Climate-induced changes in the small mammal communities of the Northern Great Lakes Region

We use museum and other collection records to document large and extraordinarily rapid changes in the ranges and relative abundance of nine species of mammals in the northern Great Lakes region (white-footed mice, woodland deer mice, southern red-backed voles, woodland jumping mice, eastern chipmunks, least chipmunks, southern flying squirrels, northern flying squirrels, common opossums). These species reach either the southern or the northern limit of their distributions in this region. Changes consistently reflect increases in species of primarily southern distribution (white-footed mice, eastern chipmunks, southern flying squirrels, common opossums) and declines by northern species (woodland deer mice, southern red-backed voles, woodland jumping mice, least chipmunks, northern flying squirrels). White-footed mice and southern flying squirrels have extended their ranges over 225 km since 1980, and at particularly well-studied sites in Michigan's Upper Peninsula, small mammal assemblages have shifted from numerical domination by northern species to domination by southern species. Repeated resampling at some sites suggests that southern species are replacing northern ones rather than simply being added to the fauna. Observed changes are consistent with predictions from climatic warming but not with predictions based on recovery from logging or changes in human populations. Because of the abundance of these focal species (the eight rodent species make up 96.5% of capture records of all forest-dwelling rodents in the region and 70% of capture records of all forest-dwelling small mammals) and the dominating ecological roles they play, these changes substantially affect the composition and structure of forest communities. They also provide an unusually clear example of change that is likely to be the result of climatic warming in communities that are experienced by large numbers of people.     

Effects of Logging Pattern and Intensity on Squirrel Demography

ABSTRACT We examined the effect of harvesting intensity and pattern on red squirrels (Tamiasciurus hudsonicus), northern flying squirrels (Glaucomys sabrinus), and yellow-pine chipmunks (Tamias amoenus) in mature inland Douglas-fir (Pseudotsuga menziesii glauca) forests in south-central British Columbia, Canada. We sampled squirrels 1 year before harvesting through 4 years after harvesting and estimated population parameters using open-population models. Relative to unharvested stands, each of the 3 species showed a strong response to tree removal. From 2 years to 4 years after logging, red squirrel density was 40% (SE = 7.1) lower in stands with 50% basal-area tree removal. From 1 year and up to 4 years after logging, northern flying squirrel density averaged 60% (SE = 5.2) lower in harvested treatments regardless of intensity or pattern of logging. In contrast, density of yellow-pine chipmunks increased markedly with increased logging intensity. Beginning 3 years after logging, yellow-pine chipmunk density was 734% (SE = 269) greater in stands with 50% basal-area tree removal. In the short term, harvesting intensity was a more important determinant of squirrel density than harvesting pattern. Retaining >10 m² per ha of live residual stand structure in mature inland Douglas-fir forests maintained habitat for forest-dependent species such as red squirrels and northern flying squirrels, albeit at lower densities.     

Fleas community in introduced Siberian chipmunks (Tamias sibiricus Laxmann) in Forest of Sénart, France

We examined the fleas community in an introduced population of Siberian chipmunks, Tamias sibiricus, between 2005 and 2007, in the Forest of Sénart (Essonne, France). We collected and identified 383 fleas on 463 chipmunks (total: 1,891 captures on 471 chipmunks). In 2005, 120 fleas were also collected on 65 bank voles, Clethrionomys glareolus, and on 25 wood mice, Apodemus sylvaticus, trapped within the same area. Ceratophyllus sciurorum sciurorum formed 73.6% of the chipmunks' flea community, with an annual prevalence (P) ranging between 8 and 13% and a mean intensity (I) ranging between 1.1 and 1.6 fleas per individual. Among the six other species infecting this Sciurid, Ctenophthalmus agyrtes impavidus constituted 17.2% (P: 1.6-2.2%; I: 1.1-2.6), and Megabothris turbidus 8.1% (P: 0.8-1.9%; I: 1.0-1.4) of the flea community, respectively. These last two species represented respectively 60.8% and 36.6% of the flea community on the bank vole and the wood mouse. Originated from Asia, chipmunks did not import any flea species to Sénart, probably because they were used as pets before their release in the wild. Abundance in C. s. sciurorum increased with adult chipmunk density and with juvenile density in summer. On adult chipmunks C. s. sciurorum tented to decrease with increasing abundance of red squirrels (Sciurus vulgaris]. Moreover, the two other flea species mainly infected young chipmunks during the fall, and their number was not related to chipmunk density. However, the distribution of species within the flea community became more balanced with increase juvenile chipmunk density. Overall, these results indicate that the close phyletic relationship between chipmunks and red squirrels contributed in the acquisition and the spread of fleas by chipmunks. Primary and secondary hosts densities, their habitat use, and more specifically burrowing activities and tree canopy use, also played a role in the spread of fleas on chipmunks.     

Experimental infection and horizontal transmission of Modoc virus in deer mice (Peromyscus maniculatus)

Deer mice (Peromyscus maniculatus) were inoculated with a sublethal dose of a field strain of Modoc virus to determine patterns of viral persistence, shedding, and transmission. Blood, serum, urine, fecal, and oral swab samples were collected at selected intervals until 63 days postinoculation (PI) after which lung, liver, spleen, kidney, and salivary glands were explanted. Viral assays were conducted by intracranial inoculations of suckling mice and antibody titers were determined by the micro-complement-fixation test. Viremias lasted for up to 4 days PI. Antibody titers were present by day 8 PI, peaked at day 13-20 PI, and persisted until day 63 PI. There was no evidence of viral shedding in urine, fecal, or oral swab samples. Virus was detected in explanted lungs only. In a separate experiment, deer mice were inoculated with virus and lungs were removed from five mice per wk for 10 wk. Indirect fluorescent antibody (IFA) techniques were used to determine the location of virus in lung tissue and to examine fixed tissue for lesions. IFA showed virus in lung parenchymal cells beginning 42 days PI and persisting at least 70 days PI. No histopathologic changes were seen. Horizontal transmission of the virus was studied by placing uninoculated mice with inoculated mice for 42 days and determining if the test animals developed antibodies or had virus in their lungs. Fifty-percent of the uninoculated mice developed antibody. One of these animals had virus in its lungs. Therefore, Modoc virus may be transmitted by direct contact.     

Diversity of Function of SCN Pacemakers in Behavior and Ecology of Three Species of Sciurid Rodents

Information on the widespread function of the suprachiasmatic (SCN) pacemaker has increased dramatically in the past decade for laboratory rodents and for humans. Interest is now also growing in the adaptive value of SCN pacemakers in wild species of free-living mammals in natural pristine habitat. The squirrel family, Sciuridae, is highly diverse in North America. Radiation into many specialized groups has evolved in response to the temporal and spatial characteristics of specific habitats. The sciurids thus offer possibilities for comparison of SCN function for closely related species from very different habitats. Results from field and laboratory investigations concerning the ecological significance of the SCN are reported here for three ground squirrel species. The semi-fossorial antelope squirrels of the American southwest deserts were arrhythmic in above ground activities after deletion of the SCN pacemaker. In a desert enclosure, predation rates rose dramatically for lesioned animals in comparison to intact controls. The semi-arborial chipmunks were prone to capture by mustelid burrow predators such as weasels after SCN-deletion, even though the chipmunks did not exhibit any above ground activity at night. The telltale cue for the predator was apparently arrhythmic restlessness within the complex home burrow. The semi-fossorial golden-mantled squirrels of the Cascade Mountains of western United States rely on profound, long-lasting hibernation for over-wintering. Periods of torpor alternate with brief euthermic arousal episodes lasting less than 24 h throughout the 5- to 8-month winter phase. Deletion of the SCN in golden-mantled squirrels resulted in aberrations of hibernation arousal patterns that were metabolically expensive and put individuals at risk of starvation immediately after emergence in spring.     

Where's my dinner? Adult neurogenesis in free‐living food‐storing rodents

Postnatal hippocampal neurogenesis in wild mammals may play an essential role in spatial memory. We compared two species that differ in their reliance on memory to locate stored food. Yellow-pine chipmunks use a single cache to store winter food; eastern gray squirrels use multiple storage sites. Gray squirrels had three times the density of proliferating cells in the dentate gyrus (determined by Ki-67 immunostaining) than that found in chipmunks, but similar density of young neurons (determined by doublecortin immunostaining). Three explanations may account for these results. First, the larger population of young cells in squirrels may increase the flexibility of the spatial memory system by providing a larger pool of cells from which new neurons can be recruited. Second, squirrels may have a more rapid cell turnover rate. Third, many young cells in the squirrels may mature into glia rather than neurons. The densities of young neurons were higher in juveniles than in adults of both species. The relationship between adult age and cell density was more complex than that has been found in captive populations. In adult squirrels, the density of proliferating cells decreased exponentially with age, whereas in adult chipmunks the density of young neurons decreased exponentially with age.     

An Experimental Test of Competition among Mice,Chipmunks, and Squirrels in Deciduous Forest Fragments

Mixed hardwood forests of the northeast United States support a guild of granivorous/omnivorous rodents including gray squirrels (Sciurus carolinensis), eastern chipmunks (Tamias striatus), and white-footed mice (Peromyscus leucopus). These species coincide geographically, co-occur locally, and consume similar food resources. Despite their idiosyncratic responses to landscape and patch variables, patch occupancy models suggest that competition may influence their respective distributions and abundances, and accordingly their influence on the rest of the forest community. Experimental studies, however, are wanting. We present the result of a large-scale experiment in which we removed white-footed mice or gray squirrels from small, isolated forest fragments in Dutchess County, New York, and added these mammals to other fragments in order to alter the abundance of these two species. We then used mark–recapture analyses to quantify the population-level and individual-level effects on resident mice, squirrels, and chipmunks. Overall, we found little evidence of competition. There were essentially no within-season numerical responses to changes in the abundance of putative competitors. Moreover, while individual-level responses (apparent survival and capture probability) did vary with competitor densities in some models, these effects were often better explained by site-specific parameters and were restricted to few of the 19 sites we studied. With only weak or nonexistent competition among these three common rodent species, we expect their patterns of habitat occupancy and population dynamics to be largely independent of one another.     

Serological evidence of bovine herpesvirus 1-related virus infection in the white-tailed deer population on Anticosti Island, Quebec

High white-tailed deer (Odocoileus virginianus) population densities and the occurrence of harsh environmental conditions are present on Anticosti Island, located in the Gulf of Saint-Lawrence (Quebec, Canada). This island is the northernmost region of white-tailed deer distribution in northeastern North America. The aim of this work was to determine whether a herpesvirus serologically related to bovine herpesvirus 1 (BHV1) may occur in a stressed white-tailed deer population. One hundred one deer sera were collected from apparently healthy animals during the hunting season from September to late November 1985. Fifty-three percent of tested deer were positive to the seroneutralization test using Colorado strain of BHV1 virus. Higher percentages of seropositivity were observed in animals of both sexes greater than 4-yr-old. Analysis of antibody titers in seropositive animals according to age suggests that BHV1-related viral infection is endemic in the Anticosti Island deer population. It is postulated that environmental stress may induce immunosuppression of certain infected and/or carrier animals in their population that shed virus for long periods of time.     

Seasonal glucocorticoid responses to capture in wild free-living mammals

We determined baseline and capture-induced glucocorticoid concentrations during two different seasons in three species of wild free-living rodents: brown lemmings (Lemmus trimucronatus), golden-mantled ground squirrels (Spermophilus saturatus), and yellow-pine chipmunks (Tamias amoenus). Initial blood samples were obtained within 3 min of capture, so that initial glucocorticoid levels reflect baseline titers of undisturbed animals. Animals were held for an additional 30 min, when a second blood sample was taken to measure stress-induced glucocorticoid titers. The primary glucocorticoid differed in each species. Lemmings secreted extremely large amounts of corticosterone (as high as 8,000 ng/ml). These high concentrations were accompanied by high corticosterone-binding globulin capacity and resistance to negative feedback. Squirrels and chipmunks secreted a mixture of cortisol and corticosterone (10-400 ng/ml). In males of all three species and female squirrels and chipmunks, glucocorticoid levels were significantly elevated 30 min after capture. Baseline and 30-min glucocorticoid levels differed seasonally in each species. Levels were higher during summer (with no snow cover) than in spring (with approximately 60% snow cover) in female lemmings, higher during breeding than before hibernation in squirrels, and higher postreproductively than during breeding in chipmunks. Together, these data indicate that glucocorticoid responses to stress in these free-living species are similar to those in laboratory species, but the magnitude of the response appears to depend on life-history features specific to each species.     

Experimental infection of nontarget species of rodents and birds with Brucella abortus strain RB51 vaccine

The Brucella abortus vaccine strain RB51 (SRB51) is being considered for use in the management of bnucellosis in wild bison (Bison bison) and elk (Cervus elaphus) populations in the Greater Yellowstone Area (USA). Evaluation of the vaccines safety in non-target species was considered necessary prior to field use. Between June 1998 and December 1999, ground squirrels (Spermophilus richardsonii, n = 21), deer mice (Peromyscus maniculatus, n = 14), prairie voles (Microtus ochrogaster, n = 21), and ravens (Corvus corax, n = 13) were orally inoculated with SRB51 or physiologic saline. Oral and rectal swabs and blood samples were collected for bacteriologic evaluation. Rodents were necropsied at 8 to 10 wk and 12 to 21 wk post inoculation (PI), and ravens at 7 and 11 wk PI. Spleen, liver and reproductive tissues were collected for bacteriologic and histopathologic evaluation. No differences in clinical signs, appetite, weight loss or gain, or activity were observed between saline- and SRB51-inoculated animals in all four species. Oral and rectal swabs from all species were negative throughout the study. In tissues obtained from SRB51-inoculated animals, the organism was isolated from six of seven (86%) ground squirrels, one of six (17%) deer mice, none of seven voles, and one of five (20%) ravens necropsied at 8, 8, 10, and 7 wk PI, respectively. Tissues from four of seven (57%) SRB51-inoculated ground squirrels were culture positive for the organism 12 wk PI; SRB51 was not recovered from deer mice, voles. or ravens necropsied 12, 21, or 11 wk, respectively, PI. SRB51 was not recovered from saline-inoculated ground squirrels, deer mice, or voles at any time but was recovered from one saline-inoculated raven at necropsy, 7 wk PI, likely attributable to contact with SRB51-inoculated ravens in an adjacent aviary room. Spleen was time primary tissue site of colonization in ground squirrels, followed by the liver and reproductive organs. The results indicate oral exposure to SRB51 does not produce morbidity or mortality in ravens, ground squirrels, deer mice, or prairie voles.     

Small-Mammal Microhabitat Associations and Response to Grazing in Oregon

Abstract: Livestock grazing is common and widespread throughout North America, yet few studies have evaluated its effects on small mammals. We studied small mammals in mixed-conifer forests and oak woodlands on the Cascade-Siskiyou National Monument in southern Oregon, USA, to 1) evaluate small-mammal microhabitat associations, 2) identify riparian-associated species, and 3) test the hypothesis that grazing does not influence small mammals after accounting for microhabitat associations. We live-trapped small mammals at 16 study sites and used logistic regression to model probability of capture on measured habitat characteristics at each trap station and to evaluate effects of grazing. Over 2 years, we trapped 1,270 individual small mammals representing 18 species. Odds of capturing western harvest mice (Reithrodontomys megalotis), dusky-footed woodrats (Neotoma fuscipes), and long-tailed voles (Microtus longicaudus) were lower (P < 0.05) on heavily versus lightly grazed sites. Odds of capture for deer mice (Peromyscus maniculatus) were lower (P < 0.05) on heavily versus lightly grazed sites in woodlands, but there was less difference in the odds of capture between grazing intensities in conifer forests. Odds of capturing Townsend's vole (Microtus townsendii) were lower on heavily versus lightly grazed riparian areas. Western harvest mice, long-tailed voles, and Townsend's voles were associated with, but not obligated to, riparian areas. Deer mice were ubiquitous, but captures were also higher (P < 0.05) in riparian areas compared with uplands. Siskiyou chipmunks (Tamias siskiyou) and piñon mice (Peromyscus truei) were associated with uplands (P < 0.05) rather than riparian areas. Trowbridge's shrews (Sorex trowbridgii), Siskiyou chipmunks, and bushy-tailed woodrats (Neotoma cinerea) were positively associated with coarse woody debris. Land managers should anticipate that small mammals associated with herbaceous or shrub cover, particularly in riparian areas, will decline when cattle remove this cover.     

Photoperiodic control of circadian activity rhythms in diurnal rodents

The responses of red squirrels(Tamiasciurus hudsonicus) and eastern chipmunks(Tamias striatus) to complete and skeleton light-dark (LD) cycles were compared. The skeletons, comprised of two 1-h pulses of light per day, effectively simulated the complete photoperiods in the squirrels, but not the chipmunks. Skeleton photoperiods greater than 12-h caused the chipmunks to shift activity from the longer to the shorter of the two intervals between the pulses. To interpret the mechanism of phase control, squirrels and chipmunks were kept in continuous darkness and exposed to 1-h light pulses every 10 days. The time-course of entrainment was also quantified. Both techniques produced light-response curves. The data suggest that the parametric and non-parametric contributions to entrainment are different in these two rodent species.Presented at the Eighth International Congress of Biometeorology, 9–14 September 1979, Shefayim, Israel.     

Bovine virus diarrhea and the vector-borne diseases Anaplasmosis and Bluetongue: a sero-surveillance in free-ranging red deer (<Emphasis Type="Italic">Cervus elaphus</Emphasis>) in selected areas of Switzerland

Due to climate changes, diseases emerging from southerly adjacent areas (Mediterranean countries) are likely to spread northward. Expanded migration of red deer harbors the risk of introducing new pathogens into a naive population of either wild or domestic animals. Little is known about the importance of red deer as a potential reservoir for diseases of domestic ruminants in Switzerland. Deer is susceptible for all three agents that were selected in this study: bovine virus diarrhea virus (BVDV), Anaplasma marginale (AM), and Bluetongue virus (BTV). The goal of this project was to establish the serological status of red deer in Switzerland concerning BVDV, AM, and BTV, and to assess the possible impact of disease dynamics with a focus on potential transmission of these diseases from red deer to cattle or vice versa. Sampling areas were selected according the following criteria: abundance of red deer, potential insect vector distribution due to climatic conditions, and traditional alpine pasture husbandry along with known migration routes of red deer. Blood samples were collected during the regular hunting season 2004 and 2005 by hunters and gamekeepers. There was no serological evidence for the presence of the vector-borne diseases AM and BT in red deer in Switzerland. Four out of 234 sera showed a positive result for BVD, corresponding to a sero-prevalence of 1.7% (95% CI 0.46–4.38). Facing the fact of the high sero-prevalence for BVD in Swiss cattle (60–80%) disease transmission from red deer to cattle in these areas under investigation is rather unlikely.     

Nest predation in forest birds: influence of predator type and predator's habitat quality

We used the introduction of a generalist nest predator, the red squirrel Tamiasciurus hudsonicus, and of a large herbivore, the Sitka black-tailed deer Odocoileus hemionus sitkensis, to the islands of Haida Gwaii (Queen Charlotte Islands, British Columbia, Canada) to study how predator assemblage and habitat quality and structure influenced nest predation in forest birds. We compared losses of natural nests to predators on islands with and without squirrels. We selected nine islands with or without squirrel or deer and used 506 artificial nests put on the ground or in shrubs to further analyse variation of nest predation with predator assemblage and habitat quality for the predators. For both natural and artificial nests predation risk was higher in presence of squirrels. But predation risk varied within island categories. In presence of squirrels it was highest in stands with mature conifers where it fluctuated from year to year, in response to fluctuations in squirrel abundance. Vegetation cover around the nest had little effect on nest predation by squirrels. Where squirrels were absent, nest predation concentrated near predictable food sources for corvids, the main native predators, and increased with decreasing vegetation cover, suggesting that removal of the vegetation by deer increased the risk of predation by native avian nest predators that use visual cues. Predation risk in these forests therefore varies in space and time with predator composition and with quality of the habitat from the predators' perspective. This temporal and spatial variation in predation risk should promote trade-offs in the response of birds to nest predation, rather than fine-tuned adaptations to a given predation pattern.     

Preparing for winter: divergence in the summer-autumn hematological profiles from representative species of the squirrel family

We examined hematological parameters in four related sciurid species in the late summer-autumn to assess the role of habitat, elevation, body size, and behavior in shaping these parameters. Red squirrels (Tamiasciurus hudsonicus) and Arctic ground squirrels (Spermophilus parryii) were sampled in southwestern Yukon, yellow-pine chipmunks (Tamias amoenus) in southern Alberta, and the eastern grey squirrel (Sciurus carolinensis) in southern Ontario. We obtained whole blood samples from each species and compared glucose levels, red blood cell characteristics (hematocrit, red blood cell count, hemoglobin concentration, mean corpuscular volume, mean corpuscular hemoglobin, and mean corpuscular hemoglobin concentration), and white blood cell counts (neutrophils, lymphocytes, monocytes, eosinophils, basophils) across species. We found species differences in glucose and red blood cell characteristics that may be a function of activity levels, phylogeny, or elevation, but not of body size, body condition, or adaptations to a semi-fossorial habitat. We also found species differences in white blood cell counts that remain unexplained by any single simple explanation and may be more useful for comparison of individuals within a given species than for interspecies comparisons.     

Prevalence of Neospora caninum and Toxoplasma gondii antibodies in wild ruminants from the countryside or captivity in the Czech Republic

In the Czech Republic, sera from 720 wild ruminants were examined for antibodies to Neospora caninum by screening competitive-inhibition enzyme-linked immunosorbent assay and confirmed by indirect fluorescence antibody test (IFAT); the same sera were also examined for antibodies to Toxoplasma gondii by IFAT. Neospora caninum antibodies were found in 14% (11 positive/79 tested) roe deer (Capreolus capreolus), 14% (2/14) sika deer (Cervus nippon), 6% (24/ 377) red deer (Cervus elaphus), 1% (2/143) fallow deer (Dama dama), 3% (3/105) mouflon (Ovis musimon), and none of 2 reindeer (Rangifer tarandus). Toxoplasma gondii antibodies were found in 50% (7/14) sika deer, 45% (169/377) red deer, 24% (19/79) roe deer, 17% (24/143) fallow deer, 9% (9/105) mouflon, and 1 of 2 reindeer. In 42 samples of wild ruminants that tested positive for N. caninum antibodies, 28 (67% of the positive N. caninum samples) reacted solely to N. caninum. This is the first evidence of N. caninum infection in mouflon, the first N. caninum seroprevalence study in farmed red deer, and the first survey of N. caninum in wild ruminants from the Czech Republic.     

Enzootic Babesia microti in Maine

Human babesiosis in the northeastern United States caused by Babesia microti (Apicomplexa: Piroplasmida) is mainly reported from coastal New England sites, where deer ticks (Ixodes dammini) are common. However, the piroplasm has been detected in microtine rodents elsewhere in association with I. angustus or other nidicolous ticks, suggesting that the agent is widely distributed but zoonotically significant only where a human-biting "bridge" vector is present. To determine whether this piroplasm may be enzootic in areas where I. dammini is absent, we surveyed small mammals collected from 2 sites in Maine, where I. angustus or I. muris is common but I. dammini is not. Of 43 chipmunks, voles, deer mice, and shrews examined, 3 (6.9, 95% confidence interval 0 to 14.5) were parasitemic, as determined by blood smear or polymerase chain reaction targeting a piroplasm-specific portion of the 18S ribosomal DNA gene. Phylogenetic analysis of the sequenced amplification products demonstrates the presence of 2 forms of B. microti. We conclude that B. microti may be enzootic in the absence of I. dammini but that human risk relates to dense infestations of this human-biting tick.     

Red Squirrel Middens Influence Abundance but Not Diversity of Other Vertebrates

Some animals modify the environment in ways that can influence the resources available to other species. Because red squirrels (Tamiasciurus hudsonicus) create large piles of conifer-cone debris (middens) in which they store cones, squirrels concentrate resources that might affect biodiversity locally. To determine whether other animals are attracted to midden sites beyond their affinity for the same resources that attract red squirrels, we assessed associations between middens, mammals, and birds at population and community levels. We surveyed 75 middens where residency rates of red squirrels varied during the previous five years; sampling along this residency gradient permitted us to evaluate the influence of resources at middens beyond the influence of a resident squirrel. At each location, we quantified vegetation, landscape structure, abundance of conifer cones, and midden structure, and used capture–recapture, distance sampling, and remote cameras to quantify presence, abundance, and species richness of mammals and birds. Red squirrels and the resources they concentrated at middens influenced mammals and birds at the population scale and to a lesser extent at the community scale. At middens with higher residency rates of red squirrels, richness of medium and large mammals increased markedly and species richness of birds increased slightly. After accounting for local forest characteristics, however, only species richness of medium-to-large mammals was associated with a red squirrel being resident during surveys. In areas where red squirrels were resident during surveys or in areas with greater amounts of resources concentrated by red squirrels, abundances of two of four small mammal species and two of four bird species increased. We conclude that the presence of this ecosystem modifier and the resources it concentrates influence abundance of some mammals and birds, which may have implications for maintaining biodiversity across the wide geographic range inhabited by red squirrels and other larderhoarding animals.     

Influence of vegetation structure on the small mammal community in a shortgrass prairie ecosystem

The structure of vegetation, and how this structure varies across a landscape, is crucial to understanding the distribution of wildlife species. Between 2002 and 2004, we sampled small mammal communities and measured vegetation structure at 185 locations across a range of disturbance regimes in a shortgrass prairie ecosystem in southeastern Colorado, USA. At each sampling location, the local disturbance regime was some combination of varying intensity of livestock grazing, military training activity, and fire. Vegetation structural characteristics measured included percent bare ground, basal cover, litter, shrub density, and mean grass and shrub height. Rodent communities were described by richness, diversity, total and per capita biomass, and species abundances. Northern grasshopper mice (Onychomys leucogaster), Ord's kangaroo rats (Dipodomys ordii), silky pocket mice (Perognathus flavus), western harvest mice (Reithrodontomys megalotis), white-footed mice (Peromyscus leucopus), southern plains wood rats (Neotoma micropus), thirteen-lined ground squirrels (Spermophilus tridecemlineatus), deer mice (Peromyscus maniculatus), and spotted ground squirrels (Spermophilus spilosoma) accounted for >99 % of all captures. Canonical correlation analysis was used to assess the relationship between small mammals and vegetation structure. The first two canonical variates explained over 50 % of the variation in vegetation structure and were related to the ratio of bare ground to basal coverage and litter accumulation. Rodent community indices were most strongly related to litter accumulation and shrub density, though the models had low explanatory power. Our results agreed with published findings regarding microhabitat associations and indicated small mammal communities benefited from a system of interacting disturbances and the resulting landscape mosaic.