Volume Four, Number One
Summer 2000

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


ARTICLES:
TITLE
AUTHOR

New GAIM on the Launchpad
John Schellnhuber
Dork Sahagian

LBA-GAIM Workshop
Dork Sahagian
Michael Keller
Carlos Nobre

TransCom 3

Kevin Gurney

EMDI

Kathy Hibbard

OCMIP-2


Biogeosciences at AGU

Dork Sahagian

CCMLP
Martin Heimann &
CCMLP Participants

GAIM Website


Open Science Conference



GAIM on the Launchpad
By John Schellnhuber and Dork Sahagian

According to the Waikiki Principles (see Research GAIM v. 3 n. 2), the New GAIM shall become a major integrating factor for Earth System Science. The GAIM Plan (1998-2002) serves as an excellent roadmap on this intellectual mission, and yet it must not be carved in stone. Worldwide research under the umbrella of IGBP as well as its sister programmes WCRP, IHDP, and DIVERSITAS is making breathtaking progress each year (or even each month). So, the body of evidence to be "integrated" is changing perpetually. As a consequence, we GAIMblers find ourselves in a permanent state of self-examination and re-orientation: Are we looking for the answers to still pertinent questions? If so, are we using the most promising approaches? Do new observations undermine our explanatory constructions?

We recently met at the Potsdam Institute (PIK) to discuss some of the progress GAIM has made to date and some of the requirements we face for further progress. In revisiting the GAIM PLAN's list of fundamental questions, we "discovered" that indeed, many of the questions were being addressed by GAIM and other activities, but that there are some important questions toward which there has been little direct progress in achieving an answer. For these more "difficult" questions, functioning Earth System models are required to provide even preliminary insights. Consequently, as we re-examine the GAIM PLAN and make adjustments each year, we face head-on the challenge of promoting the development of Earth System models in their various forms. While progress is being made in Earth System Models of Intermediate Complexity (EMICs) and the "Flying Leap" experiment is being launched (see "Research GAIM, v. 3 n.2), the specific tasks that lay ahead call for a concerted effort by dedicated scientists who have the time to spend not only on model development, but also additional model intercomparison, applications to a set of specific set of forcing scenarios, and sensitivity studies directed at determining which components and linkages play the strongest roles in the dynamics of the Earth System, and which can be safely ignored or reparameterized for the purposes of understanding the system as a whole.

Such a concerted effort is unlikely to be possible with a small Task Force of volunteers with many demands on their time placed by their "regular" jobs, as well as other volunteer work such as IPCC and other national and international committees. It is therefore necessary to consider some options that involve placement of full-time scientists in positions that will optimize their effectiveness in promoting the development of Earth System models at the range of complexity envisioned in the GAIM PLAN. This modelling venture is such an ambitious undertaking that even the coordination of existing efforts internationally requires a team of dedicated researchers. GAIM is currently considering some options to most effectively proceed in this direction. In fact, we are presently exploring the possibility of setting up a global post-doc network with multinational funding and nodes at various Global Change centres of excellence. The vision is to assemble a "gang of young savages" that will not shy away from the immense challenges posed by Earth System modelling.

At the same time, the fundamental questions of the GAIM PLAN themselves call for reconsideration. Are the questions still relevant? Have more important questions arisen recently for which near-term of long-term answers will be necessary or even feasible before further progress toward understanding the Earth System is possible? In the coming GAIM Task Force meeting (Dec, 2000), we will revisit the questions of the GAIM PLAN to ensure that we are still "on track" given new results such as the interpretations of the Vostok ice core published by Petit et al. last year as well as other important results and breakthroughs throughout the field. The GAIM PLAN will thus be revised, refined, and perhaps reformulated in GAIM's constant process of self-adjustment.

Equally important is the issue of determining who should seek the answers to the various questions in the evolving GAIM PLAN. Some questions are already being addressed by the Core Projects, and we eagerly anticipate the results. Others are being tackled by IGBP-wide initiatives such as the "IGBP Carbon Project" and the "IGBP Water Project". Others are best initially investigated by GAIM-Core Project joint activities, and then entirely taken over by the Core Projects in their implementation activities. As GAIM focuses on its integrated Earth System modelling activities, the links between the Task Force and the Core Projects will become necessarily stronger and more varied, and the ties with the SC-IGBP will naturally grow tighter.

While GAIM is rounding up adequate approaches and approachers to difficult questions, near-term results and products must not be neglected. Many of these results are "tools" such as methods for model intercomparison provided by EMDI, TransCom, and OCMIP. However, other results should be more concrete, available, and useful to the general public. One landmark product of GAIM, DIS, and indeed of all of IGBP, would be an "Atlas of the Earth System and Global Change". Such a product would be available on the web even before it is finalized and published in hard-copy so that results can be made available as they emerge. An initial Table of Contents for such a product is now being formulated by GAIM and DIS, and ultimately will require input from all IGBP/Core Projects, WCRP and IHDP for many critical aspects.

GAIM is now - and must remain - a flexible entity that can evolve in response to results brought forth by all branches of Earth System Science. As such, readjustments are a normal part of our operation, and we will continue to seek the most effective paths toward understanding the operation of the Earth System and to answering the evolving questions that are emerging from IGBP research. Top


LBA-GAIM Modelling Workshop
July 17-28, 2000
CPTEC-INPE, Cachoeira Paulista, Brazil

By Dork Sahagian, Michael Keller, and Carlos Nobre


GAIM, in conjunction with the Large-Scale Biosphere-Atmosphere Experiments in Amazonia (LBA) held a workshop at CPTEC-INPE, in Cachoeira Paulista, Brazil for the purpose of developing and integrating the modelling capacity of the LBA ecological research community.

LBA is designed to create the new knowledge needed to understand the climatological, ecological, biogeochemical, and

Instructors and students pose for a group photo in Cachoeira Paulista.
hydrological functioning of Amazonia, the impact of land use change on these functions, and the interactions between Amazonia and the Earth system. LBA is centered around two key questions being addressed through multi-disciplinary research with integrating studies in the physical, chemical, biological, and human sciences:

How does Amazonia currently function as a regional entity?

How will changes in land use and climate affect the biological, chemical and physical functions of Amazonia, including the sustainability of development in the region and the influence of Amazonia on global climate?

In contrast to other fields such as meteorology where modeling is very well established in Brazil and other countries, currently most of the ecological modeling work is being conducted outside of South America. The goals of the LBA-GAIM Modelling Workshop were thus to:

Within these overall goals, specific objectives to be met in the workshop were to:

The LBA-GAIM Modelling Workshop thus focused upon:

In the context of the LBA scientific priorities and fundamental questions, it is possible to address the role of modelling in LBA science. Modelling identifies the gaps in both understanding of ecosystems and data and as such plays a key role in the LBA program at early stages, as well as throughout the maturation of the project. Gaps in conceptual understanding can be filled by further analysis of focused scientific issues, and gaps in data can be filled by targeted field campaigns. With sufficient modelling abilities, LBA will be ready to move in both of these directions, so a workshop on modelling was both appropriate and timely.

Following the success of the African-GAIM Modelling Workshop (See Research GAIM, v. 1, #2; and GAIM Report #1), GAIM was able to organize a similar training workshop in Brazil focusing on LBA. The experience gained in Africa as well as through various activities done in
Michael Keller at the whiteboard.
conjunction with IGBP/START has provided GAIM with the wherewithal to effectively bring together a group of scientists with various, but strong, backgrounds in their respective fields, to focus on modelling for the purpose of better understanding the processes which lead to the observations obtained throughout LBA field campaigns as well as the implications regarding global change.

South American participants were selected from early to mid-career academic and governmental researchers who are in the strongest position to contribute regional data and understanding to the workshop as well as to use and disseminate the modeling capabilities obtained at the workshop to their colleagues and students. Several outstanding graduate students were also selected to paticipate, so there was a range of experience at the workshop. Following the course, the participants have established an electronic network of communications and a course web site for sharing their results -- at least for now, among themselves. This group can grow into a teaching force which will maximize the positive impact of the workshop on building modelling capacity among the Amazonian global change research community.

The Workshop involved ecosystem models at various levels applicable to Amazonia in the global context. The models were presented by the instructors, run by participants in hands-on "laboratory" sessions, and interpreted in terms of Amazonian and global applications using data contributed by all participants. Participants conducted modelling projects during the workshop, to be subsequently developed at their home institutions as part of planned collaborative follow-up activities.

The workshop was built around four identified models relevant to Amazonian ecosystems, hydrology, and climate (see box, below). These were introduced by the model developers in the mornings, and excercises were assigned for hand-on laboratories in the afternoons. The second week of the workshop was largely devoted to the development of individual and team projects based on the presented models and specific modelling needs of the participants and their involvement in LBA. This hands-on approach maximized the practical impact of the workshop. It also set the stage for follow-up modelling activities within the LBA science program.

MODEL LEAD INSTRUCTORS
Century Bill Parton
SiB (Simple Biosphere Model) Humberto Rocha
ED (Ecosystem Demography) George Hurtt
Earth System Models Marcos Costa
Workshop Organizers Dork Sahagian
Michael Keller
Carlos Nobre

The four major models were chosen because they have comlpementary approaches and emphases. Century has a very detailed treatment of soil carbon and nutrients as well as decomposition processes, while SiB treats radiative transfer and plant physiology in a more detailed manner. ED looks at the ecosystem as a whole, while incorporating many of the features necessary for tracking ecosystem function. Finally, the Earth Ssytem Models presented by Marcos Costa looked a the global system. In addition to these four major types of models, the participants were presented with Stella, a graphical box modelling software package that easily allowed them to construct their own models as well as reconstruct specific features of the four major models presented by the instructors. Many of the participants also used Stella in their own projects, which they presented to the group at the end of the workshop. A full report of the LBA-GAIM Modelling Workshop is now being prepared in the GAIM office, together with LBA and CPTEC. It will be circulated amongst the participants and other intersted parties by request to the GAIM Office.

As the second in a "series" of GAIM modelling workshops, the LBA-GAIM training workshop demonstrated the utility of modelling in the developing world, and also within specific international research projects (i.e. LBA). After Africa and Amazonia, GAIM will turn its attention to Southeast Asia for a comparable workshop to be held in 2003, in conjunction with START.

The workshop conveners are grateful to NSF (U.S.), NASA (U.S.) IAI (U.S.), and CNPq (Brazil) for funding in support of the LBA-GAIM workshop.
Top


TransCom 3 Update
By Kevin Gurney

The TransCom 3 experiment completed a very successful second meeting in Paris, France hosted by the Philippe Ciais group. The first day and a half were spent listening to talks from various groups directly or indirectly involved in TransCom. New approaches to the inversion problem, data sensitivity studies, and refinement of uncertainty estimates were some of the topics presented.

Preliminary results from the Level I (annual mean inversion) were also presented. A number of key conclusions can be drawn from this first glance at the annual mean model output. 1) The rectified neutral biosphere portion of the surface fluxes has a significant impact on the inverted fluxes. 2) The flux uncertainty generated by individual models is greater than the mean uncertainty across the participating models. 3) All the participating models appear to agree on placing more oceanic uptake in the northern oceans and less uptake in the southern oceans relative to the pre-subtracted Takahashi oceanic exchange.

These results were also presented at the American Geophysical Union meeting in Washington DC in early May.
The second half of day 2 was spent discussing the next steps in the TransCom experiment. The following is a summary of this discussion:

A) T3 Website:

B) Level II logistics:

C) Other issues:

Since the Paris meeting, Kevin Gurney and Rachel Law have taken the lead in preparing the first TransCom 3 paper for publication. This paper will focus on introducing the experiment and presenting results on a control case inversion across all the participating models. Top


EMDI Update
By Kathy Hibbard

Recent progress with the Ecosystem Model/Data Intercomparison (EMDI) activity has been primarily geared towards rectifying driver data and identification of observed Net Primary Productivity (NPP) outliers. The Potsdam Institute for Climate Study has provided the modeling groups with a corrected monthly climatology, and several of the modeling groups are collaborating to produce a common daily and sub-daily climatology (Chris Kucharik, Joerg Kaduk, Nicolas Viovy, Peter Thornton). The group at the Oak Ridge National Laboratory (Dick Olsen, Keri Johnson, Jonathon Scurlock) has been particularly active in the past few months. Data have all been reclassified into the land cover types that were agreed upon by the modeling groups at the December 1999 workshop held at the University of New Hampshire. The observed NPP datasets have also been analyzed for outliers (Table 1) and have been reclassified into three classes: Class A Sites (well documented and intensively studied); Class B (globally extensive, but less well documented and less site-specific information); and Class C, or 0.5° x 0.5° gridded datasets.

Table 1. Results from Outlier Analysis performed at Oak Ridge National Laboratory

Description A
Class
B
C
Initial number of observed NPP
162
2363
2768
Number of observed NPP after outliers removed
151
1689
2624
Percent measurements that were excluded (%)
7
29
5
Mean of observed NPP excluding outliers
81
933
1637

It was agreed at the workshop that because of problems with the driver data and mismatched land use classifications, results from individual models would be anonymous and an "ensemble model" or the average of all model results would be analyzed for publication. Some preliminary results suggest that the models appear to systematically overestimate observed NPP (Figs 1,2).

Several EMDI members were able to gather at the Ecological Society of America (ESA) Meeting held at Snowbird, Utah in August 2000. It was agreed that the current draft paper describing the input driver datasets and results from the first workshop would be a priority activity in the coming weeks. It was suggested that a follow-on workshop be held in April or May, 2001 in either Durham, New Hampshire or Madison, Wisconsin. It was also agreed that some of the time during the next workshop would be focussed upon analysis and papers from EMDI 1 and that the modelers weren't ready to do full transient runs and comparisons. It was suggested, however, to include the transient datasets (e.g. Fluxtowers) to ensure there would be some interannual comparisons.

At this time, we would like to encourage any additional modeling groups that would like to participate in the next EMDI comparison to contact Kathy Hibbard at: k.hibbard@unh.edu. The GAIM website will post EMDI highlights at:http://gaim.unh.edu/Structure/Intercomparison/EMDI/

Figure 1. Simulated and observed Net Primary Productivity across 11 models for all Class A sites (N-87). Insert shows locations of Class A Sites (intensively studied and well documented.
Figure 2. Ensemble modelled versus observed total Net Primary Productivity by biome for Class B sites (N=933).
Top


During July 2000, the OCMIP-2 Group held a meeting at Princeton, NJ, USA to discuss research results as well as plans for the future. An update of OCMIP-2 will be provided in the next issue of Research GAIM. More information about OCMIP can be seen at: http://www.ipsl.jussieu.fr/OCMIP/.


Biogeosciences at AGU
By Dork Sahagian

The American Geophysical Union (AGU) has now officially created a new section on Biogeosciences. This is the first new AGU section to be created in decades, and highlights the importance of emerging research results in the interaction between the geosphere and the biosphere. It also provides a valuable forum and outlet for all aspects of global change research, with a particular focus on the biologic aspects of concern to IGBP.

The initial forays of AGU into Biogeosciences were met with enthusiastic response and a "flood" of papers for presentation at the previous two AGU meetings. Now that Biogeosciences is an official section, we expect an even stronger showing from the global change research community at the coming Fall 2000 meeting (Abstract deadline, Sept. 7, 2000). Abstracts can be submitted on any topic related to biology and its interaction with any or all of the Earth System. There are already 19 sessions identified for the Fall Meeting, and more will be created as abstracts come in. Some sessions of interest are listed in the adjacent text box. Your abstract does NOT need to fit into any of the pre-identified sessions- new sessions will be created to fit all abstracts as they arrive. Abstract submission can be done on the web at http://www.agu.org. If you need an AGU sponsor number, contact the GAIM office (gaim@unh.edu).

Submission of Abstracts
Authors are encouraged to submit abstracts electronically via the Interactive Web Form on the AGU Web site. Abstracts submitted by postal/express mail must be received at AGU Headquarters by September 1, 2000. Abstracts submitted by the Interactive Web Form must be received at AGU by September 7, 2000. (These deadlines are firm and no exceptions will be granted.)

Abstract submission instructions are available at http://www.agu.org

Important Dates
Abstract Submission Deadlines:
September 1, 2000 (Postal/Express Mail)
September 7, 2000 (Interactive Web Form)
2000 Fall Meeting: Dec 15-19 (Fri-Tues)

For More Information, contact:
AGU Meetings Department
2000 Florida Avenue, NW
Washington, DC 20009 USA
Tel: +1-800-966-2481 or +1-202-462-6900
Fax: +1-202-328-0566
E-mail: meetinginfo@agu.org (subject: 1999 Fall Meeting)
AGU Web site: http://www.agu.org

A few "Biogeosciences" and related Special Sessions already formulated for the AGU Meeting, listed below. A full list of session descriptions is available at the AGU website.

Preliminary Biogeosciences Session List

A02 Evolution of the Atmospheric Methane Budget
B01 Remote Detection and Survey in Astrobiology and the Study of Life in Extreme Environments
B02 Microbial Processes: I. Microbial-Mineral Interactions in Deep Subsurface Environments
B03 Microbial Processes, II, Constraints on Microbial Survival in Geological Environments
B04 Halocarbons: Global Biogeochemistry and Contaminant Transformations
B05 The Influence of Hydrosphere-Biosphere Interactions on the Speciation and Transport of Metals in Surface Waters
B06 Nitrogen-Cycling Processes in Rivers and Streams
B07 Life in a Turbulent Environment: Primary Production in the Surface Mixed Layer
B08 New Breakthroughs in Field-Scale Bacterial Transport
B09 The Role of Fire in the Boreal Forest and Its Impacts on Climatic Processes
B10 Land-Atmosphere Interactions, I, Arctic Transitions
B11 Land-Atmosphere Interactions, II, North Africa
B12 Responses of Soil Processes to Elevated Atmospheric CO2
B13 Ecological Component of Large Scale Biosphere-Atmosphere Experiment (LBA)
B14 Land Use Change, I, A Paleoenvironmental Perspective on Ecosystem Sustainability
B15 Land Use Change, II, Global Land Use and Land Cover Change Over the Last 300 Years
B16 Biospheric Results from Terra, NASA’s Earth Observing System
B17 The Science of Carbon Sequestration
B18 Biogeosciences: Expanding Horizons in Understanding Earth and Planetary Systems (Biogeosciences Overview session)
B19 Iron Cycling in the Natural Environment: Biogeochemistry, Microbial Diversity, and Bioremediation
U01 Paleoenvironmental Evidence for Prehistoric Natural Hazards and Their Impact on Human Societies
U04 Quantifying Predictability in Geophysical Systems I
U07 Global Climate Change: Current Challenges in Research and Scientific Leadership
U08 Coping With Uncertainties in Climate Change Research and Policy Activities
U09 Global Change and the Nature of the Earth System (IGBP Synthesis Session)
U10 Land-Atmosphere Coupling in the Global Climate System
U11 The Changing Level of the Sea: Paleogeodesy, Space Geodesy, and Global Change

Top


The Carbon Cycle Model Linkage Project (CCMLP)
By Martin Heimann and CCMLP Participants

Overview
The global carbon cycle constitutes an important constituent of the earth system. In order to understand and predict its behavior as a function of direct anthropogenic impacts by the emissions of CO2 from fossil fuel burning and from changes in land use, but also as an interactive component of the physical climate system necessitates the development of realistic, comprehensive simulation models of the global carbon cycle. Already since the early 1960’s have simple, conceptual carbon cycle box-models been constructed, however, the establishment of process-based simulation models of the oceanic and terrestrial carbon cycle components has not begun before the late 1980’s and early 1990’s. This is substantially later than the corresponding development of comprehensive three-dimensional models of the physical climate system components. In part, this delay may be attributed to the relatively poorly known biochemical and ecological processes, which control the exchanges of carbon between inorganic and organic forms on land and in the oceans. The wealth of process studies conducted during the last decades, in part also by activities within the core projects of the IGBP, have now allowed the modelers to take up the challenge to build global process-based carbon cycle models, which are currently being coupled to climate models.
Any model development necessitates a careful evaluation of the model performance. In order to rigorously evaluate the performances of comprehensive global carbon cycle models the IGBP-GAIM task force, as part of its activity "The Coupled Atmosphere-Land-Ocean Carbon System 1980-2000", helped to initiate of the "Carbon Cycle Model Linkage Project" (CCMLP) in the early 1990’s. Since its beginning, substantial funding support for CCMLP has been provided by the U.S. Electric Power Research Institute (EPRI).
During phase I of the project, CCMLP focused particularly on a rigorous evaluation of recently developed models of the terrestrial carbon cycle (terrestrial biochemical models, TBMs). A significant effort went into the establishment of a series of model simulation tests, which permit an integrative and quantitative assessment of the model’s performance on global and regional scales. As an example, by coupling TBMs with oceanic components (prescribed model derived sea-air CO2 fluxes) and an atmospheric transport model, the simulated temporal and spatial atmospheric CO2 concentration patterns may be compared to the observations from the global monitoring networks, e.g. of the Climate Monitoring and Diagnostics Laboratory of the U.S. National Oceanic and Atmospheric Administration, (NOAA-CMDL) [Convay et al., 1994].
CCMLP is not a model intercomparison project. It was intended as a pilot study to explore model predictions in as many ways as possible, which may be compared to integral observations of the carbon cycle (e.g. measurements of atmospheric CO2 concentration, isotopes and other coupled tracers, data compilations from statistics, results from deconvolution studies etc.). Some of the experiments conducted within CCMLP, however, have now found their way into other, specific model intercomparison studies (e.g. the Potsdam NPP model intercomparison study [Cramer et al., 1999]).
CCMLP included a series of additional studies beyond the evaluation of model simulations. These included the development of tools to represent the global behavior of complex TBMs by means of simple pulse substitute models [Joos et al., 1996], the investigation of the effects of biomass burning on the atmospheric concentration variations of CO2 [Wittenberg et al., 1998] and the use of radiocarbon in the evaluation of global TBMs [Meier et al., in preparation]. In the following, a few results from two of the studies conducted within CCMLP are briefly presented.

Seasonal Cycle Study
Figure 1. Seasonal cycle of atmospheric CO2 predicted by five TBMs at Mauna Loa, Hawaii (upper panel) and Ragged Point, Barbados (lower panel) [from Heimann et al., 1998]. The observations (monthly mean and one standard deviation) are from Conway et al., 1994.
As an example of a coupled model evaluation experiment, Figure 1 [Heimann et al., 1997], shows the modeled and observed seasonal cycle of atmospheric CO2 at two monitoring stations: Mauna Loa, Hawaii and Ragged Point, Barbados. The black symbols denote monthly mean observations and standard deviation [Conway et al., 1994] while the colored curves indicate the seasonal cycle predicted by model configurations in which different TBMs were included (indicated by the acronyms – for a description of the models see Heimann et al., 1998).
Obviously, there is significant scope for improvement. Some of the models seriously mismatch amplitude and phase of the observed seasonal cycle. The panel of the Ragged Point station also displays the results of a sensitivity study: The red dotted line shows the prediction of a variant of the TEM model, in which the standard rooting depth has been significantly increased. With this modification the subtropical vegetation is less susceptible to water stress during the dry season, leading to a less pronounced seasonality of CO2 exchanges and thus reduced atmospheric seasonal CO2 amplitude in subtropics, which is more compatible with the observations.
Clearly, a single evaluation such as the seasonal cycle test cannot comprehensively validate the correctness of a complex TBM. Furthermore, the predicted atmospheric CO2 signals do not only depend on the TBM, but to some limited extent also on the specified air-sea fluxes and the atmospheric transport model. Nevertheless the test allows a quantitative assessment of the modeled net CO2 surface flux patterns of the Northern Hemisphere, where the terrestrial seasonal signal is dominant.

The "Grand Slam" Experiments
Building on the expertise from early part of CCMLP, at the end of phase I a series of so-called "Grand Slam" Experiments were conducted [McGuire et al., submitted to GBC]. In these experiments, four terrestrial biochemical models were subjected to the observed historical perturbations of three factors over the last 150 years: (i) the rising atmospheric CO2 concentration as determined from ice-core measurements, (ii) monthly spatio-temporal variations of temperature [Jones, 1994] and precipitation [Hulme, 1994] and (iii) historical changes in land use by deforestation of natural vegetation and abandonment of crop lands [Ramankutty and Foley, 1999].
These experiments built upon the results obtained in previous CCMLP studies on the effects of the rising CO2 concentration [Kicklighter et al., 1999] and on the interannual, climate driven
Figure 2. History of net land to atmosphere flux of carbon as simulated by four terrestrial ecosystem models of CCMLP driven by observed changes in atmospheric CO2, climate and land use [from McGuire et al., submitted to GBC]. The light-gray shaded region denotes the range of estimates from a global atmospheric CO2 deconvolution study [Bruno and Joos, 1997]. The dark gray areas denote estimates from atmospheric O2/N2 measurements for the 1980’s and 1990’s [updated from Keeling et al., 1996, and Langenfelds et al., 1999].
variability in TBMs [Heimann et al., in preparation]. The inclusion of the crucial effects caused by land use change in the simulation allowed a much more comprehensive evaluation against contemporary global carbon cycle data.
The "Grand Slam" experiments were performed by two TBMs with prescribed vegetation: the High Resolution Biosphere Model [HRBM, Esser et al., 1994], and the Terrestrial Ecosystem Model [TEM, Tian et al., 1999], and two dynamical global vegetation models: the Integrated Biosphere Simulator [IBIS, Foley et al., 1996] and the Lund-Potsdam-Jena Dynamic Global Vegetation Model [LPJ, Sitch, 2000].
Figure 2 shows the temporal evolution of the modeled terrestrial carbon budget over the simulation period. Because the models were initialized with inadequate climate data for the time period prior to 1900, only the results after 1920 can be regarded as significant. In Figure 2 the model predicted net land to air fluxes of CO2 are compared against independent estimates: from a CO2 deconvolution study [Bruno and Joos, 1997] and, over the 1980’s and the early 1990’s, from O2/N2 measurements [updated from R. F. Keeling et al., 1996, and Langenfelds et al., 1999]. Overall, it is seen that during the last decades the TBM simulations are broadly consistent with the independent estimates. Nevertheless, there are still large differences among the models, even when driven with the same forcing data
Figure 3 shows global maps of the simulated net CO2 fluxes during the 1980’s (1980-89). The 4 models of the study predict substantial uptake of carbon (green colors) in the tropics and temperate latitudes, which are contrasted by deforestation hotspots, in particular in the Sahel region in Africa, representing a net source of CO2 to the atmosphere (brown-red colors).
The experiment protocol included also a simulation run without changes in land-use. Hence, the effects of the latter may be assessed by difference of the model results. (It was independently verified that the effects of the three forcing factors on the predicted fluxes were almost additive.) The effects
Figure 3. Global map of predicted net land-air CO2 flux averaged over 1980-89, simulated by four terrestrial ecosystem models of CCMLP driven by observed changes in atmospheric CO2, climate and land use [from McGuire et al., submitted to GBC].
of land use changes during the 1980’s are shown in Figure 4 and as global time series in Figure 5. All the models predict substantial areas with vegetation regrowth (Figure 4, green colors), but the modeled geographical patterns are not very robust except, perhaps, in Europe, the eastern United States, parts of India and China. Overall, however, the effects of changes in land use lead to a net source of CO2 into the atmosphere. The global history of the emissions due to changes in land use (Figure 5) are compared against the independent, canonical land use CO2 flux estimates of Houghton [1999]. After the 1970’s all the TBMs significantly deviate from Houghton’s estimates. It is believed that this difference is due to the fact that the land use patterns of the present study only included conversions from natural vegetation to agriculture and vice versa, but did not include any conversion to pasture land.
By considering each perturbation factor alone, the contribution of each factor to the net terrestrial carbon budget can be inferred. Table 1 lists the resulting breakdown of the global terrestrial carbon budget as simulated by the different models, averaged over 1980-89. According to these simulations, the CO2 fertilization effect dominates the response, leading to a modest carbon uptake over
Figure 4. Global maps of net CO2 flux induced by changes in land use, averaged over 1980-89, predicted by four terrestrial ecosystem models of CCMLP [from McGuire et al., submitted to GBC].
the 1980’s. The differences between the models, however, are substantial. It is interesting to note, that the model with the smallest CO2 senstitivity, TEM, was the only one with an explicitly modeled nitrogen cycle. This results demonstrates the role of nitrogen limitation [Kicklighter et al., 1999].

Table 1. Simulated global terrestrial carbon budget averaged over 1980-89 and breakdown of balance into the three perturbation factors: impact of CO2 fertilization, climate variations and changes in land use. Units: 1012kgCa-1, negative values: carbon uptake. [McGuire et al., submitted to GBC].

TBM HRBM IBIS LPJ TEM
CO2 -1.6 -3.1 -2.1 -0.9
Climate 0.0 0.8 0.9 -0.2
Land Use 1.0 0.8 0.9 0.6
Total -0.6 -1.5 -0.3 -0.5


An intriguing feature of the atmospheric CO2 record of Mauna Loa is the observed increase of the amplitude of the seasonal cycle over the last 40 years [C. D. Keeling et al., 1996] by more than 25%. This increase, presumably caused by changes in the "breathing" of the Northern Hemisphere terrestrial biosphere, is also predicted by the models that conducted the "Grand Slam" experiments. In Figure 6 the relative changes in the seasonal amplitude as predicted by the four models are shown against the observations (black line with square symbols). Two of the models are able to predict rather convincingly the observed increase including the broad features of its interannual variability. This result demonstrates again the power of the atmospheric observations, a central aspect of the model evaluations explored within CCMLP.

Figure 5. History of globally integrated land-atmosphere CO2 flux induced by changes in land use [from McGuire et al., submitted to GBC]. Also shown for comparison are the estimates from Houghton [1999].
Figure 6. Relative change of the amplitude of the seasonal cycle of atmospheric CO2 at the Mauna Loa, Hawaii, monitoring station. The observations (from Keeling et al., [1996]) are indicated by the black squares. The CCMLP model predictions are indicated by the colored lines [from McGuire et al., submitted to GBC].

The Future of CCMLP – Phase II
After the first phase the project has been reorganized with a change in objectives and project participants. The overall objective of phase II is the development and evaluation of carbon cycle modules (land and sea) as integral components of the Earth System. This implies a close collaboration with climate model developers and a major effort directed at coupling the carbon cycle components to climate models.
At present, CCMLP II is structured among four main tasks:
The contemporary comprehensive carbon cycle. This task includes an in-depth evaluation of the terrestrial ecosystem models. This includes a revisit of the experiments conducted during CCMLP phase I but extended with a series of site specific model evaluations against local process information such as the observations from eddy covariance flux towers, FACE studies, Nitrogen fertilization studies, and regional assessments against hydrological data and forest inventories. In addition, a comprehensive model study of the cycles of the carbon and oxygen isotopes will be conducted.
Human impacts: Land-use, land cover and Nitrogen deposition. This task addresses a more comprehensive compilation and modeling of the impacts from cropland and pasture creation and abandonment, forest harvest and regrowth and Nitrogen deposition.
"Grand Slam" experiments revisited. With the new and improved models and driving data sets the "Grand Slam" experiments will be revisited.
"Great Leap" experiments and evaluation. Ultimately, the model components as developed and evaluated within CCMLP phase II will be coupled to climate models for participation in the new GAIM activity of simulation experiments with fully coupled carbon-climate earth system models (the "Flying Leap", Fung et al., 2000).

Acknowledgement
The CCMLP participants would like to acknowledge Larry Williams, project manager at EPRI, and his predecessor, Louis Pitelka, for their long term continuing support of the project.

References

Bruno M. and Joos F. (1997) Terrestrial Carbon Storage During the Past 200 Years - a Monte Carlo Analysis of CO2 Data From Ice Core and Atmospheric Measurements. Global Biogeochemical Cycles 11(1), 111-124.

Conway T. J., Tans P. P., Waterman L. S., and Thoning K. W. (1994) Evidence For Interannual Variability of the Carbon Cycle From the National Oceanic and Atmospheric Administration Climate Monitoring and Diagnostics Laboratory Global Air Sampling Network. Journal of Geophysical Research-Atmospheres 99(D11), 22831-22855.

Cramer W., Kicklighter D. W., Bondeau A., Moore B., Churkina C., Nemry B., Ruimy A., and Schloss A. L. (1999) Comparing global models of terrestrial net primary productivity (NPP): overview and key results. Item Corporate Author Participants Potsdam NPP Model Intercompariso 5(Suppl 1), 1-15.

Esser, G., J. Hoffstadt, F. Mack, and U. Wittenberg, High-resolution biosphere model (HRBM) - Documentation model version 3.00.00. Mitteilungen aus dem Institut fur Pflanzenokolgie der Justus-Liebig-Universitat Giessen, Vol. 2 (ed. G. Esser), Giessen, Germany, 70 pp., 1994.
Foley, J. A., I.C. Prentice, N. Ramankutty, S. Levis, D. Pollard, S. Sitch, and A. Haxeltine, An integrated biosphere model of land surface processes, terrestrial carbon balance, and vegetation dynamics, Global Biogeochemical Cycles, 10, 603-628, 1996.

Fung, I.Y., P. Rayner, and P. Friedlingstein, (2000) Full-Form Earth System Models: Coupled Carbon-Climate Interaction Experiment (the "Flying Leap"), IGBP Global Change Newsletter, No. 41, 7-8.

Heimann M., Esser G., Haxeltine A., Kaduk J., Kicklighter D. W., Knorr W., Kohlmaier G. H., McGuire A. D., Melillo J., Moore B., Otto R. D., Prentice I. C., Sauf W., Schloss A., Sitch S., Wittenberg U., and Wurth G. (1998) Evaluation of terrestrial Carbon Cycle models through simulations of the seasonal cycle of atmospheric CO2: First results of a model intercomparison study. Global Biogeochemical Cycles 12(1), 1-24.

Houghton, R. A., The annual net flux of carbon to the atmosphere from changes in land use 1850-1990, Tellus, 51B, 298-313, 1999

Hulme, M., Validation of Large-Scale Precipitation Fields in General Circulation Models, in Global Precipitation and Climate Change, edited by M. Desbois and F. Desalmand, pp. 387-406, NATO ASI Series, Springer-Verlag, Berlin, 1994.

Jones P. D. (1994) Hemispheric Surface Air Temperature Variations: A Reanalysis and an Update to 1993. J. Clim. 7, 1794-1802.

Joos F., Bruno M., Fink R., Siegenthaler U., Stocker T. F., and LeQuéré C. (1996) An Efficient and Accurate Representation of Complex Oceanic and Biospheric Models of Anthropogenic Carbon Uptake. Tellus Series B-Chemical & Physical Meteorology 48(3), 397-417.

Keeling C. D., Chin J. F. S., and Whorf T. P. (1996) Increased Activity of Northern Vegetation Inferred from Atmospheric CO2 Measurements. Nature 382(6587), 146-149.
Keeling R. F., Piper S. C., and Heimann M. (1996) Global and Hemispheric CO2 Sinks Deduced from Changes in Atmospheric O2 Concentration. Nature 381(6579), 218-221.

Kicklighter D. W., Bruno M., Donges S., Esser G., Heimann M., Helfrich J., Ift F., Joos F., Kaduk J., Kohlmaier G. H., McGuire A. D., Melillo J. M., Meyer R., Moore B., Nadler A., Prentice I. C., Sauf W., Schloss A. L., Sitch S., Wittenberg U., and Wurth G. (1999) A first-order analysis of the potential role of CO2 fertilization to affect the global carbon budget: a comparison of four terrestrial biosphere models. Tellus Series B-Chemical & Physical Meteorology 51(2), 343-366.

Langenfelds R. L., Francey R. J., Steele L. P., Battle M., Keeling R. F., and Budd W. F. (1999) Partitioning of the global fossil CO2 sink using a 19-year trend in atmospheric O2. Geophysical Research Letters 26(13), 1897-1900.

Ramankutty N. and Foley J. A. (1999) Estimating historical changes in global land cover: Croplands from 1700 to 1992. Global Biogeochemical Cycles 13(4), 997-1027.

Sitch, S., (2000) The role of vegetation dynamics in the control of atmospheric CO2 content. Ph. D. thesis, Lund University, Sweden, 213pp.

Tian, H., J. M. Melillo, D. W. Kicklighter, A. D. McGuire, and J. Helfrich, The sensitivity of terrestrial carbon storage to historical climate variability and atmospheric CO2 in the United States, Tellus, 51B, 414-452, 1999b.

Wittenberg U., Heimann M., Esser G., McGuire A. D., and Sauf W. (1998) On the Influence of Biomass Burning on the Seasonal CO2 Signal as Observed at Monitoring Stations. Global Biogeochemical Cycles 12(3), 531-544.

List of CCMLP Participants (1phase I, 2phase II):
1,2 Max-Planck-Institute for Biogeochemistry, Jena, Germany:
M. Heimann, C. Prentice, D. Schimel
1 University of Frankfurt, Germany:
G. Kohlmaier, G. Würth
1 University of Giessen, Germany:
G. Esser, U. Wittenberg, T. Reichenau
2 University of Wisconsin, Madison, U.S.A.:
J. Foley, N. Ramankutti
2 Stanford University, Stanford, U.S.A.:
C. Field
1,2 Marine Biological Laboratory, Woods Hole,
J. Melillo, D. Kicklighter, H. Tian
2 University of Alaska, Fairbanks, U.S.A.:
A. McGuire, R. Dargaville, R. A. Meier
1,2 University of New Hampshire, Durham, U.S.A.:
B. Moore III, A. Schloss
1,2 University of Bern, Switzerland:
F. Joos, R. Meier
2 Laboratoire des Sciences du Climat et de l’environnement, Saclay, France,
P. Ciais, P. Friedlingstein
2 Potsdam Institute for Climate Impact Research, Potsdam, Germany:
W. Cramer, S. Sitch
1,2 Electric Power Research Institute, Palo Alto, Ca. U.S.A.:
L. Williams, L. Pitelka

Top


A New Look for the GAIM Website!
The GAIM website was recently updated in an attempt to imporve both its structure and appearance. We've added what we thing is a more "user-friendly" navigation bar along with more information that we hope you find useful. The most notable changes include a "What's New" page, along with the "Structure" and "Products" pages which should lead you more directly to the information you are seeking. A working search tool has also been added. Although many changes have been made, we do realize that there is still much that can be done to improve our site. Please take some time to check out our new look, and feel free to comment on the appearance and content. Comments should be sent to the webmaster, Jennifer Boles, at: jennifer.boles@unh.edu. Any feedback we receive would be greatly appreciated. Please visit the renewed GAIM website at http://gaim.unh.edu.




Home