5.1 Background Information for Terrestrial Carbon Cycle Observation
Dr Cihlar explained the background for this topic with reference to Prof. Bretherton’s and Dr Thomas’ reports on IGOS-P (see Section 2.). Following IGOS-P meeting in June, 1999 in Rome, TOPC was requested to lead the definition of the terrestrial carbon observation theme on behalf of GTOS and with the agreement and support of GCOS (refer to Appendix 3 for terms of reference). Dr Cihlar described the approach to this task, which was intended to produce useful initial results within a short period of time. This meeting was the first opportunity to consider this subject. Thus, preliminary inputs were solicited by e-mail, and special presentations prepared as a background for discussion.
Dr Steve Running discussed issues to be considered in designing a global observing system for terrestrial carbon.
Optimum terrestrial carbon cycle monitoring would entail sites distributed across the full climatic range of vegetation, roughly a water balance range from - 3,000 to + 3,000mm/yr and annual temperature ranges from - 20 to + 30°C. In addition, all biome types, including those with perennial vegetation and a chronosequence, should be sampled within their bioclimatic subspace (Churkina and Running, 1997).
The current 80 Fluxnet sites are concentrated mostly in temperate forests, with both cold and warm low precipitation zones being particularly underrepresented. The optimum number of Fluxnet sites is probably on the order of 500, but currently there are an insufficient number of trained scientists and technicians to operate that many.
Vegetation structural sampling for LAI and NPP is not as technologically difficult as NEP from a flux tower, and need not be done continuously. Rather, it is a GHOST Tier 4 activity where sample plots can be located, measured and vacated. Again, the complete array of bioclimates, biome types and major crops need to be sampled, with a temporal frequency ranging from annual in forests to monthly in croplands. Because of the simpler and less frequent sampling requirement, a much larger number of samples should be possible than those available with flux towers. With a global vegetated area of about 1.5 x 108 km2, 1500 sites would yield a sampling density of one site for each area of approximately 300 x 300 km. The representativeness of those sample sites to the larger region to which they are to correspond, is a key question that the satellite mapping of NPP should be able to evaluate. Technical training of workers is less rigorous for vegetation structural data, although uniform methodologies are essential if measurement error is to be held below 20 percent (BigFoot web site).
The extreme microheterogeneity of soil carbon makes a global monitoring design seem unrealistic. However, in the context of the vegetation sampling suggested above, it would be highly efficient to take limited soil carbon samples. It is suggested that a measure of fast (<5 years) and medium (five to 50 years) turnover time litter/soil carbon be taken, and if possible, the tissue nitrogen. Since land use history of the previous century also impacts current soil carbon, the history of the sample site would aid significantly in interpreting the measurements attained. The litter/soil carbon data collected would still be invaluable for extrapolating the Fluxnet NEP measurements spatially.
The benefits of global NEP and NPP terrestrial monitoring networks are obvious for global science. There is no way a small number of global scientists could ever measure thousands of sites around the world themselves to test their global models and satellite algorithms. The benefit to field scientists in cooperating with this globally coordinated carbon sampling may initially be less clear, but ultimately equally valuable. The most legitimate way for field scientists to extrapolate their plot measures to overall estimates of vegetation productivity throughout their region is with satellite-derived computations of spatially georeferenced NPP. However, these satellite products cannot be offered as credible options for general use until rigorous field validation is completed. In the end, global scientists gain by having trustworthy estimates for carbon cycle research, and local land managers benefit by having a previously unavailable data stream of NPP that are continuous across the landscape, not merely plot samples, and are repeated frequently. Furthermore, local scientists have the option of enhancing these global satellite products with local weather data and crop yield factors for more specialised local products.
5.2 Kyoto Protocol
Dr Allen Solomon reviewed the Kyoto Protocol on forest sinks and its relation to the global terrestrial carbon cycle issues:
The Intergovernmental Panel on Climate Change (IPCC) is an international body of scientists convened to synthesise current information on the known characteristics of climate change and its impacts, for application by policy-makers. Since its credibility is critical to its mission, IPCC uses only scientific results published in the peer-reviewed literature. It does not commission collections of scientific or monitoring data, hence it is quite complementary to GTOS and TOPC, which address the gathering and/or organisation of the collection and the assembly of such measurements into data sets. The most likely relationship between GTOS and IPCC is through the Conference of Parties (COP) to the Framework Convention on Climate Change (FCCC). COP sponsored the Kyoto meeting in 1998 in order to operationalise the FCCC, generating the Kyoto Accords for that purpose. The scientific information for COP is provided by the Subsidiary Body for Scientific and Technical Advice (SBSTA). SBSTA has requested IPCC to produce a special report which describes how to define the carbon amounts associated with three specific types of land use: afforestation, reforestation, and deforestation (ARD). The Special Report on Land Use, Land Use Change, and Forestry, is being written by IPCC to meet this need, and is expected to be published in May 2000. The recommendations of that report, in turn, will be reviewed by SBSTA. They will form the basis for its recommendations to the COP concerning subsequent efforts to include the carbon sequestered and released by land use and forestry within the carbon abatement commitments by Annex I countries.
The Kyoto Accords contain three Articles of critical importance to efforts aimed at providing data in support of the accords:
Article 3.3. The net changes in greenhouse gas emissions from sources and removals by sinks resulting from direct human-induced land use change and forestry activities, limited to afforestation, reforestation, and deforestation since 1990, measured as verifiable changes in stocks in each commitment period shall be used to meet the commitments in this Article of each Party included in Annex I. The greenhouse gas emissions from sources and removals by sinks associated with those activities shall be reported in a transparent and verifiable manner and reviewed in accordance with Articles 7 and 8.
Note that the subject is land use change induced by direct human activity, that is, it probably does not include indirectly human-induced land use change, such as intense forest fires induced by fire suppression activities. Note also that the article is limited to forests and their appearance or disappearance, not to other land cover types. The article also specifies that forests of interest are only those undergoing a land use change since 1990; these are referred to hereafter as Kyoto Forests. Finally, note in this article that reporting requirements demand verifiable estimates of carbon fluxes; however, the values must be spatially explicit, otherwise, specific changes of land use cannot be measured and verified.
Article 3.4. Prior to the first session of the Conference of the Parties serving as the meeting of the Parties to this Protocol, each Party included in Annex I shall provide for consideration by the Subsidiary Body for Scientific and Technological Advice data to establish its level of carbon stocks in 1990 and to enable an estimate to be made of its changes in carbon stocks in subsequent years. The Conference of the Parties serving as the meeting of the Parties to this Protocol shall, at its first session or as soon as practicable thereafter, decide upon modalities, rules and guidelines as to how and which additional human-induced activities related to changes in greenhouse gas emissions and removals in the agricultural soil and land use change and forestry categories, shall be added to, or subtracted from, the assigned amount for Parties included in Annex I, taking into account uncertainties, transparency in reporting, verifiability, the methodological work of the Intergovernmental Panel on Climate Change, the advice provided by the Subsidiary Body for Scientific and Technological Advice in accordance with Article 5 and the decisions of the Conference of the Parties. Such a decision shall apply in the second and subsequent commitment periods.
Article 3.4 specifies that only Annex I countries (essentially, the developed countries) must keep track of carbon storage and emission from land use and forestry because only Annex I countries are required to provide information on their carbon stocks. However, Article 6 of the Kyoto Accords permits trading of carbon credits between Annex I and Annex II countries and with countries in transition (primarily, Eastern Europe and Russia). Hence, land use-induced carbon sequestration in those countries also must be measured if it occurs after 1990, and if it is involved with carbon offset projects claimed by Annex I countries. Note in this regard that data are required to establish spatially explicit carbon density in 1990 (because of its application to later measurements), and spatially explicit changes in carbon density in succeeding years. Finally, the article demonstrates that other land use activities which are not ARD will only apply in the second commitment period and later ones. Hence, for simplicity those changes are excluded from this discussion. This point, however, brings up the question of what commitment periods are, and which is the first one.
Article 3.7. In the first quantified emission limitation and reduction commitment period, from 2008 to 2012, the assigned amount for each Party included in Annex I shall be equal to the percentage inscribed for it in Annex B of its aggregate anthropogenic carbon dioxide equivalent emissions of the greenhouse gases listed in Annex A in 1990, or the base year or period determined in accordance with paragraph 5 above, multiplied by five. Those Parties included in Annex I for whom land use change and forestry constituted a net source of greenhouse gas emissions in 1990 shall include in their 1990 emissions base year or period the aggregate anthropogenic carbon dioxide equivalent emissions minus removals in 1990 from land use change for the purposes of calculating their assigned amount .....
Article 3.7 prescribes the first commitment period to be 2008 to 2012. While Kyoto Forests are those which undergo a change in land uses associated with ARD activities from 1990 onward, only the carbon losses or gains measured between 2008 and 2012 are to be included in the accounting of a country’s carbon sum. To date, ARD has been considered by the Special Report authors to mean that a change in land use has occurred - either between deforestation and reforestation, or following deforestation or preceding afforestation or reforestation. As a result, a forest harvested for its lumber or pulp can only be defined as “deforested” when it undergoes some other intermediate land use, such as farming or pasturing. If instead the harvested land is immediately replanted, or if it continues to be carried as forest in national land classification schemes, then it would not be Kyoto Forest, but merely land subject to forest management. However, note that the decision on whether forest harvests are deforestation or are part of forest management is a decision COP will make in the future with advice from SBSTA.
Given the above, precise definitions in the application of the Kyoto Accords will be critical in determining the nature of the monitoring variables and measuring programmes which are of value in defining the Kyoto Accords requirements. Yet, the framers of the Kyoto Accords did not define such keystone terms as “forest,” “direct human-induced land use change,” “afforestation,” “reforestation,” and “deforestation.” Instead, the authors of the IPCC special report are expected to review the more common definitions of these terms and to discuss the implications to carbon credits that each carries. This lack of definitions makes it difficult to estimate carbon amounts and especially, to avoid unintended consequences of the definitions. Examples of the complexities include:
Definition of “forest.” If by forest density, say a minimum of 50 Mt/ha of woody biomass combined with some minimum height (5 m), then countries can remove trees from their dense forests down to that value, emitting large amounts of carbon but not requiring reclassification of the decimated forests as Kyoto Forests. If one defines forest by cover values, say the 10 percent of possible canopy area and 5 m height which FAO uses, then again forests can be thinned throughout the 22 years leading to the conclusion of the first commitment period without contravening a Kyoto Forest definition.
Definition of afforestation, reforestation and deforestation. Do they require a land use change? If so, how does one detect a change from say, pasture for cows and natural regeneration after removal of the cows? Do the definitions truly exclude forest harvest and regeneration? After all, the exact words in Article 3.3 refer to changes in greenhouse gas emissions due to direct-human induced land use and forestry activities - what can be more clearly described as forestry activities than the harvest and regeneration of forests?
The need for identifying the mechanism of change. This induces the requirement for land-surface observations in addition to any remotely sensed data since the latter do not necessarily permit attribution of causes for changes in surface characteristics. Hence, despite the unavoidable emphasis on remotely sensed data, reliance on those data either available now or from instruments on upcoming satellites, will not fully meet the requirements of COP for including carbon derived from land use change and forestry.
How “direct” must human-induced changes be? Do they include only plantation forest regeneration, or, can they also include natural regeneration where that is the standard approach to replant forests, as in the Pacific Northwest of North America? Do they include carbon releases from insect infestations and wildfire, where those losses derive from forest management policies (and if so, how is such causal source proven?), or, are these to be considered natural changes and hence, of no relevance to Kyoto Forest? If the latter, are wildfires caused by the careless use of campfires by humans also irrelevant to Kyoto Forest accounting? What about fires purposely started by humans?
Other complications arise from the definition of Kyoto Forest area. Only ARD is used in carbon accounting, but the activity can occur any time from 1990 to the end of 2012. Changes which first occur in 2010 may represent the first designation of land as Kyoto Forest, which then requires carbon density measures from 1990 as well as the changes which have occurred thereafter. A related consideration is the lags built into forest changes. Deforestation can take place during a single year, but carbon emissions from associated forest soils can continue for decades. Equally, planted forests gain very little carbon during their first decade or so, but continue to gain carbon for 50, 100 or even 200 years thereafter. How does one credit the carbon resulting from these activities if the commitment period ends and the accounting occurs in e.g. 2012? Does a country take the carbon losses associated with removal of an entire mature forest with e.g. 200 Mt/ha above ground biomass, and take the credits from its newly growing replacement a few years later which may amount to only e.g. 5 or 10 Mt/ha? Or, does the expected growth of replacement forests over subsequent decades add to the credit, while loss of carbon from soils in subsequent decades reduce that credit? If the latter, does one calculate the expected future growth in the absence of effects by ongoing climate change, or does one somehow calculate the expected change in growth from climate change during the several decades of lagged carbon storage and release?
There are also complications due to the need for a continued monitoring of land parcels once they become identified as Kyoto Forest. Obviously, a country cannot accept the carbon credits derived from a parcel on which reforestation takes place in one commitment period, then remove that parcel from consideration during the next commitment period, when it can be harvested without an accounting. Hence, continuous monitoring is required once a land is declared Kyoto Forest. This appellation applies as well to land involved in carbon trading in less developed countries; once the land is defined as Kyoto Forest, it cannot be removed and used for some other purpose, such as farming, urbanisation or the like, without incurring a carbon debt to the country in which it is located (i.e., the country under whose control the used land resides). This may severely limit the amount of land less developed countries will be willing to include in trading carbon credits with Annex I countries.
The above issues address some of the complications arising from the attempt to operationalise the Kyoto Accords. The IPCC special report is designed to highlight these difficulties and unintended consequences for SBSTA, a body which will use the report as the basis for their recommendations to COP following presentation of the IPCC special report in May 2000. There is also some question as to whether GTOS remit includes provision of carbon density and forest distribution data which the Accords require and, if so, how to facilitate their generation. These data do not involve climate change; rather, they are focused entirely on carbon sources and sinks resulting from land use and forestry activities, which only indirectly relate to climate change, especially during the few years involved in the current Kyoto Accord initiation of measures, and during subsequent commitment periods.
5.3 Perspectives and Requirements for Terrestrial Carbon Observation
Dr Wolfgang Cramer discussed IGBP perspective and requirements for terrestrial carbon observation:
The IGBP Carbon Project represents an initiative by a large part of the IGBP community, across core project and other programme elements, and in cooperation with the World Climate Research Programme (WCRP) and the International Human Dimensions Programme (IHDP) to provide a synthesis of current understanding of the entire global carbon cycle (not just the biospheric component of it) on two time frames: as a fast-track synthesis report (to be completed during 1999, targeted to the needs of scientific input to the IPCC), and a more comprehensive synthesis of present knowledge (to be completed during 2000).
Part of the comprehensive synthesis will be a reassessment of currently available data sets, as well as the definition of requirements for a better assessment of the observational basis. In Appendix IV, some elements of past discussions on the data requirements for the IGBP Carbon Project are listed, but this list is preliminary and has not yet been discussed by IGBP. Many data requirements in the list are clearly not directly available from observing systems - therefore, methodologies to derive quantitative estimates from observed data sets are a separate and critical topic for further work.
Dr Cihlar presented an overview of the evolving requirements for information on terrestrial carbon cycle: from information on land use changes (mainly in the tropics) in the 1980s, to vegetation (mainly forest) inventories and their biomass and age characteristics in the 1990s (carbon pools-based approach), to data on flux exchange between the ecosystem (plants and soil) in the late 1990s (flux-based approach). While the current IPCC-specified approach to estimating national trace gas emissions is based on the inventory/pools approach as the only feasible one at the national level for many countries, the research frontier is addressing the flux-based approach and has already produced validated regional results. From the GCOS/GTOS perspective, an additional advantage of this approach is its compatibility with the GHOST observation strategy.
Dr Cihlar summarised the expectations of IGOS-P in the development of the terrestrial carbon theme, based on recent documents. Briefly, the proposed theme should have well-articulated, synthesised high level requirements, separately supported with more details; should take account of international, national, and regional requirements; should be sensitive to major issues connected with international conventions; could include operational, research requirements but aim for operational; should include the definition and inclusion of in situ requirements as a vital component; should identify requirements beyond data and products that can be met/provided from existing systems and those currently under development; and should have clearly defined objectives, roles and responsibilities of space and user agencies, milestones, outputs, success criteria, and resources required for execution.
5.6 In a breakout group, meeting participants discussed the vision, scope and content of a terrestrial carbon theme. They agreed to put forth the concept for a Terrestrial Carbon Initiative (TCI), based on the following vision:
TCI is envisioned as a functioning network of frequent observations and computer models, targeted for documenting and understanding the present state of the terrestrial component of the global carbon cycle. The input data will be provided by an optimised combination of remotely sensed and ground-based observing systems. The carbon cycling models must be sensitive to natural and human - induced changes in the environment; operate at a relatively high resolution to use optimally the input data; and produce maps of the terrestrial C sources and sinks on a seasonal, annual, interannual, and decadal basis. The resulting products must be sufficiently detailed to be useful for international conventions, policy decisions, and climate or environmental change assessments at global, regional and national levels.
TOPC also defined the objectives, scope, intended products, essential components of TCI; identified existing observation requirements identified by various groups, including TOPC (Global Climate Observing System, 1997) and IGBP (Appendix IV); discussed technical approach, gaps and problem areas; and prepared an outline of the prospectus and an initial version (Appendix V). It decided that regarding the forest biome, given the complexities and the present state of implementation of the Kyoto Accords (see Section 5.4), TCI would provide contributing information but should not endeavour to meet all the information requirements.
In response to the charge from GTOS/GCOS, TOPC agreed to carry out the following actions and schedule:
Complete the first version of the prospectus by the end of August, 1999
Distribute the prospectus for comments among scientists in G3OSs, IGBP, IPCC, and in other forums to ensure the soundness of the overall goals and strategy
Revise the prospectus, using email and/or additional meeting if necessary
Prepare a presentation for the November 1999 meeting of IGOS-P, along with a plan for the completion of the theme document (with a full meeting in the spring of 2000 as a key step)
Present to IGOS-P and obtain comments
Modify approach based on comments, carry out the rest of the plan.
Regarding the terrestrial carbon theme, TOPC made the following recommendation:
Recommendation 5.1: TOPC recommends that GCOS/GTOS support the development of the TCI concept along the lines described Appendix V, including interactions with key agencies and interest groups as well as additional meetings as required.
