Darren J. Bender

Habitat loss and population decline: A meta-analysis of the patch size effect

We evaluated the conditions under which patch size effects are important determinants of local population density for animals living in patchy landscapes. This information was used to predict when patch size effects will be expected to occur following habitat loss and fragmentation. Using meta-analysis, we quantitatively reviewed the results of 25 published studies that tested for a relationship between patch size and population density. Patch size effects were strong for edge and interior species (negative and postive patch size effects, respectively), but negligible for generalist species that use both edge and interior habitat. We found significant differences in mean patch size effects between migratory and residential species, between herbivores and carnivores, and among taxonomic groups. We found no evidence that patch size effects were related to landscape character- istics such as the proportion of landscape covered by habitat, median patch size, or the scale at which a study was conducted. However, species in the Western Hemisphere tended to have larger absolute effect sizes, and eastern species tended to be more variable in their response. For landscapes undergoing habitat loss and fragmentation, our results predict the fol- lowing: (1) among generalist species that use both the edge and the interior of a habitat patch, the decline in population size associated with habitat destruction should be accounted for by pure habitat loss alone; (2) for interior species, the decline in population size as- sociated with habitat fragmentation per se will be greater than that predicted from pure habitat loss alone; (3) for edge species, the decline in population size will be less than that predicted by pure habitat loss alone; (4) these relative effects will not be influenced by the extent of habitat loss, but they will be affected by the pattern of habitat when large or small patches are preferentially removed; and (5) as loss and fragmentation increase within a landscape, migratory species will generally suffer less of a decline in population size than resident species.

Using patch isolation metrics to predict animal movement in binary landscapes

Habitat isolation can affect the distribution and abundance of wildlife, but it is an ambiguous attribute to measure. Presumably, isolation is a characteristic of a habitat patch that reflects how spatially inaccessible it is to dispersing organisms. We identified four isolation metrics (nearest-neighbor distance, Voronoi polygons, proximity index, and habitat buffers) that were representative of the different families of metrics that are commonly used in the literature to measure patch isolation. Using simulated data, we evaluated the ability of each isolation metric to predict animal dispersal. We examined the simulated movement of organisms in two types of landscapes: an artificially-generated point-pattern landscapes where patch size and shape were consistent and only the arrangement of patches varied, and realistic landscapes derived from a geographic information system (GIS) of forest-vegetation maps where patch size, shape, and isolation were variable. We tested the performance of the four isolation metrics by examining the strength of the correlation between observed immigration rate in the simulations and each patch isolation metric. We also evaluated whether each isolation metric would perform consistently under varying conditions of patch size/shape, total amount of habitat in the landscape, and proximity of the patch to the landscape edge. The results indicate that a commonly-used distance-based metric, nearest-neighbor distance, did not adequately predict immigration rate when patch size and shape were variable. Area-informed isolation metrics, such as the amount of available habitat within a given radius of a patch, were most successful at predicting immigration. Overall, the use of area-informed metrics is advocated despite the limitation that these metrics require parameterization to reflect the movement capacity of the organism studied.

MATRIX STRUCTURE OBSCURES THE RELATIONSHIP BETWEEN INTERPATCH MOVEMENT AND PATCH SIZE AND ISOLATION

Metapopulation models assume that local population size (in a habitat patch) and therefore local extinction probability, is a function of patch size, and that interpatch movement rate and therefore recolonization of local extinctions is a function of both patch size and patch isolation. We hypothesized that the predictive power of models that relate patch immigration rate to patch size and isolation will be reduced when spatial structure in the landscape matrix (the nonhabitat portion of the landscape) affects organism movement through the landscape. We used a simulation model to evaluate this hypothesis for three different aspects of matrix spatial structure (contrast among matrix cover types, number of matrix cover types, and grain of matrix spatial pattern) and for two different types of movement behavior, representing a specialist and a generalist species. We tested the hy- pothesis for one aspect of matrix structure (number of matrix cover types) in a translocation field study of the eastern chipmunk ( Tamias striatus, a habitat specialist) and the white- footed mouse (Peromyscus leucopus, a habitat generalist). When the matrix was composed of a small number of cover types, patch size and isolation accounted for up to 75% of the variation in patch immigration rate in the simulation study, and for 61% of the variation in interpatch movement in the field study. However, when the matrix was composed of a large number of cover types, the amount of explained variation dropped to as little as 33% for the simulation study and to 17% in the field study. Our results suggest that patch characteristics, such as patch size and isolation, may be poor predictors of interpatch movement when the landscape matrix is heterogeneous and when the organism responds to boundaries between different matrix cover types. These results imply that habitat patch- based models, such as those based on current metapopulation theory, will perform poorly in these situations.

Evaluation of patch isolation metrics in mosaic landscapes for specialist vs. generalist dispersers

We examined the effects of matrix structure and movement responses of organisms on the relationships between 7 patch isolation metrics and patch immigration. Our analysis was based on simulating movement behaviour of two generic disperser types (specialist and generalist) across mosaic landscapes containing three landcover types: habitat, hospitable matrix and inhospitable matrix. Movement, mortality and boundary crossing probabilities of simulated individuals were linked to the landcover and boundary types in the landscapes. The results indicated that area-based isolation metrics generally predict patch immigration more reliably than distance-based isolation metrics. Relationships between patch isolation metrics and patch immigration varied between the two generic disperser types and were affected by matrix composition or matrix fragmentation. Patch immigration was always affected by matrix composition but not by matrix fragmentation. Our results do not encourage the generic use of patch isolation metrics as a substitute for patch immigration, in particular in metapopulation models where generic use may result in wrong projections of the survival probability of metapopulations. However, our examination of the factors affecting the predictive potential of patch isolation metrics should facilitate interpretation and comparison of existing patch isolation studies. Future patch isolation studies should include information about landscape structure and the dispersal distance and dispersal behaviour of the organism of interest.

Botfly (Diptera:Oestridae) parasitism of Ord's kangaroo rats (Dipodomys ordii) at Suffield National Wildlife Area, Alberta, Canada.

During field study of Ords kangaroo rat (Dipodomys ordii) at Suffield National Wildlife Area, Alberta, Canada, a high prevalence of parasitism by botfly (Diptera: Oestridae) larvae was observed. Botflies have not previously been documented as parasites of kangaroo rats. Botfly parasitism could have a significant impact on the growth, survival, and reproduction of Ords kangaroo rat, which is considered a vulnerable species in Canada. Therefore, it is important to investigate how botfly parasitism varies with season and with gender or age of host. In 1995, 525 individual kangaroo rats were caught by nightlighting and live trapping for a total of 952 capture records. Upon capture, each kangaroo rat was ear-tagged and thoroughly examined for parasites and wounds. Third-instar botfly (Cuterebra polita) larvae were observed in kangaroo rats between 16 June and 23 August. Prevalence was 34% based on 454 kangaroo rats sampled during that time, whereas the mean intensity was 2.3 larvae per infested host (n = 156, range = 1-11). In contrast to some other studies of botfly parasitism of rodents, there were no gender or age biases in either prevalence or intensity of infestation. The index of dispersion was 2.8, indicating that the parasites were aggregated in hosts. Botfly parasitism could be an important factor affecting northern populations of kangaroo rats; future investigations into the potential effects of botfly larvae on host fitness are warranted.

Assessing Landscape Relationships for Habitat Generalists

Abstract: The importance of landscape heterogeneity for the abundance and distribution of wildlife is well recognized. General relationships have been developed to link landscape pattern to demographic processes, although these relations are best demonstrated for species with specialized habitat requirements and often in landscapes that can be generalized to a simple habitat-matrix structure. Habitat generalists may interact in more complex ways with a mosaic of landscape features. A novel method for quantifying the habitat relationships of generalist species using thematic vegetation maps was proposed by Brotons et al. (2005) and based on a theoretical model by Andrén, Delin, and Seiler (1997). We tested the efficacy of this approach on moose (Alces alces) distribution in the heterogeneous landscapes of the Foothills Natural Region, Alberta, Canada, using 8 broad vegetation types. Fecal pellet group data, an index of moose occurrence, was compared across pre-selected sites. Sites were selected to represent the variable amounts and combinations of the different vegetation types available in the study area. Moose habitat preference was determined using a Chi-square test and Bonferroni confidence intervals. Moose preferred shrublands and deciduous forests. Shrubland was considered primary moose habitat as it had the highest observed proportion of pellet groups of the preferred habitats. Each vegetation type was assessed regarding its role in habitat amount, habitat compensation, supplementation, complementation, and fragmentation models using general linear modelling. Habitat amount and fragmentation were related to moose pellet occurrence. However, there was no indication of supplementation, compensation, or complementation. This mosaic approach effectively revealed habitat relationships and the potential impacts of habitat change for a generalist species at the landscape scale. Nomenclature: Kays & Wilson, 2002.

Divergence in sink contributions to population persistence.

Population sinks present unique conservation challenges. The loss of individuals in sinks can compromise persistence; but conversely, sinks can improve viability by improving connectivity and facilitating the recolonization of vacant sources. To assess the contribution of sinks to regional population persistence of declining populations, we simulated source-sink dynamics for 3 very different endangered species: Black-capped Vireos (Vireo atricapilla) at Fort Hood, Texas, Ords kangaroo rats (Dipodomys ordii) in Alberta, and Northern Spotted Owls (Strix occidentalis caurina) in the northwestern United States. We used empirical data from these case studies to parameterize spatially explicit individual-based models. We then used the models to quantify population abundance and persistence with and without long-term sinks. The contributions of sink habitats varied widely. Sinks were detrimental, particularly when they functioned as strong sinks with few emigrants in declining populations (e.g., Albertas Ords kangaroo rat) and benign in robust populations (e.g., Black-capped Vireos) when Brown-headed Cowbird (Molothrus ater) parasitism was controlled. Sinks, including ecological traps, were also crucial in delaying declines when there were few sources (e.g., in Black-capped Vireo populations with no Cowbird control). Sink contributions were also nuanced. For example, sinks that supported large, variable populations were subject to greater extinction risk (e.g., Northern Spotted Owls). In each of our case studies, new context-dependent sinks emerged, underscoring the dynamic nature of sources and sinks and the need for frequent re-assessment. Our results imply that management actions based on assumptions that sink habitats are generally harmful or helpful risk undermining conservation efforts for declining populations.

Generalizing matrix structure affects the identification of least-cost paths and patch connectivity

Understanding and assessing landscape connectivity is often a primary goal when studying patchy or spatially structured populations. It is commonly accepted that the matrix plays a role in determining connectivity; however, it is not clear how the process of assessing connectivity is impacted by different ways in which the matrix may be represented, particularly if matrix structure is generalized to expedite analysis. We conducted a controlled experiment using computer simulations to evaluate the impact of increasing levels of matrix generalization on connectivity assessment using a constant arrangement of habitat patches. We varied matrix generalization for six simulated landscape patterns by adjusting the number of classes and level of pattern grain in 16 ways to yield sets of landscapes for which the matrix ranged from not generalized (i.e., heterogeneous) to completely generalized (i.e., homogeneous) while habitat placement was held constant. Least-cost paths were calculated for each landscape, and a spatial interaction model (SIM) was implemented to model the counts of patch inflows (immigration) and pairwise exchange. Applying a SIM allowed us to generate absolute outputs and explicitly compare the effect of changes to matrix generalization on connectivity. We found that both assessment of connections (i.e., measured distance and spatial delineation of least-cost paths) and patch inflows/pairwise exchange were highly sensitive to matrix generalization and that effects were inconsistent and unpredictable across the range of matrix representations, especially when estimating connectivity for individual patches. We conclude that matrix pattern may have an underappreciated effect on connectivity and that least-cost path delineation and connectivity assessment may be very sensitive to generalizations of the matrix. We suggest that sensitivity analysis of the matrix representation should be performed when conducting connectivity analyses.