Bruce A. Pond

ELIMINATION OF RABIES FROM RED FOXES IN EASTERN ONTARIO

The province of Ontario (Canada) reported more laboratory confirmed rabid animals than any other state or province in Canada or the USA from 1958–91, with the exception of 1960–62. More than 95% of those cases occurred in the southern 10% of Ontario (≈100,000 km2), the region with the highest human population density and greatest agricultural activity. Rabies posed an expensive threat to human health and significant costs to the agricultural economy. The rabies variant originated in arctic foxes: the main vector in southern Ontario was the red fox (Vulpes vulpes), with lesser involvement of the striped skunk (Mephitis mephitis). The Ontario Ministry of Natural Resources began a 5 yr experiment in 1989 to eliminate terrestrial rabies from a ≈30,000 km2 study area in the eastern end of southern Ontario. Baits containing oral rabies vaccine were dropped annually in the study area at a density of 20 baits/km2 from 1989–95. That continued 2 yr beyond the original 5 yr plan. The experiment was successful in eliminating the arctic fox variant of rabies from the whole area. In the 1980s, an average of 235 rabid foxes per year were reported in the study area. None have been reported since 1993. Cases of fox rabies in other species also disappeared. In 1995, the last bovine and companion animal cases were reported and in 1996 the last rabid skunk occurred. Only bat variants of rabies were present until 1999, when the raccoon variant entered from New York (USA). The success of this experiment led to an expansion of the program to all of southern Ontario in 1994. Persistence of terrestrial rabies, and ease of elimination, appeared to vary geographically, and probably over time. Ecological factors which enhance or reduce the long term survival of rabies in wild foxes are poorly understood.

Woodland Caribou Extirpation and Anthropogenic Landscape Disturbance in Ontario

Abstract The decline of woodland caribou (Rangifer tarandus caribou) has been attributed to anthropogenic landscape disturbances, but critical distance thresholds and time lags between disturbance and extirpation are unknown. Using a database of caribou presence and extirpation for northern Ontario, Canada, geo-coded to 10 × 10-km cells, we constructed logistic regression models to predict caribou extirpation based on distance to the nearest of each of 9 disturbance types: forest cutovers, fires, roads, utility corridors, mines, pits and quarries, lakes, trails, and rail lines. We used Akaikes Information Criterion to select parsimonious models and Receiver-Operating Characteristic curves to derive optimal thresholds. To deal with the effects of spatial autocorrelation on estimates of model significance, we used subsampling and restricted randomizations. Forest cutovers were the best predictor of caribou occupancy, with a tolerance threshold of 13 km to nearest cutover and a time lag of 2 decades between disturbance by cutting and caribou extirpation. Management of woodland caribou should incorporate buffers around habitat and requires long-term monitoring of range occupancy.

Spatial occupancy models for large data sets

Since its development, occupancy modeling has become a popular and useful tool for ecologists wishing to learn about the dynamics of species occurrence over time and space. Such models require presence–absence data to be collected at spatially indexed survey units. However, only recently have researchers recognized the need to correct for spatially induced overdisperison by explicitly accounting for spatial autocorrelation in occupancy probability. Previous efforts to incorporate such autocorrelation have largely focused on logit-normal formulations for occupancy, with spatial autocorrelation induced by a random effect within a hierarchical modeling framework. Although useful, computational time generally limits such an approach to relatively small data sets, and there are often problems with algorithm instability, yielding unsatisfactory results. Further, recent research has revealed a hidden form of multicollinearity in such applications, which may lead to parameter bias if not explicitly addressed. Com...

Combining direct and indirect genetic methods to estimate dispersal for informing wildlife disease management decisions

Epidemiological models are useful tools for management to predict and control wildlife disease outbreaks. Dispersal behaviours of the vector are critical in determining patterns of disease spread, and key variables in epidemiological models, yet they are difficult to measure. Raccoon rabies is enzootic over the eastern seaboard of North America and management actions to control its spread are costly. Understanding dispersal behaviours of raccoons can contribute to refining management protocols to reduce economic impacts. Here, estimates of dispersal were obtained through parentage and spatial genetic analyses of raccoons in two areas at the front of the raccoon rabies epizootic in Ontario; Niagara (N = 296) and St Lawrence (N = 593). Parentage analysis indicated the dispersal distance distribution is highly positively skewed with 85% of raccoons, both male and female, moving < 3 km. The tail of this distribution indicated a small proportion (< 4%) moves more than 20 km. Analysis of spatial genetic structure provided a similar assessment as the spatial genetic correlation coefficient dropped sharply after 1 km. Directionality of dispersal would have important implications for control actions; however, evidence of directional bias was not found. Separating the data into age and sex classes the spatial genetic analyses detected female philopatry. Dispersal distances differed significantly between juveniles and adults, while juveniles in the Niagara region were significantly more related to each other than adults were to each other. Factors that may contribute to these differences include kin association, and spring dispersal. Changes to the timing and area covered by rabies control operations in Ontario are indicated based on these dispersal data.

RURAL/URBAN LAND CONVERSION I:ESTIMATING THE DIRECT AND INDIRECT IMPACTS

This paper provides an accounting methodology for estimating the direct: indirect urban/rural land conversion ratio. This methodology is then used in two case studies (in central Canada) for four different years over a 15-year time period. The results indicate that the direct:indirect land conversion ratios fluctuate considerably in a specific region during urbanization. The results are used to suggest a stage model of rural/urban land conversion that highlights the sensitivity of different planning questions in the various stages of conversion.

Landscape modelling spatial bottlenecks: implications for raccoon rabies disease spread

A landscape genetic simulation modelling approach is used to understand factors affecting raccoon rabies disease spread in southern Ontario, Canada. Using the Ontario Rabies Model, we test the hypothesis that landscape configuration (shape of available habitat) affects dispersal, as indicated by genetic structuring. We simulated range expansions of raccoons from New York into vacant landscapes in Ontario, in two areas that differed by the presence or absence of a landscape constriction. Our results provide theoretical evidence that landscape constriction acts as a vicariant bottleneck. We discuss implications for raccoon rabies spread.

Spatial patterns of neutral and functional genetic variations reveal patterns of local adaptation in raccoon (Procyon lotor) populations exposed to raccoon rabies

Local adaptation is necessary for population survival and depends on the interplay between responses to selective forces and demographic processes that introduce or retain adaptive and maladaptive attributes. Host–parasite systems are dynamic, varying in space and time, where both host and parasites must adapt to their ever‐changing environment in order to survive. We investigated patterns of local adaptation in raccoon populations with varying temporal exposure to the raccoon rabies virus (RRV). RRV infects approximately 85% of the population when epizootic and has been presumed to be completely lethal once contracted; however, disease challenge experiments and varying spatial patterns of RRV spread suggest some level of immunity may exist. We first assessed patterns of local adaptation in raccoon populations along the eastern seaboard of North America by contrasting spatial patterns of neutral (microsatellite loci) and functional, major histocompatibility complex (MHC) genetic diversity and structure. We explored variation of MHC allele frequencies in the light of temporal population exposure to RRV (0–60 years) and specific RRV strains in infected raccoons. Our results revealed high levels of MHC variation (66 DRB exon 2 alleles) and pronounced genetic structure relative to neutral microsatellite loci, indicative of local adaptation. We found a positive association linking MHC genetic diversity and temporal RRV exposure, but no association with susceptibility and resistance to RRV strains. These results have implications for landscape epidemiology studies seeking to predict the spread of RRV and present an example of how population demographics influence the degree to which populations adapt to local selective pressures.

Modelling the effect of landscape heterogeneity on the efficacy of vaccination for wildlife infectious disease control

Summary Zoonotic disease control presents significant costs and challenges in human and wildlife populations. Although spatial variability and temporal variability in host populations play a significant role influencing the spread and persistence of pathogens, their impact on the effectiveness of disease control are not well understood. Field studies are impractical for many zoonotic diseases; thus, simulation modelling is an alternative. Some research has experimented with metapopulation models of host–pathogen systems, with discrete host populations distributed on a network of connections or on a one-dimensional transect of contiguous cells. Little attention has been paid to treating geographic space as a fine-grained two-dimensional continuum, a more appropriate spatial model for many generalist host and vector species. Using raccoon rabies as an example, we apply an individual-based spatially explicit stochastic simulation model to evaluate effectiveness of vaccination barrier strategies to control rabies. Barrier width and immunization levels are varied over landscapes with habitats of varying quality and spatial heterogeneity, resulting in varying degrees of host connectivity. Our results demonstrate that spatial heterogeneity in the landscape does affect vaccination efficacy. The probability that rabies will breach a vaccination barrier is greater and rabies incidence is higher in landscapes with (i) overall good-quality homogeneous habitat and (ii) overall poor-quality habitat with high spatial heterogeneity, than in landscapes with overall good-quality habitat and high spatial heterogeneity. The influence of landscape conditions on disease dynamics decreases with increasing population immunity. Synthesis and applications. Using a spatially explicit stochastic simulation model, we demonstrated that landscape spatial heterogeneity and vaccination control will interact to influence the success of controlling infectious disease outbreaks. Further, under some landscape conditions, insufficient vaccination is counter-productive because immunized individuals (i) reduce the number of disease transmitting contacts, preventing the disease from growing rapidly thus depleting the susceptible population; and (ii) survive to replenish the stock of susceptible animals through reproduction, facilitating disease persistence.

Using multiple imputation to estimate missing data in meta-regression

Summary There is a growing need for scientific synthesis in ecology and evolution. In many cases, meta-analytic techniques can be used to complement such synthesis. However, missing data are a serious problem for any synthetic efforts and can compromise the integrity of meta-analyses in these and other disciplines. Currently, the prevalence of missing data in meta-analytic data sets in ecology and the efficacy of different remedies for this problem have not been adequately quantified. We generated meta-analytic data sets based on literature reviews of experimental and observational data and found that missing data were prevalent in meta-analytic ecological data sets. We then tested the performance of complete case removal (a widely used method when data are missing) and multiple imputation (an alternative method for data recovery) and assessed model bias, precision and multimodel rankings under a variety of simulated conditions using published meta-regression data sets. We found that complete case removal led to biased and imprecise coefficient estimates and yielded poorly specified models. In contrast, multiple imputation provided unbiased parameter estimates with only a small loss in precision. The performance of multiple imputation, however, was dependent on the type of data missing. It performed best when missing values were weighting variables, but performance was mixed when missing values were predictor variables. Multiple imputation performed poorly when imputing raw data which were then used to calculate effect size and the weighting variable. We conclude that complete case removal should not be used in meta-regression and that multiple imputation has the potential to be an indispensable tool for meta-regression in ecology and evolution. However, we recommend that users assess the performance of multiple imputation by simulating missing data on a subset of their data before implementing it to recover actual missing data.

Raccoon ecology database: A resource for population dynamics modelling and meta-analysis

Abstract The value of scientific studies increases and is extended when their data are stored in a manageable and accessible format. This is demonstrated through development of a raccoon ecology database (REDB) to store, manage and disseminate available peer-reviewed and unpublished data on raccoon ( Procyon lotor ) biology, ecology and raccoon rabies, including citations for data sources. Over 800 documents were identified and citations for them entered into the database as literature references. Approximately 1000 trait values were entered from almost 200 of these sources. These data included estimates of population density, survival rates, rabies incubation period, litter size, body weight, dispersal distance and home-range size, often by age or sex class. Each datum is linked to a citation for its source, and to information about location and land use in the study area, time of year the study was undertaken, sample size, and variance. The relational database design enables querying and easy updating and manipulation of data. The relational data model is presented, as is its application in further developing an individual-based, spatially-explicit population model of raccoon rabies. Using information queried from the REDB benefits model development by: i) assessing the appropriateness of input parameter values, ii) providing sources for citing input values, iii) parameterising the model to different geographic regions, iv) enabling meta-analyses for evaluating model structure, as well as further contributing to parameterisation at specific locations, and v) providing biologically appropriate parameter input values for model sensitivity testing. The REDB is a useful research resource that will increase in value with ongoing inclusion of data from future raccoon and raccoon rabies studies and serves as a model for database design and research applications to other species. The database and an empty database for use with other species are available online ( http://redb.nrdpfc.ca ).