The fossil forest on Axel Heiberg Island in the Canadian High Arctic, discovered in 1985 by the Geological Survey of Canada, is a unique site due to the amount of exposed tree material — some of which is 40 million years old, and dates from the Eocene period. On hillside terraces formed by glacial rivers and the action of wind on thawed soil, tree stumps up to several metres in diameter poke through the forest floor detritus. The site is 1 km long x 5 km wide, and is located near 80^o north latitude.

Over a vertical climb of 80 m, one passes through nearly 30 fossil forests that grew on the site at different times. Separate plant communities can be identified — from thick lower deposits containing giant metasequoia, to thin upper layers of larch and birch. In all, over 60 plant species have been identified. No animal or insect remains, however, have been found. Fossil forest wood, notable for its lack of mineralization, is similar to archaeological wood found in marine deposits, but is more compressed.

CCI began a map-inventory of the stumps and other features in 1989 accompanied by efforts to preserve the fragile specimens that were excavated from the site and subsequently treated in CCI laboratories. Over seven years, mapping surveys were accomplished despite the challenges presented by navigating the narrow terraces that required frequent movement of the observer's survey station. As well, short stays on the site, often accompanied by foul weather, restricted most of the work to single sightings rather than using triangulation or measurement tape. The many individual surveys were integrated to produce a distribution map that was published in 1995, and to present estimates of the rate at which stumps were lost and new ones came into view.

In August 1995, CCI's David Grattan and I went to the fossil forest to gather a final set of erosion readings, cut off metal erosion markers that might pose hazards to landing helicopters, and to survey specific stump fields. We replaced the theodolite method used in prior mapping surveys with Global Positioning System (GPS) technology because it promised several advantages:

• The survey could be carried out by one person.
• The base station only had to be set up in one location, because measurement is coordinated by radio contact between the base and roving GPS units.
• The positions were automatically available as three-dimensional coordinates, residing in a rugged, hand-held computer. This eliminated error-prone hand recording and calculation.
• The day's work was easily viewed on a laptop computer and readily converted into one of many possible conventional map projections by the GPS mapping software.

The resolution of the GPS equipment we used (Trimble Systems 4000ssi) was about 1 cm horizontal and 2 cm vertical for static positions, and 4 to 6 cm for measurements taken while walking, all in relation to the base station's position. This GPS system determines the relative position of the rover by tracking continuous broadcasts from GPS satellites 20,000 km out in space, and passing data from a static base station via radio. Another useful capability was the GPS rover's ability to establish its position "on the fly", which meant I did not have to return to a previously surveyed point to initialize the instrument's location after losing radio or satellite contact when the antennae were obstructed. This is very useful when negotiating awkward terrain.

The 1995 survey expedition experienced several difficulties, the greatest being snow cover that obscured stumps, and cold weather that drained batteries. Despite these problems, we learned enough about logistical difficulties with GPS to propose a complete survey of the hill in 1996.

In 1995 we had to use 30 watt radios, far too powerful for the size of project and very greedy on batteries. In 1996, I chose to use 4 half-watt radios (Trimble Trimtalk), two for the GPS receivers, and two repeaters to bend communications over the hilltop in order to eliminate a radio shadow discovered in 1995. Without a repeater, the base station would have had to be moved far from the camp in order to complete the survey. We also rigged the repeater station with a small solar panel to replenish its battery, however an Arctic hare adopted it as a warm place to leave deposits, necessitating recharging the battery every week.

For the 1996 survey, my CCI colleagues, Carl Bigras, Malcolm Bilz, and I put together a 66 amp hour, 12 volt gel cell battery bank, supplied by a 72 watt solar panel that recharged batteries and ran the laptop computer. The solar powered system also preserved the beautiful quietness of the Arctic — we were continually rewarded by the sight of caribou and muskox coming close to our campsite.

Malcolm Bilz, Tom Strang and Carl Bigras with one of the exposed stumps.

Tom Strang beside a repeater station rigged with a solar panel.

Measuring soil loss at a stake prior to mapping with GPS. Floating markers are visible in the background.

An Arctic hare adopted this solar panel as a warm place to leave deposits.

Orienting our solar panel to power the camp.

Malcolm Bilz (left) and Tom Strang cross a flooded stream at the base of a hill.

GPS Database

While Carl carried out his long list of photographic tasks, Malcolm and I collected points by GPS that included, among other data:

• Approximately 1,000 exposed stumps in growth position.
• Latitude, longitude, and elevation of the site to an accuracy of one half metre, and precise control points for topographic map production.
• All remaining floating erosion markers and steel stakes.
• Recording accurately spaced elevations in a 20 m wide swath along our longest erosion study line to map a precise cross section of the hill.
• Kite aerial stereophotography control points.
• One metre square markers laid down for, and visible in, the 1989 aerial photographs.
• Four vertical profiles to measure the tilt and thickness of layers across the site.
• Fully exposed logs on terraces.
• Separation of logs by permafrost polygon activity.
• Perimeters of silicaceous hardpan subject to undercutting and erosion, often used as helicopter landing sites and scenic lookouts by visitors.
• Contours of likely walking approaches to the hill, and spot heights.
• River bank contours to compare with 1989 aerial photographs.
• Location of all optical survey stations to tie in previous optical surveys that had recorded vanished stumps.
• Notes on feature conditions.

We also mapped and recorded the impact of human activities on the site:

• The many cairns erected by departed scientific parties.
• Visible remains of scientific excavations.
• Footpath duration from previous seasons of occupation.
• Tourist and visitor damage (bonfire, large-scale unauthorized excavation, graffiti spelled out in stones that was large enough to be read from the air, leaf-mat trampling, stump removal, modern trees transplanted to the site).
• Helicopter skid and wheel marks on the exposed forest floors.
• Loss of fossils lying on the ground surface.

The resulting 5,000 GPS records were easily combined with a topographic map, which was produced for us by Base Mapping Co. Ltd. of Ottawa from 1:2000 aerial photographs taken for CCI in 1989.

Initially intended as a study in natural loss, the focus of CCI's involvement has changed somewhat because ease of access and fame have drawn many visitors to the fossil forest, some of whose actions have damaged this site. Ultimately, CCI's map and photographs will contribute to a site management plan and to the efforts toward attaining protected status for the site. An information pamphlet will be produced by CCI that can be used by helicopter pilots to avoid landing on the often difficult to see stumps, and by visitors to avoid trampling sensitive features or creating erosion gullies. Above all, creating the most complete and accurate record of this site has allowed CCI to establish a thorough, long-term study of site use and abuse for one of the most sensitive environments on Earth.

Acknowledgements

We wish to thank the following for their assistance:

Nunavut Research Institute, Iqaluit, NWT
Polar Continental Shelf Project, Department of Natural Resources, Ottawa
Cansel Survey Equipment, Ottawa
Base Mapping Co. Ltd., Ottawa

Further Reading

Carl Bigras, Malcolm Bilz, D.W. Grattan, and C. Gruchy. "Erosion of the Geodetic Hills Fossil Forest, Axel Heiberg Island, Northwest Territories," Arctic, vol. 48, no. 4 (1995), pp. 342-353.

R.L. Christie, and N.J. McMillan. "Tertiary Fossil Forests of the Geodetic Hills, Axel Heiberg Island, Arctic Archipelago," Geological Survey of Canada Bulletin, no. 403 (1991).
Ottawa: Energy Mines and Resources.