Tanya M. Shenk

SAMPLING-VARIANCE EFFECTS ON DETECTING DENSITY DEPENDENCE FROM TEMPORAL TRENDS IN NATURAL POPULATIONS

Monte Carlo simulations were conducted to evaluate robustness of four tests to detect density dependence, from series of population abundances, to the addition of sampling variance. Population abundances were generated from random walk, stochastic exponential growth, and density-dependent population models. Population abundance es- timates were generated with sampling variances distributed as lognormal and constant coefficients of variation (CV) from 0.00 to 1.00. In general, when data were generated under a random walk, Type I error rates increased rapidly for Bulmers R, Pollard et al.s, and Dennis and Tapers tests with increasing magnitude of sampling variance for n . 5y r and all values of process variation. Bulmers R* test maintained a constant 5% Type I error rate for n . 5 yr and all magnitudes of sampling variance in the population abundance estimates. When abundances were generated from two stochastic exponential growth models (R 5 0.05 and R 5 0.10), Type I errors again increased with increasing sampling variance; magnitude of Type I error rates were higher for the slower growing population. Therefore, sampling error inflated Type I error rates, invalidating the tests, for all except Bulmers R* test. Comparable simulations for abundance estimates generated from a density-dependent growth rate model were conducted to estimate power of the tests. Type II error rates were influenced by the relationship of initial population size to carrying capacity ( K), length of time series, as well as sampling error. Given the inflated Type I error rates for all but Bulmers R*, power was overestimated for the remaining tests, resulting in density depen- dence being detected more often than it existed. Population abundances of natural popu- lations are almost exclusively estimated rather than censused, assuring sampling error. Therefore, because these tests have been shown to be either invalid when only sampling variance occurs in the population abundances (Bulmers R, Pollard et al.s, and Dennis and Tapers tests) or lack power (Bulmers R* test), little justification exists for use of such tests to support or refute the hypothesis of density dependence.

SMALL‐MAMMAL DENSITY ESTIMATION: A FIELD COMPARISON OF GRID‐BASED VS. WEB‐BASED DENSITY ESTIMATORS

Statistical models for estimating absolute densities of field populations of animals have been widely used over the last century in both scientific studies and wildlife management programs. To date, two general classes of density estimation models have been developed: models that use data sets from capture–recapture or removal sampling techniques (often derived from trapping grids) from which separate estimates of population size (N) and effective sampling area (Â) are used to calculate density (D = N/Â); and models applicable to sampling regimes using distance-sampling theory (typically transect lines or trapping webs) to estimate detection functions and densities directly from the distance data. However, few studies have evaluated these respective models for accuracy, precision, and bias on known field populations, and no studies have been conducted that compare the two approaches under controlled field conditions. In this study, we evaluated both classes of density estimators on known densities of e...

Population Estimation with Radio-Marked Animals

Publisher Summary This chapter discusses three uses of radio-marked animals to estimate population size. Direct mark–resight estimation of a given number of individuals ( N ) is considered first, where the probability of observing an individual in the population is estimated from observation rates of radio-marked individuals that are present in the population. Second, the chapter discusses the use of sight ability models developed from radio-marked animals, where trials are conducted to estimate the probability of sighting given that the radio-marked animal is present on a survey unit. Covariates, such as size of the group of animals containing the radio-marked individual at the time of the survey or the percent vegetation cover of the animals location during the survey, are used to develop models that predict the probability of sighting individuals in the population. Finally, the chapter considers how radio-marked animals can resolve the issue of estimating animal density from a trapping grid of size A with N individuals estimated from mark–recapture methods. By radio-marking a sample of individuals captured on the grid and tracking them shortly after the grid trapping procedure is finished, the proportion of their locations that occurs on the former trapping grid can be used to correct the bias of the naive estimate of density.

Using simulation to compare methods for estimating density from capture–recapture data

Estimation of animal density is fundamental to wildlife research and management, but estimation via mark–recapture is often complicated by lack of geographic closure of study sites. Contemporary methods for estimating density using mark–recapture data include (1) approximating the effective area sampled by an array of detectors based on the mean maximum distance moved (MMDM) by animals during the sampling session, (2) spatially explicit capture–recapture (SECR) methods that formulate the problem hierarchically with a process model for animal density and an observation model in which detection probability declines with distance from a detector, and (3) a telemetry estimator (TELEM) that uses auxiliary telemetry information to estimate the proportion of animals on the study site. We used simulation to compare relative performance (percent error) of these methods under all combinations of three levels of detection probability (0.2, 0.4, 0.6), three levels of occasions (5, 7, 10), and three levels of abundanc...

Using auxiliary telemetry information to estimate animal density from capture–recapture data

Estimation of animal density is fundamental to ecology, and ecologists often pursue density estimates using grids of detectors (e.g., cameras, live traps, hair snags) to sample animals at a study site. However, under such a framework, reliable estimates can be difficult to obtain because animals move on and off of the site during the sampling session (i.e., the site is not closed geographically). Generally, practitioners address lack of geographic closure by inflating the area sampled by the detectors based on the mean distance individuals moved between trapping events or invoking hierarchical models in which animal density is assumed to be a spatial point process, and detection is modeled as a declining function of distance to a detector. We provide an alternative in which lack of geographic closure is sampled directly using telemetry, and this auxiliary information is used to compute estimates of density based on a modified Huggins closed-capture estimator. Contrary to other approaches, this method is free from assumptions regarding the distribution and movement of animals on the landscape, the stationarity of their home ranges, and biases induced by abnormal movements in response to baited detectors. The estimator is freely available in Program MARK.

Population Estimates of Snowshoe Hares in the Southern Rocky Mountains

Abstract In 1999 Canada lynx (Lynx canadensis) were reintroduced to the southern Rocky Mountains and in 2000 the species was listed as threatened under the Endangered Species Act in the contiguous United States (Colorado Division of Wildlife 2000, U.S. Fish and Wildlife Service 2000). To better evaluate the progress of this reintroduction, we conducted field studies to estimate population densities of snowshoe hares (Lepus americanus), the primary prey of lynx in Colorado, USA. We conducted our field studies in southwestern Colorado in winters 2002 and 2003. We estimated population densities in forested stands of mature Engelmann spruce (Picea engelmannii)–subalpine fir (Abies lasiocarpa) and mature lodgepole pine (Pinus contorta) using mark–recapture data and 3 methods for estimating effective area trapped: half trap interval, mean maximum distance moved (MMDM), and half MMDM. In Engelmann spruce–subalpine fir, we found density estimates ranged from 0.08 ± 0.03 (SE) hares/ha to 1.32 ± 0.15 hares/ha and in lodgepole pine, density estimates ranged from 0.06 ± 0.01 hares/ha to 0.34 ± 0.06 hares/ha, depending on year and method used for estimating effective area trapped. Our density estimates are similar to those reported at the low phase of the hare cycle in populations to the north (<0.1–1.1 hares/ha), where Canada lynx persist (Hodges 2000a). Although density estimates are a useful comparative tool, they depend upon methods used to estimate effective area trapped. Therefore, we urge caution in comparing our density estimates with those from other areas, which may have used dissimilar methods. We also examined effects of temperature and moon phase on capture success of snowshoe hares; extremely low temperatures affected capture success but moon phase did not. Capture success can be improved by trapping snowshoe hares at temperatures above their lower critical temperature (Tlc). If abundance estimates are derived from mark–recapture studies then effects of temperature should be included when modeling capture probabilities.

Assessing release protocols for Canada lynx reintroduction in colorado

ABSTRACT Reintroductions are a common strategy to restore ecosystem integrity, especially when top predators are involved. Reintroductions are often time consuming, expensive, and controversial, and thus understanding what aspects are important for a successful program is needed. Focusing on the example of the reintroduction of Canada lynx (Lynx canadensis) to Colorado, we investigated how different release protocols (RP) affected mortality within the first year post-release. We found that average monthly mortality in the study area during the first year decreased with time in captivity from 0.205 (95% CI = 0.069, 0.475) for lynx having spent up to 7 days in captivity to 0.028 (95% CI = 0.012, 0.064) for lynx spending >45 days in captivity before release. Our results also suggest that keeping lynx in captivity beyond 5–6 weeks accrued little benefit in terms of monthly survival. We found that, on a monthly average basis, lynx were as likely to move out (P = 0.196, SE = 0.032) as well as back onto (P = 0.143, SE = 0.034) the reintroduction area during the first year after release. Mortality was 1.6 times greater outside of the study area, suggesting that permanent emigration and differential mortality rates on and off reintroduction areas should be factored into sample size calculations for an effective reintroduction effort. A post-release monitoring plan is critical to providing information to assess aspects of RP and to improve survival of individuals. Future lynx and other carnivore reintroductions may use our results to help design reintroduction programs including both the release and post-release monitoring protocols.

Microhabitat Characteristics of Preble's Meadow Jumping Mouse High-Use Areas

Abstract Riparian wetlands are complex ecosystems containing species diversity that may easily be affected by anthropogenic disturbances. Prebles meadow jumping mouse (Zapus hudsonius preblei) is a federally threatened subspecies dependent upon riparian wetlands along the Front Range of Colorado and southeastern Wyoming, USA. Although habitat improvements for Prebles meadow jumping mouse are designed at multiple spatial scales, most knowledge about its habitat requirements has been described at a landscape scale. Our objective was to improve our understanding of Prebles meadow jumping mouse microhabitat characteristics within high-use areas (hotspots), which are essential for the development of effective management and conservation strategies. We evaluated Prebles meadow jumping mouse habitat by describing areas of high use and no use as determined from monitoring radiocollared individuals. A comparison of microhabitat characteristics from random samples of high-use and no-use areas indicated that mice use areas closer to the center of the creek bed and positively associated with shrub, grass, and woody debris cover. Distance to center of the creek bed, and percent of shrub and grass cover also had the greatest relative importance of habitat variables modeled when describing high-use areas. High-use areas contained 3 times more grass cover than forb cover, and overall had a greater proportion of wetland shrub and grass cover. However, proportion of cover type (shrub or grass) did not vary greatly between high-use and no-use areas. Our results suggest that management and conservation efforts should continue to focus on establishment of native wetland vegetation near streams and creeks. For example, vegetation should include shrubs such as willow (Salix spp.), narrowleaf cottonwood (Populus angustifolia), alder (Alnus incana), grasses such as fescue (Fescue spp.), sedges (Carex spp.), and rush (Juncus spp).

Habitat use by cross-fostered whooping cranes in Colorado

No intensive studies have been conducted on the migration habitat requirements of whooping cranes (Grus americana). Consequently, we characterized habitats used by cross-fostered whooping cranes during the 1986 and 1987 autumn migration in the San Luis Valley (SLV), Colorado. Water depths <30cm, extensive horizontal visibility, and close proximity to feeding sites, loafing areas, and similar wetlands were common attributes of roost sites. Characteristics that showed no consistency among roost sites included wetland size, emergent plant composition and density, and distances to power lines, fences, and potential disturbances such as roads and residences