Biodiversity is intrinsically important for ecosystem structure and function. While ecologists are uncovering the relationships between species and their environments, and how these relationships are important for maintaining ecosystems in their current states, we are still just beginning to understand them. One of the first steps in this type of research is to examine the relationships between biodiversity and the biophysical environment along gradients.

Gradients in soil moisture and elevation will cause changes in biodiversity on the local scale (e.g. within a few metres), while gradients in climate and perhaps broad changes in geology (which influence soil types) will affect biodiversity changes on a regional scale ( e.g. over 1-100 km).

This project considers the biodiversity of plant community types across Arctic landscapes of various spatial scales. The broad latitudinal and longitudinal gradients covered by the Tundra Northwest 1999 expedition make it possible to observe the relationship between biodiversity and environmental variables along local moisture gradients and regional climate gradients. Remote sensing from aerial photography and satellite images combined with ground truthing methods at each site will elucidate regional differences in plant community diversity. The results obtained by other members of the Biodiversity Theme (”Yellow Camp”) will be used to supplement the community descriptions obtained in this project.

It has been predicted that the relationship between biodiversity and net primary production (NPP, the amount of new plant biomass produced per unit of area per year) will follow a ”normal” curve when NPP is plotted against biodiversity (e.g. species density). Both low and high levels of NPP are associated with reduced diversity, while intermediate levels of NPP are associated with greater diversity. This hypothesis has been shown to be generally true across many temperate vegetation types, but it has not been extensively tested in the Arctic. Hence, a second objective of this project is to test the hypothesis linking biodiversity and NPP using the data collected a cross the Canadian Arctic Archipelago on Tundra Northwest 1999.

Methods

Site selection was based on the presence of a representative moisture gradient in order to accommodate sampling in dry (xeric), mesic and wet communities. In general, south or south-east facing gentle slopes were chosen. Vegetation communities were delineated visually based on the homogeneity of species composition and surface characteristics. In each plant community a combination of sampling methods was used: plant cover measurements using point quadrats, biomass harvesting, and descriptive environmental observations. The ground truthing methods enabled us to distinguish plant community characteristics which will be linked with patterns observed on higher scales (aerial photos from helicopter, satellite images) in order to build a vegetation map of each site.

We attempted to survey a maximum of 6 and a minimum of 3 quadrats in each of a mesic, xeric and wet community at each site. There was not always enough time to complete 6 quadrats in each community and priority therefore was given first to a mesic community, followed by a xeric community and if time allowed, a wet community. Given the time constraints at each site, biomass sampling was only carried out at the mesic and xeric sites.

Within each of the three communities, quadrats (50 cm x 50 cm) were randomly located and the area pegged and marked for further photography and sampling. One hundred points were surveyed within each quadrat area and all plants under each point were recorded following typical ITEX protocol (Walker 1996). Following the point quadrat surveys, all above-ground vegetation was removed and placed in paper bags to be taken back to the ship for drying. The dried biomass will be sorted according to species and further divided in to current year’s growth and then weighed to give an indication of net above-ground production.

All randomly selected areas were photographed. The quadrat photographs were consistently taken vertically with a common orientation to ensure that the photographed images would correlate with the point quadrat data. An additional 10 photos were taken of quadrats randomly placed in each community to increase the number of samples. These photos will be analysed using a digital analysis system after correlation with the 3-6 quadrats that were both sampled and photographed.

Aerial photographs of each site were taken from a helicopter using a hand held digital camera. The photographer captured sequential images of the ground over a prearranged flight path representing a typical moisture gradient for the site. The altitude, distance and speed were recorded throughout each flight. Large, visible markers of known size were placed on the ground under the flight line for scale.

Three flights were made over the study area whenever possible, generally in a straight line and along the chosen gradient. The first sequence of images was taken at 500 feet (ca 150 m) above ground with the digital camera, the second at 300 feet (ca 90 m) with the digital camera and the third also from 90 m with an automatic SLR camera as a back-up in the event of digital camera or computer failure. When time permitted, an overall view of the entire site area was taken at 300m. Upon return to the ship the digital photographs of both the quadrats and the flight lines were downloaded, labelled and saved.

Laboratory and data analysis

The point cover data have nearly all been entered into spreadsheet programs. Species identifications will be confirmed by the taxonomic specialists in Theme B. Biomass samples will be sorted in to species, and then into current year’s growth, and live and dead standing crop. The samples will be dried to constant weight and weighed. Total weights of current year’s growth will be used to calculate net above-ground production. In some cases (e.g. early season and late season sites), adjustments will be made for the phenological state of the plants.

Once cover and biomass data have been entered and checked, the data will be analysed using multivariate ordination and classification techniques, which will allow us to examine patterns in biodiversity across a range of spatial scales and with all measured environmental variables. The relationship between biodiversity and net above-ground production will be analysed using regression analysis to determine if the relationship follows a normal distribution.