Assessing landscape functional connectivity in a forest carnivore using path selection functions

Context

Understanding connectivity patterns in relation to habitat fragmentation is essential to landscape management. However, connectivity is often judged from expert opinion or species occurrence patterns, with very few studies considering the actual movements of individuals. Path selection functions provide a promising tool to infer functional connectivity from animal movement data, but its practical application remains scanty.

Objectives

We aimed to describe functional connectivity patterns in a forest carnivore using path-level analysis, and to explore how connectivity is affected by land cover patterns and road networks.

Methods

We radiotracked 22 common genets in a mixed forest-agricultural landscape of southern Portugal. We developed path selection functions discriminating between observed and random paths in relation to landscape variables. These functions were used together with land cover information to map conductance surfaces.

Results

Genets moved preferentially within forest patches and close to riparian habitats. Functional connectivity declined with increasing road density, but increased with the proximity of culverts, viaducts and bridges. Functional connectivity was favoured by large forest patches, and by the presence of riparian areas providing corridors within open agricultural land. Roads reduced connectivity by dissecting forest patches, but had less effect on riparian corridors due to the presence of crossing structures.

Conclusions

Genet movements were jointly affected by the spatial distribution of suitable habitats, and the presence of a road network dissecting such habitats and creating obstacles in areas otherwise permeable to animal movement. Overall, the study showed the value of path-level analysis to assess functional connectivity patterns in human-modified landscapes.

     

Measuring landscape connectivity: On the importance of within-patch connectivity

Context

Many connectivity metrics have been used to measure the connectivity of a landscape and to evaluate the effects of land-use changes and potential mitigation measures. However, there are still gaps in our understanding of how to accurately quantify landscape connectivity.

Objectives

A number of metrics only measure between-patch connectivity, i.e. the connectivity between different habitat patches, which can produce misleading results. This paper demonstrates that the inclusion of within-patch connectivity is important for accurate results.

Methods

The behavior of two metrics is compared: the Connectance Index (CONNECT), which measures only between-patch connectivity, and the effective mesh size (m_eff), which includes both within-patch and between-patch connectivity. The connectivity values of both metrics were calculated on a set of simulated landscapes. Twenty cities were then added to these landscapes to calculate the resulting changes in connectivity.

Results

We found that when using CONNECT counter-intuitive results occurred due to not including within-patch connectivity, such as scenarios where connectivity increased with increasing habitat loss and fragmentation. These counter-intuitive results were resolved when using m_eff. For example, landscapes with low habitat amount may be particularly sensitive to urban development, but this is not reflected by CONNECT.

Conclusions

Applying misleading results from metrics like CONNECT can have detrimental effects on natural ecosystems, because reductions in within-patch connectivity by human activities are neglected. Therefore, this paper provides evidence for the crucial need to consider the balance between within-patch connectivity and between-patch connectivity when calculating the connectivity of landscapes.

     

Quantifying landscape connectivity through the use of connectivity response curves

Habitat connectivity is an essential component of biodiversity conservation. Simulated landscapes were manipulated to quantify the impact of changes to the amount, fragmentation and dispersion of habitat on a widely applied landscape connectivity metric, the probability of connectivity index. Index results for different landscape scenarios were plotted against the dispersal distances used for their calculation to create connectivity response curves for each scenario. Understanding index response to controlled changes in landscape structure at a range of spatial scales can be used to give context to comparison of alternative landscape management scenarios. Increased amounts of habitat, decreased fragmentation and decreased inter-patch distances resulted in increased connectivity index values. Connectivity response curves demonstrated increases in assessed connectivity for scenarios with continuous corridors or “stepping stone” connectors. The sensitivity of connectivity response curves to controlled changes in landscape structure indicate that this approach is able to detect and distinguish between different types of landscape changes, but that delineation of habitat and method of quantifying dispersal probability incorporate assumptions that must be recognized when interpreting results to guide landscape management. Representing landscape connectivity in this manner allows for the impacts of alternative landscape management strategies to be compared visually through comparative plots, or statistically through the parameters that describe connectivity response curves.     

Improving habitat and connectivity model predictions with multi-scale resource selection functions from two geographic areas

Landscape Ecology - Habitat loss and fragmentation are the most pressing threats to biodiversity, yet assessing their impacts across broad landscapes is challenging. Information on habitat...     

Habitat suitability modelling to correlate gene flow with landscape connectivity

Landscape connectivity is important in designing corridor and reserve networks. Combining genetic distances among individuals with least-cost path (LCP) modelling helps to correlate indirect measures of gene flow with landscape connectivity. Applicability of LCP modelling, however, is reduced if knowledge on dispersal pathways or routes is lacking. Therefore, we integrated habitat suitability modelling into LCP analysis to avoid the subjectivity common in LCP analyses lacking knowledge on dispersal pathways or routes. We used presence-only data and ecological niche factor analysis to model habitat suitability for the spiny rat, Niviventer coninga, in a fragmented landscape of western Taiwan. We adapted the resultant habitat suitability map for incorporation into LCP analyses. Slightly increased Mantel correlations indicated that a class-weighted suitability map better explained genetic distances among individuals than did geographical distances. The integration of habitat suitability modelling into LCP analysis can thus generate information on distribution of suitable habitats, on potential routes of dispersal, for placement of corridors, and evaluate landscape connectivity. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Examining speed versus selection in connectivity models using elk migration as an example

Context

Landscape resistance is vital to connectivity modeling and frequently derived from resource selection functions (RSFs). RSFs estimate relative probability of use and tend to focus on understanding habitat preferences during slow, routine animal movements (e.g., foraging). Dispersal and migration, however, can produce rarer, faster movements, in which case models of movement speed rather than resource selection may be more realistic for identifying habitats that facilitate connectivity.

Objective

To compare two connectivity modeling approaches applied to resistance estimated from models of movement rate and resource selection.

Methods

Using movement data from migrating elk, we evaluated continuous time Markov chain (CTMC) and movement-based RSF models (i.e., step selection functions [SSFs]). We applied circuit theory and shortest random path (SRP) algorithms to CTMC, SSF and null (i.e., flat) resistance surfaces to predict corridors between elk seasonal ranges. We evaluated prediction accuracy by comparing model predictions to empirical elk movements.

Results

All connectivity models predicted elk movements well, but models applied to CTMC resistance were more accurate than models applied to SSF and null resistance. Circuit theory models were more accurate on average than SRP models.

Conclusions

CTMC can be more realistic than SSFs for estimating resistance for fast movements, though SSFs may demonstrate some predictive ability when animals also move slowly through corridors (e.g., stopover use during migration). High null model accuracy suggests seasonal range data may also be critical for predicting direct migration routes. For animals that migrate or disperse across large landscapes, we recommend incorporating CTMC into the connectivity modeling toolkit.

     

A landscape connectivity approach to mitigating the urban heat island effect

Landscape Ecology - Urban integration has exacerbated the spreading of urban heat islands (UHIs) across cities. Blue/green landscapes embedded within urban areas, behaving as cool islands...     

Making the connection: combining habitat suitability and landscape connectivity to understand species distribution in an agricultural landscape

Landscape Ecology - The current biodiversity crisis has intensified the need to predict species responses to landscape modification and has renewed attention on the fundamental question of what...     

Genetic analysis of landscape connectivity in tree populations

Genetic connectivity in plant populations is determined by gene movement within and among populations. When populations become genetically isolated, they are at risk of loss of genetic diversity that is critical to the long-term survival of populations. Anthropogenic landscape change and habitat fragmentation have become so pervasive that they may threaten the genetic connectivity of many plant species. The theoretical consequences of such changes are generally understood, but it is not immediately apparent how concerned we should be for real organisms, distributed across real landscapes. Our goals here are to describe how one can study gene movement of both pollen and seeds in the context of changing landscapes and to explain what we have learned so far. In the first part, we will cover methods of describing pollen movement and then review evidence for the impact of fragmentation in terms of both the level of pollen flow into populations and the genetic diversity of the resulting progeny. In the second part, we will describe methods for contemporary seed movement, and describe findings about gene flow and genetic diversity resulting from seed movement. Evidence for pollen flow suggests high connectivity, but it appears that seed dispersal into fragments may create genetic bottlenecks due to limited seed sources. Future work should address the interaction of pollen and seed flow and attention needs to be paid to both gene flow and the diversity of the incoming gene pool. Moreover, if future work is to model the impact of changing landscapes on propagule movement, with all of its ensuing consequences for genetic connectivity and demographic processes, we will need an effective integration of population genetics and landscape ecology.     

Sensitivity of landscape resistance estimates based on point selection functions to scale and behavioral state: pumas as a case study

Estimating landscape resistance to animal movement is the foundation for connectivity modeling, and resource selection functions based on point data are commonly used to empirically estimate resistance. In this study, we used GPS data points acquired at 5-min intervals from radiocollared pumas in southern California to model context-dependent point selection functions. We used mixed-effects conditional logistic regression models that incorporate a paired used/available design to examine the sensitivity of point selection functions to the scale of available habitat and to the behavioral state of individual animals. We compared parameter estimates, model performance, and resistance estimates across 37 scales of available habitat, from 250 to 10,000 m, and two behavioral states, resource use and movement. Point selection functions and resistance estimates were sensitive to the chosen scale of the analysis. Multiple characteristic scales were found across our predictor variables, indicating that pumas in the study area are responding at different scales to different landscape features and that multi-scale models may be more appropriate. Additionally, point selection functions and resistance estimates were sensitive to behavioral state; specifically, pumas engaged in resource use behavior had an opposite selection response to some land cover types than pumas engaged in movement behavior. We recommend examining a continuum of scales and behavioral states when using point selection functions to estimate resistance.     

Dispersal traits determine plant response to habitat connectivity in an urban landscape

Identification of trait syndromes that make species vulnerable to habitat fragmentation is essential in predicting biodiversity change. Plants are considered particularly vulnerable if their capacities for persistence in and for dispersal among local habitats are low. Here we investigated the influence of easily measured functional traits on the presence of 45 plant species in an urban landscape in north-west Germany where patches were separated by distances consistent with regular plant dispersal range. To describe the spatial configuration of patches we calculated species-specific patch connectivities. Then we assessed plant connectivity responses in distribution models calculated from connectivities and environmental predictors. Twenty (45%) of the analysed species showed a positive connectivity response after accounting for species-specific habitat requirements. These species differed from non-responsive species by functional traits associated with dispersal, including reduced seed numbers and higher terminal velocities relative to non-responsive species. Persistence traits played however no role which we attribute to the environmental conditions of urban habitats and their spatiotemporal characteristics. Our study underlines that even ruderal plants experience dispersal limitation and demonstrates that easily measured functional traits may be used as indicators of fragmentation vulnerability in urban systems allowing generalizations to larger species sets.     

Signal detection theory clarifies the concept of perceptual range and its relevance to landscape connectivity

Understanding functional connectivity is critical for several issues in ecology and conservation. When animals actively search for habitat across landscapes, their perceptual range of habitats can profoundly influence connectivity. Nonetheless, conceptual development and estimation of perceptual ranges and their influence on connectivity have been limited. Signal detection theory (SDT) has a long tradition in several disciplines to address the problem of detecting stimuli in noisy and uncertain environments. SDT is particularly useful for understanding perceptual ranges because it acknowledges uncertainty in the detection process and distinguishes between two key parameters that have previously been confounded when interpreting the perceptual range of animals: signal detectability and response bias of individuals. Here we extend SDT to the concept of perceptual range, provide approaches for estimating patch detectability and response bias, and apply this framework to interpreting the perceptual range of cactus bugs (Chelinidea vittiger). We find that signal detectability of habitat by C. vittiger, and thus their perceptual range, is between 1 and 2 m, based on generalized linear models aimed at estimating signal detection parameters. SDT provides new insights into perceptual ranges and functional connectivity, which may help understand intra and interspecific variation in animal responses to modified landscapes.     

Considering landscape connectivity and gene flow in the Anthropocene using complementary landscape genetics and habitat modelling approaches

Landscape Ecology - A comprehensive understanding of how rapidly changing environments affect species gene flow is critical for mitigating future biodiversity losses. While recent methodological...     

Sensitivity of ecosystem goods and services projections of a forest landscape model to initialization data

Projections of indicators of forest ecosystem goods and services (EGS) based on process-based landscape models are critical for adapting forest management to climate change. However, the scarcity of fine-grained, spatially explicit forest data means that initializing these models is both a challenge and a source of uncertainty. To test how different initialization approaches influence the simulation of forest dynamics and EGS indicators we initialized the forest landscape model LandClim with fine resolution empirical data, coarse empirical data, and simulation-derived data, and evaluated the results at three spatial scales (stand, management area and landscape). Simulations were performed for a spruce (Picea abies) dominated landscape in the Black Forest, Germany, under current climate and a climate change scenario. We found that long-term (>150 years) projections are robust to initialization uncertainty. In contrast, shorter-term projections are sensitive to initialization uncertainty, with sensitivity increasing when EGS are assessed at smaller spatial scales, and when the EGS indicators depend on the spatial distribution of individual species. EGS dynamics are strongly influenced by interactions between the density, species composition, and age structure of initialized forests and simulated forest management. If EGS dynamics are strongly influenced by climate change, such as when climate change induces mortality in drought-sensitive species, some of the initialization uncertainty can be masked. We advocate for initializing landscape models with fine-grained data in applications that focus on spatial management problems in heterogeneous landscapes, and stress that the scale of analysis must be in accordance with the accuracy that is warranted by the initialization data.     

Coupling ecosystem and landscape models to study the effects of plot number and location on prediction of forest landscape change

A fundamental but unsolved dilemma is that observation and prediction scales are often mismatched. Reconciling this mismatch largely depends on how to design samples on a heterogeneous landscape. In this study, we used a coupled modeling approach to investigate the effects of plot number and location on predicting tree species distribution at the landscape scale. We used an ecosystem process model (LINKAGES) to generate tree species response to the environment (a land type) at the plot scale. To explore realistic parameterization scenarios we used results from LINKAGES simulations on species establishment probabilities under the current and warming climate. This allowed us to design a series of plot number and location scenarios at the landscape scale. Species establishment probabilities for different land types were then used as input for the forest landscape model (LANDIS) that simulated tree species distribution at the landscape scale. To investigate the effects of plot number and location on forest landscape predictions, LANDIS considered effects of climate warming only for the land types in which experimental plots were placed; otherwise inputs for the current climate were used. We then statistically examined the relationships of response variables (species percent area) among these scenarios and the reference scenario in which plots were placed on all land types of the study area. Our results showed that for species highly or moderately sensitive to environmental heterogeneity, increasing plot numbers to cover as many land types as possible is the strategy to accurately predict species distribution at the landscape scale. In contrast, for species insensitive to environmental heterogeneity, plot location was more important than plot number. In this case, placing plots in land types with large area of species distribution is warranted. For some moderately sensitive species that experienced intense disturbance, results were different in different simulation periods. Results from this study may provide insights into sample design for forest landscape predictions.     

Linking movement behavior and fine-scale genetic structure to model landscape connectivity for bobcats (Lynx rufus)

Spatial heterogeneity can constrain the movement of individuals and consequently genes across a landscape, influencing demographic and genetic processes. In this study, we linked information on landscape composition, movement behavior, and genetic differentiation to gain a mechanistic understanding of how spatial heterogeneity may influence movement and gene flow of bobcats in the agricultural landscape of Iowa (USA). We analyzed movement paths of 23 animals to parameterize landscape resistance surfaces, applied least cost path analysis to generate measures of effective geographic distance between DNA collection locations of 625 bobcats, and tested the correlation between genetic distance and the different models of geographic distance. We found that bobcats showed a strong preference for forest over any other habitat type, and that incorporating information on habitat composition both along the path and in the surrounding landscape provided the best model of movement. Measures of effective geographic distance were significantly correlated with genetic distance, but not once the effects of Euclidean distance were accounted for. Thus, despite the impact of habitat composition on movement behavior, we did not detect a signature of a landscape effect in genetic structure. Our results are consistent with the issue of limiting factors: the high uniformity of forest fragmentation across southern Iowa, the primary study area, results in a landscape resistance pattern virtually indistinguishable from the isolation-by-distance pattern. The northern portion of the state, however, is predicted to pose a high level of resistance to bobcat movement, which may impede the regional genetic connectivity of populations across the Midwest.     

Sensitivity and effectiveness and of landscape metric scalograms in determining the characteristic scale of a hierarchically structured landscape

Landscape metric scalograms (the response curves of landscape metrics to changing grain size) have been used to illustrate the scale effects of metrics for real landscapes. However, whether they detect the characteristic scale of hierarchically structured landscapes remains uncertain. To address this question, the scalograms of 26 class-level metrics were systematically examined for a simple random landscape, seven hierarchical neutral landscapes, and the real landscape of the Xilin River Basin of Inner Mongolia, China. The results show that when the fraction of the focal patch type (P) is below a critical value (P _c), most metric scalograms are sensitive to change in single-scale and lower-level hierarchical structure and insensitive to change in higher-level hierarchical structure. The scalograms of only a few metrics measuring spatial aggregation and connectedness are sensitive to change in intermediate-level hierarchical structure. Most metric scalograms explicitly identify the characteristic scale of a single-scale landscape and fine or intermediate characteristic scales of a multi-scale landscape for both simulated and real landscapes. When P exceeds P _c, only some metrics detect scale and change in structure. The scalograms of total class area and Euclidean nearest-neighbor distance cannot detect scale or change in structure in either case. Landscape metric scalograms are useful for addressing scale issues, including illustrating the scale effects of spatial patterns, detecting multi-scale patterns, and developing possible scaling relations.

     

Optimizing landscape selection for estimating relative effects of landscape variables on ecological responses

Empirical studies of the relative effects of landscape variables may compromise inferential strength with common approaches to landscape selection. We propose a methodology for landscape sample selection that is designed to overcome some common statistical pitfalls that may hamper estimates of relative effects of landscape variables on ecological responses. We illustrate our proposed methodology through an application aimed at quantifying the relationships between farmland heterogeneity and biodiversity. For this project, we required 100 study landscapes that represented the widest possible ranges of compositional and configurational farmland heterogeneity, where these two aspects of heterogeneity were quantified as crop cover diversity (Shannon diversity index) and mean crop field size, respectively. These were calculated at multiple spatial extents from a detailed map of the region derived through satellite image segmentation and classification. Potential study landscapes were then selected in a structured approach such that: (1) they represented the widest possible range of both heterogeneity variables, (2) they were not spatially autocorrelated, and (3) there was independence (no correlation) between the two heterogeneity variables, allowing for more precise estimates of the regression coefficients that reflect their independent effects. All selection criteria were satisfied at multiple extents surrounding the study landscapes, to allow for multi-scale analysis. Our approach to landscape selection should improve the inferential strength of studies estimating the relative effects of landscape variables, particularly those with a view to developing land management guidelines.     

Selection of scale for Everglades landscape models

This article addresses the problem of determining the optimal “Model Grain” or spatial resolution (scale) for landscape modeling in the Everglades. Selecting an appropriate scale for landscape modeling is a critical task that is necessary before using spatial data for model development. How the landscape is viewed in a simulation model is dependent on the scale (cell size) in which it is created. Given that different processes usually have different rates of fluctuations (frequencies), the question of selection of an appropriate modeling scale is a difficult one and most relevant to developing spatial ecosystem models. The question of choosing the appropriate scale for modeling is addressed using the landscape indices (e.g., cover fraction, diversity index, fractal dimension, and transition probabilities) recently developed for quantifying overall characteristics of spatial patterns. A vegetation map of an Everglades impoundment area developed from SPOT satellite data was used in the analyses. The data from this original 20 × 20 m data set was spatially aggregated to a 40 × 40 m resolution and incremented by 40 meters on up to 1000 × 1000 m (i.e., 40, 80, 120, 160 … 1000) scale. The primary focus was on the loss of information and the variation of spatial indices as a function of broadening “Model Grain” or scale. Cover fraction and diversity indices with broadening scale indicate important features, such as tree islands and brush mixture communities in the landscape, nearly disappear at or beyond the 700 m scale. The fractal analyses indicate that the area perimeter relationship changes quite rapidly after about 100 m scale. These results and others reported in the paper should be useful for setting appropriate objectives and expectations for Everglades landscape models built to varying spatial scales.