Annual and Decadal Variability of the Carbon Cycle and Detection and Attribution of Carbon and Associated Biogeochemical Trends
Summary of Carbon Meeting
Chaired by Jorge L. Sarmiento
This document summarizes the results of a 2 ½ day workshop that was organized by the Cooperative Institute on Climate Science (CICS) on the topic:
The meeting was held on 1 to 3 June, 2005, at Princeton University. The purpose of the meeting was to review progress in our understanding of the global carbon cycle, and to discuss future research priorities, with the emphasis on variability and detection and attribution of trends. Attached is a list of attendees and the final agenda of the meeting. The primary organizers were Richard Feely, Jorge Sarmiento and Pieter Tans.
The meeting began with a set of science highlight presentations on the first morning, followed in the early afternoon by a discussion of NOAA programmatics led by Dave Hofmann. The rest of the first day and all of the second day were dedicated to a set of presentations on inverse modeling, atmospheric observations, ocean observation, and prognostic modeling, and included plenty of time set aside for discussion. The purpose of these talks was to expose the group to the most recent observational, inverse modeling, and forward modeling results, and to where the most important gaps in our understanding and opportunities for scientific breakthroughs lie. In addition to the more senior investigators and program managers present, the audience included a large number of graduate students and post-docs who later reported that they found the presentations to be highly instructive and that they appreciated the opportunity to get an overview of where the field is at present.
At the end of the presentations on the second day, and continuing through the morning of the third day, the focus of the meeting turned to a broad ranging discussion of the major research issues, data and modeling gaps, and what is needed for a better understanding of the carbon cycle and how we can take advantage of our joint capabilities within NOAA and associated groups to push forward on resolving them. The discussion began with a review of milestones that NOAA must achieve within the next year or so. These include:
- An annual update on the magnitude of the US carbon sink, the first of which is due in September, 2005. NOAA scientists have been asked to develop some metric that provides a measure of the uncertainty in the carbon sink. The carbon sink is reported to be about ±0.6 Pg C/yr at present and it is the goal that this will be improved by ~50% during the lifetime of the NOAA Carbon Cycle Research Program.
- The decadal uptake of carbon by the North Atlantic and North Pacific, with the uncertainty in this reduced by ~50% during the lifetime of the NOAA Carbon Cycle Research Program.
- Analysis of the interannual variability in the North Atlantic and North Pacific.
- An assessment of the carbon component of the global ocean observing system with a proposed metric for the success and improvement of the measurements achievable by the year 2012.
- An earth system model ready to run scenarios during FY06 (October ’05 to September ’06).
The discussion of NOAA milestones was followed by a more specific discussion of North American flux maps and what will be needed in order to improve our ability to develop such maps. Among the items discussed were the following:
- Plans by Wouter to produce the first NOAA North American flux map by September 2005 are on track.
- A discussion of the importance of improved information on ignition frequency and fires needed to explain interannual variability.
- The need for fine resolution data sets on fossil fuel emissions, with a promise by Gloor and Crevoisier to produce a virtual data set using information such as from night time lighting.
- A discussion of plans for an effort led by Pacala and others to develop a new inverse modeling approach aimed at improving land biosphere models.
- Identification of major scientific issues that hinder the development of land models including an improved understanding of tree mortality, the water-CO2 feedback and its link to evapotranspiration, and the frequency of ignition already mentioned above.
There was also an interesting discussion on the human dimensions problem, including the importance of assessing vulnerabilities, the tie between fossil fuel emissions and air quality, and the importance of linking our large-scale research interests to local issues.
The discussion turned next to air-sea flux maps. The overriding concern of the observationalists was the need for a closer interaction between them and the ocean modeling community. It was felt that ocean modeling needed to be higher on the priority list as a tool for interpretation of observations through inverse modeling and improving our understanding of processes through forward modeling. A particular need is for an ocean model with high resolution forced by realistic surface boundary conditions. The major foci of the NOAA observational program are:
- Fluxes on a decadal time scale based primarily on repeat hydrographic sections. The first set of such measurements obtained as part of the CLIVAR program show a considerably greater spatial variability in the decadal uptake than had been expected. Decisions need to be made about the appropriate way to use such observations in order to do global extrapolation.
- Seasonal and interannual variability based on ships of opportunity and buoys and including the use of satellite observations for extrapolation. Model simulations, including assimilation models, are needed in order to understand how anomalies develop and are propagated. Present models disagree greatly with each other and appear to underestimate the observed level of variability.
- Improving our understanding of gas exchange, with a particular focus on the Southern Ocean. This is presently being tackled by a mix of in situ observations (Wanninkhof et al.) as well as large scale inverse modeling of radiocarbon observations (Sweeney et al.).
- Coastal carbon. This program is presently in the initial stages of a basic research mode focused on understanding how carbon functions in these regions.
The next topic discussed was global flux maps of both the ocean and land. Items covered included:
- The value of coupling ocean flux maps with the corresponding atmospheric CO2 observations as in the joint inverse.
- The need for a clear strategy to improve our ability to better predict the future trajectory of CO2 in the atmosphere including a better understanding of the putative large CO2 land fertilization sink now being called more and more into question, and the need to better understand how changes in ocean circulation and biology will affect the future ocean carbon sink. Strategies for dealing with this include:
(a) Improving our understanding of the fundamental mechanisms
(b) The need to pin down the tropical carbon budget, including a search for additional constraints on the Jacobson et al. joint inverse model finding of no CO2 fertilization sink in the tropics.
(c) A closer examination of the distribution of the ocean carbon sink between the Southern Ocean and southern hemisphere temperate ocean, which is critical to the result giving a low CO2 fertilization sink in the tropics.
(d) The need to do carbon inventories in the Amazon, particularly given that the flux tower and other methods have not resolved the issue of how large the carbon sink is in tropical rain forests.
The final topic discussed was the remarkable recent findings regarding temporal variability of atmospheric greenhouse gases during the Holocene, with evidence that there may have been considerable anthropogenic impacts for thousands of years prior to the industrial revolution. Jim White’s presentation on his carbon-13 measurements in CH4 during this period excited considerable interest. White commented during the discussion that the paleo-community is already looking at this issue and that it would be good to get the carbon community together with them. Some of the scientific issues/challenges that came up during the discussion included:
- The overriding issue is to determine whether or not the response of the carbon system to climate change is capable of non-linear abrupt change that could have a significant impact on the greenhouse gas budgets. There is clear evidence of such changes during the ice ages, but during the Holocene, the challenge is to understand how the signals could have been so small despite the evidence for significant climate change.
- A way of focusing research on the above issue would be to ask what it would take in order to reproduce the observed CH4 signal and other observations from this time such as charcoal in soils, pollen maps, etc. It was felt that taking on this challenge would likely require a good size group of observationalists and modelers.
In conclusion, the greatest value of this meeting was that it brought together a wide range of observationalists and modelers working in the ocean and atmosphere, as well as on land; and on all time scales from seasonal to glacial/interglacial. Correspondingly, the most urgent future needs that were identified by the discussions were for a closer coordination of efforts between the scientists working in these different areas. Specific needs include:
- Scheduling regular annual meetings to bring together all the scientists working on the carbon cycle within NOAA and in associated research programs such as in the cooperative institutes, and the need for a carbon steering team to maintain continuity between the annual meetings.
- Development of a closer coordination between those working from the ocean side to estimate air-sea fluxes and those working with atmospheric observations to produce CO2 flux maps over both the ocean and land. Results shown at the meeting give dramatically different land flux distributions when oceanic constraints are used along with atmospheric constraints.
- Development of high-resolution ocean carbon models and data analysis methods to assist in improving ocean sampling design and to aid in the interpretation of the sparse oceanic observations.
- Closer coordination between those working with terrestrial models and observations, and the atmospheric observationalists, in order to improve estimates of land-atmosphere fluxes; and to improve our understanding of critical terrestrial processes such as tree mortality, fire ignition frequency, and the water-CO2 feedback.
- Initiate research on the Holocene and ice age carbon cycle in order to determine the causes of abrupt carbon cycle changes during the ice ages, and to determine why the atmospheric CO2 and other carbon cycle parameters have been so remarkably constant during the Holocene.
J. L. Sarmiento
4 July, 2005