Katrine S. Hoset

Mammalian herbivores confer resilience of Arctic shrub-dominated ecosystems to changing climate.

Climate change is resulting in a rapid expansion of shrubs in the Arctic. This expansion has been shown to be reinforced by positive feedbacks, and it could thus set the ecosystem on a trajectory toward an alternate, more productive regime. Herbivores, on the other hand, are known to counteract the effects of simultaneous climate warming on shrub biomass. However, little is known about the impact of herbivores on resilience of these ecosystems, that is, the capacity of a system to absorb disturbance and still remain in the same regime, retaining the same function, structure, and feedbacks. Here, we investigated how herbivores affect resilience of shrub-dominated systems to warming by studying the change of shrub biomass after a cessation of long-term experimental warming in a forest-tundra ecotone. As predicted, warming increased the biomass of shrubs, and in the absence of herbivores, shrub biomass in tundra continued to increase 4 years after cessation of the artificial warming, indicating that positive effects of warming on plant growth may persist even over a subsequent colder period. Herbivores contributed to the resilience of these systems by returning them back to the original low-biomass regime in both forest and tundra habitats. These results support the prediction that higher shrub biomass triggers positive feedbacks on soil processes and microclimate, which enable maintaining the rapid shrub growth even in colder climates. Furthermore, the results show that in our system, herbivores facilitate the resilience of shrub-dominated ecosystems to climate warming.

Diet Quality Limits Summer Growth of Field Vole Populations

Marked variation occurs in both seasonal and multiannual population density peaks of northern European small mammal species, including voles. The availability of dietary proteins is a key factor limiting the population growth of herbivore species. The objective of this study is to investigate the degree to which protein availability influences the growth of increasing vole populations. We hypothesise that the summer growth of folivorous vole populations is positively associated with dietary protein availability. A field experiment was conducted over a summer reproductive period in 18 vegetated enclosures. Populations of field voles (Microtus agrestis) were randomised amongst three treatment groups: 1) food supplementation with ad libitum high protein (30% dry weight) pellets, 2) food supplementation with ad libitum low protein (1% dry weight; both supplemented foods had equivalent energy content) pellets, and 3) control (no food supplementation), n = 6 per treatment. Vole density, survival, demographic attributes and condition indicators were monitored with live-trapping and blood sampling. Highest final vole densities were attained in populations that received high protein supplementation and lowest in low protein populations. Control populations displayed intermediate densities. The survival rate of voles was similar in all treatment groups. The proportion of females, and of those that were pregnant or lactating, was highest in the high protein supplemented populations. This suggests that variation in reproductive, rather than survival rates of voles, accounted for density differences between the treatment groups. We found no clear association between population demography and individual physiological condition. Our results demonstrate that dietary protein availability limits vole population growth during the summer growing season. This suggests that the nutritional quality of forage may be an underestimated source of interannual variation in the density and growth rates of widely fluctuating populations of herbivorous small mammals.

Changes in the Spatial Configuration and Strength of Trophic Control Across a Productivity Gradient During a Massive Rodent Outbreak

Understanding the determinants of spatial and temporal differences in the relative strength of consumer–resource interactions is an important endeavour in ecology. Here, we explore the necessary conditions for temporal shifts in the relative strength of rodent–plant interactions in an area characterised by profound spatial differences in trophic control, with predator–prey interactions prevailing in productive habitats and rodent–plant interactions dominating unproductive habitats of the forest–tundra ecotone. We report data obtained during the exceptionally massive rodent outbreak of 2010–2012 in northernmost Fennoscandia, including an experimental manipulation of herbivore access to vegetation plots across a large-scale productivity gradient, multiple observational measures of plant–rodent interactions linked to rodent abundance data and a large-scale survey of breeding avian predators and mammalian predator activity. Unexpectedly, rodent grazing impacts documented during the rodent outbreak were uniformly strong across the landscape, regardless of habitat productivity. The runaway response in rodent populations was facilitated by a high population growth rate in the early phase of the outbreak due to the extended absence of predators in productive habitats, concomitant with an exceptionally long-lasting lemming outbreak in unproductive habitats. Our results showed that spatio-temporal variation in trophic control also occurs in ecosystems structured according to the exploitation ecosystems hypothesis and emphasises the importance of long-term studies to capture nonlinear and stochastic features that shape ecosystem functioning. In this context, the temporary release from top–down regulation in productive habitats caused strong grazing impacts that may be crucial for the resilience of tundra ecosystems under the threat of climate change-driven shrub encroachment.

Pulsed food resources, but not forest cover, determine lifetime reproductive success in a forest‐dwelling rodent

The relative contributions of habitat and food availability on fitness may provide evidence for key habitat features needed to safeguard population persistence. However, defining habitat quality for a species can be a complex task, especially if knowledge on the relationship between individual performance and habitat quality is lacking. Here, we determined the relative importance of the availability of suitable forest habitat, body mass and food from masting tree species on female lifetime reproductive success (LRS) of Siberian flying squirrels (Pteromys volans). We calculated LRS of 500 female flying squirrels based on a 22-year-long longitudinal dataset of two populations from western Finland. We assessed with generalised additive models the potential effects of availability of suitable habitat and cumulative lifetime availability of food from masting tree species on female LRS, longevity and fecundity. On a reduced dataset, we evaluated the importance of female winter body mass and conducted a piecewise path analysis to determine how variables were connected. According to generalised additive models female longevity, fecundity and LRS were mainly determined by variation in cumulative lifetime availability of food from masting alder and birch. Instead, habitat and body mass had a smaller role. The path analysis indicated that lifetime food availability had a direct effect on longevity and fecundity, and these had an equal effect on LRS at both study sites. Our results on LRS show that the occurrence of tree masting events during a female flying squirrels lifetime has a profoundly larger effect on LRS than the cover of suitable forest habitat. Furthermore, this study emphasises the importance of both fecundity and longevity, and the indirect effects of food availability via those components, as determinants of lifetime fitness in female flying squirrels.

Herbivore Effects on Ecosystem Process Rates in a Low-Productive System

Mammalian herbivores shape the structure and function of many nutrient-limited or low-productive terrestrial ecosystems through modification of plant communities and plant–soil feedbacks. In the tundra biome, mammalian herbivores may both accelerate and decelerate plant biomass growth, microbial activity and nutrient cycling, that is, ecosystem process rates. Selective foraging and associated declines of palatable species are known to be major drivers of plant–soil feedbacks. However, declines in dominant plants of low palatability often linked with high herbivore densities may also modify ecosystem process rates, yet have received little attention. We present data from an island experiment with a 10-year vole density manipulation, to test the hypothesis that herbivores accelerate process rates by decreasing the relative abundance of poorly palatable plants to palatable ones. We measured plant species abundances and community composition, nitrogen contents of green plant tissues and multiple soil and litter variables under high and low vole density. Corroborating our hypothesis, periodic high vole density increased ecosystem process rates in low-productive tundra. High vole density was associated with both increasing relative abundance of palatable forbs over unpalatable evergreen dwarf shrubs and higher plant N content both at species and at community level. Changes in plant community composition, in turn, explained variation in microbial activity in litter and soil inorganic nutrient availability. We propose a new conceptual model with two distinct vole–plant–soil feedback pathways. Voles may drive local plant–soil feedbacks that either increase or decrease ecosystem process rates, in turn promoting heterogeneity in vegetation and soils across tundra landscapes.