Alice J. Kenney

Limitation of collared lemming population growth at low densities by predation mortality

Populations of the collared lemming (Dicrostonyx kilangmiutak) and the tundra vole (Microtus oeconomus) have been at consistently low densities, and non-cyclic, at Pearce Point, Northwest Territories, Canada, for six yr. In most summers population densities decline, or only increase marginally, despite ongoing reproduction. We investigated the hypothesis that predation mortality is sufficient and necessary to curtail lemming population growth in summer. To test predictions of the hypothesis, we compared lemming demography, using mark-recapture and radiotelemetry, on a population from which predators were excluded (PE), using a perimeter fence and aerial mesh of nylon (11.4 ha), with demographies of three control populations (18-25 ha). Predation was the proximate cause of the majority of adult and neo-natal mortality, and was not replaced in a compensatory fashion by any other mortality factor in the PE. Significantly fewer adult lemmings died in the PE, and consequently survival inside the PE was significantly enhanced. Recruitment of juvenile lemmings to the adult population was significantly higher in the PE on a per unit area basis. The lemming population in PE followed a significantly different trajectory than the control populations in 1990 and 1991, remaining fairly stable while controls declined. However, the protected population did not grow, apparently because of juvenile dispersal: telemetered juveniles dispersed at an average rate of 53 m/d within the first ten d after weaning. We believe that the protected area was too small to encompass such dispersal, and that emigrants were not replaced by immigrating juveniles since the latter faced heavy mortality outside the exclosure. In 1992, numbers on PE and all controls grew, in conjunction with a regional absence of rough-legged hawks (Buteo lagopus) and a scarcity of red foxes (Vulpes vulpes), the two principal microtine predators. Tundra vole population growth was also limited by predation mortality, but to a lesser extent. We conclude that predation mortality is sufficient and necessary to limit summer population growth in these microtine species.

Terrestrial trophic dynamics in the Canadian Arctic

The Swedish Tundra Northwest Expedition of 1999 visited 17 sites throughout the Canadian Arctic. At 12 sites that were intensively sampled we estimated the standing crop of plants and the densities of herbivores and predators with an array of trapping, visual surveys, and faecal-pellet transects. We developed a trophic-balance model using ECOPATH to integrate these observations and determine the fate of primary and secondary production in these tundra ecosystems, which spanned an 8-fold range of standing crop of plants. We estimated that about 13% of net primary production was consumed by herbivores, while over 70% of small-herbivore production was estimated to flow to predators. Only 9% of large-herbivore production was consumed by predators. Organization of Canadian Arctic ecosystems appears to be more top-down than bottom-up. Net primary production does not seem to be herbivore-limited at any site. This is the first attempt to integrate trophic dynamics over the entire Canadian Arctic.

Synchrony in lemming and vole populations in the Canadian Arctic

Population fluctuations may occur in synchrony among several rodent species at a given site, and they may occur in synchrony over large geographical areas. We summarize information on synchrony in lemmings and voles from the Canadian Arctic for the past 20 years. The most detailed available information is from the central Canadian Arctic, where snap-trap samples have been taken annually at several sites for periods of up to 15 years. Geographical synchrony in the same species among different sites was strong, especially for the central and eastern Canadian Arctic. Synchrony among different species at a given site was also generally high. When one species is at high density, densities of all species at that site tend to be high. These results do not easily fit the mobile-predator hypothesis proposed to explain regional synchrony, and are more consistent with the weather hypothesis, which we suggest both entrains synchrony among sites and enforces synchrony among species within a site. We tentatively support the weather hypothesis for geographical synchrony in lemmings, and recommend the establishment of a circumpolar program to monitor lemming cycles and predator movements that would advance our understanding of these large-scale patterns of cyclic synchrony.

Expansion of Canopy-Forming Willows Over the Twentieth Century on Herschel Island, Yukon Territory, Canada

Canopy-forming shrubs are reported to be increasing at sites around the circumpolar Arctic. Our results indicate expansion in canopy cover and height of willows on Herschel Island located at 70° north on the western Arctic coast of the Yukon Territory. We examined historic photographs, repeated vegetation surveys, and conducted monitoring of long-term plots and found evidence of increases of each of the dominant canopy-forming willow species (Salix richardsonii, Salix glauca and Salix pulchra), during the twentieth century. A simple model of patch initiation indicates that the majority of willow patches for each of these species became established between 1910 and 1960, with stem ages and maximum growth rates indicating that some patches could have established as late as the 1980s. Collectively, these results suggest that willow species are increasing in canopy cover and height on Herschel Island. We did not find evidence that expansion of willow patches is currently limited by herbivory, disease, or growing conditions.

Movements of feral house mice in agricultural landscapes.

From September 1992 to May 1993 we radio-collared 155 house mice (Mus domesticus) on agricultural fields in southern Queensland to measure movements and to determine social organisation. During the breeding season most individuals were site-attached and home ranges of both sexes overlapped extensively. There was no sign of exclusive space use for breeding individuals. Breeding males had home ranges that were larger than those of breeding females (0.035 ha v. 0.015 ha), and moved about more. After breeding ended, home ranges increased over tenfold in area, and most mice became nomadic and not site-attached.

Patterns of predation on noncyclic lemmings

Noncyclic populations of microtine rodents may be limited within a relatively constant range of densities by generalist predators with a prey base sufficiently diverse to sustain them when rodents are scarce (generalist predator hypothesis). Collared lemmings (Dicrostonyx kilangmiutak) at Pearce Point, Northwest Territories, Canada, are noncyclic and limited to fairly constant low densities in summer by predation, principally by red fox and Rough-legged Hawks. We tested four predictions of the generalist predator hypothesis as a possible explanation for relatively constant lemming densities: (1) predators do not show strong numerical responses to lemming density; (2) the proportion of lemming biomass in predator diets declines with declining lemming abundance, compensated for by increased consumption of alternative prey; (3) predators show a type-III functional response to lem- ming density; and (4) at low densities, predation on lemmings ceases. The first prediction was not satisfied by the principal predators: at very low lemming densities, Rough-legged Hawks did not settle, and breeding success of red foxes and hawks was limited by lemming abundance. However, a number of generalist predators (Golden Eagle, grizzly bear, arctic ground squirrel, Peregrine Falcon, and Gyrfalcon) did not respond numerically to the lowest lemming densities. The second prediction was partly supported: all predators consumed lemmings at a lower rate as lemming densities declined. However, Rough-legged Hawks were not able to compensate fully for the declining consumption by increasing their use of alternative prey, and red foxes were able to do so in one of three years. Regarding the third prediction, foxes showed some evidence of a type-III functional response but hawks did not. As for the fourth prediction, most predators still consumed lemmings at very low densities; lemmings lacked a secure refuge. The Pearce Point system differs from those where microtine dynamics are relatively constant and nonirruptive because of persistent predation by generalists. In terms of biomass, lemmings are the principal prey for their dominant predators. These predators (the semi- generalist red fox and the specialist Rough-legged Hawk) rely on lemmings to breed, but drive them to densities too low to sustain breeding by these same predators in the subsequent spring. In this regard, the system is similar to one driven by specialists. In some winters, however, populations recover because lemmings breed under the snow and most summer predators are absent. As a result, lemming densities in spring are often high enough for specialists and semigeneralists to initiate breeding. When winter breeding and survival fail to allow population growth, hawks and foxes may fail to breed and then leave the system. Even so, summer generalists still persist and continue to consume lemmings, curtailing potential irruptive growth. In this regard, the system is similar to one where prey are relatively constant because of generalists. Community dynamics at Pearce Point can best be understood as a combination of three dominant processes. Summer predation by specialists and semigeneralists results in desta- bilizing declines. Winter breeding, coupled with good survival, can lead to destabilizing growth. However, this growth is curtailed in the following summer by either destabilizing specialist predation or the stabilizing influence of generalist predation. When lemmings are scarce, the semigeneralist red fox and some generalist predators rely on arctic ground squirrels as their primary prey or their principal alternative prey. The ground squirrel appears to be the critical species maintaining this relatively diverse arctic tundra predator community and the relatively constant lemming densities.

Population dynamics of the collared lemming and the tundra vole at Pearce Point, Northwest Territories, Canada

From 1987 to 1989 we monitored population changes during summer of the collared lemming (Dicrostonyx groenlandicus) and the tundra vole (Microtus oeconomus) at Pearce Point, Northwest Territories, Canada (69° 48′ N, 122° 40′ W). Populations on four study areas did not cycle but remained at low density (<3/ha) each year and continued at low numbers for the following 3 years (Reid et al. 1995). Lemming numbers often declined throghout the summer in spite of continous reproduction, and population recovery occurred overwinter. Heavy predation losses of radio-collared lemmings occurred each summer and this lemming population may be trapped in a predator-pit. Collared lemmings breed in winter and only because of winter population growth do these populations persist. Tundra vole numbers increased rapidly in most summers but usually declined overwinter. Tundra voles do not seem able to sustain winter reproduction in this extreme environment and this prevents them from reaching high density because of the short summer. Population growth in both these rodents could be prevented by poor food or by predation losses, and landscape patchiness may also help to prevent population growth. For lemmings we do not think that a shortage of shelter or intrinsic limitations could be restricting population increase at Pearce Point. This is the first detailed study of a non-cyclic collared lemming population.

Phylogeography and mitochondrial DNA (mtDNA) diversity in North American collared lemmings ( Dicrostonyx groenlandicus )

Variation in the nucleotide sequence of the mitochondrial control region (250 bp) and the cytochrome b region (870 bp) was examined in collared lemmings (Dicrostonyx groenlandicus) from 19 localities in northern Alaska and the Canadian Arctic. The division of D. groenlandicus in two phylogeographical groups with limited divergence across the Mackenzie River is consistent with the separation of this species in more than one refugial area located to the northwest of the Laurentide ice sheet during the last glaciation. Populations of D.groenlandicus from formerly glaciated areas are no less variable than those in nonglaciated areas. Instead, the low intrapopulation and intraregional diversity estimates in D. groenlandicus are probably a result of regional bottleneck events due to range contractions during Holocene warming events. These results are consistent with findings previously reported on collared lemmings (D. torquatus) from the Eurasian Arctic.

Do changes in berry crops drive population fluctuations in small rodents in the southwestern Yukon

Abstract Small mammals in boreal forest ecosystems fluctuate dramatically in abundance and 1 possible mechanism to explain these changes is the bottom-up hypothesis of variation in food supplies. Here we ask if variation in berry crops produced by 6 major species of dwarf shrubs and herbs, epigeous mushroom crops, and white spruce seeds allow us to predict changes in the abundance of the red-backed vole (Myodes [ =  Clethrionomys] rutilus), the deer mouse (Peromyscus maniculatus), and field voles (Microtus oeconomus and M. pennsylvanicus combined) over 13 years (1997–2009) in the Kluane Lake region of the southwestern Yukon, Canada. M. rutilus is the dominant rodent in these forests, comprising 64% of the catch. Overwinter survival is a key demographic variable in all these rodents, and the winter food supply—principally berries produced the previous summer—may be 1 key to overwinter survival. We predicted that berry, mushroom, and tree seed crops in year t would produce changes in rodent density in year t + 1. We could explain statistically 78–98% of the variation in May and August abundance of all 3 rodent species with indices of berry crops and mushrooms in the previous summer. For M. rutilus the critical predictor was berry crops of Empetrum nigrum. For P. maniculatus, the critical species were Arctostaphylos uva-ursi, A. rubra, and mushrooms. Spruce seed crops were not significantly correlated with rodent densities or changes in density. A large fraction of the variation in rodent numbers in this ecosystem is explained by a simple bottom-up model of population limitation.

Lemming winter habitat choice: a snow-fencing experiment

The insulative value of early and deep winter snow is thought to enhance winter reproduction and survival by arctic lemmings (Lemmus and Dicrostonyx spp). This leads to the general hypothesis that landscapes with persistently low lemming population densities, or low amplitude population fluctuations, have a low proportion of the land base with deep snow. We experimentally tested a component of this hypothesis, that snow depth influences habitat choice, at three Canadian Arctic sites: Bylot Island, Nunavut; Herschel Island, Yukon; Komakuk Beach, Yukon. We used snow fencing to enhance snow depth on 9-ha tundra habitats, and measured the intensity of winter use of these and control areas by counting rodent winter nests in spring. At all three sites, the density of winter nests increased in treated areas compared to control areas after the treatment, and remained higher on treated areas during the treatment. The treatment was relaxed at one site, and winter nest density returned to pre-treatment levels. The rodents’ proportional use of treated areas compared to adjacent control areas increased and remained higher during the treatment. At two of three sites, lemmings and voles showed significant attraction to the areas of deepest snow accumulation closest to the fences. The strength of the treatment effect appeared to depend on how quickly the ground level temperature regime became stable in autumn, coincident with snow depths near the hiemal threshold. Our results provide strong support for the hypothesis that snow depth is a primary determinant of winter habitat choice by tundra lemmings and voles.