Scale dependence of felid predation risk: identifying predictors of livestock kills by tiger and leopard in Bhutan

Context

Livestock predation by tiger and leopard in Bhutan is a major threat to the conservation of these felids. Conflict mitigation planning would benefit from an improved understanding of the spatial pattern of livestock kills by the two predators.

Objectives

We aimed to identify the landscape features that predict livestock kills by tiger and leopard throughout Bhutan. Our goals were to: (1) identify the predictors that have the largest influence in determining livestock kills, (2) assess the influence of scale across the different predictors evaluated and identify the scale at which each was most important.

Methods

We used livestock kills obtained from compensation records of tiger (n = 326) and leopard (n = 377) across Bhutan between 2003 and 2012 to run predation risk models with MaxEnt algorithm, using a multi-scale modeling approach (1, 2, 4, 8 and 16 km).

Results

Human-presence (density of settlements and roads) and land-cover (percentage of tree cover and meadow patches) were the main variables contributing to livestock kills by both species. Livestock kills were likely driven by a trade-off between livestock density and predator ecology, and the balance of this trade-off varied with scale. Risk maps revealed different hotspots for tiger and leopard kills, and analysis showed both species preferentially killed equids over other livestock types.

Conclusions

Our results highlight the importance of evaluating scale when investigating the spatial attributes of livestock kills by tiger and leopard. Our findings provide guidance for reducing conflict between humans and large felids throughout the country.

     

multifit: an R function for multi-scale analysis in landscape ecology

Context

Multi-scale analyses are a common approach in landscape ecology. Their aim is to find the appropriate spatial scale for a particular landscape attribute in order to perform a correct interpretation of results and conclusions.

Objectives

I present an R function that performs statistical analysis relating a biological response with a landscape attribute at a set of specified spatial scales and extracts the statistical strength of the models through a specified criterion index. Also, it draws a plot with the value of these indexes, allowing the user to choose the most appropriate spatial scale. This paper introduces the usage of multifit and demonstrates its functionality through a case study.

Conclusions

The spatial scale at which ecologists conduct studies may change study outcomes and conclusions. Because of this, landscape ecologists commonly conduct multi-scale studies in order to establish an appropriate spatial scale for particular biological or ecological responses. The tool presented here allows ecologists to simultaneously run several statistical models for a response variable and a specified set of spatial scales, automating the process of multi-scale analysis.

     

Multi-scale Mexican spotted owl (<Emphasis Type="Italic">Strix occidentalis lucida</Emphasis>) nest/roost habitat selection in Arizona and a comparison with single-scale modeling results

Context

Organisms commonly respond to their environment across a range of scales, however many habitat selection studies still conduct selection analyses using a single-scale framework. The adoption of multi-scale modeling frameworks in habitat selection studies can improve the effectiveness of these studies and provide greater insights into scale-dependent relationships between species and specific habitat components.

Objectives

Our study assessed multi-scale nest/roost habitat selection of the federally “Threatened” Mexican spotted owl (Strix occidentalis lucida) in northern Arizona, USA in an effort to provide improved conservation and management strategies for this subspecies.

Methods

We conducted multi-scale habitat modeling to assess habitat selection by Mexican spotted owls using survey data collected by the USFS. Each selected covariate was included in multi-scale models at their “characteristic scale” and we used an all-subsets approach and model selection framework to assess habitat selection.

Results

The “characteristic scale” identified for each covariate varied considerably among covariates and results from multi-scale models indicated that percent canopy cover and slope were the most important covariates with respect to habitat selection by Mexican spotted owls. Multi-scale models consistently outperformed their analogous single-scale counterparts with respect to the proportion of deviance explained and model predictive performance.

Conclusions

Efficacy of future habitat selection studies will benefit by taking a multi-scale approach. In addition to potentially providing increased explanatory power and predictive capacity, multi-scale habitat models enhance our understanding of the scales at which species respond to their environment, which is critical knowledge required to implement effective conservation and management strategies.

     

What determines the spatial extent of landscape effects on species?

Context

Landscape ecologists are often interested in measuring the effects of an environmental variable on a biological response; however, the strength and direction of effect depend on the size of the area within which the environmental variable is measured. Thus a central objective is to identify the optimal spatial extent within which to measure the environmental variable, i.e. the “scale of effect”.

Objectives

Our objectives are (1) to provide a comprehensive summary of the hypotheses concerning what determines the scale of effect, (2) to provide predictions that can be tested in empirical studies, and (3) to show, with a review of the literature, that most of these predictions have so far been inadequately tested.

Methods

We propose 14 predictions derived from five hypotheses explaining what determines the scale of effect, and review the literature (if any) supporting each prediction. These predictions involve five types of factors: (A) species traits, (B) landscape variables, (C) biological responses (e.g. abundance vs. occurrence), (D) indirect influences, and (E) regional context of the study. We identify methodological issues that hinder estimation of the scale of effect.

Results

Of the 14 predictions, only nine have been tested empirically and only five have received some empirical support. Most support is from simulation studies. Empirical evidence usually does not support predictions.

Conclusions

The study of the spatial scale at which landscape variables influence biological outcomes is in its infancy. We provide directions for future research by clarifying predictions concerning the determinants of the scale of effect.

     

Seasonal and temporal changes in species use of the landscape: how do they impact the inferences from multi-scale habitat modeling?

Context

Multi-scale approaches to habitat modeling have been shown to provide more accurate understanding and predictions of species-habitat associations. It remains however unexplored how spatial and temporal variations in habitat use may affect multi-scale habitat modeling.

Objectives

We aimed at assessing how seasonal and temporal differences in species habitat use and distribution impact operational scales, variable influence, habitat suitability spatial patterns, and performance of multi-scale models.

Methods

We evaluated the environmental factors driving brown bear habitat relationships in the Cantabrian Range (Spain) based on species presence records (ground observations) for the period 2000–2010, LiDAR data on forest structure, and seasonal estimates of foraging resources. We separately developed multi-scale habitat models for (i) each season (spring, summer, fall and winter) (ii) two sub-periods with different population status: 2000–2004 (with brown bear distribution restricted to the main population nuclei) and 2005–2010 (with expanding bear population and range); and (iii) the entire 2000–2010 period.

Results

Scales of effect remained considerably stable across seasonal and temporal variations, but not the influence of certain environmental variables. The predictive ability of multi-scale models was lower in the seasons or periods in which populations used larger areas and a broader variety of environmental conditions. Seasonal estimates of foraging resources, together with LiDAR data, appeared to improve the performance of multi-scale habitat models.

Conclusions

We highlight that the understanding of multi-scale behavioral responses of species to spatial patterns that continually shift over time may be essential to unravel habitat relationships and produce reliable estimates of species distributions.

     

Using step and path selection functions for estimating resistance to movement: pumas as a case study

Context

GPS telemetry collars and their ability to acquire accurate and consistently frequent locations have increased the use of step selection functions (SSFs) and path selection functions (PathSFs) for studying animal movement and estimating resistance. However, previously published SSFs and PathSFs often do not accommodate multiple scales or multi-scale modeling.

Objectives

We present a method that allows multiple scales to be analyzed with SSF and PathSF models. We also explore the sensitivity of model results and resistance surfaces to whether SSFs or PathSFs are used, scale, prediction framework, and GPS collar sampling interval.

Methods

We use 5-min GPS collar data from pumas (Puma concolor) in southern California to model SSFs and PathSFs at multiple scales, to predict resistance using two prediction frameworks (paired and unpaired), and to explore potential bias from GPS collar sampling intervals.

Results

Regression coefficients were extremely sensitive to scale and pumas exhibited multiple scales of selection during movement. We found PathSFs produced stronger regression coefficients, larger resistance values, and superior model performance than SSFs. We observed more heterogeneous surfaces when resistance was predicted in a paired framework compared with an unpaired framework. Lastly, we observed bias in habitat use and resistance results when using a GPS collar sampling interval longer than 5 min.

Conclusions

The methods presented provide a novel way to model multi-scale habitat selection and resistance from movement data. Due to the sensitivity of resistance surfaces to method, scale, and GPS schedule, care should be used when modeling corridors for conservation purposes using these methods.

     

Assessing wild bees in perennial bioenergy landscapes: effects of bioenergy crop composition,landscape configuration,and bioenergy crop area

Context

Wild bee populations are currently under threat, which has led to recent efforts to increase pollinator habitat in North America. Simultaneously, U.S. federal energy policies are beginning to encourage perennial bioenergy cropping (PBC) systems, which have the potential to support native bees.

Objectives

Our objective was to explore the potentially interactive effects of crop composition, total PBC area, and PBC patches in different landscape configurations.

Methods

Using a spatially-explicit modeling approach, the Lonsdorf model, we simulated the impacts of three perennial bioenergy crops (PBC: willow, switchgrass, and prairie), three scenarios with different total PBC area (11.7, 23.5 and 28.8% of agricultural land converted to PBC) and two types of landscape configurations (PBC in clustered landscape patterns that represent realistic future configurations or in dispersed neutral landscape models) on a nest abundance index in an Illinois landscape.

Results

Our modeling results suggest that crop composition and PBC area are particularly important for bee nest abundance, whereas landscape configuration is associated with bee nest abundance at the local scale but less so at the regional scale.

Conclusions

Strategies to enhance wild bee habitat should therefore emphasize the crop composition and amount of PBC.

     

Sensitivity of resource selection and connectivity models to landscape definition

Context

The definition of the geospatial landscape is the underlying basis for species-habitat models, yet sensitivity of habitat use inference, predicted probability surfaces, and connectivity models to landscape definition has received little attention.

Objectives

We evaluated the sensitivity of resource selection and connectivity models to four landscape definition choices including (1) the type of geospatial layers used, (2) layer source, (3) thematic resolution, and (4) spatial grain.

Methods

We used GPS telemetry data from pumas (Puma concolor) in southern California to create multi-scale path selection function models (PathSFs) across landscapes with 2500 unique landscape definitions. To create the landscape definitions, we identified seven geospatial layers that have been shown to influence puma habitat use. We then varied the number, sources, spatial grain, and thematic resolutions of these layers to create our suite of plausible landscape definitions. We assessed how PathSF model performance (based on AIC) was affected by landscape definition and examined variability among the predicted probability of movement surfaces, connectivity models, and road crossing locations.

Results

We found model performance was extremely sensitive to landscape definition and identified only seven top models out of our suite of definitions (<1%). Spatial grain and the number of geospatial layers selected for a landscape definition significantly affected model performance measures, with finer grains and greater numbers of layers increasing model performance.

Conclusions

Given the sensitivity of habitat use inference, predicted probability surfaces, and connectivity models to landscape definition, out results indicate the need for increased attention to landscape definition in future studies.

     

Multi-level,multi-scale habitat selection by a wide-ranging,federally threatened snake

Context

Although multi-scale approaches are commonly used to assess wildlife-habitat relationships, few studies have examined selection at multiple spatial scales within different hierarchical levels/orders of selection [sensu Johnson’s (1980) orders of selection]. Failure to account for multi-scale relationships within a single level of selection may lead to misleading inferences and predictions.

Objectives

We examined habitat selection of the federally threatened eastern indigo snake (Drymarchon couperi) in peninsular Florida at the level of the home range (Level II selection) and individual telemetry location (Level III selection) to identify influential habitat covariates and predict relative probability of selection.

Methods

Within each level, we identified the characteristic scale for each habitat covariate to create multi-scale resource selection functions. We used home range selection functions to model Level II selection and paired logistic regression to model Level III selection.

Results

At both levels, EIS selected undeveloped upland land covers and habitat edges while avoiding urban land covers. Selection was generally strongest at the finest scales with the exception of Level II urban edge which was avoided at a broad scale indicating avoidance of urbanized land covers rather than urban edge per se.

Conclusions

Our study illustrates how characteristic scales may vary within a single level of selection and demonstrates the utility of multi-level, scale-optimized habitat selection analyses. We emphasize the importance of maintaining large mosaics of natural habitats for eastern indigo snake conservation.

     

Resource depletion versus landscape complementation: habitat selection by a multiple central place forager

Context

The spatial distribution of non-substitutable resources implies diverging predictions for animal movement patterns. At broad scales, animals should respond to landscape complementation by selecting areas where resource patches are close-by to minimize movement costs. Yet at fine scales, central place effects lead to the depletion of patches that are close to one another and that should ultimately be avoided by consumers.

Objectives

We developed a multi-scale resource selection framework to test whether animal movement is driven by landscape complementation or resource depletion and identify at which spatial scale these processes are relevant from an animal’s perspective.

Methods

During the dry season, surface water and forage are non-substitutable resources for African elephants. Eight family herds were tracked using GPS loggers in Hwange National Park, Zimbabwe. We explained habitat selection during foraging trips by mapping surface water at two scales with gaussian kernels of varying widths placed over each waterhole.

Results

Unexpectedly, elephants select areas with low waterhole density at both fine scales (< 1 km) and broad scales (5–7 km). Selection is stronger when elephants forage far away from water, even more so as the dry season progresses.

Conclusions

Elephant selection of low waterhole density areas suggests that resource depletion around multiple central places is the main driver of their habitat selection. By identifying the scale at which animals respond to waterhole distribution we provide a template for water management in arid and semi-arid landscapes that can be tailored to match the requirements and mobility of free ranging wild or domestic species.

     

Modeling local spatial patterns of wild bee diversity in Pennsylvania apple orchards

Context

The positive correlation between landscape area of semi-natural habitat and wild pollinator richness and abundance in agroecosystems has been well studied. However, we lack a deep understanding of local scale floral resource and nest provisioning for wild bees necessary to optimize implementation of pollinator conservation practices.

Objectives

The primary objective of this study was to use a spatially interactive landscape pollination model (hereafter, the Lonsdorf model) to represent field scale spatial patterns of wild bee abundance and richness within a heterogeneous landscape in the mid-Atlantic USA.

Methods

We parameterized the Lonsdorf model with high resolution aerial imagery and insight from a previously published floristic study. To test the Lonsdorf model predictions, field studies were conducted to measure wild bee abundance and species richness in apple orchards as a function of distance from a forest edge.

Results

Field measurements indicated apple pollinator abundance and species richness significantly decreased with increasing distance from the forest edge. The Lonsdorf model pollination service score was highly sensitive to changes in resource provisioning in orchard and non-crop areas, and including resource rich forest and forest edge habitats in the model significantly improved pollination service estimates.

Conclusions

We demonstrated a novel application of the Lonsdorf model at a field scale to predict trends in pollination service provisioning as a factor of local habitat features. With sufficiently detailed inputs, the Lonsdorf model is a promising tool to quantify field scale pollination service deficits, guiding more cost effective habitat supplementation and other conservation efforts.

     

The past and future of modeling forest dynamics: from growth and yield curves to forest landscape models

Context

Quantitative models of forest dynamics have followed a progression toward methods with increased detail, complexity, and spatial extent.

Objectives

We highlight milestones in the development of forest dynamics models and identify future research and application opportunities.

Methods

We reviewed milestones in the evolution of forest dynamics models from the 1930s to the present with emphasis on forest growth and yield models and forest landscape models We combined past trends with emerging issues to identify future needs.

Results

Historically, capacity to model forest dynamics at tree, stand, and landscape scales was constrained by available data for model calibration and validation; computing capacity; model applicability to real-world problems; and ability to integrate biological, social, and economic drivers of change. As computing and data resources improved, a new class of spatially explicit forest landscape models emerged.

Conclusions

We are at a point of great opportunity in development and application of forest dynamics models. Past limitations in computing capacity and in data suitable for model calibration or evaluation are becoming less restrictive. Forest landscape models, in particular, are ready to transition to a central role supporting forest management, planning, and policy decisions.

Recommendations

Transitioning forest landscape models to a central role in applied decision making will require greater attention to evaluating performance; building application support staffs; expanding the included drivers of change, and incorporating metrics for social and economic inputs and outputs.

     

Multi-scale mismatches between urban sprawl and landscape fragmentation create windows of opportunity for conservation development

Context

Urban sprawl and the expanding transportation infrastructure drive land consumption and landscape fragmentation, causing environmental deterioration and loss of species. Current understanding of how these drivers interact to shape landscape fragmentation is still poor. However, a strong correlation between urban sprawl and landscape fragmentation patterns is commonly assumed.

Objectives

Our main objective was to test the strength, non-stationarity, and scale-dependency of the relationship between urban sprawl and landscape fragmentation patterns (‘sprawl-fragmentation relationship’). Subsequently, we propose an extended framework for the links between urban sprawl, expansion of transport infrastructure, and landscape fragmentation.

Methods

We quantified spatial patterns of urban sprawl and landscape fragmentation for mainland Spain at multiple scales. We then fitted global regression models and geographically weighted regression models with metrics of landscape fragmentation and urban sprawl.

Results

Most variation in landscape fragmentation values (almost 80 % on average) is not explained by urban sprawl metrics through global modeling. Local models show substantial improvements in model performance, with an average of 37 % of the variance remaining unexplained. The contribution of urban sprawl to landscape fragmentation patterns varies locally and depends on scale, with higher contributions at coarser scales and at higher organizational levels.

Conclusions

Our investigation revealed three critical characteristics of the sprawl-fragmentation relationship: it does not prevail, is non-stationary, and scale-dependent. We propose four mechanisms that may have resulted in this mismatch: scale, time-lagged development, spatial arrangement of development, and other external variables including teleconnections. These spatial mismatches provide windows of opportunity for conservation through better development strategies.

     

Surface metrics for landscape ecology: a comparison of landscape models across ecoregions and scales

Context

The patch-mosaic model is lauded for its conceptual simplicity and ease with which conventional landscape metrics can be computed from categorical maps, yet many argue it is inconsistent with ecological theory. Gradient surface models (GSMs) are an alternative for representing landscapes, but adoption of surface metrics for analyzing spatial patterns in GSMs is hindered by several factors including a lack of meaningful interpretations.

Objectives

We investigate the performance and applicability of surface metrics across a range of ecoregions and scales to strengthen theoretical foundations for their adoption in landscape ecology.

Methods

We examine metric clustering across scales and ecoregions, test correlations with patch-based metrics, and provide ecological interpretations for a variety of surface metrics with respect to forest cover to support the basis for selecting surface metrics for ecological analyses.

Results

We identify several factors complicating the interpretation of surface metrics from a landscape perspective. First, not all surface metrics are appropriate for landscape analyses. Second, true analogs between surface metrics and patch-based, landscape metrics are rare. Researchers should focus instead on how surface measures can uniquely measure spatial patterns. Lastly, scale dependencies exist for surface metrics, but relationships between metrics do not appear to change considerably with scale.

Conclusions

Incorporating gradient surfaces into landscape ecological analyses is challenging, and many surface metrics may not have patch analogs or be ecologically relevant. For this reason, surface metrics should be considered in terms of the set of pattern elements they represent that can then be linked to landscape characteristics.

     

Multi-scale habitat selection modeling: a review and outlook

Context

Scale is the lens that focuses ecological relationships. Organisms select habitat at multiple hierarchical levels and at different spatial and/or temporal scales within each level. Failure to properly address scale dependence can result in incorrect inferences in multi-scale habitat selection modeling studies.

Objectives

Our goals in this review are to describe the conceptual origins of multi-scale habitat selection modeling, evaluate the current state-of-the-science, and suggest ways forward to improve analysis of scale-dependent habitat selection.

Methods

We reviewed more than 800 papers on habitat selection from 23 major ecological journals published between 2009 and 2014 and recorded a number of characteristics, such as whether they addressed habitat selection at multiple scales, what attributes of scale were evaluated, and what analytical methods were utilized.

Results

Our results show that despite widespread recognition of the importance of multi-scale analyses of habitat relationships, a large majority of published habitat ecology papers do not address multiple spatial or temporal scales. We also found that scale optimization, which is critical to assess scale dependence, is done in less than 5 % of all habitat selection modeling papers and less than 25 % of papers that address “multi-scale” habitat analysis broadly defined.

Conclusions

Our review confirms the existence of a powerful conceptual foundation for multi-scale habitat selection modeling, but that the majority of studies on wildlife habitat are still not adopting multi-scale frameworks. Most importantly, our review points to the need for wider adoption of a formal scale optimization of organism response to environmental variables.

     

Synthetic Aperture Radar (SAR) images improve habitat suitability models

Context

The ability to detect ecological networks in landscapes is of utmost importance for managing biodiversity and planning corridors.

Objectives

The objective of this study was to evaluate the information provided by a synthetic aperture radar (SAR) image for landscape connectivity modeling compared to aerial photographs (APs).

Methods

We present a novel method that integrates habitat suitability derived from remote sensing imagery into a connectivity model to explain species abundance. More precisely, we compared how two resistance maps constructed using landscape and/or local metrics derived from AP or SAR imagery yield different connectivity values (based on graph theory), considering hedgerow networks and forest carabid beetle species as a model.

Results

We found that resistance maps using landscape and local metrics derived from SAR imagery improve landscape connectivity measures. The SAR model is the most informative, explaining 58% of the variance in forest carabid beetle abundance. This model calculates resistance values associated with homogeneous patches within hedgerows according to their suitability (canopy cover density and landscape grain) for the model species.

Conclusions

Our approach combines two important methods in landscape ecology: the construction of resistance maps and the use of buffers around sampling points to determine the importance of landscape factors. This study was carried out through an interdisciplinary approach involving remote sensing scientists and landscape ecologists. This study is a step forward in developing landscape metrics from satellites to monitor biodiversity.

     

Estimating the spatial scales of landscape effects on abundance

Context

Spatial variation in abundance is influenced by local- and landscape-level environmental variables, but modeling landscape effects is challenging because the spatial scales of the relationships are unknown. Current approaches involve buffering survey locations with polygons of various sizes and using model selection to identify the best scale. The buffering approach does not acknowledge that the influence of surrounding landscape features should diminish with distance, and it does not yield an estimate of the unknown scale parameters.

Objectives

The purpose of this paper is to present an approach that allows for statistical inference about the scales at which landscape variables affect abundance.

Methods

Our method uses smoothing kernels to average landscape variables around focal sites and uses maximum likelihood to estimate the scale parameters of the kernels and the effects of the smoothed variables on abundance. We assessed model performance using a simulation study and an avian point count dataset.

Results

The simulation study demonstrated that estimators are unbiased and produce correct confidence interval coverage except in the rare case in which there is little spatial autocorrelation in the landscape variable. Canada warbler abundance was more highly correlated with site-level measures of NDVI than landscape-level NDVI, but the reverse was true for elevation. Canada warbler abundance was highest when elevation in the surrounding landscape, defined by an estimated Gaussian kernel, was between 1300 and 1400 m.

Conclusions

Our method provides a rigorous way of formally estimating the scales at which landscape variables affect abundance, and it can be embedded within most classes of statistical models.

     

Integrated models that unite local and regional data reveal larger-scale environmental relationships and improve predictions of species distributions

Context

The scale of environmental relationships is often inferred through the use of species distribution models. Yet such models are frequently developed at two distinct scales. Coarse-scale models typically use information-poor (e.g., presence-only) data to predict relative distributions across geographic ranges, whereas fine-scale models often use richer information (e.g., presence–absence data) to predict distributions at local to landscape scales.

Objectives

We unite presence–absence and presence-only data to predict occurrence of species, what we refer to as integrated distribution models. We determine if integrated models improve predictions of species distributions and identification of characteristic spatial scales of environmental relationships relative to presence–absence modeling and ensemble modeling that averages predictions from separate presence-only and presence–absence models.

Methods

We apply recent advances in integrated distribution models to predict Sherman’s fox squirrel (Sciurus niger shermani) distribution in north-central Florida. Presence-only data were collected through a citizen-science program across its geographic range, while presence–absence data were collected using camera trapping surveys across 40 landscapes.

Results

Integrated models estimated environmental relationships with greater precision and identified larger characteristic scales for environmental relationships than using presence–absence data alone. In addition, integrated models tended to have greater predictive performance, which was more robust to the amount of presence–absence and presence-only data used in modeling, than presence–absence and ensemble models.

Conclusions

Integrated distribution models hold much potential for improving our understanding of environmental relationships, the scales at which environmental relationships operate, and providing more accurate predictions of species distributions. Many avenues exist for further advancement of these modeling approaches.

     

Anuran responses to spatial patterns of agricultural landscapes in Argentina

Context

Amphibians are declining worldwide and land use change to agriculture is recognized as a leading cause. Argentina is undergoing an agriculturalization process with rapid changes in landscape structure.

Objectives

We evaluated anuran response to landscape composition and configuration in two landscapes of east-central Argentina with different degrees of agriculturalization. We identified sensitive species and evaluated landscape influence on communities and individual species at two spatial scales.

Methods

We compared anuran richness, frequency of occurrence, and activity between landscapes using call surveys data from 120 sampling points from 2007 to 2009. We evaluated anuran responses to landscape structure variables estimated within 250 and 500-m radius buffers using canonical correspondence analysis and multimodel inference from a set of candidate models.

Results

Anuran richness was lower in the landscape with greater level of agriculturalization with reduced amount of forest cover and stream length. This pattern was driven by the lower occurrence and calling activity of seven out of the sixteen recorded species. Four species responded positively to the amount of forest cover and stream habitat. Three species responded positively to forest cohesion and negatively to rural housing. Two responded negatively to crop area and diversity of cover classes.

Conclusions

Anurans within agricultural landscapes of east-central Argentina are responding to landscape structure. Responses varied depending on species and study scale. Life-history traits contribute to responses differences. Our study offers a better understanding of landscape effects on anurans and can be used for land management in other areas experiencing a similar agriculturalization process.

     

Exploring dispersal barriers using landscape genetic resistance modelling in scarlet macaws of the Peruvian Amazon

Context

Dispersal is essential for species persistence and landscape genetic studies are valuable tools for identifying potential barriers to dispersal. Macaws have been studied for decades in their natural habitat, but we still have no knowledge of how natural landscape features influence their dispersal.

Objectives

We tested for correlations between landscape resistance models and the current population genetic structure of macaws in continuous rainforest to explore natural barriers to their dispersal.

Methods

We studied scarlet macaws (Ara macao) over a 13,000 km² area of continuous primary Amazon rainforest in south-eastern Peru. Using remote sensing imagery from the Carnegie Airborne Observatory, we constructed landscape resistance surfaces in CIRCUITSCAPE based on elevation, canopy height and above-ground carbon distribution. We then used individual- and population-level genetic analyses to examine which landscape features influenced gene flow (genetic distance between individuals and populations).

Results

Across the lowland rainforest we found limited population genetic differentiation. However, a population from an intermountain valley of the Andes (Candamo) showed detectable genetic differentiation from two other populations (Tambopata) located 20–60 km away (F _ST = 0.008, P = 0.001–0.003). Landscape resistance models revealed that genetic distance between individuals was significantly positively related to elevation.

Conclusions

Our landscape resistance analysis suggests that mountain ridges between Candamo and Tambopata may limit gene flow in scarlet macaws. These results serve as baseline data for continued landscape studies of parrots, and will be useful for understanding the impacts of anthropogenic dispersal barriers in the future.