Réseau mondial de surveillance terrestre du pergélisol
GHOST: Global Hierarchical Observing Strategy
(seulement en anglais)
The five-tier, hierarchical system for surface observations is briefly described below, as it would apply to permafrost; a more complete generic description can be found in Version 2.0 of the GCOS/GTOS Plan for Terrestrial Climate-related Observations (GCOS Report #32). It is also important to note that the tier structure is a classification system to aid implementation, not a rigid formula for implementation. All the tiers may not be necessary for a permafrost monitoring system. Rather, they are intended as a general guide for structuring the system.
The Tier Structure
Tier 1: These are major assemblages of experimental sites subject to intensive activities over large areas, are organized to emphasize detailed measurements and process understanding across environmental gradients. They should be located with a primary emphasis on spatial diversity. Capturing the range of the major types of permafrost terrain is a critical priority, but the location within the regions will be opportunistic.
Although all Tier 1 data and research findings are important to GCOS/GTOS, special attention should be given to long-term measurements. Tier 1 assemblages encompass sites distributed over large areas, and various adjustments are required before they can become part of a long-term monitoring programme. The long-term measurements will be a subset of those made during the initial experimental period, but the transition from intensive field studies to continuous monitoring requires careful planning. Several examples of climatic/landscape gradients exist within the CALM network: Kuparuk River basin in arctic Alaska (U.S.), Mackenzie River (Canada), Lena River and Kolyma River basins (Russia). The PACE transect across the mountains of Europe to Svalbard is another example. These regional networks or transects are composed of nested sets of other observing sites that individually and collectively make up lower-level tiers.
Tier 2: Tier 2 sites encompass permafrost-active layer studies within the major climatic zones. Ideally, Tier 2 sites should be located near the centre of the range of environmental conditions (though not necessarily near the geographical centre) of the region they represent. Actual locations will depend more on existing infrastructure and logistical feasibility rather than on strict spatial guidelines, but there is a need to capture a broad range of climatic zones. The Toolik and Bonanza Creek LTERs in the U.S. and Zackenberg (ZERO) site in Greenland are examples. The "Tier 2 Facilities" are commonly surrounded by other sites in the same ecoregion or physiographic region, and may also function quasi-independently as Tier 3 facilities (see below)
Tier 3: Collectively, Tier 3 sites are intended to sample the range of environmental variation present in the permafrost system within a climatic zone or region. They are chosen to be representative "integrated mosaics" of local or landscape scale conditions (topography, vegetation, soils, etc.). They are well-located (georeferenced and surveyed) areas in and around which intensive monitoring and critical field experiments can be performed. The "well-located" aspect relates to the remote sensing applications, and therefore selection criteria include emphasis on size and position with respect to the environmental range, particularly in regions of discontinuous permafrost and mountains. This means that some of them will be close to the modal or midrange of the various environmental factors which make up the environmental range of the system, while others will be closer to the extremes. As a result, some permafrost terrain types may have more potential Tier 3 sites than are needed for GCOS/GTOS. Other types may have too few sites, or none at all and thus GCOS/GTOS will need to work to stimulate efforts to enhance and balance the network. Agricultural research stations and experimental watersheds located in the permafrost regions are examples of sites to be included. The Murtel/Corvatsch high altitude experimental area in Switzerland and the Caribou-Poker Creek Research Watersheds in central Alaska can be considered examples.
Tier 4: At this level, spatial representativeness is of the highest priority. The locations of Tier 4 sites should be based on statistical considerations. It is impractical to prescribe one statistical design for all regions or countries. Hence, individual participating organizations would be responsible for locating the sites based on general guidelines, and may choose either a systematic or a stratified-random approach (or a combination, depending on the permafrost conditions and terrain types). The latter approach requires an a priori location specification, but permits rejection of the site and resampling out of the same population if the site is inaccessible or if sampling at the site would compromise national interests.
Sites falling within Tier 4 are areas of very limited extent (perhaps one hectare or less). Land-cover category (vegetation and soil) is the best criterion for selection and can be "satellites" of the Tier 3 constructs. Rock glaciers could be considered in this tier. For active layer measurements to be most effective, they should be actual components of the Tier 3 facility, for example 100 x 100 m grid cells or smaller plots on the larger CALM grids. Individual plots in study areas would also qualify as Tier 4 sites, with the entire collection of plots constituting a Tier 3 entity.
For the permafrost borehole network, long-term observations of temperatures in boreholes (25 to 125 m depths) within major lowlands, uplands and mountain regions are required and should be selected ideally according to permafrost occurrence based on existing maps or models that predict distribution of permafrost.
Tier 5: To detect changes over an area (or region), satellite remote sensing may be the only practical means to bridge the gap between in situ point measurements and areal averages, eventually expanded to regional scale studies. Satellite observations are usually for area averages (for areas 20-100 m depending on the sensor and the surface variable), while ground observations are essentially point values. Passive microwave remote sensing data can be used to detect surface soil freeze/thaw status, thus providing information on the onset dates and the length of surface soil thawing and freezing seasons for both permafrost and non-permafrost affected regions. Using a combination of conventional synthetic aperture radar (SAR) backscatter and more sophisticated interferometric SAR (INSAR) techniques, the magnitude of surface differential deformation (frost heave and thaw settlement) can be determined on an order of less than a centimetre. Combined with in situ measurements, such as at CALM sites, regional active layer thickness can be estimated from these INSAR data. AVHRR and other visible remote sensing data can be used to detect thaw-lake surface change including the formation of new thaw lakes or thermokarst development as ground ice melts. Ground-truth measurements are required to validate the results from these remote sensing techniques, as well as for aerial photography. Remotely sensed observations at multi-year intervals can be used to measure changes in permafrost slope stability by monitoring frequency and extent of landslides triggered or re-activated in relation to climate or due to increasing forest fire activity. Observations of active layer detachment slides in the Mackenzie Valley, Canada, are examples of this application.
Preparation of data products from satellite measurements must be based on a long-term program of data acquisition, archiving, product generation, and quality control. Discussions are now underway in the Committee on Earth Observation Satellites (CEOS) to set up such a system. In particular, coordination is needed with the Global Land Ice Monitoring System (GLIMS) project which will map changes in the areas of selected glaciers worldwide and thus contribute observations of adjacent permafrost areas.