Abstract

Active layer monitoring in Greenland was started in 1996 and 1997, and forms part ofthe Circumpolar Active Layer Monitoring (CALM) Network of the International Permafrost Association (IPA). The results of the first years ofthis monitoring of thaw progression and maximum active layer thickness in two Greenlandic permafrost areas åre presented. Two sites åre in the continuous permafrost zone at Zackenberg in NE Greenland (74 °N), and one at Disko Island in W Greenland (69 °N), at the border between discontinuous and continuous permafrost.

The data collected at Zackenberg demonstrate interannual variation in the timing of thaw progression in the monitoring grid holding a seasonal snowpatch, while there is less variation in the horizontal grid without a snowpatch. The maximum active layer thickness for the two Zackenberg grids is more or less consi_stentfor the first three years with averages from 58 to 66 cm in mid and late August. At Disko the active layer reached 71 cm in mid August 1998. Spatially the distribution ofthe maximum, annual active layer thickness within the grids is concordant.

Keywords

Active layer monitoring, permafrost, Greenland, Zackenberg, Disko
Island.

Hanne H. Christiansen: Institute ofGeography, University ofCopenhagen,
Øster Voldgade 10, 1350 Copenhagen K., Denmark.
Email: hhc@geogr.ku.dk

GeografiskTidsskrifr, DanishJournalofGeography99,ll7-121,1999

The active layer is the surface layer of ground which is subject to winter freezing and summer thawing in areas underlain by permafrost (Muller, 1947). This makes the active layer thickness mainly climatically controlled. The active layer is critical to the ecology and hydrology of permafrost terrain, providing a rooting zone for plants and acting as a seasonal aquifer for near-surface ground water (Burn, 1998). Likewise it enables the exchange of trace

gasses such as carbon dioxide and methane, heat and moisture
between the atmosphere and the soil in arctic areas.

Changes in the active layer thickness and timing of thaw progression åre likely consequences of climate change. Therefore annual measurements of the active layer thickness, as performed in the Circumpolar Active Layer Monitoring (CALM) network, provide documentation for interannual fluctuations, and register the environmental impact of global climate change in permafrost areas (Nelson & Brown, 1996). Also active layer monitoring represents an important basis for verifying active layer modelling.

In Greenland monitoring of permafrost borehole temperatures and mapping of permafrost åre sparse. However, in areas of intensive monitoring and research of the physical environment, such as at Zackenberg, where Zackenberg Ecological Research Operations (ZERO) (Meltofte, 1996) has operated since 1995, and on Disko Island at the Arctic Station in Qeqertarsuaq, where meteorological data has been collected since 1990 (Nielsen et al., 1996), participation in the CALM network is now initiated.

Circumpolar active layer monitoring, CALM

The CALM network currently consists of 69 research sites in the Northern Hemisphere (Figure 1) operated by researchers in 10 different countries. Seasonal thaw data has been collected since 1991 using a standard measuring protocol in the CALM grids. The network consists of permanently marked measuring grids, mainly in fine-grained soils in the Arctic. CALM grids generally measures 100 x 100 m or 1000 x 1000 m. The traditional method of soil thaw measurement is pushing a narrow-diameter metal probe into the soil to the point of refusal (Nelson & Oultcalt, 1982). In some grids also soil temperature and moisture content åre logged within the active layer.

Currently, all CALM data åre quality controlled by the
individual researchers who established the grids and control
data collection. Data åre then reported to the Department
of Geography at University of Cincinnati, where a
group of researchers has a 5-year U.S. National Science
Foundation grant for running CALM, including making
the data available on the CALM web site:
http://www.geography.uc.edu/~kenhinke/CALM.

The first two Greenlandic CALM sites, ZEROCALMI
and ZEROCALM2 were established in 1996 in Zackenberg,
NE Greenland (Christiansen, 1997). In 1997 a third

CALM site, DISKOCALMI was established near Qeqertarsuaq
at Disko Island, central W Greenland.

CALM in Zackenberg, NE Greenland

Both ZEROCALM monitoring grids åre located in the centre of the Zackenberg Valley at 74°28'N, 20°30'W. The long-term terrestrial monitoring commitment of the Geoßasis programme of ZERO (Christiansen & Humlum, 1996) ensures the annual late summer measurements of both grids (Christiansen, 1997), and makes it an important cooperating partner in the annual collection of CALM seasonal thaw progression data from both grids.

The ZEROCALMI (ZCI) grid is located on a slightly SW sloping (0-2°) marine abraded plain, at 36-37 m a.s.l. The grid system measures 100 x 100 m. With a grid size of 10 x 10 m, there åre 121 measuring points. Texturally, the grid is mainly sandy with some gravel. There is continuous vegetation dominated by Cassiope tetragona, but also with Salix, Vaccinium, mosses and lichens. Immediately south of ZCI, the ZERO meteorological station is located, and here soil temperatures åre collected at 10 le vels from the terrain surface through the active layer into the top permafrost, at a maximum depth of 150 cm. Air temperature, precipitation, wind speed, wind direction and other meteorological parameters åre also logged at this station (Christiansen & Humlum, 1996). In the southern part of ZCI a snow depth sensor was installed in 1997, continuously logging the winter snow depth (Rasch, 1998). Likewise in 1997 a permanent, double vehicle track was made crossing through the ZCI grid.

The ZEROC ALM2 (ZC2) monitoring grid is placed from 11-22 m a.s.l. and measures 120 x 150 m, also with a 10 x 10 m grid size, which gives 208 measuring points. It covers a seasonal snowpatch on a south-facing slope, maximally 15 ° inclined, surrounded by more level areas in the northern and southern parts. This grid is located in the Zackenberg Delta, and it includes two fluvial terraces. Texturally, the grid consists of both sand and silt-clay dominated sediments. The grid has an almost complete vegetation cover. There is a distinctive vegetation zonation controlled mainly by the snow cover duration in the snowpatch area, such as described by Christiansen (1996,1998). In the areas outside the snowpatch Dryas dominate on the dry, coarse-grained upper fluvial level, whereas grasses and mosses dominate in the fine-grained, water-soaked fen area downslope the snowpatch. Some metres south of the grid there is a Geoßasis soil temperature monitoring profile, logging the temperature in 4 levels from the terrain surface, through the active layer down to 60 cm in the top permafrost.

CALM at Disko Island, W Greenland

At 69°16'N, 53°30'W about 2 km NE of the village Qeqertarsuaq/Godhavn on Disko Island the third Greenlandic CALM grid, called DISKOCALMI (DCI), was established in 1997 on top of a broad-crested E-W extending moraine ridge, Pjeturssons Moræne, at about 125 m a.s.l. This grid measures 90 x 90 m, with 100 measuring points. Morphologically the grid slopes slightly (maximally 3°) towards the north in the northem end, slightly (maximally 3°) towards the south in the southern end, and is almost horizontal in the middle part. This reflects the moraine ridge top location. Vegetation, primarily mosses and Salix, covers about 80 % of the grid surface. The sediment is till, with some boulders. In the centre of the grid a soil temperature monitoring profile was installed in 1997, logging the temperature at 3 levels from the terrain surface down to 74 cm. In this grid it has only been possible to measure thaw thickness annually in the late summer.

Results

All measured maximum active layer thickness for the Greenlandic CALM grids åre presented in Table l, together with dates of the measurements and standard deviation.

The ZCI grid results can be compared with the soil temperature data of the ZERO meteorological station, close to ZCI. These data show the maximum active layer depth to be about 92 cm in 1996 at 7 September, and about 92 cm in 1997 at l September. The relatively large difference between the measured maximum thaw depth in ZC l and at the temperature profile close by, can be due to difference in texture, but since both ZC l and the meteorological sta

tion åre located on the same landform, this does not seem to be the main cause. It is, however, possible that the installation of the soil temperature sensors in a vertical tube in the soil may enable local downward water penetration in the profile, which might locally increase the active layer thickness. Soil temperature data from the Geoßasis profile immediately south of 2C2 has failed for the sensors closest to the permafrost top in both 1997 and 1998, so there is no other information on maximum active layer thickness from there.

The first three years of observation in both ZEROCALM grids show similarly shaped thaw progression curves, indicating a spatially almost equal thaw development in each grid (Figure 2). However, there was a temporal delay in the ZC2 grid thawing between 1996 and the following two years of 18-23 days for most of the mid summer period. This was not seen in ZC l. The main reason for this difference was an extended snow cover duration particularly downwind of the seasonal snowpatch in ZC2 in 1997 and 1998 compared to 1996. The increased snowpatch size is caused mainly by increased winter wind speed in the later two years, in combination with a more equal distribution of the snow precipitation during these winters. These informations åre basedon data from the ZERO meteorological station.

About 2 km away from the DISKOCALMI grid the maximum active layer thickness is generally found in early September to be 170-195 cm at 20 m a.s.l. This is basedon soil temperature data in sand and gravel sediments at the permanent meteorological station at the Arctic Station in Qeqertarsuaq (Humlumet al., 1999, Humlum, 1998). Presumably, the exposed location of the DC l grid on top of the moraine ridge enables a reduced active layer thickness, suitable for the CALM probing technique. No thaw progression data have so far been obtained from this grid. Soil temperature monitoring in the DC l grid partly failed in the first year, so there is no information on the maximum late summer thaw this way.

Perspective

Obviously, it would be relevant with more CALM grids in selected areas of Greenland, such as in North Greenland, e.g. at the manned stations. This would give a regional overview of the thermal state and thickness of the active layer, and thus the permafrost conditions. Permafrost

monitoring should also be carried out in the discontinuous and sporadic permafrost areas of particularly SW Greenland, where climatic changes might have the largest socioeconomic effects, by causing permafrost growth if the present climatic cooling in this area continues. Likewise if warming occurs, permafrost might disappear in certain areas, also affecting society by increased slope instability as is presently occurring in the European Alps (Haeberli & Beniston, 1998).

Snow fence manipulation is now also initiated at Zackenberg to study the effect of a prolonged snow cover duration on active layer thaw progression and associated physical processes (Jakobsen et al., 1999). This in combination with the presented active layer monitoring of natural conditions will allow a better understanding of how changes in key physical climatic change parameters such as snow precipitation and wind can affect the permafrost environment.

Acknowledgements

The establishment and use of the Greenlandic C ALM grid data is part of the periglacial research of the programme The Arctic Landscape: Interactions and feedbacks among physical geographical and biological processes' 1995-1999. The Danish Science Research Councils special programme for Polar Science has funded this research project. The ZEROCALM grids åre run in cooperation with the

Geoßasis programme of ZERO. Thanks to Claus Nordstrøm and Ole Humlum, both from the Institute of Geography, University of Copenhagen, who respectively measured the ZEROCALM grids medio August 1996, and the DISKOCALMI grid medio August 1998.

References

Burn, C.B. (1998): The Active Layer: Two Contrasting Definitions.
Permafrost and Periglacial Processes, 9: 411-416.

Christiansen, H.H. (1996): Nivation Forms, Processes and Sediments
in Former and Recent Periglacial Areas. Geographica
Hafniensia, A4.

Christiansen, H.H. (1997): Periglacial and glacial geomorphological research. Pp 57-59 in Meltofte, H. & Thing, H. (eds.) Zackenberg Ecological Research Operations, 2"* Annual Report, 1996. Danish Polar Center, Ministry of Research & Information Technology.

Christiansen, H.H. (1998): Nivation forms and processes in
unconsolidated sediments, NE Greenland. Earth Surface Processes
and Landforms, 23: 751-760.

Christiansen. H.H. & Humlum. O. (1996): The Geoßasis Programme.ln Meltofte, H. & Thing, H. (eds.) Zackenberg Ecological Research Operations, l^st Annual Report, 1995. Danish Polar Center, Ministry of Research and Technology, 18-32.

Haeberli, W. & Beniston, M. (1998): Climate Change and Its Impacts
on Glaciers and Permafrost in the Ambio, 27: 258-265.

Alps. Humlum.O. (1998): ActiveLayerThermal Regime 1991-1996 at Qeqertarsuaq, Disko Island, Central West Greenland. Arctic and Alpine Research, 30: 295-305.

Humlum, 0., Nielsen, N., & Hansen, B.U. (1999): Meteorological Observations 1998 at the Arctic Station, Qeqertarsuaq (69° 15 N), Central W. Greenland. Danish Journal of Geography, 99: 113-115.

Jakobsen, 8.H., Christiansen, H.H., Elberling, 8., Hagedorn, B. & Sletten, R. (1999): Dynamics of High Arctic soils: Physical and chemical processes in the active layer - permafrost system. In Rasch, M. (ed.) 1999: Zackenberg Ecological Research Operations, 4* Annual Report, 1998: 48-49. Danish Polar Center, Ministry of Research and Information Technology, Copenhagen.

Meltofte, H. (1996): Introduction. In Meltofte, H. and Thing, H. (eds.) Zackenberg Ecological Research Operations, l^st Annual Report 1995: 7-8. Danish Polar Center, Ministry of Research and Technology.

Muller, S.W. (1947): Permafrost and Permanently Frozen Ground
and RelatedEngineering Problems, Ann Arbor, Michigan, J. W.
Edwards.

Nelson, F.E. & Brown, J. (1996): Circumpolar active layer monitoring
(CALM). Frozen Ground, 21: 21-24.

Nelson, F.E. & Oultcalt, S.I. (1982): Anthropogenic geomorphology
in northern Alaska. Physical Geography, 3: 17-48.

Nielsen, N., Hansen, 8.U.; Humlum, O. & Rasch, M. (1996): Meteorological Observations at Arctic Station, Qeqertarsuaq (Godhavn), Central West Greenland. Danish Journal of Geography, 96: 97-104.

Rasch, M. (1998): The meteorologica! station. Pp 9-10 in Meltofte, H. & Rasch, M. (eds.): Zackenberg Ecological Research Operations, 3"¹ Annual Report, 1997. Danish Polar Center, Ministry of Research & Information Technology.