The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes

Habitat quality and metapopulation effects are the main hypotheses that currently explain the disproportionate decline of insects in cultivated Holarctic landscapes. The former assumes a degradation in habitat quality for insects within surviving ecosystems, the latter that too few, small or isolated islands of ecosystem remain in landscapes for populations to persist. These hypotheses are often treated as alternatives, and this can lead to serious conflict in the interpretations of conservationists. We present the first empirical demonstration that habitat quality and site isolation are both important determinants of where populations persist in modern landscapes. We described the precise habitat requirements of Melitaea cinxia, Polyommatus bellargus and Thymelicus acteon, and quantified the variation in carrying capacity within each butterfly's niche. We then made detailed surveys to compare the distribution and density of every population of each species with the size, distance apart and quality of their specific habitats in all their potential habitat patches in three UK landscapes. In each case, within-site variation in habitat quality explained which patches supported a species' population two to three times better than site isolation. Site area and occupancy were not correlated in any species. Instead of representing alternative paradigms, habitat quality and spatial effects operate at different hierarchical levels within the same process: habitat quality is the missing third parameter in metapopulation dynamics, contributing more to species persistence, on the basis of these results, than site area or isolation. A reorientation in conservation priorities is recommended.     

Developing a functional connectivity indicator to detect change in fragmented landscapes

Biodiversity indicators are increasingly used to assess progress towards conservation targets. Particular indicators are required to assess the impacts of habitat fragmentation on landscape connectivity and biodiversity value. This paper recognises that connectivity is best defined by the interaction between species and the landscape in which they occur, and proposes a functional approach to assess connectivity. The approach utilises an incidence function model (IFM) as a spatially explicit method to assess potential species-level connectivity. The standard IFM connectivity measure is modified to account for the influence of the surrounding landscape matrix on edge impacts (through a weighted internal edge buffer) and ecological isolation (through an assessment of least-cost distance to account for landscape permeability). It has been recognised that such patch-based connectivity measures can provide misleading results when used to examine change, as they only focus on between patch movements. As a result, a modified hybrid IFM, based on a combination of patch and cell-based approaches, is developed to account for both within (intra) and between (inter) patch connectivity. The resulting probability of functional connectivity (PFC) indicator was evaluated, alongside a patch-based connectivity measure, through the application to four model landscapes based on changes (2 negative and 2 positive) to a control landscape. The four model landscapes illustrate the impact of landscape change on habitat area, edge impacts and matrix permeability. The proposed PFC indicator successfully discriminated between the two negative and the two positive changes to the control landscape, whereas, the patch-based connectivity measure detected change successfully within three of the four landscapes. The PFC indicator predicted a decrease in intra and inter-patch connectivity following habitat loss and fragmentation (negative change 1), whereas patch-based connectivity measures indicate an increase in connectivity between fragmented patches. The proposed PFC indicator offers the opportunity to take the necessary species-based perspective to examine functional connectivity, incorporating habitat preference, dispersal probability, edge impacts and ecological isolation/permeability. The urgency to assess changes in connectivity and support conservation policy means that there is little time to wait for more complete data. We believe the proposed approach provides a robust balance between the data required and the biologically meaningful indicator produced.     

Habitat selection by Canada lynx: making do in heavily fragmented landscapes

Habitat loss and fragmentation result in landscapes where high quality habitat patches are surrounded by matrix habitats of low and variable quality. For mobile species to persist in such landscapes, individual animals often rely on the high quality habitats but also use matrix habitats for supplemental resources or while moving between higher quality patches. Determining what habitat features animals select when in these matrix areas is important, as retaining desirable features in lower quality habitats may enable species persistence. We examine a population of US federally threatened Canada lynx (Lynx canadensis) in northcentral Washington, near the southwestern range limit, where lynx habitat is fragmented by topography, wildfires, and human impacts. We used Global Positioning System radio-collar data from 17 lynx in the North Cascade Mountains during 2007–2013 to explore lynx habitat use. We used Random Forest models to analyze core hunting, resting, and denning habitat, and the habitats lynx select while between patches of core habitat. While selecting core habitat, lynx used spruce (Picea engelmannii)-fir (Abies lasiocarpa), lodgepole pine (Pinus contorta), and mixed sub-boreal-Douglas fir (Pseudotsuga menziesii) forests, and avoided dry forests and forest openings including new burns. When not in core habitat, lynx used a wider range of habitats, including new burns where fire skips and residual trees offered cover. Our results show clearly that Canada lynx tolerate a wider range of habitats where they occupy fragmented landscapes. Consequently, maintaining animals in fragmented landscapes requires that we identify and conserve not only the core habitats a particular species selects, but also the habitat features animals use while in less suitable environments.     

Habitat quality and connectivity in agricultural landscapes: The role of land use systems at various scales in time

Connectivity is a key concept of landscape ecology as it relates to flows and movements of organisms as driven by landscape structure. More and more aspects of landscape heterogeneity are considered in measuring connectivity, as the diversity of crops in agricultural landscapes. In this paper, we explored the value of considering changes and cumulated effects of connectivity over time. As an example, we analysed connectivity among patches influenced by maize over 7 years in an agricultural landscape in Brittany, France.Clear temporal patterns appeared: maize is concentrated in certain parts of the landscape, but over the period the whole area, 70% of the landscape, used for maize was connected. Instead of discrete patches, maize may produce large clusters allowing movement from patch to patch from year to year. This reinforces the importance of understanding land use allocation rules within farms and landscapes to evaluate the ecological effects of agriculture.     

Patch-occupancy dynamics in fragmented landscapes

Recent work on the dynamics of species living In fragmented landscapes has produced much Information on patterns of habitat patch occupancy in a wide range of organisms. Building on an elementary Markov chain model of patch occupancy, a family of Incidence-function models can be constructed for particular kinds of metapopulations. These models can be parameterized with field data on patch occupancy, and the models can be used to make quantitative predictions about specific metapopulations. This approach provides a potentially powerful tool for the management of reserve networks and species living in fragmented landscapes.     

Neutral communities in fragmented landscapes

Habitat fragmentation displays a crucial role in conservation biology. Despite this, little is known about the detailed ecological consequences of habitat fragmentation due to the scarce number of controlled experimental surveys. The species–area relationship, a fundamental concept in ecology, requires the understanding of the fragmentation effects in a long term perspective, which turns this task even harder. Here we address the spatial patterns of species distribution in fragmented landscapes, assuming a neutral community model. We study the species area relationship and how its shape changes as the landscape becomes more fragmented. Recent investigations, based on extensive computer simulation, have contributed to establish some definite conclusions in the study of non‐fragmented landscapes: the existence of a three‐regime or two‐regime scenario for the species–area relationship, the emergence of a power‐law regime at intermediate scales and the augment of the species–area exponent z with the speciation rate. Despite the recent efforts, some other questions remain, such as the dependence of z in the whole range of the speciation rate. Questions like these are currently debated but generalizations cannot be drawn. This is the first paper, to our knowledge, that uses the coalescence method and neutral theory to examine biodiversity on more complex spatial structures. Our simulation results corroborate that the fragmentation plays a crucial role in shaping the species–area relationship, by determining the existence and extension of the power‐law regime associated with small and intermediate areas. On the other hand, when individuals are allowed to disperse over longer distances the species–area relationship now displays the classic triphasic pattern, and the intermediate regime, which is well described by a power‐law, is established even for highly fragmented landscapes.     

Dispersal and extinction in fragmented landscapes

Evolutionary and population dynamics models suggest that the migration rate will affect the probability of survival in fragmented landscapes. Using data for butterfly species in the fragmented British landscape and in immediately adjoining areas of the European continent, this paper shows that species of intermediate mobility have declined most, followed by those of low mobility, whereas high-mobility species are generally surviving well. Compared to the more sedentary species, species of intermediate mobility require relatively large areas where they breed at slightly lower local densities. Intermediate mobility species have probably fared badly through a combination of metapopulation (extinction and colonization) dynamics and the mortality of migrating individuals which fail to find new habitats in fragmented landscapes. Habitat fragmentation is likely to result in the non-random extinction of populations and species characterized by different levels of dispersal, although the details are likely to depend on the taxa, habitats and regions considered.     

Contrasting effects of habitat quantity and quality on moth communities in fragmented landscapes

Habitat loss is commonly identified as a major threat to the loss of global biodiversity. In this study, we expand on our previous work by addressing the question of how lepidopteran species richness and composition vary among remnants of North American eastern deciduous forest located within agricultural or pastoral landscapes. Specifically, we tested the relative roles of habitat quantity (measured as stand area and percent forest in the greater landscape) and habitat quality (measured as tree species diversity) as determinants of moth species richness. We sampled >19 000 individuals comprising 493 moth species from 21 forest sites in two forested ecoregions. In the unglaciated Western Allegheny Plateau, the species richness of moths with woody host plants diminished as forest stand size and percent forest in the landscape decreased, but the total species richness and abundance of moths were unaffected by stand size, percent forest in the landscape, or tree species diversity. In contrast, the overall species richness and abundance of moths in the glaciated North Central Tillplain were affected primarily by tree species diversity and secondarily by forest size. Higher tree species diversity may reduce species loss from smaller forest stands, suggesting that small, diverse forests can support comparable numbers of species to those in less diverse, large stands. Smaller forests, however, contained a disproportionate number of moth species that possess larvae known to feed on herbaceous vegetation. Thus, although woody plant feeding moths are lost from forests with changes in stand area, new species appear capable of recolonizing smaller fragments from the surrounding habitat matrix. Our study further suggests that when species replacement occurs, local patch size and habitat quality may be more important than landscape context in determining the community structure of forest Lepidoptera.     

Forest-climate interactions in fragmented tropical landscapes

In the tropics, habitat fragmentation alters forest-climate interactions in diverse ways. On a local scale (less than 1 km), elevated desiccation and wind disturbance near fragment margins lead to sharply increased tree mortality, thus altering canopy-gap dynamics, plant community composition, biomass dynamics and carbon storage. Fragmented forests are also highly vulnerable to edge-related fires, especially in regions with periodic droughts or strong dry seasons. At landscape to regional scales (10-1000 km), habitat fragmentation may have complex effects on forest-climate interactions, with important consequences for atmospheric circulation, water cycling and precipitation. Positive feedbacks among deforestation, regional climate change and fire could pose a serious threat for some tropical forests, but the details of such interactions are poorly understood.     

Making statistics biologically relevant in fragmented landscapes

The biological impacts of habitat fragmentation are routinely assessed using standard statistical modelling techniques that are used across many ecological disciplines. However, to assess the biological relevance of fragmentation impacts, we must consider an extra, spatial dimension to the standard statistical model: the biological importance of a significant and well supported model with large effect sizes crucially depends on the configuration of habitat within the study area. We argue that mapping the outputs from statistical models across a study area generates biologically meaningful estimates of fragmentation impacts. Integrating traditional statistical approaches with geographic information systems will facilitate rigorous comparisons of fragmentation impacts between taxa, studies and ecosystems.     

Habitat quality and population density drive occupancy dynamics of snowshoe hare in variegated landscapes

The influence of habitat quality and population density on occupancy dynamics may surpass that of traditional metrics of area and isolation, but often this is not considered explicitly in studies of spatially structured populations. In landscapes that are not easily characterized as binary habitat/non‐habitat (e.g. variegated landscapes), this influence may be even more important and occur at both local and landscape levels. It follows that occupancy dynamics may be driven by disparate processes depending on how extinction or colonization relate to habitat quality and population density. We examined the relative influence of area, structural isolation, habitat quality, local population density, and neighborhood population density (i.e. population density in the landscape around a site) on the probability of extinction and colonization of snowshoe hare Lepus americanus across an expansive forest mosaic landscape (encompassing the northern third of Idaho). Habitat quality and population density were highly influential in determining extinction and colonization, whereas patch area and isolation were much less important. Sites with heavier vegetative cover at the site or landscape‐level were more likely to be colonized and less likely to go extinct, and sites with greater local population density in the previous time step had lower probability of extinction. Sites embedded in high density neighborhoods also were less likely to go extinct, but not more likely to be colonized. We found a significant interaction between local and neighborhood population density on extinction in 1 yr, suggesting that the strength of demographic rescue may vary dependent on local site densities. Our results add to a growing literature showing that factors outside of structural metrics of area and isolation are important drivers of occupancy dynamics. Given the multi‐scaled influence of habitat quality and population density on occupancy dynamics, our work also indicates that research on snowshoe hare must extend beyond simply assessing local factors to understand the spatial dynamics of populations.     

Distribution and abundance of dung beetles in fragmented landscapes

Related species utilising similar resources are often assumed to show similar spatial population structures and dynamics. This paper reports substantial ecological variation within a set of Aphodius dung beetles occurring in the same patchily distributed resource, livestock dung in pastures. We show how variation in habitat and resource selectivity, in the rate of movements between pastures, and in the distribution of local population sizes all contribute to interspecific differences in spatial population structures. Local dung beetle assemblages are compared between two landscapes with different densities of pastures. In one of the landscapes, we contrast the abundances and regional distributions of Aphodius before and after 15 years of rapid habitat loss. Different species show very dissimilar responses to changes in the structure of the landscape. Our results suggest that generalist Aphodius species, and specialist species with high dispersal powers, occur as large "patchy" populations in the landscape. In contrast, a strict pasture specialist species with limited dispersal powers (A. pusillus) forms classical metapopulations. At the community level, interspecific differences in spatial population structures make the local community composition a function of the structure of the surrounding landscape.     

Finding habitat patches and directional connectivity

For animal species inhabiting patchy environments, the search behavior of individuals and the distance from which they can detect suitable habitat (perceptual range) are key determinants of the functional connectivity of landscapes. We examined the movement behavior and perceptual range of adult cactus bugs ( Chelinidea vittiger ), which are habitat specialists that feed and reproduce on Opuntia cactus. Movement pathways of walking individuals released into unsuitable matrix habitat (30–3000 m from Opuntia ) were highly directional. These results supported predictions of optimal search behavior from published simulation models. A release experiment within natural patch networks indicated that the perceptual range of C. vittiger depended on size of the target patch, matrix structure, and direction of the target patch relative to prevailing winds. A strong effect of wind direction on orientation behavior (and presumed perceptual range) was evident in a release experiment using 'artificial' patches of potted Opuntia . In these two experiments, individuals released 50–100 cm from Opuntia patches were more likely to orientate toward patches located upwind than to those located crosswind or downwind. A reexamination of the pathways of individuals walking in the matrix also revealed a strong bias for movement upwind. Such upwind movement by individuals, both within and outside of patch networks, suggests that C. vittiger uses olfaction to navigate and it complicates our ability to interpret search behavior and to estimate perceptual range. Current techniques for assessing perceptual range have limitations for olfactory-based species. Furthermore, we need to broaden our view of perceptual range and of patch and landscape connectivity. Perceptual range may be anisotropic and connectivity may be directional. An organism-based approach to spatial ecology requires that we consider the dominant senses used by species when navigating around patchy landscapes.     

Emergent properties of conspecific attraction in fragmented landscapes

Attraction to conspecifics may have wide-ranging implications for habitat selection and metapopulation theory, yet little is known about the process of attraction and its effects relative to other habitat selection strategies. Using individual-based simulations, I investigated the emergent properties of conspecific attraction during habitat selection on survival, fecundity, short-term fitness (survival x fecundity), and distributions in fragmented landscapes. I simulated conspecific attraction during searching and settlement decisions and compared attraction with random, habitat-based (searching for the presence of habitat), and habitat quality sampling strategies (searching for and settling in high-quality habitat). Conspecific attraction during searching or settlement decisions had different consequences for animals: attraction while searching increased survival by decreasing time spent in nonsuitable habitat, whereas attraction during settlement increased fecundity by aggregating animals in high-quality habitats. Habitat-based sampling did not improve fitness over attraction, but directly sampling habitat quality resulted in the highest short-term fitness among strategies. These results suggest that attraction can improve fitness when animals cannot directly assess habitat quality. Interestingly, conspecific attraction influenced distributions by generating patch size effects and weak edge effects, highlighting that attraction is one potential, yet previously unappreciated, mechanism to explain the widespread patterns of animal sensitivity to habitat fragmentation.     

Forces structuring tree squirrel communities in landscapes fragmented by agriculture: species differences in perceptions of forest connectivity and carrying capacity

The North American red squirrel ( Tamiasciurus hudsonicus ) has expanded its range into the central hardwoods of the United States in conjunction with increasing forest fragmentation and declining gray squirrel ( Sciurus carolinensis ) populations. We used translocation experiments and patch occupancy data to test for interspecific differences in mobility and sensitivity to habitat loss and modification by agriculture. We released squirrels in fencerows to test the hypothesis that gray squirrels display inferior mobility relative to red and fox ( S. niger ) squirrels. Elapsed time to movement from fencerows for 76 individuals increased with distance to forest patches and harvesting of crops. Gray and red squirrels took longer to move from fencerows than fox squirrels, and gray squirrels were less successful at moving from fencerows than red and fox squirrels. Ecologically scaled landscape indices revealed the degree to which interspecific differences in mobility and individual area requirements accounted for the occurrence of these species across landscapes. Gray squirrel distribution was constrained both by individual area requirements and dispersal ability. Occurrence of red and fox squirrels was related to patch size but was unaffected by landscape connectivity.     

Relative importance of quantity, quality and isolation of patches for butterfly diversity in fragmented urban forests

The majority of forests in urban areas are small and isolated. Improving habitat quality of small forests instead of increasing habitat size and connectivity could be an effective means of conserving the biodiversity of such highly fragmented landscapes. In this study, we investigated the relative importance of habitat quantity, quality and isolation on butterfly assemblages in urban fragmented forests in Tokyo, Japan. We used four habitat geographic parameters: (1) fragment size, (2) shape index, (3) isolation (distance to the mainland), and (4) connectivity; and three habitat quality parameters: (1) herbaceous nectar plant abundance, (2) herbaceous nectar plant diversity, and (3) larval host plant diversity. We surveyed butterfly assemblages along transects in 20 forest fragments that ranged in size from 1 to 122 ha. We used generalized linear models to relate the number of species in a fragment to four habitat geographic parameters and three habitat quality parameters. The averaged models based on AIC_c showed that fragment size had a strong positive effect on butterfly species richness. There was also a positive effect of herbaceous nectar plant abundance on species diversity. These findings suggest that improving the habitat quality of small and isolated forests in highly fragmented landscapes may be capable of maintaining levels of butterfly diversity comparable to those of large fragments.     

Mobility-dependent effects on species richness in fragmented landscapes

In fragmented landscapes, mobility is an important trait for population persistence but the predictions on the relationship between habitat fragmentation and extinction risk are contradictory. Here, we test the effects of the two main aspects of fragmentation, patch area and isolation, on the species richness of groups of butterflies associated with semi-natural grasslands, differing in mobility. Total species richness increased with increasing patch area and with decreasing isolation, but the strength of these effects differed between mobility classes. The effect of patch area was strongest for the sedentary species, while the effect of isolation was only statistically significant for the mobile species. We interpret these results as evidence for a predominant influence of local processes on sedentary species, and an increasing influence of regional compared to local processes with increasing mobility. When groups of species respond differently to habitat loss and fragmentation this affects community composition, with potential implications for ecosystem processes. Similar effects can be expected for other traits than mobility, and this should be an important question for future studies.     

Density dependence slows invader spread in fragmented landscapes

Patchiness is a defining characteristic of most natural and anthropogenic habitats, yet much of our understanding of how invasions spread has come from models of spatially homogeneous environments. Except for populations with Allee effects, an invader's growth rate when rare and dispersal determine its spread velocity; intraspecific competition has little to no influence. How this result might change with landscape patchiness, however, is poorly understood. We used simulation models and their analytical approximations to explore the effect of density dependence on the spread of annual plant invaders moving through heterogeneous landscapes with gaps in suitable habitat. We found that landscape patchiness and discrete invader population size interacted to generate a strong role for density dependence. Intraspecific competition greatly slowed the spread of invasions through patchy landscapes by regulating how rapidly a population could produce enough seeds to surpass habitat gaps. Populations with continuously varying density showed no such effect of density dependence. We adapted a stochastic dispersal model to approximate spread when gap sizes were small relative to the mean dispersal distance and a Markov chain approximation for landscapes with large gaps. Our work suggests that ecologists must consider reproduction at both low and high densities when predicting invader spread.