Appendix VIII: Descriptions of Existing GTOS/GCOS Networks

A. GCOS Terrestrial Network for Glaciers (GTN-G)

Overview

Changes over time that can be detected in the mass, volume, area and length of the world's glaciers provide some of the clearest signals in nature of climate change.^[1] Unlike the permafrost network, which is only now being established, systematic observations of glacier fluctuations have been made in some parts of the world, notably in the Alps and Scandinavia, for over 100 years. The responsibility to collect and publish standardised data has been assumed since 1986 by the World Glacier Monitoring Service (WGMS), which is supported by the University of Zurich and the Swiss Federal Institute of Technology. WGMS now manages the recently created GCOS Terrestrial Network for Glaciers (GTN-G).

The mass balance of a glacier, which is a measure of the mass of water a glacier as a whole is gaining or losing, is a direct and undelayed signal of climate change. Some 60 glaciers throughout the world are currently monitored for mass balance as part of GTN-G. About 15 of these are being intensively studied with extensive stake/pit networks for improved process understanding and numerical modelling, while the remaining ones are monitored in a simpler way (index points and repeated mapping) to obtain information on regional changes. The changing length of a glacier also provides a key climate change signal. It is important both for enabling global coverage and for understanding the past evolution of a glacier. Although the climate change signal provided by glacier length is delayed and indirect, length is more easily measured than mass balance. The length of some 800 glaciers worldwide is measured approximately every five to 10 years.

Several publications document glacier data. The Glacier Mass Balance Bulletin is published by the WGMS every two years and reports measured values from selected reference glaciers.^[2] The Bulletin complements the publication series Fluctuations of Glaciers, where the full collection of digital data, including the more numerous observations of glacier length variation can be found. The WGMS maintains data exchange with the ICSU World Data Center A (WDC-A) for Glaciology in Boulder, Colorado, and the UNEP Global Resource Information Database (GRID).

Network details

Subnetworks

None

Sites

60 glaciers monitored for mass balance, 15 intensively

Some 800 glaciers monitored for length

Inventory data (not complete) for some 67,000 glaciers

See http://www.geo.unizh.ch/wgms/for pdf files of Mass Balance Bulletins and http://www-nsidc.colorado.edu/NOAA/wgms_inventory/, a WGMS mirror site, for inventory information.

Measurements taken

Mass balance

Length

Distribution

Elevation

Areal extent

Best practices

See the WGMS web site (http://www.geo.unizh.ch/wgms/) for a description of the monitoring strategy for mass balance, length change and inventory data.

Data quality control

Collection of standardised glacier fluctuation data follows recommendations (and regularly updated instructions) first published by UNESCO in 1969.

The WGMS does not guarantee the correctness of inventory data. The accuracy of the data is the responsibility of the data collectors in the individual countries. All data are subjected to plausibility checks while being loading into the database

Data availability

Mass Balance Bulletins published every two years. See web site. Starting in 1999, mass balance measurements will be available for selected glaciers one year after the end of the measurement year. Length measured annually or every five to 10 years; report published every five years.

Data archiving

See the series Fluctuations of Glaciers, published by WGMS, for mass balance and length data. See http://www-nsidc.colorado.edu/noaa/wgms_inventory/for inventory information.

Other useful web sites

National Snow and Ice Data Center (http://www-nsidc.colorado.edu/). Also see links on WGMS web site.

Deficiencies

The total number and worldwide distribution of glaciers monitored for mass balance provides important information on general trends, but it is far from ideal. For example, almost one-third of the world's glaciers are in the Alps and Scandinavia, where practical applications with respect to landscape change, tourism or hydropower production are obvious. Other areas are either underrepresented or have no sites at all in the network. New or intensified observations are most urgently needed in New Zealand, with its Southern Alps. Chile and Argentina, with major glaciers in the Andes and Patagonia, have no observation sites. In the former Soviet Union, the number of glaciers monitored has declined in recent years. Even the United States and Canada, with many glaciers in Alaska, the Canadian Rockies and Canadian Arctic region, have been monitoring only a few glaciers on a long-term basis. Some regions/countries where needs are acute include: Kenya, where important measurements of a long-studied tropical glacier have recently been terminated; Argentina, where mass balance observations in Patagonia have started, but for which funding for continuing measurements must still be assured; Tajikistan, where an important research station at the Abramova glacier was recently destroyed in a war action; Peru, where funding, monitoring, and scientific activity is in danger of being completely lost; and Mongolia, where observations have only recently begun.

Continuity and amount of funds at the national level is a major problem for the operation of glacier observing sites. Mass balance measurements require a team of three to four people for only a few days each year. The funds required for drilling and surveying equipment and salaries are therefore modest but must be available on a long-term basis. Moreover, even such small amounts can be beyond the means of many developing countries, in which scarce resources are needed for higher priority purposes. In at least one case, an offer of assistance was unsuccessful because the assistance was deemed to be inadequate. More typically, however, the low priority of glacier observations as perceived at the national level has meant inadequate funds for long-term monitoring. This situation is as true in countries such as Canada and the United States, where budgets for survey work have been reduced, as it is in developing countries. In some developing countries, lack of scientific capacity is also a problem. Adequate training and provision of field equipment can remedy the problem, but this too requires funding. Tensions in remote mountainous border areas in certain parts of the world (e.g., Central Asia) have limited mass balance observations to some extent. Finally, some potentially valuable data (including survey records, maps and aerial photographs) exist in countries of the former Soviet Union that have not been recorded and are in danger of being lost. High staff turnover in many of these countries as a result of general instability, as well as lack of funds, is at the root of this problem.

In addition to mass balance and length observations of a representative subset of the world's glaciers, the WGMS collects inventory information on the distribution, elevation, and areal extent of glaciers.^[3] Among others, such information is valuable for assessing climate change effects at the regional or national levels. However, detailed inventory information is currently not available for remote glaciated areas. A substantial number of aerial photographs exist for glaciers in the Alps, Scandinavia, and the former Soviet Union. Landsat satellite images are available for glaciers in other countries. However, Landsat images do not furnish elevation information. To provide global coverage of the world's glaciers, many of which are in remote areas, use of remote sensing techniques will be essential. New remote sensing technologies, such as synthetic aperture radar or laser altimetry used in connection with kinematic GPS and combined with digital terrain information, provide the basis for periodic detailed inventories.^[4]

B. GCOS Terrestrial Network for Permafrost (GTN-P)

Overview

The GCOS Terrestrial Network for Permafrost (GTN-P) was established to organize and manage a global network of permafrost observations, most importantly of temperature changes in frozen ground. Some 25 percent of the landmass of the Northern Hemisphere is underlain by permafrost, with important permafrost zones in Canada, China, Russia and the United States, and with smaller permafrost areas in many other countries in both the Northern and Southern Hemispheres. Permafrost observations are an important element of the mission of GCOS because variations in permafrost temperature can be a sensitive indicator of climate change and climate variability. For these purposes, observations are required in both the ‘active layer’ - the upper or surface layer which thaws and freezes seasonally - and the lower permafrost layer of perennially frozen ground.

Important observations in the active layer include the maximum thickness of the seasonal thaw and, when possible, a record of ground temperature. In the underlying permafrost zone, temperature profiles are required, with the desired frequency of observations decreasing with depth. For example, monthly observations may be desirable in the upper 5-15 m, where ground temperatures follow the annual air temperature wave (although with a phase lag and amplitude attenuation). Annual observations are sufficient at greater depths in shallow boreholes (up to 50 m), while observations at five- to 10-year intervals are acceptable at the greatest depths (up to several hundred metres) where temperature changes very slowly.

The Circumpolar Active Layer Monitoring (CALM) network is part of the GTN-P that is already in place for obtaining active layer measurements at about 80 sites in the Northern Hemisphere. For the deeper permafrost layer, temperature measurements are obtained through boreholes, using one of several types of thermistor sensor and measurement systems. A survey of existing boreholes has indicated that at least several hundred locations throughout both hemispheres are candidate sites for future, long-term observations of permafrost temperatures and related climatic variables. Most of these boreholes were drilled for either research or hydrocarbon exploration purposes. A European Community project, the Permafrost and Climate Change in Europe project (PACE), is drilling and instrumenting a series of nine permafrost boreholes in mountains from Spain and Italy to Svalbard. However, a globally comprehensive network of permafrost borehole temperature measurements, using the candidate holes, has not yet been developed.

Network details

Subnetworks

Circumpolar Active Layer Monitoring (CALM)

Permafrost and Climate Change in Europe (PACE)

Borehole

Sites

CALM: About 80 stations in the Northern Hemisphere. See information at http://www-nsidc.colorado.edu/nsidc/catalog/entries/g01175.html. A CD-Rom containing data sets is available. See also http://www.geography.uc.edu/~kenhinke/calm/. Expansion of network to Southern Hemisphere by 2001.

PACE: Nine sites in Europe. See http://www.cf.ac.uk/uwcc/earth/pace/fieldsites/index.html. Potential borehole sites: For inventory of accessible boreholes that are potentially eligible to GTN-P see http://sts.gsc.nrcan.gc.ca/permafrost/gtn-p.htm

Measurements taken

Active layer depth (maximum thickness of seasonal thaw)

Ground temperature

Borehole temperature

Best practices

See http://www.geography.uc.edu/~kenhinke/calm/for CALM.

Data quality control

See CALM and PACE web sites for metadata forms, observing protocol, and quality control. Same will be available for borehole temperatures in 2000.

Data availability

See web site for annual updates

Data archiving

Annual reports for CALM and PACE subnetworks and for borehole temperatures

Five-year synthesis for CALM

Final archiving in World Data Center A for Glaciology

Other useful web sites

International Permafrost Association (http://www.geodata.soton.ac.uk/ipa/)Deficiencies Observation deficiencies are most prominent in permafrost areas of five countries - Argentina, China, Kazakhstan and Russia - although regional gaps exist elsewhere. Fundamentally, the problems have to do with lack of resources for installation and continuing operations. This is why the International Permafrost Association has requested funding from the United States National Science Foundation. The funding would assist in the development of observational programmes in these countries and be used to maintain a web site on which data could be regularly reported and subsequently archived. Lack of standardised instrumentation is a problem in all countries. Additional funds would thus be used in part to provide the thermistors and other equipment needed for temperature measurements at some 15 new sites in the five countries identified above. A second problem is that many existing boreholes are in remote and often inaccessible locations; hence, there may be logistical difficulties in reaching sites, leading to higher expenses. A major problem in Russia is that while a substantial amount of useful historical data already exist, they are dispersed throughout the country, generally not in usable digital form, and often are not subjected to quality control procedures. To avoid losing these valuable records, the data need to be located, processed and archived in an accessible form. Additional resources would also be required for data management.

Additional boreholes are desirable in some locations to ensure adequate coverage for a global observation network, but new holes are expensive to drill. It is estimated that 20 new holes are required initially in underrepresented regions to provide appropriate global coverage. High priority areas include Argentina, which currently has no boreholes, and in Antarctica, where there are existing boreholes, but data recovery and reporting will depend on coordination with the Scientific Committee on Antarctic Research.

An administrative problem that might be easily addressed is that the appropriate ministries in many permafrost countries are unaware of the existence of the Global Terrestrial Network for Permafrost or of the importance of permafrost data for the assessment of climate change and its impacts. In addition, national permafrost monitoring activities have not always been clearly incorporated into the mandate of government agencies. Recognition of the importance of these monitoring activities through SBSTA could lead to more direct involvement in GTN-P by these approximately 15 countries, and potentially also make available some internal funds for implementation and long-term monitoring. The deficiencies should also be addressed in the development of national GCOS plans.

C. Global Flux Tower Network (Fluxnet)

Overview

The Global Terrestrial Network - Carbon (GTN - Fluxnet) was established in 1997 to coordinate existing regional networks and independent sites monitoring long-term carbon and water vapour fluxes and the surface energy budget. Monitoring of these and associated variables, such as site vegetation and soil, hydrologic, and meteorological characteristics, enables researchers to assess the annual net uptake of carbon dioxide from particular biomes, to quantify year-to-year differences in canopy carbon exchange, and to understand the environmental and biological factors controlling trace gas fluxes.^[5] Such information is crucial for a better understanding the Earth's carbon cycle and human perturbations to this cycle, and thus has direct relevance to political decision-making on global climate change.^[6] The creation of Fluxnet is intended to advance site-to-site comparability and coordination of enhancements to regional network plans.

At present, there are about 80 stations in the global Fluxnet network, but this number is expected to expand. Sites are heavily concentrated in North America and Western Europe. The xxaaxxaaxaaxAMERIFLUX network (including stations in Canada, Costa Rica and the United States) is the largest regional network with 35 sites. The EUROFLUX regional network has 18 sites in northwestern Europe, while MEDEFLU has seven sites bordering the northern Mediterranean Sea. JapanNet consists of some five sites, an expanding network in Brazil currently has two sites, and some 13 independent sites around the globe make up the rest of the network. A regional network is currently under development in Australia.

· Network details

Subnetworks

AmeriFlux

EUROFLUX

MEDEFLU

JapanNet

Sites

35 Ameriflux sites in the United States, Canada, and Costa Rica. See http://cdiac.esd.ornl.gov/programs/ameriflux/

18 EUROFLUX sites. See http://www.unitus.it/eflux/euro.html

Seven MEDEFLU sites. Posted site (http://www.iata.fi.cnr.it/medeflu/minute.htm) no longer active

Five JapanNet sites. See http://daacl.ESD.ORNL.Gov/Fluxnet/japan.txt for list of Japanese teams.

Measurements taken

Carbon dioxide, water vapour, sensible heat and momentum flux densities measured at a certain height above and within the canopy;

Air temperature and humidity measured above and within the canopy;

CO2 profiles measured above and within the canopy;

Soil temperature measured at certain distances above the soils;

Net radiation, PAR and solar radiation (direct and diffuse) measured above the canopy;

PAR measured below the canopy at 1m along a 20 m transect;

Precipitation and soil moisture.

Best practices

For Ameriflux see the science plan at http://cdiac.esd.ornl.gov/programs/ameriflux/scif.htm

For EUROFLUX see http://www.unitus.it/eflux/for project methodology.

Data quality control

See Fluxnet Overview at http://daacl.ESD.ORNL.Gov/Fluxnet/FLUX_Plan.html

For Ameriflux, see appropriate section in science plan.

Data availability

In general, unrestricted within two years after the date of collection.

For Ameriflux, investigators required to submit fully documented data sets “shortly after acquisition,” allowing for time needed to check and assure data quality.

Data archiving

Periodically, copies of finalised data in Fluxnet will be submitted to an archive (e.g., Oak Ridge National Laboratory Distributed Active Archive Center). For Ameriflux, data to be archived at the Carbon Dioxide Information Analysis Center. See http://cdiac.esd.ornl.gov/.

Other useful web sites

Oak Ridge National Laboratory Distributed Active Archive Center: http://www-eosdis.ornl.gov/

Deficiencies

Many parts of the world are either not represented in Fluxnet or are poorly represented. Thus far, there are no Fluxnet sites in Africa, none in most of Asia (i.e., China, Siberia, India, the Near East, Indonesia, etc.), none in Eastern Europe, none in South America outside Brazil, and none in Mexico. Also, although Canada has seven sites, none are located in northern Canada. Gaps in the system are inevitable with an ad hoc volunteer network, and not all facilities have equal capabilities or levels of scientific activity.

Gaps in the Fluxnet network can also be characterized by biome type. The current network is concentrated on forests, particularly temperate broad-leaved forests, temperate conifer forests, and semi-arid woodlands. However, many important biomes, including tundra, peatlands, wetlands, deserts, and savannas are underrepresented. Tropical forests may be underrepresented as well. Additional sites are also needed over intensively cultivated areas, and disturbed systems, such as burned, grazed, and logged areas also need to be investigated (e.g., we do not know how NEP varies with time since disturbance). Finally, forest age class deserves greater attention: e.g., young forest stands are underrepresented in some regional networks. Mountain sites have not been considered due to their complexity, but focused studies are warranted in mountains to resolve thorny issues related to flux measurements in suboptimal terrain.

As with other global networks, funding can be an important constraint to establishing and operating a Fluxnet site. A typical costs would include the purchase of instruments to make core measurements, spare sensors, data telemetry, and data archiving hardware. The cost of site infrastructure is extra and will vary according to the remoteness of the site (e.g., if a road or power line is needed), the height of vegetation (i.e., whether or not a tall tower must be built), and the existence of other facilities.^[7] Also, at minimum, a team of two individuals is required to operate a flux system. Day-to-day chores include calibration, instrument and computer maintenance, and data archiving. Such tasks require skilled operators, not readily available in many developing countries, to interpret data and detect subtle problems. Operations would be enhanced by measurement standardisation, as it would increase data quality and the ability to compare between sites and would also provide support and standards for emerging groups.