around 100,000 is viewed in some models as a non-linear response to orbital forcing. High-frequency variability at approximately 2 and 4-7 years is interpreted as internal variations in the stratosphere - troposphere system (the Quasi Biannual Oscillations) and atmosphere - ocean system (the El Nino - Southern Oscillation), respectively. Of particular interest are the Vostok ice core records that provide a wealth of information about the mode of operation of the natural Earth system in the most recent geological epoch.

Exploration of characteristic time scales and regimes of natural planetary variability requires a coordinated joint effort of paleoclimatic reconstruction and modelling. Once models of the natural Earth system are validated by use of paleo data, these models can be used to explore the phase space in order to access the possibility of alternative regimes of operation.

5. Which are the vital ecosphere organs and critical planetary elements that can actually be transformed by human action?
One general characteristic distinguishing humans from other species is their ability to choose to significantly modify their environment and other life forms to gain a short-term advantage. Human action is transforming a number of aspects of the Earth system, such as the global radiation budget and the character of some sectors of the biota. Many of these transformations may be large from a human perspective, but may exert only a small impact on the overall operation of the Earth System. While a tightly-
coupled system might allow a perturbation of one component to propagate to other components, the survival of the biota through the past 3.6 billion years suggests that the system possesses self-regulatory properties that act to damp perturbations in the long term. However, since humans are modifying many aspects of the Earth System, it might be possible for large system-level perturbations to occur through synergisms between several independent anthropogenic transformations. Furthermore, if some human activities modify those aspects of the Earth System which contribute to self-regulation, this could affect the resilience of the system to other human perturbations and also to non-anthropogenic perturbations such as solar variability or volcanic eruptions.

Progress has been made in understanding certain individual perturbations in isolation. For example, atmospheric chemistry and surface climate are observed to be changing in a manner attributable largely to human activity, and extensive research effort is focused on predicting further such changes within human lifetimes. While many such modifications are already under investigation for their wider implications, the significance of others may not yet have been realized. Rather than simply distinguishing between elements that can be transformed and those that cannot be transformed, it may be useful to consider a classification of perturbations according to magnitude, ranging from no anthropogenic effect to complete anthropogenic obliteration of the planetary element in question. For each element, it could then be informative to consider what thresholds exist between negative and positive feedback behavior, and what would be necessary for these thresholds to be exceeded?

To identify the critical Earth System components vulnerable to human transformation, it will be necessary to identify the links between these and the parts of the Earth System which can be directly modified by humans. Integrated Earth-System models are likely to be valuable tools for investigating links between critical and directly-transformable components of the system. The directly-transformable components of the system will include those which are already implicated as mechanisms of environmental change, but may also include other means of humans interaction with the rest of the system that have not yet been identified as being of particular significance.

6. What are the anthropogenic disturbance regimes and teleperturbations that matter at the Earth-System level?
In counterpoint to Question #5, which addresses the elements that can be anthropogenically perturbed, it is also necessary to examine the suite of known anthropogenic perturbations to assess which ones have Earth-system level effects, and which ones merely affect the sustainability of human population or other individual species without fundamentally altering the state of Earth system operation. Evidence is accumulating that suggests that anthropogenic carbon emissions are affecting global mean temperature, but there is still not much conclusive scientific evidence that human perturbations can effect the mode of operation of the Earth System.

Many connections between human disturbances and Earth-System features can, and have been studied on a more or less ad hoc basis, and rarely with a research design that directly permits one to establish or refute the attribution. Making a convincing case for any such connection is important, since it means establishing an important aspect of Earth system behavior, as well as providing more solid underpinnings for decision making, particularly in terms of preemptive adaptation in regions or sectors that are likely affected by such disturbances (See Question #7, below). Whole-system perturbations may be unlikely the time scale of human lifespan, but one concern is the possibility that anthropogenic perturbations could induce large positive feedbacks such as the release of natural reservoirs of methane. Other potential human-induced radiation-balance changes could severely impact many "higher" life forms but still may not significantly other organisms such as bacteria or even insects. Many known perturbations are still only just beginning to be investigated. For example, while it is known that biodiversity is being reduced through a number of anthropogenic causes, the importance of this for the stability of the Earth System has only recently come under investigation.

Earth-System modelling needs to be taken to a level of development where signal-to-noise relationships in model experiments are assessed in a more consistent way. It is clear that many comprehensive coupled climate model experiments cannot provide this due to high computational costs. In addition to the expectation of ever more powerful computers and numerically more efficient model implementations, it will be necessary to investigate hypothesized cause-effect relationships of human impacts on Earth-System features in direct relation to model complexity. This could be done by asking, for example, "what level of process resolution is required for a climate model to adequately resolve the effect of deforestation on broad-scale circulation features?".

7. Which are the most vulnerable regions under global change?
The vulnerability of natural and social systems to global change depends on the degree to which the systems respond to a given type of environmental stress and the ability of the systems to adapt to global changes. Although all regions are vulnerable, the extent of vulnerability will vary across regions, and even within each region, because of differences in local environmental conditions, stresses to ecosystems, resource-use patterns and policy frameworks. An understanding of the extent of vulnerability is necessary for implementation of mitigation policy and of adaptive responses that would ensure the continued ecosystem health as well as the provision of key goods and services for successful economic and social development.

Recent efforts directed towards understanding vulnerability of different regions to global change have produced in a number of key publications (e.g. IPCC Third Assessment Report, 2001). Focus has been placed on certain sectors that are identified as vital to sustainable development of social systems and that are also known to be sensitive to global change and climate variability. These assessments indicate that the degree of vulnerability of different regions depends on their capability to implement appropriate adaptive responses, as a result of suitable resource allocation and relevant policy frameworks. There appears to be consensus, however, that the regions most vulnerable to global change are the tropical and polar regions.

Accurate assessment of vulnerability is currently constrained by a number of factors, namely: uncertainties regarding the extent and impact of global change, inadequate understanding of the interaction between different systems and sectors in the event of global change, the diversity of policy frameworks and resource-use patterns. To address these issues, therefore, there is need to focus on vulnerability research with the following objectives:

8. How are abrupt and extreme events processed through nature-society interactions?
Nature-society systems have many components that interact in a great many ways, some of which are feedbacks. There are also redundant components in such systems; or perhaps more accurately, components are duplicated. The strength of the feedbacks and the degree of duplication contribute to the dynamics and resilience of the system, but in ways that are not always clear. Empirical studies of the impact of abrupt and extreme events on nature-society systems are needed, along with analysis of feedbacks, redundancies, and resilience.

It is noteworthy that perturbations reveal the dynamics of a system better than any other method. The ringing of a bell is a useful example - even though a bronze bell and a clay bell might look the same, a single blow from a hammer will reveal the nature of the bell. While it is important to understand how perturbations reverberate through nature-society systems for policy formulation and institutional design purposes, the study of such perturbations will also be methodologically very instructive. Although there are no precedents of extreme Earth system events since the onset of human society, exploration of the processing of various minor perturbations may reveal at least qualitative insights regarding the role and response to more extreme events.


Operational Questions

9. What are the principles for constructing "macroscopes", i.e., representations of the Earth System that aggregate away the details while retaining all systems-order items? In the same sense that a microscope represents an instrument that attempts to magnify the details of a complex structure to a degree determined by the available technology and by the choice of the observer, we can speak conceptually about a macroscope that steps back to observe the Earth system in total so as to allow the observer to identify those aspects of the system that are crucial to overall functioning, and those that, in the range of their possible variability, do not affect system behavior. It is useful to explore the possible aspects of the Earth System macroscope, since this exploration could provide guidelines for "complexity design" for streamlining and optimizing Earth System models (i.e. "what needs to be included, and what may be left out"), observation strategies, and experimental studies.

Although there is no established macroscope for the Earth system (a "geoscope") some progress may be made by exploring observational, experimental, and modelling approaches. While it is difficult at present to define the kinds of necessary observations and experiments that would highlight system level behavior by "filtering" away the details of component subsystems, this can more readily be accomplished by targeted modelling activities. One such type of activity is through the development of models of intermediate complexity, described with relation to Question #10, below.

10. What levels of complexity and resolution have to be achieved in Earth System modelling?
Earth system models are constructed using various levels of temporal and geographic resolution and complexity in terms of components, processes and linkages within the system. Here-tofore, this has been done in a generally ad hoc manner applying the highest resolution possible from calibration data as modulated by computational resources. However, optimal model efficiency can only be achieved by a balance of resolution between component processes on the basis of rates and magnitude of interactions through energy and material exchanges that control the functioning of the Earth system as a whole.

Range of Earth System models in three dimensional complexity space (After Claussen et al. 2002)
There is a spectrum of models used in Earth system analysis which can, to a first approximation, be categorized into three groups: conceptual models, Earth-system Models of Intermediate Complexity (EMICs), and comprehensive models. Comprehensive models are designed to provide the most detailed treatment of the climate system and its dynamics thereby serving as the most re-alistic representation of nature. The major limitation in the application of comprehensive mod-els to studies of long-term climate variability arises from their high computational cost. At the other end of the complexity spectrum of natural Earth system models are the conceptual or tu-torial models. These mostly inductive models are simple mechanistic or statistical models that are designed to demonstrate the plausibility of processes. They are formulated based on a gross understanding of the feedbacks that are likely to be involved. To bridge the gap between con-ceptual and comprehensive models, EMICs are being developed. These models explicitly sim-ulate the interactions among as many components of the natural Earth system as possible. They include most of the processes described in comprehensive models, albeit in a more reduced, i.e., a more parameterized form, and they operate at a much coarser spatial and temporal resolution than comprehensive models.

In order to investigate the appropriate levels of resolution for components of Earth system mod-els, it is first necessary to search through functional space for spatio-temporal regimes. In so do-ing, the structure of the relationship between the amplitude and spatio-temporal range of each modelled process will be revealed. It may be then possible to simplify the analysis by lineariza-tion of some components of the system. There is no "standard" resolution within the Earth sys-tem, or even within specific ecosystems. For example, trophic complexity in the oceans is far greater than its terrestrial counterpart. Consequently, models must be devel-oped that can accommodate the differences in appropriate resolution within the myriad compo-nents of the Earth system.

11. Is it possible to describe the Earth System as a composition of weakly coupled organs and regions, and to reconstruct the planetary machinery from these parts?
An "Earth System organ approach" focuses on key functional entities in order to provide a quick and qualitative understanding of the system's dynamics. It may be possible to characterize these functional entities in such a way that they can be studied individually, and then recombined by highlighting the interactions and feedbacks between each and the whole.

Atmospheric scientists and oceanographers identified "organs" long before the development of Earth system models. For example, decades of observations and conceptual modeling led to the identification of structures (telecorrelations), culminating in the notion of "teleconnections" like North Atlantic Oscillation (NAO), El Niño Southern Oscillation (ENSO), and Thermo-Haline Circulation in general. The rich literature on teleconnections dealing with the attempt to understand them as semi-isolated phenomena provides a hint that indeed they could qualify as "organs". However, up to now, attempts to construct a global dynamics from these organs have not been successful. On the other hand, some EMICs are successfully operating on the basis of a few dozen key processes that have been identified by a bottom-up procedure of multi-scale analysis of the fundamental equations of motion. Teleconnections are the result of the non-linear, sometimes strong, interaction of those key processes.

In using weakly coupled "organs", it may be possible to clarify qualitative influences to facilitate a proper modeling integration of anthroposphere and ecosphere, elucidate a phase space structure of switches & thresholds, and ultimately obtain a suite of Earth system models with quantitative predictive capability.

12. What might be the most effective global strategy for generating, processing and integrating relevant Earth System data sets?
The scientific community has not yet devised an effective global strategy for the production of the data archives that are required to develop a better understanding of Earth System processes. New data archives are usually created for specific purposes only, or they reach only a fraction of the scientific community. In addition, the integration of data sets, e.g. by rescaling of spatial information to common grids or surfaces, is presently done on an ad hoc basis by most researchers, and the influence of the chosen method on the end result is rarely investigated. In many cases, public funds are expended first to generate data, and then expended again to purchase the same data due to "cost recovery" policies that have emerged in recent years. It would be beneficial to generate true state-of-the-art data sets for basic quantities of the Earth System (such as climate, atmospheric trace gases, human land use etc.), while ensuring free and open access to these data by the scientific community as well as the lay public and educational institutions.

The current level of development is well illustrated by ISLSCP (International Satellite Land Surface Climatology Programme), where multiple global temporal and spatial data sets are assembled in a set of CD-ROMs which are distributed free of charge to all scientists. This activity has, however, grossly insufficient resources to ensure optimal quality of all involved data sets, and it includes by necessity only a limited set of quantities. Several similar web- or CD-based activities exist elsewhere but none contain the global and topically comprehensive data sets needed for Earth System Science.

It is possible that some improvement could be achieved through collaborative agreements, e.g. on the use of common data documentation standards and collaborative ways of quality assessment. IGBP/GAIM is now launching an effort to construct an "Earth System Atlas" on the basis of compilation, evaluation, and standardization of a comprehensive suite of Earth system level data sets. While this effort may serve to enable access to diverse communities to the data, it will likely reveal gaps in existing data resources, leading, perhaps, to generation of key additional data in the future.

13. What are the best techniques for analyzing and possibly predicting irregular events?
Three modes of thermohaline circulation during the last ice age. Center: Prevailing stable cold conditions; Top: Heinrich event; Bottom: Dansgaard-Oeschger event. (Modified from Rahmstorf, 2001)
The term irregular event here refers to a non-linear change in the Earth system, i.e. a change that is not smooth, gradual and proportional to its forcing, in the way the global mean temperature is expected to rise gradually as the atmospheric greenhouse gas concentration rises. A typical, simple mechanism for an irregular event is that of crossing a stability threshold. There is little response for a forcing below the critical threshold, but when the threshold is crossed a qualitative change occurs. This is often the case in systems with positive feedback. Non-linear and irregular changes present a particular challenge, since the response of a system near a threshold (or other non-linearity) can be complex, and depends very much on how close the system is to the threshold, i.e. it depends on the mean state.

Reasonable simulations of past irregular events, such as Dansgaard-Oeschger events, have been obtained with climate models in recent years. Irregular events in response to global warming are found in some coupled climate model simulations in some scenarios. These include, for example, a shutdown of either Labrador Sea convection or Greenland-Norwegian Sea convection in the first half of this century. However, other models do not show these responses. This is perhaps not surprising given the sensitive dependence of thresholds on the details of the mean state and the exact strength of various feedbacks, but the causes for model differences are not yet fully understood.

The challenge that confronts the scientific community lies in a systematic assessment of the risk of crossing critical thresholds and triggering irregular events for a given change in climate. Clearly, the traditional approach of computing a few 'best guess' climate change scenarios is not an appropriate strategy for this. In addition to the classic methods of narrowing down the uncertainty in observations, performing process studies and improving models, this will require novel ways to deal with uncertainty, and to scan a broad range of possible futures within the uncertainty limits of our current knowledge. It will also require assessment of the probabilities for events that are neither the most likely outcome of anthropogenic warming, nor can be ruled out with confidence: the so-called "low probability - high impact" risks.

14. What are the most appropriate methodologies for integrating natural-science and social-science knowledge?
There has traditionally been a wide gap between social scientists and natural scientists. It is now widely agreed that attempts to understand environmental change, and finding solutions to problems caused by this change, require insights from the natural and social sciences, and from the humanities. The environmental changes that are within the ability of humanity to control, have been, for the most part, created by people. Understanding of human values, beliefs and institutions is therefore just as necessary as an understanding of biogeochemistry and other natural sciences. Solutions require economics and political science, as well as natural science. The process by which integrated knowledge is created should be applied equally to the research team as well as to the research problem.

In order to create intergrated research programs, it will be necessary to: identify the controlling models used by members of the research team, both to explain the world and to integrate the worldviews of the disciplines in the team; analyze social, and biological learning and adaptation in human-environment systems and their management agents, and in the research team; chart the history of both the human-environment systems and their management agents, and in the research team to discover how it arrived at its current state, and to document its dynamics; focus on management and the policy context of human-environment systems, and reflect on the management of the research team; structure the research and research team to fit the organization and scale of the human-environment system.


Normative Questions

15. What are the general criteria and principles for distinguishing non-sustainable and sustainable futures?
There has been considerable discourse within and beyond the scientific community regarding "sustainability issues" and the types of human actions that would lead to more or less sustainable futures. However, a generally accepted definition of sustainability does not yet exist, and even within presently proposed definitions, there are seldom any quantitative bounds on the geographic, temporal, and conceptual extent of the "system" with whose sustainability we are concerned. As such, it is necessary at this point to determine these bounds so that specific sustainability criteria can be developed. At the Earth system level, geographic extent is obviously global. Temporal sustainability is a more complex issue, and while some consider a process sustainable if it can be maintained without significant change for a duration of one human lifetime, others limit sustainability strictly to processes that can be maintained indefinitely. A geological view lies between these two extremes because Earth history has shown us that environmental conditions and the ecosystems that depend on them have evolved throughout the last few billion years at various time scales, and are sure to continue to do so in the future. The conceptual extent of a sustainable system encompasses the number and nature of other systems with which it interacts. Consequently, the criteria for identifying potential sustainable futures depend on the geographic, temporal and conceptual bounds that define the system of interest.

For the purpose of Earth System Analysis, we will define a sustainable system as one that can be maintained in its present form, with input and output of energy and materials in steady state equilibrium with the other systems with which is interacts. This exchange of energy and materials must be at quantitative levels that do not lead to a significant change in any reservoirs of energy or materials within the supporting systems, but the systems themselves may evolve over time that are long relative to the rate of exchanges between systems. With this definition in mind, is becomes clear that most segments of human society are distinctly UNsustainable and at present rates of consumption and emissions, will collapse in a very short time relative to the time scales of natural environmental variability. A second aspect of sustainability involves social systems and their ability to adapt to ecologically sustainable futures. If a "sustainable future" is politically or socially unpalatable it will not be maintained, and is thus unsustainable on social and political grounds. In some cases, what is sustainable socially or politically may not be sustainable ecologically and vice-versa.

In order to determine criteria for identifying sustainable futures, it will be necessary to:

16. What is the carrying capacity of the Earth?
The carrying capacity of the Earth can be defined by the number of humans that can be supported by the Earth's life support system that consists of interlinked biological, chemical and geophysical subsystems. Carrying capacity can only be estimated within the context of long-term sustainability of the delivery of ecosystem goods and services (see Question #15 above). The concept of carrying capacity derives from terrestrial ecology. Within such systems there is no single value for carrying capacity of an ecosystem. For example, in grazed savannas, there are multiple stable combinations of trees, grasses, and grazing animals. For humans in the Earth system, the situation is even more complex. The carrying capacity will in fact be a multidimensional surface influenced technology, lifestyle, social organization, distribution of wealth, and other socioeconomic factors. It is important to determine the Earth's carrying capacity so that the multidimensional range of thresholds can be identified, which when exceed, will cause the stock from which the Earth's life support system is derived to become depleted. In part, carrying capacity must be -defined as much as determined. Given a fixed level of ecosystem goods and services, a value judgment can be made regarding how much should be allocated to each person, and population numbers calculated from that point.

Some attempts are being made to estimate some aspects of carrying capacity, including application of ecological footprints at the global scale, the Millennium Ecosystem Assessment, resource assessments such as done by the World Resource Institute (WRI), and various joint research programs such as Carbon, Food Systems, and Water.

In any attempt to quantify the Earth's carrying capacity it is necessary to first define the operating parameters and limiting constraints. Constraints are provided by biological, geophysical, and socioeconomic considerations. For example, the availability of fresh water cannot be calculated on the basis of mining of fossil groundwater because of its nonrenewable nature. As another example, a socioeconomic constraint is provided by societal decisions regarding landscape amenity, rural vs. urban lifestyles, transportation systems, and other normative issues pertaining to "quality of life." While these are only examples, a comprehensive analysis of such constraints must be compiled so that the rate of production and thus availability of ecosystem goods and services can be compared to the complexities of demand defined by human needs and desires. Once this is accomplished, a multidimensional carrying capacity surface can be constructed expressing the tradeoffs between availability, consumption, and human population.

17. What are the accessible but intolerable domains in the co-evolution space of nature and humanity?
If we construct a "phase space" that follows the coevolution of the ecosphere and anthroposphere, "catastrophe domains" emerge that can be defined as (i) accessible, (ii) intolerable, and (iii) inescapable subsets of that space. Due to effect of accelerated Global Change on global environmental, economic and social conditions, the potential for non-sustainable developments ending in catastrophe domains is growing dramatically. In view of retardation effects, i.e., big time lags between causes and impacts, science is confronted with the need to anticipate those developments as soon as possible to provide decision makers with mitigation options. A characteristic feature of catastrophe domains is that there are no acceptable adaptation measures, so mitigation, however unpalatable, becomes the only alternative.

When it comes to assessing the intolerability of domains, both systemic and normative criteria can be employed. A judgment becomes rather easy if critical ecosphere thresholds can be identified where exceedence leads to global or regional collapse of some aspect of the system. IGBP and WCRP research is discovering more and more of those critical elements in coevolution space (switches and chokes). Equivalent criteria are much harder to identify in the socioeconomic realm. The attempts made so far focus on the definition of basic human needs (like a minimum daily allowance of clean water per capita) or the delineation of individual "environmental spaces."

Several steps will need to be taken in order to identify accessible but intolerable domains. The critical Earth System elements that can be affected by humanity have to be identified (see Question #5, above), methodologies for inverse approaches have to be advanced, and a systematic stakeholder discourse involving all societal strata has to be devised for exploring the socioeconomic "limits of tolerability".

18. What kind of nature do societies want?
Part of the motivation for current studies in Earth System analysis comes from interest in global-scale conservation, i.e. from the desire to preserve specific aspects of global "nature", such as the potential of the biosphere to provide food, fiber and various services for human society. The vulnerability of this life support system has so far mostly been assessed in relation to the view that an optimal solution would be the stabilization of current conditions, or even the restoration of conditions some time ago. Agreement regarding the optimal configuration of nature is lacking globally, and even within some sectors of society.
Further, there is presently no scientific evidence that this conservationist view is the only view possible. Instead, a number of societies may decide that they might be better served by a different management that tries to optimize services provided by nature in such a way that human needs (such as the production of food) were satisfied in the most efficient way according to some criteria (e.g., economic aspects, or nutritional value).

This question involves ethical as well as practical issues. Should "wanted" relationships with natural systems that are unsustainable be disqualified from consideration? Should humanity play a "stewardship" role in the maintenance of natural ecosystems? Does humanity, as a whole, have the "right" to displace natural ecosystems for the sole purpose of increasing human population, or increasing resource consumption of the existing population? Value-laden issues such as these render the resolution of this question problematical in a way that cannot be addressed by the scientific community, so it has been largely ignored.

In order to even begin to explore this question, a dialog must be established between scientists and investigators in sociology, ethics, philosophy, cultural anthropology, and religion. While it could be agreed that peoples' values change in response to changing environmental conditions, we may be limited to using our present systems of values to address this question. Ultimately, addressing these questions would help in understanding the broader options that exist for the achievement of greater sustainability at the broad-scale Earth System level.

19. What are the equity principles that should govern global environmental management?
The Rio Declaration on Environment and Development consists of a number of principles designed to guide national and international actions on environment, development and social issues. One of the principles is that of intergenerational equity, which reflects the view that as members of the present gen
Global average sea level rise 1990 to 2100 for the IPCC Special Report Emission Scenarios (SRES). Thermal expansion and land ice changes were calculated using a simple climate model calibrated separately for each of seven AOGCMs, and contributions from changes in permafrost, the effect of sediment deposition and the long-term adjustments of the ice sheets to past climate change were added. Each of the six lines appearing in the key is the average AOGCMs for one of the six illustrative scenarios. The region in dark shading shows the range of the average of AOGCMs for all 35 SRES scenarios. The region in light shading shows the range of the average of AOGCMs for all 35 scenarios. The region delimited by the outermost lines shows the range of all AOGCMs and scenarios including uncertainty in land-ice changes, permafrost changes and sediment deposition. Note that this range does not allow for uncertainty relating to ice-dynamic changes in the West Antartic ice sheet. (from IPCC Third Assessment Report, 2001)
eration, we hold the Earth in trust for future generations and therefore we should not preclude the options of future generations. Partially as a result of the IPCC assessment reports, the emission of green house gases (GHGs) has assumed added urgency and international negotiations have been taking place on how nations can share the reduction in emission of GHGs in an equitable way. By the same token, there is inequity in presentutilization of ecosystem goods and services, both between regions (e.g. developing vs. industrialized world) and within regions (wealthy vs. poor). It could be argued that these inequities do not impact the Earth system directly, but if they create conflict, they could prevent the establishment of global environmental policies that could lead to intergenerational equity.

The equity debate hinges on burden-sharing (or resource-sharing), and the focus is on "differentiation of future commitments" among various nations, based on the equity principles of responsibility, capability and need. Three different approaches of this differentiation have emerged- multi-stage, contraction and convergence, and triptych. The methods use prescribed emissions scenarios and make assumptions on economic and industrial activities of nations. No specific approach has achieved universal acceptability and negotiations on differentiation have been on shaky ground until recently.

Difficulty in reaching consensus on burden/resource sharing stems from various factors such as historical inequality in the industrial and economic development of nations, perceived restrictions on the mode of future development by certain nations, and inequality in resource availability and use. In addition, the emissions scenarios that are designed to influence the policies of all nations are developed using simple climate models rely on data inventories from a limited number of countries, and have a wide range of uncertainty. To address these issues, it will be necessary to develop a research strategy on the equity principles of global environmental management that has the following objectives:


Strategic Questions

20. What is the optimal mix of adaptation and mitigation measures to respond to global change?
Mitigation measures are directed to reducing the impact of human activities on the Earth system, while adaptation measures are directed to reducing the impacts of global change on human society. There has been a significant focus recently on adaptation side of the climate policy debate. However, some observers fear that advancing our understanding of how to cope with exposure and sensitivity to climate change and to climate variability will diminish our will to alleviate the sources of climate related stress on our social, economic, political and ecological systems. They fear, in short, that adaptation will be used as a large-scale substitute for mitigation in the policy continuum.

Some studies of mitigation work from the top down with aggregate models of economic activity. These studies explore the relative efficacy of applying macro-level policy levers to reducing the emission of greenhouse gases; they seek to describe trajectories of various policies designed to achieve specific targets (emissions caps, concentration caps, temperature caps, etc…) under a variety of assumptions about international participation. Other studies work from the bottom up with highly disaggregated models of specific economic sectors. These studies confront the same sort of issues, and but they can offer more detailed descriptions of incidence (Who would pay for how much mitigation, and when would they incur those costs?)
The literature is also beginning to fill with analyses of adaptation to the impacts of climate change and climate variability. While they are quite different, they can be assessed in terms of a unifying model of vulnerability that depends, for any system, on exposure, sensitivity, and adaptive capacity.

A few integrated studies have tried to integrate the two sides of the policy calculus. These integrated assessments apply criteria like welfare maximization or the precautionary principle to weigh the relative values of mitigation and adaptation. All work from reduced forms, though, so details are typically missing when, for example, adaptation reduces the cost of exposure to anticipated climate change or relaxes the constraint imposed by a predetermined window of "tolerance". Progress in evaluating the relative roles of adaptation and mitigation will be most rapid when it becomes clear that geographically specific factors determine mitigative capacity as well as adaptive capacity. Coping with the climate problem is not a question of mitigating and then adapting. Nor is it a question of adapting and then mitigating. It is a more holistic question of doing both at the same time, and focusing attention on the common determinants of mitigative and adaptive capacities can lead productively to understanding exactly how to meet these coincident challenges. Indeed, even a cursory look at the determinants of mitigative and adaptive capacity make it clear that the best climate policies for some nations over the foreseeable future may have nothing specific to do with climate.

21. What is the optimal decomposition of the planetary surface into nature reserves and managed areas?
Much of the Earth's land surface is managed by human society, and there are large variations in the intensity of management, from forestry, to intensive agriculture and irrigation, to urban development. In general, land management affects continental and global scale systems through alterations of biogeochemical cycles, biophysical feedbacks, and interactions with the climate system. More specifically, the spatial pattern of land use affects the exchange of water, energy, and momentum between the land surface and the atmosphere leading to alterations to regional and global climate. The size, shape, connectivity and location of nature reserves and the intensity of management in manages areas are important for the maintenance of biodiversity. In order to optimize the spatial structure of managed areas and relative proportions of nature reserves for purposes of maintaining biodiversity as well as various ecosystem goods and services, it is necessary to explore the sensitivity of the Earth system to land use regimes, spatial scales of land cover variability, and the extent to which nature preserves can be used as storehouses of biodiversity.

Model "experiments" of land cover changes have been explored, shedding some light on the implications of land use. One type of model study involves the effects of deforestation in tropical regions, through which it has been determined that the scale of patchiness of deforestation strongly controls the regional climate response. Another type of model study simulates the change in continental scale climate due to historical land cover change and compares that with the observed records of climate change, thus demonstrating the importance of anthropogenic alterations of the land surface with respect to regional climate dynamics. Biodiversity studies have been conducted and have revealed that species richness within a ecosystem is affected by landscape fragmentation.

While various case studies have been conducted, demonstrating the effects of land use on climate, a mechanistic understanding founded on general principles is still lacking. Consequently we must go beyond case studies to more systematically aim at understanding the basic principles operating in the exchanges between the land surface and climate. Analyses of strategies for biodiversity conservation will have to be extended to account for changing climatic conditions on decadal to century time scales.

Representativeness of conservation units and Indian reservations of the Amazonian biomass 'ecoregions' (from C. Nobre)

22. What are the options and caveats for technologic fixes like geoengineering and genetic modification?
Description and justification
As anthropogenic perturbations lead to changes in global environmental conditions, many have been led to consider "fixing" the problems caused by emissions and land use by the burgeoning human population. Severe or catastrophic change in Earth’s climate system could potentially be averted by intentional modification of Earth’s radiation balance, for instance. Various methods have been proposed to effect this modification including placing scattering particles (such as sulfates) or small reflective objects (such as micro-balloons) in the stratosphere or low-earth orbit, coating the ocean surface with reflective films or objects, placing sun-blocking satellites in space, or producing whiter roads, rooftops, and other land surfaces.

Optimistically, such strategies could be an emergency backup system that would be used only to avoid catastrophic global change that might occur despite our best efforts to diminish CO2 emissions. Indeed, some have proposed researching these geoengineering options as an emergency climate catastrophe response system. Pessimistically, deployment of such schemes could constitute reckless interference in the Earth system, merely compounding problems produced by reckless release of CO2 to the atmosphere. Research into such systems could diminish the pressure to reduce fossil-fuel CO2 emissions (see Question #20, above). Indeed, many of these geoengineering schemes may be cheaper than weaning ourselves from fossil fuels, and have been proposed by some as an alternative to reducing fossil-fuel emissions.

Genetic modification has also been proposed as a means of diminishing the climate impacts of human activities. For example, it is possible that more reflective forests could be engineered, or that plankton could be developed that would more effectively export carbon to the deep sea. Some fear that such organisms released into the environment could dangerously spread and multiply, irreversibly altering the Earth in unforeseen ways. Others suggest that such organisms would not compete well in a Darwinian sense, as nature has not already selected for such organisms on its own. The least controversial of these proposals is the idea that we should make roads and rooftops more reflective. Such an approach is unlikely to have significant adverse effects and could slightly diminish global climate change. However, this would only very partially offset the large areas of high-albedo desert that are being irrigated to grow low-albedo crops.

In a preliminary study of geoengineering Earth’s radiation balance, it was found that geoengineering schemes could potentially ameliorate not only global mean temperature change, but regional and seasonal temperature change as well. This is despite the fact that the radiative forcing from a change in solar flux differs both spatially and seasonally from the radiative forcing from a change in atmospheric CO2 content. This seems to be the case because feedbacks in Earth’s climate system are strong relative to the forcing factors being applied to that system. For example, a doubling of CO2 provides a radiative forcing of about 4 W m–2, whereas the presence or absence of sea-ice can affect the air-sea heat flux by 50 W m–2 or more. Hence, the character of the climate response is largely (but not exclusively) governed by feedbacks in the climate system.

For each proposed geoengineering scheme, several issues must be addressed:

The political, social, and ethical questions may be the most difficult. What would happen if several countries implemented a geoengineering scheme over the objection of other nations? Would the existence of these schemes result in additional anthropogenic greenhouse gas emissions? What is the ethical (and legal) difference between knowingly altering climate (as through fossil-fuel burning) and intentionally altering climate (as through geoengineering schemes)? How do we make cost-benefit analysis of the scheme when the value systems of various affected parties differ, and costs and benefits are not easily monetarized? How do we assess the risks associated with our uncertainties? Is it ethical to research such systems, or is geoengineering research a Frankensteinian program from the start?

23. What is the structure of an effective and efficient system of global environment & development institutions?

A striking development of the period since the close of World War II has been the rise of international and transnational institutions - or regimes as they are often called - dealing with a range of largescale environmental issues and, more broadly, problems of sustainable development. The resultant collection of regimes encompasses arrangements addressing a variety of substantive concerns, taking the form of legally binding arrangements and informal practices, and extending to arrangements involving non-state actors as well as purely interstate arrangements. All these institutional arrangements have several features in common. They are sets of rights, rules, and decision making procedures that define social practices, assign roles to the participants in these practices, and govern the interactions among occupants of these roles. The issue here centers on the extent to which regimes matter in the sense that their activities account for some significant proportion of the variance in the course of human/environment relations.

Early efforts to identify the determinants of regime effectiveness featured in-depth analyses of individual cases (e.g. the long-range transboundary air pollution regime, the ozone regime, the climate regime). More recent work in this field has turned to efforts to develop and test generalizations across larger numbers of cases. These studies seek to evaluate the relative importance of variables relating to power, interests, and ideas and to problem structure in contrast to regime attributes as determinants of effectiveness. A central challenge facing this line of inquiry arises from the fact that assessments of effectiveness treated as the dependent variable rely on causal judgments.

A consensus has arisen on the need to employ a suite of strategies at the same time as a means of addressing problems arising from the need to make causal judgments and the difficulties in constructing large universes of cases. A basic distinction separates strategies that take effectiveness as the dependent variable and seek to identify drivers of effectiveness from strategies that take some other factor (e.g. atmospheric concentrations of greenhouse gases) as the dependent variable and seek to determine the role of institutions as determinants of change in this factor. Among those focusing on effectiveness as the dependent variable, interest is growing in procedures, including qualitative comparative analysis and studies of the functional requisites of regime success, as supplements to standard statistical procedures. The limits of inductive procedures in this field have also given rise to a significant move toward efforts to illuminate the behavioral mechanisms through which regimes affect the actions of actors participating in regime-governed activities.


Upcoming Conferences of Interest

AGU Fall Meeting, Dec. 6-10, 2002, San Francisco, USA
Several sessions that pertain to Earth system science are being organized in the Biogeosciences and other sections at the Fall meeting. See the full program at http://agu.org/meetings/fm02. A few examples from the preliminary program are as follows- the final program is yet to be determined.
Carbon Sinks and Carbon Management:...
Carbon Cycle Science: The North American Carbon Program
Water, Energy, and Carbon Exchange in Forest Systems
The Large Scale Biosphere-Atmosphere Experiment in Amazonia (LBA)
Impacts of Ecosystem Restoration on C, N and P Cycling in Estuaries
Unintended Consequences of Carbon Sequestration
Geophysical Disturbance and Ecosystem Patterns
Application and Validation of Land Surface Products from the MODIS Sensor
Coupled behavior of biotic systems and Climate
Impacts of Air/Sea Exchange on Biogeochemical Processes in the Ocean

AGU-EGU(EGS/EUG) Joint meeting, April 7-11, 2003, Nice, France
This meeting represents the first joint meeting of AGU, EGS and EUG (EGS and EUG have recently joined force to become the EGU). This joint meeting replaces the regular AGU Spring meeting and the Annual EGS meeting. Abstract deadline, January 15, 2003. For more information see http://www.copernicus.org/egsagueug/index.html
With the addition of Biogeosciences to EGS, the joint meeting has special significance IGBP and the biogeosci. community in general. The B program can be viewed as it evolves at http://www.cosis.net/members/meetings/skeleton/view.php?p_id=45
A few of the preliminary Biogeosciences sessions include:
Merging marine and continental approaches to biogeochemical cycles
Atmosphere-biosphere exchanges: a comprehensive approach to sinks and sources
Role of prokaryotes and microalgae in the biogeochemical cycling within soils, marine and freshwater ecosystems
Anthropogenic perturbations to the Earth System: climate, water cycle
Data assimilation for carbon cycle studies
Methane fluxes on Earth: budgets and biological controls
Biogeophysics of land cover change, the hydrologic cycle, and climate
Emergent properties of the carbon-climate-human system: insights from modelling
Earth System Modelling


International Conference On Earth System Modelling

International Conference on Earth System Modelling, September 15-19, 2003, Hamburg, Germany
This conference is jointly sponsored by IGBP/GAIM, WCRP/WGCM, and the Max Planck Institute for Meteorology. It will address global, regional, and reduced complexity modelling, and will provide an opportunity to present new results as well as plan for the future of Earth system modelling. Contributions are invited in any of the following general areas:
A. Development and evaluation of comprehensive Earth system models
B. Variability of the coupled Earth system at different time scales
C. Anthropogenic climate change
Further information, abstract submission, and registration can be obtained at http://www.mpimet.mpg.de/mpi-conference2003 or by contacting the GAIM office.



Claussen, M., Mysak, L.A., Weaver, A.J., Crucifix, M., Fichefet, T., Loutre, M.-F., Weber, S.L., Alcamo, J., Alexeev, V.A., Berger, A., Calov, R., Ganopolski, A., Goosse, H., Lohman, G., Lunkeit, F., Mokhov I.I., Petoukhov, V., Stone, P., Wang, Zh., 2002: Earth System Models of Intermediate Complexity: Closing the Gap in the Spectrum of Climate System Models, Climate Dyn., 18, 579-586

Petit, J.R; J. Jouzel, D. Raynaud, N.I. Barkov, J.M. Barnola, I. Basile, M. Bender, J. Chappellaz, M. Davis, G. Delaygue, M. Delmotte, V.M. Kotlyakov, M. Legrand, V.Y. Lipenkov, C. Lorius, L. Pepin, C. Ritz, E. Saltzman, & M. Stievenard. 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice Core, Antartica. Nature, 399:429-436.

Rahmstorf, S.: Warum das Eiszeitklima Kapriolen schlug, Spektrum der Wissenschaft, September 2001, pp. 12-14..