John R. Cary

Interactive effects of sublethal nematodes and nutritional status on snowshoe hare vulnerability to predation

1. We studied the effect of parasitism on snowshoe hare (Lepus americanus) survival through a field experiment that reduced natural burdens of sublethal nematodes in a free-ranging hare population. We treated half the hares on each of six study areas year-round with an anthelminthic drug, and the other half with a placebo. Potential interactive effects of parasitism and nutrition were examined by supplementing the natural food supply on three of the six study areas during two winters. Survival was measured by radio-collaring a total of 612 hares with mortality-sensitive transmitters and monitoring survival daily between April 1991 and June 1993. 2. Overall, nematode burdens in hares were highest between March and October, and Obeliscoides cuniculi was the most abundant of the five species present. The proximate cause of 95% of mortalities (n = 318) during the study was predation. Predators killed hares with heavy burdens of O. cuniculi disporportionately during May-June, but burdens of the four other species (Nematodirus triangularis, Trichuris leporis, Dirofilaria scapiceps, Protostrongylus boughtoni) were similar between predator-killed hares and the live population. 3. During May-October 1991, the anthelminthic treatment did not affect hare survival, but during May-October 1992 survival of parasite-reduced hares was 2.4-times higher than that of controls. During November 1991-April 1992 survival of food supplemented hares was 21% higher than food-normal animals, but survival was similar during November 1992-April 1993. 4. The effect of parasite-reduction on hare survival was apparently contingent on overwinter food supplies between November 1991 and April 1992, with survival being highest in hares subjected to both treatments, intermediate in those receiving only food supplementation, and lowest in unfed hares. We conclude that the effect of sublethal nematode parasitism on hare survival can be important if interactive with predation, and that synergistic effects of parasitism and nutrition may affect hares if food availability is limited.

DO PARASITISM AND NUTRITIONAL STATUS INTERACT TO AFFECT PRODUCTION IN SNOWSHOE HARES

We evaluated the hypothesis that the effects of parasitism on animal body condition are accentuated when host nutritional status is compromised. In a field experiment we manipulated natural burdens of five nematode species via anthelmintic treatment in one- half the free-ranging snowshoe hares (Lepus americanus) on each of six study areas and supplemented overwinter food level on three areas, to assess whether treatments interacted synergistically. We predicted that the effects of parasitism would be aggravated in mal- nourished hares (i.e., hares not administered supplemental food), and that animals receiving both treatments would have (1) lower parasite burdens, and (2) higher body mass, fat storage, and reproduction, than those receiving only one or no treatments. Mean numbers of Trichuris leporis were lower in hares subjected to food supplementation, suggesting that nutritional status affected immunity to that species. Obeliscoides cuniculi, Nematodirus triangularis, Dirofilaria scapiceps, and Protostrongylus boughtoni numbers were not directly affected by supplementation. Body mass of parasite-normal controls (i.e., not administered the anthelmintic treatment) was negatively correlated to nematode abundance during May- June, implying that parasitism had a deleterious effect on hare mass at that time. However, the anthelmintic treatment failed to directly affect hare body mass, fat storage, or repro- duction. Food supplementation had a positive effect on body mass, fat storage, and repro- duction, although effects of food were neither strong nor consistent among study areas and body condition parameters being measured. Only 2 of 22 analyses of hare production yielded marginally significant (i.e., 0.05 < P < 0.10) interactions between food and parasitism; therefore, we concluded that the two factors were not strongly linked. However, because our snowshoe hare population was at low densities and likely exposed to neither severe food shortage nor high rates of nematode parasitism, failure to detect an interaction between treatments may not comprise a general refutation of the hypothesis. Rather, hare populations subject to more serious food limitation (i.e., during a cyclic peak in numbers) may be apt to display stronger interactions between nutrition and parasitism.

PARASITISM IN A DECLINING POPULATION OF SNOWSHOE HARES

Prevalence and intensity of six endoparasites were determined in 346 snowshoe hares (Lepus americanus) obtained at Rochester, Alberta, during December–April 1981–1982, the second winter of a cyclic population decline. The data were analyzed for (1) differences among host sex and age classes, and among months and sample sources, and (2) evidence that parasitism was of demographic significance to the hare population. Prevalence and intensity of Obeliscoides cuniculi were consistently highest among adult hares, but rose most sharply from February to March among juveniles. In contrast, prevalence and intensity of Nematodirus triangularis were highest among juveniles; prevalence reached 90–100% by January, whereas intensity continued to rise through April. Prevalence and intensity of both Trichuris leporis and Protostrongylus boughtoni were highest also among juvenile hares; neither parameter exhibited a definite trend over time. Prevalences of Taenia pisiformis (cysticerci) and Eimeria spp. were unrelated to sex, age or month; but Taenia intensity was highest among juveniles, and Eimeria intensity tended to decrease from December to April. Intensities of Nematodirus, Protostrongylus and Eimeria were higher in male hares than in females. Prevalence and intensity were correlated directly in Obeliscoides, Nematodirus, Trichuris and Eimeria. Hares that died during trapping and handling, or from natural predation, had greater intensities of Obeliscoides than did animals killed on purpose. There was no indication, however, that risk of death was increased by the other parasitic infections. Age-related immune responses to parasitism (except Obeliscoides) were evidenced by reduced or stabilized prevalence and/or intensity among older hares. A multiple-regression model predicted depressed body weight with increasing intensities of Nematodirus, Trichuris or Protostrongylus. Other body-condition and reproductive indices were unassociated with parasite intensities. Within the hare population, Obeliscoides, Trichuris, Protostrongylus and Taenia had overdispersed distributions (typical of many endoparasites) that did not differ from a negative binomial. The frequency with which each possible combination of helminth species occurred within individual hares was consistent with the assumption that such infections occurred independently. There was no compelling reason to believe parasitism was a significant factor in the overwinter decline of this population of snowshoe hares.

PREVALENCE OF HELMINTHS IN A CYCLIC SNOWSHOE HARE POPULATION

Five species of helminths were monitored in a population of snowshoe hares (Lepus americanus) near Rochester, Alberta, during 1961-1977. Prevalence of both Obeliscoides cuniculi and Protostrongylus boughtoni among young hares averaged about 50% by age 2 mo, then tended to level off. Prevalence of Taenia pisiformis (cysticerci) and Dirofilaria scapiceps rose more slowly, but continued to increase steadily beyond their mean levels of 8% and 1% at age 2 mo. There were well denned seasonal (within-year) cycles in prevalence of O. cuniculi and P. boughtoni that were generated evidently to a major degree by arrested development of larvae in fall and renewed development in late winter. It was hypothesized that renewed larval development was triggered (in February) in O. cuniculi by the seasonal rise of circulating pituitary gonadotropins, and (in April) in P. boughtoni by the seasonal rise of gonadal androgens and estrogens. Indices to gonadal hormone levels in hares indicated that these increased most rapidly among males, and may have accounted for the higher prevalences of P. boughtoni in males during April-May. Neither T. pisiformis nor D. scapiceps exhibited conspicuous seasonal changes in prevalence. Maximum prevalence of T. pisiformis was attained at about 1 yr of age, whereas D. scapiceps increased among adult snowshoes through age 2 yr before stabilizing. Long-term (between-year) changes in prevalence of O. cuniculi, T. pisiformis, and D. scapiceps were correlated significantly with the cyclic hare population which declined from a peak in fall 1961 to a low in 1965–1966, rose to another peak by fall 1970, and declined again to a low in 1975. There was no detectable time lag between this “10-yr” cycle in hare density and the cycles of parasite prevalence among juveniles (<1 yr of age). Among adult hares, the cycle of O. cuniculi prevalence was likewise synchronous with that of the hare population, but the cycles of D. scapiceps and T. pisiformis lagged by approximately 1 and 2 yr, respectively. This lag in T. pisiformis prevalence was largely inexplicable to us. Our data on P. boughtoni were not suitable for analyses of between-year trends; nor were those for the fifth helminth, Taenia serialis (coenuri), because mean prevalence was less than 1% among both juveniles and adults. An apparent decline in T. serialis after the early 1950s, and its continued scarcity thereafter, paralleled a major change in numbers of one important definitive host—the red fox (Vulpes vulpes). Lighter weights of young hares at age 37–96 days, and of adults and fully grown juveniles, were associated with P. boughtoni infections. There was no demonstrable relationship between snowshoe hare reproductive parameters and helminth parasitism.

Mustelid, Squirrel, and Porcupine Population Trends during a Snowshoe Hare Cycle

Rates of capture in live traps set for snowshoe hares ( Lepus americanus ) were used to index population trends of four mustelids ( Mustek frenata, M. erminea, M. vision, Mephitis mephitis ), three sciurids ( Tamiasciurus hudsonicus, Glaucomys sabrinus, Spermophilus franklinii ), and the porcupine ( Erethizon dorsatum ) at Rochester, Alberta, during 1965–1975. The Mustela species (weasels and mink) had a 10-year cycle synchronous with that of the snowshoe hare. About 81% of the annual variation in weasel numbers, as indexed by capture rates, was attributable to fluctuations of hares, voles ( Microtus and Clethrionomys ), and mice ( Peromyscus ). Onset of the cyclic decrease in hares was followed within 2 years by declines in sciurid, skunk ( Mephitis ), and porcupine populations that may have been a consequence of predator switching as hares became scarce.

Response of a Scarlet Macaw Ara macao population to conservation practices in Costa Rica

Summary The Central Pacific Conservation Area contains one of Costa Rica’s two viable, but threatened Scarlet Macaw Ara macao populations. For 14 years (1990–2003), we monitored the size of this population. Non-linear models fitted to 1990–1994 observations revealed seasonal and longterm changes in population size. The peak of annual population size occurred in August, with a cyclic range of about 90 birds between the lowest and highest points of the annual cycle. The best model also revealed a decline in population size of approximately eight birds counted per year or 4% of the total population per year (1990–1994). Young-to-adult ratios calculated for the month of August during this study fluctuated around a baseline that averaged 6.1% a year (1990–2003). These ratios exceeded 8% for three different years (1995, 1996 and 2000). All three of these “good” recruitment years occurred after management began, and two of them were associated with zealous anti-poaching efforts that ultimately could not be sustained. After intensive management practices began in 1995, the August counts increased by about 37 individuals in two years (1995–1996) to an average 243 individuals, which remained almost constant up to 2003. Management practices included creation of a local conservation organization that coordinated environmental education, artificial nest construction, networking among stakeholders and with governmental authorities, and artificial and natural nest protection. Although Scarlet Macaw conservation efforts have been inconsistent since 1997, our study demonstrates that collaborative conservation by local stakeholders increased the population of this threatened species in 1995–1996, and from 1996 to present the population has sustained itself.

LANDSCAPE FEATURES OF RED-TAILED HAWK NESTING HABITAT IN AN URBAN/SUBURBAN ENVIRONMENT

ABSTRACT We described urban/suburban Red-tailed Hawk (Buteo jamaicensis) nesting habitat and compared nesting habitat to unused habitat based on presence and absence of Red-tailed Hawks. We developed a landscape-scale logistic model of nesting habitat occupancy, then applied it to other locations to determine whether unoccupied patches of Red-tailed Hawk nesting habitat existed in urban locations. Red-tailed Hawk nesting habitat in urban/suburban Milwaukee usually included large areas of grassland and other herbaceous cover types. Urban/suburban Red-tailed Hawk nesting habitat was comprised of more than three times as much grasslands and woodlands, and had greater land-cover diversity and patch richness than unused habitat. Characteristics of unused habitat indicated that Red-tailed Hawks avoided areas of heaviest urbanization, perhaps because of insufficient hunting habitat. The logistic regression model developed from our data demonstrated that suitable, unoccupied nesting habitat existed in this urban area. As the Red-tailed Hawk population expanded into urban locations in this study area, the birds apparently were adjusting well to urbanization. Additional studies of urban raptor populations may provide valuable insight into future management considerations for wildlife in human-influenced landscapes.

Discrimination of mallard strains on the basis of morphology

We detected differences in morphology among strains of wild, urban, and game-farm mallards (Anas platyrhynchos) from the Mississippi Flyway. A 3-strain discriminant function analysis (DFA) model (to discriminate 3 strains) and a 2-strain DFA model (to discriminate wild mallards from urban mallards, the strain that most closely resembled wild mallards) were developed to predict group memberships on the basis of 6 morphological variables. We used a stepwise procedure to develop a reduced 2-strain model that discriminated wild mallards on the basis of 3 variables : tarsus width, total tarsus length, and culmen length. Accuracy of the models ranged from 87.3 to 96.4% for males and 95.7 to 100% for females. Canonical coefficients are provided for use with unstandardized measurements to classify mallards

INTERACTION OF THE TICK (HAEMAPHYSALIS LEPORISPALUSTRIS) WITH A CYCLIC SNOWSHOE HARE (LEPUS AMERICANUS) POPULATION

Prevalence and intensity of the tick, Haemaphysalis leporispalustris, were monitored during 1963 to 1976 in a cyclic snowshoe hare (Lepus americanus) population near Rochester, Alberta, Canada. Prevalence was near zero from December through March, and near 100% among adult hares from May through September. Prevalence among juvenile hares approached 100% by age 2 mo. Intensity peaked for both adults and juveniles during May–June and again in August. Mean intensities were significantly higher among adult males than adult females in 5 of 13 yr, and almost significant in two others. Tick intensities were lowest during 3 yr, 1969 to 1971, when hare densities were highest. Tick intensities in spring were correlated with intensities the previous fall. Survival of marked adult and juvenile hares was unrelated to intensities of infestation. Mean numbers of corpora lutea and embryos tended to be lower among adult females with heavy tick infestations, and intra-uterine losses rose steadily from about 3 to 13% as tick intensities increased from none to heavy. Comparison of average tick intensities on adults 1-, 2-, and ≥3-yr-old yielded no evidence of increased immunity with age.

Dispersal of Ruffed Grouse: A Large-Scale Individual-Based Model

We present an efficient data structure enabling us to model, in a microcomputer environment, as many as 10 000 individual ruffed grouse on a landscape subdivided into 9216 contiguous, 10-acre (4 ha) sites. The model is fully stochastic both in a demographic and in an environmental sense. It features use of geographic information about habitat quality and hunting pressure to define site-specific natural and hunting mortality. It combines factors describing the time-specific social landscape, i.e., local density constraints and other social interactions, together with habitat quality, into a site-suitability function. This information is then used to guide the individual movements that in aggregate simulate the dispersal process. Values for parameters controlling the dispersal submodel were determined interactively by comparing model predictions of dispersal distance and duration with actual research findings from a study conducted in central Wisconsin. Additional geographic inputs to the model specify the starting distribution of grouse on the landscape and identify potential metapopulations (habitat islands) and other geographic units such as parcels of land ownership or study areas. Storage of the population state of each subarea enables later analyses of population dynamics in terms of patch attributes. Complete life histories of the individuals in the model are available as model outputs, as are snapshots of the distribution of grouse on the landscape at any user-specified time.