Kenneth F. Kellner

Accounting for Imperfect Detection in Ecology: A Quantitative Review

Detection in studies of species abundance and distribution is often imperfect. Assuming perfect detection introduces bias into estimation that can weaken inference upon which understanding and policy are based. Despite availability of numerous methods designed to address this assumption, many refereed papers in ecology fail to account for non-detection error. We conducted a quantitative literature review of 537 ecological articles to measure the degree to which studies of different taxa, at various scales, and over time have accounted for imperfect detection. Overall, just 23% of articles accounted for imperfect detection. The probability that an article incorporated imperfect detection increased with time and varied among taxa studied; studies of vertebrates were more likely to incorporate imperfect detection. Among articles that reported detection probability, 70% contained per-survey estimates of detection that were less than 0.5. For articles in which constancy of detection was tested, 86% reported significant variation. We hope that our findings prompt more ecologists to consider carefully the detection process when designing studies and analyzing results, especially for sub-disciplines where incorporation of imperfect detection in study design and analysis so far has been lacking.

Reducing Baylisascaris procyonis Roundworm Larvae in Raccoon Latrines

Baylisascaris procyonis roundworms, a parasite of raccoons, can infect humans, sometimes fatally. Parasite eggs can remain viable in raccoon latrines for years. To develop a management technique for parasite eggs, we tested anthelmintic baiting. The prevalence of eggs decreased at latrines, and larval infections decreased among intermediate hosts, indicating that baiting is effective.

Backyard raccoon latrines and risk for Baylisascaris procyonis transmission to humans.

To the Editor: Raccoons (Procyon lotor) are abundant in urban environments and carry a variety of diseases that threaten domestic animals (1) and humans (2,3). A ubiquitous parasite of raccoons, Baylisascaris procyonis causes a widely recognized emerging zoonosis, baylisascariasis (3). Although only 14 human cases of severe B. procyonis encephalitis have been reported over 30 years (4), prevention is still a priority for public health and wildlife officials because of the seriousness of the resulting neurologic disease (5). Raccoons prefer to defecate at latrines they create. Infected animals shed ≈20,000 eggs/g of feces (3), so latrines serve as the foci of parasite transmission (6). When latrines occur in close proximity to humans, the risk for zoonotic transmission increases (2). Because B. procyonis are transmitted by the fecal–oral route, young children have the greatest risk for zoonotic infection because of their tendency to put objects into their mouths (1,2). Many human cases have occurred in environments where latrines were near children’s play areas. Our objective was to determine which factors encourage raccoons to create latrines in human habitats. This information will allow public health officials and wildlife managers to develop strategies to educate the public and to ultimately prevent zoonotic transmission. We surveyed 119 backyards for raccoon latrines in the suburbs of Chicago, Illinois, USA, near the Ned Brown Forest Preserve (n = 38; 42°01′55.05′′N, 88°00′00.62′′W, Cook County) and Lincoln Marsh (n = 81; 41°51′4.54′′N, 88°5′39.019′′W, DuPage County). Yards were selected on the basis of proximity to forest preserves and willingness of homeowners to participate in the study. We located latrines by systematically searching yards, giving special attention to horizontal substrates, such as piles of wood and the bases of large trees (6). We removed all fecal material to test for B. procyonis and stored it in plastic bags at –20oC until analysis. Composite samples that were at least 2 g underwent fecal flotation in Sheather solution (7) (at least 1 g of every fecal deposit at a latrine) (n =131). We identified B. procyonis eggs by microscopic examination on the basis of their size and morphologic appearance (2). Multiple slides were examined for ≈10% of the samples (randomly selected) to validate our results. Prevalence was considered the proportion of positive samples from all sampled yards. Each yard was additionally surveyed for potential latrine substrates (8) and factors believed to attract or deter raccoons. The distance of each yard from the nearest forested habitat was calculated by using ArcGIS 9.0 (Geographic Information Systems, Redlands, CA, USA). We used homogeneity tests to identify differences in the proportion of yards with latrines present and to compare the prevalence of B. procyonis between study areas. Logistic regression and odds ratios were used to evaluate a main effect model composed of 10 yard attributes, including the presence of a pet, birdfeeders, garbage cans, and sandboxes, and to evaluate a simplified model in which attributes were combined to reflect the presence of food and latrine substrates, such as pet food, birdfeed, garbage and piles of wood or logs, respectively. Latrines occurred in 61/119 yards (51%; 95% confidence interval [CI] 0.42%–0.60%). There was no significant difference in the proportion of backyards with latrines in proximity to Ned Brown (23/38, 82%) and Lincoln Marsh (38/81, 46%). The number of latrines per backyard ranged from 1 to 6 ( = 2.15). B. procyonis eggs were found at 14/61 latrines sampled (23%; 95% CI 12%–34%), and no significant difference in prevalence was found between the Ned Brown (6/23, 26%; 95% CI 8%–44%) and Lincoln Marsh areas (8/38, 21%; 95% CI 8%–34%). Evaluation of the main effect model identified a decreasing probability of latrine occurrence with increasing distance from the nearest forested area and the presence of an outdoor pet, although these relationships were only marginally significant (p = 0.07 and 0.08, respectively). No other variables were closely associated with the presence of raccoon latrines (p>0.20). When evaluated alone, distance from the forest preserve was significantly related to latrine occurrence (p = 0.03); probability decreased with increasing distance. Evaluation of the simplified model identified a weakly positive association with the presence of a food source (p = 0.09) and no association with the presence of latrine substrate (p = 0.35). Although the findings were not statistically significant, raccoon latrines did appear to be associated with the availability of a food source such as bird feed (odds ratio [OR] 1.9, 95% CI 0.9–4.1); the presence of an outdoor pet (OR 0.27, 95% CI 0.06–1.2) and increasing distance from the nearest forested area reduced the likelihood of latrines. No other variables were associated with the presence of raccoon latrines; however, low statistical power may have precluded adequate assessment. Our results suggest that when humans live close to protected forests or natural areas, they are more likely to attract raccoons into their yards. In addition, anthropogenic food sources such as pet food, garbage, and bird feed may increase the likelihood that a raccoon will create a latrine, and the presence of outdoor pets appears to be a deterrent. In areas of high raccoon density, these attractants should be removed. Homeowners with small children should remove latrines as quickly as they are discovered (2). The risk of children acquiring potentially fatal baylisascariasis can be reduced if parents understand how to reduce the likelihood that children will come into contact with raccoon latrines.

THE RELATIONSHIP BETWEEN BAYLISASCARIS PROCYONIS PREVALENCE AND RACCOON POPULATION STRUCTURE

Abstract Parasite transmission is a dynamic process that can be affected by factors including host and parasite population dynamics. Raccoons (Procyon lotor) are the definitive host of Baylisascaris procyonis, an intestinal roundworm. Transmission of this parasite has been linked to raccoon behavior and human land-use patterns; however, we do not know the importance of host population structure. Therefore, the objective of this study was to determine the relationship between raccoon population attributes and prevalence of B. procyonis. We necropsied 307 trapped or road-killed raccoons collected during 2000–2006 from the Chicago area. In addition, we examined, via fecal samples (n  =  433), the patterns of B. procyonis prevalence as they relate to population dynamics among 3 subpopulations within the larger study. Baylisascaris procyonis was seen in 39% of 307 necropsied raccoons. There were differences in prevalence as a function of host age and sex. Baylisascaris procyonis was observed in 18% of 433 fecal samples obtained from live-trapped raccoons, and there were differences according to age, but not by sex. We found that the host populations consistently differed in density across study areas, but were similar regarding sex and age structure. Differences in host density were associated with differences in prevalence, suggesting that possible differences between populations, as well as ecological differences in sites and raccoon behavior, may have influenced parasite prevalence.

Changes in Small Mammal Microhabitat Use Following Silvicultural Disturbance

Abstract Silvicultural disturbance can lead to major shifts in biotic and abiotic characteristics of forests, with significant implications for wildlife. Many studies have demonstrated site occupancy and abundance of small mammals changes following silviculture, but few have identified the habitat characteristics associated with small mammal responses. We conducted a study in the central hardwood forest region of the U.S. to examine the relationship between small mammal habitat use and habitat variables at the microsite scale, before and after silvicultural disturbance, while explicitly accounting for imperfect detection. Following the creation of harvest openings, vegetation cover and coarse woody debris increased, and leaf litter depth decreased. Fewer changes occurred following the midstory removal stage of a shelterwood harvest. Eastern chipmunks (Tamias striatus) and white-footed mice (Peromyscus leucopus) were positively correlated with woody vegetation cover and coarse woody debris, and short-tailed shrews (Blarina brevicauda) were positively correlated with coarse woody debris and leaf litter depth. Careful consideration of experimental scale and incorporation of detection probability are necessary for successful identification of small mammal associations within microhabitats, as these relationships help to explain shifts in small mammal communities following silviculture.

Effects of silvicultural disturbance on acorn infestation and removal

Oak (Quercus spp.) regeneration is an important forest management goal in the central hardwood forest region of the United States. Silvicultural methods that target oak regeneration, such as the creation of canopy openings and removal of shade-tolerant midstory trees, are complicated by the impact of pre- and post-dispersal predators of oak acorns like acorn weevils (Curculio spp.) and granivorous small mammals. Understanding the effects of forest disturbance created by silviculture on these predators is important to promote successful oak regeneration. We conducted a 6-year study of acorn production, weevil infestation, and acorn removal at 113 black (Q. velutina) and white oak (Q. alba) trees impacted by three types of silvicultural treatments: adjacent to harvest openings, midstory removal (first-stage shelterwood harvest), and control (no harvest) in south-central Indiana, USA. We observed highly variable acorn production across years, but minimal impact of silvicultural treatments. Levels of weevil infestation and acorn removal by small mammal seed predators also varied greatly, and predation pressure was highest in years when acorns were scarce. Weevil infestation was reduced following midstory removal in shelterwood harvests, but probability of acorn removal by small mammals was unchanged following harvest. Damaged, germinated, and weevil-infested acorns were less likely to be removed by seed predators, suggesting additive effects of pre- and post-dispersal predators. This study emphasizes the importance of considering acorn predators in oak regeneration silviculture, and timing harvests to follow large acorn crops in order to reduce predation pressure and generate a high number of seedlings.

Simulation of oak early life history and interactions with disturbance via an individual-based model, SOEL

Early tree life history and demography are driven by interactions with the environment such as seed predation, herbivory, light availability, and drought. For oak (Quercus) in the eastern United States, these interactions may contribute to oak regeneration failure. Numerous studies have examined the impact of individual factors (like seed predation) on the oak regeneration process, but less information is available on the relative and combined impacts of multiple intrinsic and extrinsic factors on early oak life history. We developed an individual-based, spatially explicit model to Simulate Oak Early Life history (SOEL). The model connects acorn survival, acorn dispersal, germination, seedling growth, and seedling survival submodels based on empirical data with an existing gap model (JABOWA). Using SOEL, we assessed the sensitivity of several metrics of oak regeneration to different parameters associated with early oak life history. We also applied the model in three individual case studies to assess: (1) how variable acorn production interacts with timing of timber harvest; (2) the effect of shelterwood harvest-induced differences on seed predation; and (3) the consequences of interactions between drought, seedling growth and survival, and timber harvest. We found that oak regeneration metrics including percent emergence, seedling density, and sapling density were most sensitive to the amount of acorn production, acorn caching probability by scatterhoarders, and seedling growth rates. In the case studies, we found that timing harvest to follow large acorn crops can increase the density of oak regeneration in the short term following harvest, at least under some conditions. Following midstory removal, lower weevil infestation probability and lower post-dispersal acorn survival resulted in a modest decline in seedling density, but the decline did not persist to the sapling life stage class. Drought frequency had a powerful negative impact on both growth and survival for individual seedlings, which resulted in large reductions in both seedling and sapling density. The case studies presented here represent only a few examples of what could be accomplished within the SOEL modeling framework. Further studies could focus on different early life history parameters, or connect the parameter values to different predictor variables based on field data.

Herbivory on planted oak seedlings across a habitat edge created by timber harvest

Edge habitats create environmental gradients that affect plant community composition and herbivore behavior. Silvicultural disturbance creates edge habitat with direct (via changes in light) and indirect (via changes in herbivore behavior) consequences for the growth and survival of tree seedlings, and thus, the composition of the future forest stands. Herbivores, particularly ungulates, can be a major limiting factor in oak regeneration, and silvicultural disturbance may alter the abundance or behavior of herbivores following harvest. We measured the severity of herbivory on experimentally planted white (Quercus alba) and black oak (Quercus velutina) seedlings by white-tailed deer (Odocoileus virginianus) and eastern cottontail rabbits (Sylvilagus floridanus), as well as foliar damage from insects, across gradients created by clearcuts in a deciduous forest in Indiana, USA. Overall browse pressure on oaks was low in our study. Nonetheless, spatial variation in herbivory depended on herbivore taxa; herbivory by rabbits was highest inside harvest openings, whereas foliar damage by insects peaked in the forest. Intensity of deer herbivory was constant across the edge. In addition, we observed indirect interactions among herbivore species mediated by a seedling’s browsing history. Herbivore damage by deer was positively related to past browsing by rabbits, and foliar damage from insects was positively related to past browsing by both deer and rabbits. Increasing woody plant competition reduced herbivory on seedlings by both deer and rabbits. Given the lack of spatial variability in deer herbivory and low overall herbivory by rabbits, we suspect that interactions between timber harvesting and herbivory did not have a strong impact on oak seedlings at our study sites.

Performance Benchmarks for Scholarly Metrics Associated with Fisheries and Wildlife Faculty.

Research productivity and impact are often considered in professional evaluations of academics, and performance metrics based on publications and citations increasingly are used in such evaluations. To promote evidence-based and informed use of these metrics, we collected publication and citation data for 437 tenure-track faculty members at 33 research-extensive universities in the United States belonging to the National Association of University Fisheries and Wildlife Programs. For each faculty member, we computed 8 commonly used performance metrics based on numbers of publications and citations, and recorded covariates including academic age (time since Ph.D.), sex, percentage of appointment devoted to research, and the sub-disciplinary research focus. Standardized deviance residuals from regression models were used to compare faculty after accounting for variation in performance due to these covariates. We also aggregated residuals to enable comparison across universities. Finally, we tested for temporal trends in citation practices to assess whether the “law of constant ratios”, used to enable comparison of performance metrics between disciplines that differ in citation and publication practices, applied to fisheries and wildlife sub-disciplines when mapped to Web of Science Journal Citation Report categories. Our regression models reduced deviance by ¼ to ½. Standardized residuals for each faculty member, when combined across metrics as a simple average or weighted via factor analysis, produced similar results in terms of performance based on percentile rankings. Significant variation was observed in scholarly performance across universities, after accounting for the influence of covariates. In contrast to findings for other disciplines, normalized citation ratios for fisheries and wildlife sub-disciplines increased across years. Increases were comparable for all sub-disciplines except ecology. We discuss the advantages and limitations of our methods, illustrate their use when applied to new data, and suggest future improvements. Our benchmarking approach may provide a useful tool to augment detailed, qualitative assessment of performance.

Correction to: Runs of homozygosity have utility in mammalian conservation and evolutionary studies

The fourth author’s name was incorrect in the original publication and correct author name is given in this Correction.