1 main objective and 5 particular objectives

Abstract and key-words of the project

An ecosystem-based approach: linking ecosystem functioning to predator populations

Main objective

To establish robust monitoring and assessment methods for terrestrial arctic predators to be employed as predictors of tundra ecosystem functionning in a changing climate

5 particular objectives

1 - To elucidate the relation between predator guild structure and food web structure under varying climates and trophic interaction cycles

2 - To establish the arctic fox as a key indicator of tundra food web structure and functioning by developing protocols for determining fox trophic position in time and space

3 - To engage Russian scientists and PhD students as to improve on ecosystem monitoring and research in the vast Russian sector of the circumpolar tundra biome

4 - To provide a legacy from IPY in terms of robust methods for large-scale and long-term monitoring of tundra ecosystems

5 - To initiate what is intended to be a lasting Norwegian-Russian cooperation concerning research and management of tundra ecosystems

BACK TO TOP

Abstract of the project

Strongly cyclic interaction between plants, herbivores and predators typically drives the food web dynamics over large tracts of the arctic tundra biome. These interaction cycles, however, now appear to be fading out at the southern edge of tundra, possibly due to climate warming. The Arctic Climate Impact Assessment (ACIA) highlights this change, if it was to spread further north into the tundra biome, to be one of the key processes by which terrestrial arctic biodiversity and ecosystem functioning could be lost. However, ACIA also acknowledged that there is currently very little research and monitoring that could detect dampened food web cyclicity and predict its consequences. The present IPY project is devoted to improve on this state of affair. Specifically, we will conduct research so as to establish arctic predators as predictors (indicators) of change in food web dynamics, based on the principle that species at the top of food webs often are most vulnerable and/or sensitive to changes. Our main research effort will be to calibrate a number of indicator parameters against data on food web structure from a number of sites distributed circumpolar within the international ArcticWOLVES initiative. The aim is then to establish a robust set of indicators to be used in large-scale/long-term research and monitoring of tundra ecosystem state. We intend to cover a large part of the Russian sector of the tundra biome and to involve (and fund) a large team of Russian ecosystem ecologists and students. The project should represent a boost for Norwegian-Russian cooperation on research and monitoring of tundra ecosystems.

Key-words

System ecology, food webs, indicators of change, monitoring, trophic interaction cycles, top-down cascades, rain-on-snow events, validation, robustness

BACK TO TOP

BACK TO TOP

INDICATOR PARAMETERS
1- ARCTIC FOX

We will establish the arctic fox as a key indicator because it is the only mammalian predator exclusive to the tundra biome, where it is omnipresent and a key predator in several respects. Due to the high degree of opportunism (Elmhagen et al. 2000), fox diets and trophic position can possibly be treated as a “blue print” of food web structure. Finally, the retreat of the arctic fox from the southern margin of the tundra biome was one of the first biotic indications of continent wide climate change (Hersteinsson and Macdonald 1992).

We estimate population density and reproduction by monitoring arctic fox dens. At several of our project sites, time series of arctic fox breeding and den use are available. We will use harvesting statistics to obtain supplementary information about population dynamics and to build predictive models based on climatic scenarios for forecasting arctic fox population trends and providing advice for harvesting management.

For each site, we will establish the time-specific trophic position of arctic fox through diet analysis (analysis of faeces and stomach contents of harvested foxes) and stable carbon (d13C) and nitrogen (d15N) isotope signatures in different tissues (Kelly 2000; Post 2002).

2- PREDATOR GUILD STRUCTURE
Tundra ecosystems harbour two main trophic guilds of predators with one specialised on small rodent prey (lemmings and voles) and consists mainly of bird species (raptors) such as the snowy owl, rough-legged buzzard, and three species of skuas (Wiklund et al. 1999). Mammals belonging to this guild are mainly the two small mustelids; i.e. weasel and stoat. It is this guild of “lemming predators” that is most likely to be sensitive to the dampening of the lemming cycle. The other trophic guild consists of generalist predators such as corvids, eagles, wolverine and wolves that frequently scavenge on carcasses of large herbivores (e.g. reindeers) or use subsidies from the marine food webs (i.e. sea birds). Being truly opportunistic predators, foxes (red and arctic foxes) may belong to the lemming predators guild in inland tundra, while in more maritime areas, foxes will be generalists depending more on marine subsidies (Roth 2003). Thus a combination of indicators based on composition and guild position of the entire predator assemblage is likely to reflect the site and time-specific structure of the food web.

We developed a standard protocol to mesure the density of raptor pairs and their breeding propension and success. Densities of small mustelids are assessed by the use of tracking tunnels. Automatic digital cameras will record relative densities of scavengers at 10 established “carcass stations” per field site (see the ecosystem-finnmark website). Predator abundance estimates will provide the primary data for establishing site and time specific predator community structure.

Wolverine at "carcass station" in the varanger pensula

3- VALIDATION OF INDICATOR PARAMETERS

For assessing the predictive power of the different indicator parameters, we will outline food web structure and climate at each field site. We will focus on the two upper trophic levels in the food web (herbivores and predators). We expect to depict food webs at least qualitatively for each site because all sites has a history of ecological research. Moreover, we will estimate the main food web compartments in quantitative terms. For some of the sites (such as Svalbard, Wrangel Island and Varanger peninsula) quantitative biomass and production estimates are already available for some temporally stable food web compartment. To highlight what is crucial missing information, a “gap analysis” has been conducted in a workshop at UiT with all project members being present. Identified gaps will then be filled by supplementary field work.

For all sites temporally varying compartments of the food web need to be quantified each year:

Small rodents: standardized index trapping and lemming winter nest counts

Ptarmigan, geese and waders: Colony censuses, line transect counts.

Large herbivore: Population size and carcass counts

Site-specific past signatures climate and primary productivity will be established by dendrochronological analyses of Cassiope tetragona and use of remote sensing data such as NDVI (Callaghan et al. 1989; Aanes et al. 2002; Pettorelli et al. 2005).

Fox functional responses to prey availability will be estimated in order to adjust for non-linearities between diet and food web structure. Prey biomass will be obtained based on year-specific quantitative information about food web structure at each field site and diets will be obtained from faeces-stomach and stable isotopes analysis.

Stable isotope signatures are used increasingly in ecosystem studies because this method has advantages over conventional diet analysis. It avoids biases resulting from differential digestion of soft versus hard-bodied prey items and integrates information over relatively long window of time rather than just a time snapshot as is the case for stomach or faeces contents. In order to validate the relationship between diets and isotope signatures at different time scales, we will conduct an experiment in which farmed arctic foxes are subjected to experimental diets simulating different food web structures. These validation experiments will be conducted at facilities certified to conduct such experiments at the Norwegian University of Life Sciences (UMB) in cooperation with Dr. Ø. Ahlstrøm.

Each of the indicators will be validated with respect to:

Reliability: Predictability with respect to food web structure in time and space

Cost efficiency: Information per effort sampling/analysis

Complementarity/Redundancy: Degree of exclusive information contents
Based on principles of parsimony and robustness we will jointly assess these indicator criteria to identify adequate set of indicators for future monitoring and research purposes (Yoccoz et al. 2001). This assessment will be based on data from all project field sites obtained from two years (2007 and 2008) of sampling from the 7 primary field sites in this proposal as well as relevant data from sites of the Canadian ArcticWOLVES.

BACK TO TOP

References

ACIA. 2004. Cambridge University Press

Aanes, R., B. E. Sæther, F. M. Smith, et al. 2002. Ecology Letters 5:445-453

Callaghan, T. V., L. O. Bjorn, Y. Chernov, T. Chapin, T. R. Christensen, B. Huntley, R. A. Ims, M. Johansson, D. Jolly, S. Jonasson, N. Matveyeva, N. Panikov, W. Oechel, and G. Shaver. 2004. Ambio 33:448-458

Chernov, Y. I., and N. V. Matveyeva. 1997 in Wielgolaski, ed. Ecosystems of the World. Elsevier Amsterdam

Elmhagen, B., M. Tannerfeldt, P. Verucci, and A. Angerbjörn. 2000. Journal of Zoology, London 251:139-149

Hersteinsson, P., and D. W. Macdonald. 1992. Oikos 64:505-515

Ims, R. A., and E. Fuglei. 2005. Bioscience 55:311-322

Kelly, J. F. 2000. Canadian Journal of Zoology 78:1-27

Pettorelli, N., J. O. Vik, A. Mysterud, et al. 2005. Trends Ecology and Evolution 20:503-510

Post, D. M. 2002. Ecology 83:703-718

Roth, J. D. 2003. Journal of Animal Ecology 72:668-676

Wiklund, C. G., A. Angerbjörn, E. Isakson, N. Kjellen, and M. Tannerfeldt. 1999. Ambio 28:281-286

Yoccoz, N. G., J. D. Nichols, and T. Boulinier. 2001a. Trends in Ecology and Evolution 16:446-453