THE VISION FOR A CARBON CYCLE OBSERVING SYSTEM IS TO CONTRIBUTE TO THE INTEGRATED UNDERSTANDING AND HUMAN MANAGEMENT OF THE CARBON CYCLE THROUGH SYSTEMATIC, LONG-TERM MONITORING OF THE EXCHANGES OF GREENHOUSE GASES BETWEEN THE LAND, ATMOSPHERE AND OCEANS, AND THE ASSOCIATED CHANGES IN CARBON STOCKS. To achieve this vision, an observing system is required which synthesises information from several types of measurements: concentration of atmospheric CO2 and other gases, surface flux observations and other in situ measurements, and satellite remote sensing. The combined monitoring system will yield estimates of CO2 sources and sinks at multiple spatial and temporal scales from global to those relevant to land-use policy and resource management. These estimates should be provided with greatly reduced uncertainty relative to current practice, by designed expansions of current measurement networks and by systematic cross-checking of independent approaches.
The elements of such a system include measurement programmes on land, in the ocean, and in the atmosphere (Figure 1). It is based on local-scale measurements of the processes that control CO2 exchanges between the atmosphere and the Earths surface, and recognizes the crucial importance of models and other scaling algorithms to extrapolate to representative regional and global scales. The system includes multiple comparisons between predictions made by process models and larger scale observations, making it possible to disprove results of models that diagnose and predict large scale fluxes. This falsifiability is a necessary condition for a confident prediction of future atmospheric CO2 levels. At the global scale, changes in atmospheric CO2 are the benchmark against which all process models must be tested. This constraint can improve quantitative process model estimates only if applied regionally. An integrated observing strategy includes the application of multiple constraints at various spatial scales, taking advantage of process-based research, the ability of satellite data to map heterogeneous properties of the surface features, and the averaging properties of the atmosphere to quantify CO2 fluxes over large areas.
Figure 1. Observing the Global Carbon Cycle
Ultimately, an integrated observation strategy should provide timely diagnosis of carbon sources and sinks at high resolution in both space and time that simultaneously satisfies all the observations/data constraints at multiple scales. Such an observing system will be more than a set of observations. Observations alone can characterize processes at local scales, and can constrain overall mass balances at the largest scales. Models ingesting data from spaceborne sensors must be used to extrapolate local understanding to regional scales. The goal is to build a system of data and models at multiple scales that is designed to contain enough overlap to allow sufficient cross-checking to permit robust estimates of fluxes and their underlying causes. The results obtained in this manner will be the best possible at a given time, constrained by the quality of the observations and the understanding of the carbon cycle processes that is embedded in the models. The evolution of the overall system will necessarily be a gradual process which achieves two principal tasks: serving the current needs of the user communities, and improving the comprehensiveness and quality of the observations and of the models so that the future needs of users may be met more effectively.
To achieve the above vision, an approach encompassing all three domains of the global carbon cycle must be adopted, as envisioned in IGCO (Steffen, 2000). The present report focuses on the atmospheric and terrestrial components, both of which are necessary to meet the terrestrial domain objectives (p.12). The two principal links between the terrestrial and oceanic domains are the atmosphere (covered in this report), and water transport (surface and subsurface) which will be fully developed within IGCO.
Many of the elements of a terrestrial carbon observing strategy are now in place or under devel-opment. The challenges are to ensure that important existing observations continue and key new observations are initiated; to identify activities and agencies willing to contribute to establishing global carbon observations; to build in appropriate overlaps and leverage among the disparate data sets, thus filling important data gaps; to design and implement linkages among components, activities and contributions; and to link observation and research programmes so that the ongoing improvements in observations and products are made in an optimum fashion. An overarching requirement is the integration of the terrestrial/atmospheric observations with ocean observations within the IGCO; in this respect, this report will provide the input on behalf of the two domains. Rolling requirements review (WMO, 2000) can be used as an effective tool to guide future evolution of the integrated system.
Previous interests in the carbon cycle have been primarily economic in nature and is mostly related to land or ocean ecosystem productivity. In recent decades, the scientific interest in the carbon cycle has intensified, driven by the need for improved understanding of the role of carbon in the total earth system. Thus, the existing observing systems are primarily a combination of national resource-based inventories and scientific programmes with observation components at various levels, from national to global. Traditional national inventories have evolved generally independently, based on perceived national resource development needs. The strong interest by countries in the broader aspects of the carbon cycle is recent, and has been stimulated by the concerns about the role of trace gases in climate change.
Questions related to components of the carbon budget (such as land productivity) have been important to international reporting organizations such as UNEP, FAO and others. Traditionally, these agencies have relied on summarized data obtained within countries. In some cases, additional data were obtained directly to meet specific needs, e.g. by FAO for forest resources assessment. In the 1990s, global observing systems have been established for climate (GCOS), terrestrial ecosystems (GTOS) and oceans (GOOS) that require carbon cycle information to meet the needs of their clients. They all have, or are in the process of establishing initial observing systems based on current capabilities (Appendix 2). The global observing systems benefit from previous developments regarding existing systematic, long-term observations in two areas: observations of some processes and variables that affect the carbon cycle (e.g. weather/state of the atmosphere); and certain aspects of the carbon cycle which have been subject to research for a long time (e.g. plant growth).
In this section, the current status of carbon observing systems is briefly reviewed, including background information on products and users of the observing systems outputs.
Current atmospheric trace gas concentration measurements are sponsored by numerous countries, in most cases as part of research programmes. These data have made pivotal contribution to the awareness and understanding of the climate change issue; in particular, the Mauna Loa data series is now arguably the best known data set in earth sciences. The impetus for the work done by the many cooperating organizations and institutions is to make atmospheric measurements of trace gas species that will lead to better understanding of the processes controlling their abundance. Aiming to overcome accuracy and consistency problems in these measurements, GLOBALVIEW-CO2 was established as a cooperative atmospheric data integration project. GLOBALVIEW - CO2 is a data construct presently involving approximately 17 organizations from 13 countries. An internally consistent 21-year global time series has been compiled so far.
Monthly average data from the individual networks are also available via the WMO World Data Center for Greenhouse Gases, Tokyo and the Carbon Dioxide Information Analysis Center, Oak Ridge. The key users of these data are global carbon cycle modellers who derive CO2 sources and sinks distribution through inversion methods (top down approach; p.12). Among the most significant impacts of the network to date has been the discovery of unexpectedly large uptake of CO2 by terrestrial ecosystems at temperate latitudes in the northern hemisphere. However, due to the sparseness of the current network (p.19), there is potential to misinterpret the derived source/sink scenarios.
In addition to CO2, the observing system includes measurements of d13C and d18O in CO2, CO, and the O2/N2 ratio. These measurements are used for deducing the mechanisms responsible for fluxes in inverse modelling: d13C and O2/N2 are used to distinguish terrestrial from air-sea exchange, d18O is used to estimate gross primary production (as opposed to net ecosystem exchange). Measurements of other trace gases related to combustion (e.g. CO, CH4, H2, and VOCs) are used to estimate the anthropogenic contribution. Careful inter-calibration among measurement laboratories and sampling programmes is essential for these data to be most valuable.
The current atmospheric observing networks focus on measurements in the remote marine boundary layer, to avoid contamination by local sources and sinks. The data are invaluable and an essential starting point, but many studies have highlighted the need for additional measurements over the continents. These are problematic because of the strong variability in both space and time, so new sampling strategies will need to be formulated (p.23). Observing system simulation with models will be required for network optimization.
Traditionally, the exploitation of biomass resources has been the main reason for terrestrial carbon observations, motivating many countries to establish operational inventory or monitoring in support of sustained use of forests, cropland and grasslands. These programmes differ in purpose, coverage and duration. National forest inventories, underway for decades (up to 1930s) in many temperate and northern countries, have emphasised harvestable timber (above ground woody biomass) as the main component of interest and have focused on forests with commercial potential. They have been carried out at national, subnational and local (by commercial operators) levels. Various designs have been implemented, employing permanent or variable sample plots, aerial photographic interpretation, and cruising techniques. For the most part, these programmes have not been coordinated internationally. In spite of these differences and other deficiencies, the inventory data bases provide unique historical information on forest and land use carbon dynamics, especially in view of the decadal or longer time scales.
For agricultural and rangeland ecosystems, regular yield/aboveground biomass production surveys have been established in various countries, either nationally or sub-nationally for specific crops. National reports have been provided to FAO for many years. Coarse estimates of soil carbon content can be derived from data embedded in national and regional soil mapping programmes, but the quantitative accuracy of this information is inadequate for carbon inventories (p.12 and p.20). Geo-referenced quantitative soil carbon data are available from SOTER (e.g. 4000 estimates in the global WISE soil profile database, 1700 estimates are available for Latin America, 800 for Eastern Europe) but these are too sparse to allow precise global or regional estimates.
In parallel to operational inventories, national research programmes have also initiated long-term observations addressing various aspects of terrestrial ecosystems, and these programmes have gradually become correlated at the international level through the efforts of the scientific community or international organizations. At the present time, numerous large-scale programmes exist that are relevant to carbon cycle observation at global to national levels. In general, such programmes have been based on site measurements, but larger area coverage has become more frequent over time and satellite observations have recently become a key observation tool.
Appendix 1 provides a sample of current programmes that require information on the terrestrial carbon cycle. Website references to specific programmes have been included to document the actual existing needs and the breadth of the user community. The requirements differ in coverage (global, continental, national); type of product; and the user group. It should be noted that for some activities, national agencies also require consistent information beyond their territories.
Appendix 2 lists some of the existing observing networks and programmes. Only examples of international or worldwide in situ networks are included. For satellite observations, Appendix 2 also lists products generated at the present or planned for the near future. So far, most of these satellite products have been prepared for research purposes or are experimental in nature. Nevertheless, they are frequently produced in multi-year series, and retrospective reprocessing is employed to improve the product characteristics.
In addition to the above acquisition and product generation programmes, a number of projects have been undertaken that contribute to the development of systematic global observing capabilities. They include the GOFC project initiated by IGOS in 1997; World Fire Web, designed to provide biomass burning products; the IGBP NPP inter-comparison project, contributing to the improvement of algorithms for ecosystem productivity estimation (Cramer and Field, 1999); the GTOS NPP project; and others. Also important are numerous research projects undertaken as part of satellite programmes (Appendix 3), many of which are global in scope and include product development, validation, and product generation activities.
The development of consistent in situ observation programmes has been a significant challenge for GTOS and other IGOS Partners. FLUXNET and ILTER (Appendix 2) are presently the most consistent and quantitative networks at the global level, and the expansion of ecosystem flux measurements has been advocated by GCOS at the SBSTA/COP forum. Harmonization of regional in situ observation programmes is being addressed by GTOS as part of the GT-Net (Appendix 2). Some research programmes are addressing harmonization of data collected nationally, e.g. a N.A./Asia comparison of national forest inventories is being carried out.