Research sites within the radius of three kilometers from the Joatkanjávri field base, and some highlights of results (for location & geography, see Research sites –overview)

Highland: cold, dry and nutrient-poor – and most of the time under snow; the picture on the right taken during the autumn migration of reindeer in early October.

Slope: favorable microclimate (S-exposure), moist + nutrient-rich - woodland, meadow, scrubland

Lowland: warmer than Highland but dominated by dry and nutrient-poor heaths; productive habitats only in sites influenced by running water; cover only a few per cents of the landscape.

Main actors: Highland & most of Lowland: lemmings, gray-sided voles; foxes and long-tailed jaegers in peak years - even then in low densities; small mustelids are practically absent (Oksanen and Oksanen 1992, Oksanen et al. 1992), and the invading predators act as outbreak croppers, unable to prevent massive habitat devastation in winters following lemming peaks, when we see lemmings making their last desperate try to find food – most end up dying on the snow, but a few survive and as the vegetation recovers, a new build up starts.

In these habitats, exclusion of herbivorous mammals does make a difference (Moen & Oksanen 1998, Oikos 82: 333-346, see also Olofsson et al. 2002, Olofsson et al. 2004)

exclosure control

Slope and the most productive Lowland habitats: Root voles, red voles stoats, weasels, rough-legged hawks; after 1990 even American mink

Predators abound periodically (Oksanen and Oksanen 1992, Oksanen et al. 1992) and regulate rodents; even short-term exclusion of predators in summer had clear impact on vole dynamics (below, right; upper figure = Microtus agrestis; lower figure = all voles: from Ekerholm et al. 2004. Exclusion of herbivores (left) has no impact; compare the two sides of the exclosure, Moen & Oksanen 1998). The winter nests of lemmings found in this area typically carry the “visit card” of a mustelid, which has rested there after a good meal and made the nest even cozier by lining it with lemming hairs

Rodent dynamics are entirely different between the sub-areas. In the Lowland-Slope complex (left), voles prevail and display a regular 5 year cycle, with lows in years ending in -5 or -0 (most trapping performed in relatively productive habitats). In the Highland, lemmings prevail and have sharp outbreaks at irregular intervals, followed by prolonged, deep lows. The difference in fluctuation patterns corresponds to the difference predicted by models for predation-controlled and food-limited rodent populations (see Turchin et al. 2000, Ekerholm et al. 2001, Oksanen & Oksanen 1992). Along with the experimental results summarized above, the fluctuation patterns of microtine rodents corroborate the prediction of the Exploitation Ecosystems Hypothesis (EEH; Oksanen et al. 1981, Oksanen & Oksanen 2000) that herbivorous endotherms are regulated by predators in relatively productive areas (= Slope and the most productive part of the Lowland), whereas in less productive tundra landscapes, the intense interaction is between herbivorous endotherms and plants.

Invertebrates do not display similar contrasts in trophic dynamics between different sub-areas. If anything, herbivorous invertebrates are most abundant in the woodland patches of the Slope and uncommon on the open tundra. The likely reasons are (1) the higher eco-energetic efficiencies of invertebrates (2) their ability to pass the unfavorable season in dormancy, and (3) the fact that the predators of folivorous insects even get considerable amounts of energy from detritus-based prey, present in the same habitat. Anyway, The invertebrate data clearly show that EEH is not applicable to this part of the food web (see Oksanen et al. 1997).

The Joatkanjávri area has even been focal in our studies on plant-plant interactions, though these have extended even to more benign and more extreme environments. In these studies, we found that the intensity of plant-plant competition decreases along gradients of decreasing primary productivity (Sammul et al. 2000). However, the trend is not linear; rather there seems to be a threshold at which competitive interactions cease to be important (Sammul et al. 2006). Moreover, when herbivorous mammals are excluded, competition becomes important in unproductive areas, too (Olofsson et al. 2002). In extreme high alpine environments, where the “vegetation” just consists of scattered patches of Ranunculus glacialis (picture below), and grazing mammals are absent, competition is substantial again - now in the form of pre-emptive competition for the few spots where plants can grow at all (Olofsson et al. 1999, Oikos 86: 539-543).

The bottom line is thus that the harshness of the arctic-alpine environment per se does not have much to do with the intensity of plant-plant competition. Rather, the intense grazer-plant interaction typical for the tundra prevents competition from becoming important. If grazers are excluded or if the environment is too barren to support them, competition can be just as intense in stressful as in benign environments.