Species inhabiting fresh waters are severely affected by climate change and other anthropogenic stressors. Effective management and conservation plans require advances in the accuracy and reliability of species distribution forecasts. Here, we forecast distribution shifts of Salmo trutta based on environmental predictors and examine the effect of using different statistical techniques and varying geographical extents on the performance and extrapolation of the models obtained.

Watercourses of Ebro, Elbe and Danube river basins (c. 1,041,000 km²; Mediterranean and temperate climates, Europe).

The occurrence of S. trutta and variables of climate, land cover and stream topography were assigned to stream reaches. Data obtained were used to build correlative species distribution models (SDMs) and forecasts for future decades (2020s, 2050s and 2080s) under the A1b emissions scenario, using four statistical techniques (generalised linear models, generalised additive models, random forest, and multivariate adaptive regression).

The SDMs showed an excellent performance. Climate was a better predictor than stream topography, while land cover characteristics were not necessary to improve performance. Forecasts predict the distribution of S. trutta to become increasingly restricted over time. The geographical extent of data had a weak impact on model performance and gain/loss values, but better species response curves were generated using data from all three basins collectively. By 2080, 64% of the stream reaches sampled will be unsuitable habitats for S. trutta, with Elbe basin being the most affected, and virtually no new habitats will be gained in any basin.

More reliable predictions are obtained when the geographical data used for modelling approximate the environmental range where the species is present. Future research incorporating both correlative and mechanistic approaches may increase robustness and accuracy of predictions.

Assessing the influence of land cover in species distribution modelling is limited by the availability of fine-resolution land-cover data appropriate for most species responses. Remote-sensing technology offers great potential for predicting species distributions at large scales, but the cost and required expertise are prohibitive for many applications. We test the usefulness of freely available raw remote-sensing reflectance data in predicting species distributions of 40 commonly occurring bird species in western Oregon.

Central Coast Range, Cascade and Klamath Mountains Oregon, USA.

Information on bird observations was collected from 4598 fixed-radius point counts. Reflectance data were obtained using 30-m resolution Landsat imagery summarized at scales of 150, 500, 1000 and 2000 m. We used boosted regression tree (BRT) models to analyse relationships between distributions of birds and reflectance values and evaluated prediction performance of the models using area under the receiver operating characteristic curve (AUC) values.

Prediction success of models using all reflectance values was high (mean AUC = 0.79 ± 0.10 SD). Further, model performance using individual reflectance bands exceeded those that used only Normalized Difference Vegetation Index (NDVI). The relative influence of band 4 predictors was highest, indicating the importance of variables associated with vegetation biomass and photosynthetic activity. Across spatial scales, the average influence of predictors at the 2000 m scale was greatest.

We demonstrate that unclassified remote-sensing imagery can be used to produce species distribution models with high prediction success. Our study is the first to identify general patterns in the usefulness of spectral reflectances for species distribution modelling of multiple species. We conclude that raw Landsat Thematic Mapper data will be particularly useful in species distribution models when high-resolution predictions are required, including habitat change detection studies, identification of fine-scale biodiversity hotspots and reserve design.

To establish the robustness of two alternative methods for predicting the future ranges and abundances for two wild-harvested abalone species (Haliotis rubra Donovan 1808 and H. laevigata Leach 1814): single atmosphere–ocean general circulation model (GCM) or ensemble-averaged GCM forecasts.

South Australia.

We assessed the ability of 20 GCMs to simulate observed seasonal sea surface temperature (SST) between 1980–1999, globally, and regionally for the Indian and Pacific Oceans south of the Equator. We used model rankings to characterize a set of representative climate futures, using three different-sized GCM ensembles and two individual GCMs (the Parallel Climate Model and the Community Climate System Model, version 3.0). Ecological niche models were then coupled to physiological information to compare forecast changes in area of occupancy, population size and harvest area based on forecasts using the various GCM selection methods, as well as different greenhouse gas emission scenarios and climate sensitivities.

We show that: (1) the skill with which climate models reproduce recent SST records varies considerably amongst GCMs, with multimodel ensemble averages showing closer agreement to observations than single models; (2) choice of GCM, and the decision on whether or not to use ensemble-averaged climate forecasts, can strongly influence spatiotemporal predictions of range, abundance and fishing potential; and (3) comparable hindcasting skill does not necessarily guarantee that GCM forecasts and ecological and evolutionary responses to these forecast changes, will be similar amongst closely ranked models.

By averaging across an ensemble of seven highly ranked skilful GCMs, inherent uncertainties stemming from GCM differences are incorporated into forecasts of change in species range, abundance and sustainable fishing area. Our results highlight the need to make informed and explicit decisions on GCM choice, model sensitivity and emission scenarios when exploring conservation options for marine species and the sustainability of future harvests using ecological niche models.

Global conservation planning is often oriented around vertebrates and plants, yet most organisms are invertebrates. To explore the potential conservation implications of this bias, we assessed how well patterns of diversity for an influential group of invertebrates, the ants, correspond with those of three vertebrate groups (birds, mammals and amphibians).

Global.

We compiled data on the number of genera of ants and the three vertebrate groups for 370 political regions across the world. We then compared their correlations both for overall diversity and between subsets of genera likely to be of conservation concern. We also developed generalized additive models (GAM) to identify regions where vertebrates and ants diverged in their diversity patterns.

While ant and vertebrate diversity do positively correlate, the correlations are substantially weaker for the ant lineages of the greatest conservation concern. Vertebrates also notably fail to predict ant diversity in specific geographic areas, including Australia and Southeast Asia, parts of Africa and Madagascar, and south-western China. These failures may be genuine differences in diversity patterns, or they may indicate important gaps in our knowledge of ant and vertebrate diversity.

We conclude that it is currently unwise to assume that global conservation priorities based on vertebrates will conserve ants as well. We suspect that this also applies to other invertebrates.

In a highly endemic, threatened amphibian assemblage, we measure and describe the geographic and taxonomic distribution of the amphibian chytrid fungus, Batrachochytrium dendrobatidis (Bd), with a view to identifying those sites and species most at risk of infection and its negative consequences. Additionally, we aimed to determine the potential for direct transmission events between two known carriers of Bd infection.

The island of Sardinia.

We collected swab and tissue samples of amphibians from a wide range of geographic sites and species. We used epidemiological and statistical techniques to quantify deviations from a random distribution. We used random forests to investigate habitat use and overlap in two species known to be infected by Bd to quantify a surrogate measure of the contact rates between these two species.

Both geographic and taxonomic distributions of Bd were highly non-random: we identified a cluster of infections in the north of the island and found that two species, Euproctus platycephalus and Discoglossus sardus, had a relatively high prevalence of infection within this cluster. Our analyses suggest that, on the basis of their fine-scale habitat use, they have relatively little opportunity for direct transmission and could maintain Bd infection independently.

Our results illustrate how obtaining detailed information on the geographic and taxonomic distribution of infection is a useful first step in assessing the risk of infection for species within the region. Attempting to quantify possible routes of transmission amongst species further aids us in identifying mechanisms of pathogen persistence within the host community. Within this assemblage, we have identified two hosts that carry infection and may be at risk. Further, our research suggests that these two species may be able to maintain infection independently of one another, which has implications for attempts to mitigate this parasite.

Accurately documenting and predicting declines and shifts in species’ distributions is fundamental for implementing effective conservation strategies and directing future research; species distribution models (SDM) have become a powerful tool for such work. Nevertheless, much of the data used to create these models are opportunistic and often violate some of their basic assumptions. We use amphibian declines and extinctions linked to the fungus Batrachochytrium dendrobatidis (Bd) to examine how sampling biases in data collection can affect what we know of this disease and its effect on amphibians in the wild.

Queensland, Australia.

We developed a distribution model for Bd incorporating known locality records for Bd and a subset of climatic variables that should correctly characterize its distribution. We tested this (original) model with additional surveys, recorded new Bd observations in novel environments and reran the distribution model. We then investigated the difference between the original and new models, and used frog abundance and infection status data from two of these new localities to look at the susceptibility of the torrent frog Litoria nannotis to chytridiomycosis.

While largely correct, the original SDM underestimated the distribution of Bd; sampling in ‘unsuitable’ drier environments discovered abundant populations of susceptible frogs with pathogen prevalences of up to 100%. The validation surveys further uncovered a new population of the frog Litoria lorica coexisting with the pathogen; this species was previously believed to be an extinct rain forest endemic.

Our results indicate that SDMs constructed using opportunistically collected data can be biased if species are not at equilibrium with their environment or because environmental gradients have not been adequately sampled. For disease ecology, the better estimations of pathogen distribution may lead to the discovery of new populations persisting at the edge of their range, which has important implications for the conservation of species threatened by chytridiomycosis.

Biogeographical theory and conservation valuation schemes necessarily involve assessing how biodiversity is distributed through space and ‘biodiversity’ encapsulates many different aspects of biological organization and information. While biogeography may try to explain biodiversity patterns, successful conservation strategies should attempt to maximize different aspects of diversity. Ultimately, diversity patterns are the product of evolutionary history, and research and conservation efforts seek to understand the unequal distribution of evolutionary history. For conservation efforts, results have been inconsistent as to whether species richness (SR) provides sufficient surrogacy for evolutionary history. Here, we provide a conceptual framework allowing for the direct comparison of taxonomic richness and phylogenetic diversity (PD), both in terms of their mechanistic relationship and the relationship between their spatial distributions.

Global.

We present a framework that relates regional SR, PD, biogeographically weighted evolutionary distinctiveness and biogeographically weighted SR. Further, we use simulations to illustrate how the size of the species pool, topological patterns within the phylogeny and autocorrelation in spatial distributions affect the correlation among metrics.

In regions that include both recently diversified groups and ancient species poor lineages, large species pools and low spatial autocorrelation, the correlation between biodiversity measures is lower than regions with low richness, balanced phylogenetic trees and high spatial autocorrelation.

We can now understand and predict when regional richness and PD should be strongly correlated. This congruency is the product of evolutionary and ecological processes that determine species pool membership and community assembly. Further, in regions where SR is not expected to be congruent with phylogenetic distinctiveness, re-examining how existing reserve networks protect the multiple aspects of biodiversity is critically important.

Those locations and features providing refugia for species during unfavourable climatic phases may be important to identify and conserve to help protect biodiversity in the near future. During the Pleistocene, climates oscillated between glacial and interglacial periods. In the Northern Hemisphere, the impact from glacial ice sheets caused species to disperse to warmer southern refugia, but less is known about changes to species' distributions during these periods in the Southern Hemisphere.

In Australia, the climate alternated between highly arid and wetter periods during the Pleistocene. It has been proposed that arid refugia may be associated with the inland ranges (areas of higher relief), the mesic east, or areas that maintained favourable species-specific ecological conditions. We tested these hypotheses with a phylogeographical analysis of a widely distributed tree-dwelling gecko, Gehyra variegata (2n = 40a chromosomal race) throughout the central and eastern regions of arid Australia.

We generated a mtDNA sequence and microsatellite dataset by sampling 740 G. variegata^40a throughout its known distribution. We also use species distribution modelling to predict the species' likely past, present and future distribution.

The majority of G. variegata^40a lineages diverged during the Pleistocene, and those located in regions of arid Australia where mean annual water run-off is highest, displayed higher levels of genetic diversity in comparison locations with lower run-off. We also show that genetic diversity increased with proximity to water sources.

It is likely that G. variegata^40a contracted to refugia associated with stable water sources during Pleistocene arid phases. However, modelling suggests that unfavourable climate conditions will be present in this region by 2070. Therefore, Pleistocene refugia for G. variegata^40a are unlikely to be refugia in the future. More generally, our results suggest that water run-off could be a useful predictor to identify favourable conditions for some arid species.

Apical pelagic species forage in predictable habitats, and their movements should signal biologically and ecologically significant areas of the marine ecosystem. Several countries are now engaged in identifying these areas based on animal tracking, but this is often limited to a few individuals from one breeding population, which may result in biased portrayals of the key marine habitats. To help identify such foraging areas, we compiled tracking data of a marine top predator from the main breeding colonies in the Central Macaronesia.

North-east Atlantic Ocean.

Over seven years, we tracked the foraging movements of Cory's shearwaters (Calonectris borealis) from several populations during the chick-rearing period using global positioning system and platform terminal transmitter devices.

We obtained foraging trips from 174 shearwaters breeding on six important colonies representative of the range occupied in the Macaronesian Archipelagos of Madeira, Salvages and Canaries. Our results show that birds orient and move rapidly towards the closest neritic waters over the African continental shelf. Birds from different colonies show substantial spatial segregation in their foraging grounds but consistently overlap in some specific foraging areas along the Canary Current characterized by high productivity. By weighting the use of foraging grounds according to the size of each study population, we inferred the overall exploitation of such areas.

Our meta-population approach provides a more comprehensive picture of space use from both perspectives: the studied species and the Canary Current system. Foraging grounds consistently used by several populations may not be adequately identified by tracking a single population, and therefore, multiple population tracking studies are needed to properly delineate key conservation areas and inform conservation planning in the marine ecosystem. Finally, we highlight the long-term stability and sustainability of identified foraging areas and propose that countries with geographical jurisdictions over the Canary Current area should work towards multilateral agreements to set management plans for this key marine ecosystem.

To analyse spatial trade-offs in conservation of mammals, considering benefits of biodiversity conservation and socioeconomic costs.

Cerrado Biodiversity Hotspot, Brazil.

We built ecological niche models (ENMs) for 154 mammals inhabiting the cerrado. We combined model projections to produce consensus maps of species' distributions for three types of models (envelope, statistical and machine-learning models). We used these range summaries as input data in spatial prioritization analyses. We compared six conservation scenarios, considering human population density, land cost, anthropogenic land use, political willingness to act and the distribution of species using trade-off analyses. We complemented the current network of protected areas, aiming to achieve a target of 17% of land protection in the cerrado.

Species-rich sites coincide with regions with high human population, high land cost, high anthropogenic land use and diverse levels of political willingness to act. We found a significant change in spatial priorities when we included socioeconomic and political dimensions in analyses: top-priority sites moved north, reduced by 68% potential conservation conflicts with human population, reduced by 72% the total monetary cost of actions and reduced by 68% conflict with other anthropogenic land uses. It also increased by 80% the beneficial effect of political willingness to act.

Our results reinforce the idea that using only biological criteria for proposing spatial conservation priorities can undermine conservation plans given increases in socioeconomic and political conflicts.

The global database of invasive trees and shrubs (Richardson & Rejmánek, 2011; Diversity Distrib. 17, 788-809) has been updated, resulting in a total of 751 species (434 trees and 317 shrubs) from 90 families. Ten originally listed species were deleted (synonyms, inconclusive identification, etc.) and 139 additional invasive species (86 trees and 53 shrubs) are now included in the database. For many species, new records on their adventive distributions are added. The updated database also includes the native ranges for all listed species.

We used a coupled biophysical ecology (BE)-physiological mechanistic modelling approach based on the Dynamic Energy Budget theory (DEB, Dynamic energy budget theory for metabolic organisation, 2010, Cambridge University Press, Cambridge; DEB) to generate spatially explicit predictions of physiological performance (maximal size and reproductive output) for the invasive mussel, Brachidontes pharaonis.

We examined 26 sites throughout the central Mediterranean Sea.

We ran models under subtidal and intertidal conditions; hourly weather and water temperature data were obtained from the Italian Buoy Network, and monthly CHL-a data were obtained from satellite imagery.

Mechanistic analysis of the B. pharaonis fundamental niche shows that subtidal sites in the Central Mediterranean are generally suitable for this invasive bivalve but that intertidal habitats appear to serve as genetic sinks.

A BE-DEB approach enabled an assessment of how the physical environment affects the potential distribution of B. pharaonis. Combined with models of larval dispersal, this approach can provide estimates of the likelihood that an invasive species will become established.

We assessed the influence of species non-detection in modelling species distributions with an ensemble consensus approach that did not account for imperfect detection, compared with an occupancy model that did.

The hydrographic network of France.

We compared range maps of 35 stream fish species with differing degrees of detectability predicted using a consensus approach combining eight species distribution models (SDMs) to maps produced using an occupancy model. Using a spatially and temporally extensive monitoring database of fish populations (France), we modelled the occurrence of species as a function of several climatic and habitat variables and projected species distributions across the whole of the French hydrographic network. The benefits of occupancy models were then assessed from the differences in both predictive performance and species distribution.

We found that although the occupancy models enhanced the performance for difficult to detect species, consensus models outperformed occupancy models for highly detectable species. In contrast to the minor differences observed in performance measures, estimates of species distributions were severely affected by whether or not imperfect detection was accounted for and varied linearly according to species detectability.

This study demonstrated that false absences could have major consequences in estimating species distribution ranges. However, accounting for imperfect detection may not be enough to improve conventional SDMs. These findings could have important implications for conservation, notably in developing large-scale distribution models and documenting species range shifts in the context of recent climate change.

Recent, rapid population declines in many Afro-Palaearctic migratory bird species have focussed attention on changing conditions within Africa. However, processes influencing population change can operate throughout the annual cycle and throughout migratory ranges. Here, we explore the evidence for impacts of breeding and non-breeding conditions on population trends of British breeding birds of varying migratory status and wintering ecology.

Great Britain (England & Scotland).

Within- and between-species variation in population trends is quantified for 46 bird species with differing migration strategies.

Between 1994 and 2007, rates of population change in Scotland and England differed significantly for 19 resident and 15 long-distance migrant species, but were similar for 12 short-distance migrant species. Of the six long-distance migrant species that winter in the arid zone of Africa, five are increasing in abundance throughout Britain. In contrast, the seven species wintering in the humid zone of Africa are all declining in England, but five of these are increasing in Scotland. Consequently, populations of both arid and humid zone species are increasing significantly faster in Scotland than England, and only the English breeding populations of species wintering in the humid zone are declining.

Population declines in long-distance migrants, especially those wintering in the humid zone, but not residents or short-distance migrants suggest an influence of non-breeding season conditions on population trends. However, the consistently less favourable population trends in England than Scotland of long-distance migrant and resident species strongly suggest that variation in the quality of breeding grounds is influencing recent population changes. The declines in humid zone species in England, but not Scotland, may result from poorer breeding conditions in England exacerbating the impacts of non-breeding conditions or the costs associated with a longer migration, while better conditions in Scotland may be buffering these impacts.

Artificial coastal defence structures are proliferating in response to rising and stormier seas. These structures provide habitat for many species but generally support lower biodiversity than natural habitats. This is primarily due to the absence of environmental heterogeneity and water-retaining features on artificial structures. We compared the epibiotic communities associated with artificial coastal defence structures and natural habitats to ask the following questions: (1) is species richness on emergent substrata greater in natural than artificial habitats and is the magnitude of this difference greater at mid than upper tidal levels; (2) is species richness greater in rock pools than emergent substrata and is the magnitude of this difference greater in artificial than natural habitats; and (3) in artificial habitats, is species richness in rock pools greater at mid than upper tidal levels?

British Isles.

Standard non-destructive random sampling compared the effect of habitat type and tidal height on epibiota on natural rocky shores and artificial coastal defence structures.

Natural emergent substrata supported greater species richness than artificial substrata. Species richness was greater at mid than upper tidal levels, particularly in artificial habitats. Rock pools supported greater species richness than emergent substrata, and this difference was more pronounced in artificial than natural habitats. Rock pools in artificial habitats supported greater species richness at mid than upper tidal levels.

Artificial structures support lower biodiversity than natural habitats. This is primarily due to the lack of habitat heterogeneity in artificial habitats. Artificial structures can be modified to provide rock pools that promote biodiversity. The effect of rock pool creation will be more pronounced at mid than upper tidal levels. The challenge now is to establish at what tidal height the effect of pools becomes negligible and to determine the rock pool dimensions for optimum habitat enhancement.

Tracking the dispersal patterns and habitat use of migratory species is necessary to delineate optimal areas for protection, with large sample sizes being more representative of the population. Here, we examine the dispersal patterns of a key Mediterranean loggerhead turtle (Caretta caretta) breeding population to identify priority foraging sites for protection.

Zakynthos Island, Greece and the wider Mediterranean.

We examined the dispersal patterns and foraging sites of 75 adult loggerheads (n = 38 males and 37 females) tracked from the breeding area of Zakynthos Island (Greece) from 2004 to 2011. We then combined our data with published sea turtle literature to identify key foraging sites for protection.

While both males and females exhibited similar dispersal patterns, about 25% males remained < 100 km of Zakynthos, whereas all females (except one) migrated > 200 km. Integration of our data with the wider literature isolated 10 core sites in proximity to existing protected areas, which could potentially protect 64% of the Zakynthos population, while five sites support individuals from at least 10 other loggerhead breeding populations.

Due to the widespread availability of neritic foraging grounds across the Mediterranean, sea turtles from Zakynthos exhibit disparate dispersal patterns. However, protecting only a few objectively defined important sites can encompass a large proportion of the foraging areas used and hence have considerable conservation benefit.

Conservation efforts in Mediterranean-climate regions are complicated by species' variability in response to multiple threats. Functional type classifications incorporating life history traits with disturbance response strategies provide a framework for predicting groups of species' response to fire, but it is unclear whether these classifications will be useful when species are exposed to multiple threats or differ in spatial context. We evaluate whether species of the same fire-response functional type exhibit similar responses to disturbance relative to, and in combination with, climate and land-use change and whether the dominant threat depends on spatial context.

Mediterranean southern California.

We developed species distribution models under current and future climate conditions for two fire-obligate seeding native shrub species that differ in geographical location and area of occupancy. Dynamic habitat maps representing alternative scenarios of climate change and urban growth were coupled with population models and simulated stochastic fire regimes.

The disturbance that defines their classification, fire, is projected to be the most serious threat to both species when fire frequency is high. At longer fire return intervals, however, the projected ranking of threats differed between the species, and spatial context played an important role in defining vulnerability.

Considering ongoing increases in fire frequency in Mediterranean-climate regions worldwide, functional type classification based on disturbance response may continue to provide a useful framework for biodiversity conservation efforts, but spatial context should also be accounted for. It may be most useful to consider the distribution of vulnerable species with regard to urban development patterns, areas of ‘high-velocity’ climate shifts, and places where altered fire regimes are likely to interact with other threats.

Correlative species distribution models (SDMs) combined with spatial layers of climate and species' localities represent a frequently utilized and rapid method for generating spatial estimates of species distributions. However, an SDM is only as accurate as the inputs upon which it is based. Current best-practice climate layers commonly utilized in SDM (e.g. ANUCLIM) are frequently inaccurate and biased spatially. Here, we statistically downscale 30 years of existing spatial weather estimates against empirical weather data and spatial layers of topography and vegetation to produce highly accurate spatial layers of weather. We proceed to demonstrate the effect of inaccurately quantified spatial data on SDM outcomes.

The Australian Wet Tropics.

We use Boosted Regression Trees (BRTs) to generate 30 years of spatial estimates of daily maximum and minimum temperature for the study region and aggregate the resultant weather layers into ‘accuCLIM’ climate summaries, comparable with those generated by current best-practice climate layers. We proceed to generate for seven species of rainforest skink comparable SDMs within species; one model based on ANUCLIM climate estimates and another based on accuCLIM climate estimates.

Boosted Regression Trees weather layers are more accurate with respect to empirically measured temperature, particularly for maximum temperature, when compared to current best-practice weather layers. ANUCLIM climate layers are least accurate in heavily forested upland regions, frequently over-predicting empirical mean maximum temperature by as much as 7°. Distributions of the focal species as predicted by accuCLIM were more fragmented and contained less core distributional area.

Combined these results reveal a source of bias in climate-based SDMs and indicate a solution in the form of statistical downscaling. This technique will allow researchers to produce fine-grained, ground-truthed spatial estimates of weather based on existing estimates, which can be aggregated in novel ways, and applied to correlative or process-based modelling techniques.

Refugia play a key ecological role for the persistence of biodiversity in areas subject to natural or human disturbance. Temporary freshwater ecosystems regularly experience dry periods, which constrain the availability of suitable habitats. Current and future threats (e.g. water extraction and climate change) can exacerbate the negative effects of drying conditions. This could compromise the persistence of a large proportion of global freshwater biodiversity, so the identification and protection of refugia seems an urgent task.

Northern Australia.

We demonstrate a new approach to identify and prioritize the selection of refugia and apply it to the conservation of freshwater fish biodiversity. We identified refugia using estimates of water residency time derived from satellite imagery and used a systematic approach to prioritize areas that provide all the fish species inhabiting the catchment with access to a minimum number of refugia while maximizing the length of stream potentially accessible for recolonization after the dry period. These priority refugia were locked into a broader systematic conservation plan with area-based targets and direct connectivity. We accounted for current threats during the prioritization process to ensure degraded areas were avoided, thus maximizing the ecological role of priority refugia.

Priority refugia were located in lowland reaches, where the incidence of threats was less prominent in our study area and headwaters in good condition. An additional set of 106 planning units (6500 km²) were required to represent 10% of each species' distribution in the broad conservation plan. A hierarchical management zoning scheme was applied to demonstrate how these key ecological features could be effectively protected from the major threats caused by aquatic invasive species and grazing.

This new approach to identifying priority refugia and incorporating them into the conservation planning process in a systematic way would help enhance the resilience of freshwater biodiversity in temporary systems.

To investigate the contribution to range filling, range extent and climatic niche space of species of information contained in the largest databank of digitized biodiversity data: the global biodiversity information facility (GBIF). We compared such information with a compilation of independent distributional data from natural history collections and other sources.

Europe.

We used data for the hawkmoths (Lepidoptera, family Sphingidae) to assess three aspects of range information: (1) observed range filling in 100 km × 100 km grid cell squares, (2) observed European range extent and (3) observed climatic niche. Range extents were calculated as products of latitudinal and longitudinal extents. Areas derived from minimum convex polygons drawn onto a 2-dimensional niche space representing the two main axes of a principal component analysis (PCA) were used to calculate climatic niche space. Additionally, record-based permutation tests for niche differences were carried out.

We found that GBIF provided many more distribution records than independent compilation efforts, but contributed less information on range filling, range extent and climatic niches of species.

Although GBIF contributed relevant additional information, it is not yet an effective alternative to manual compilation and databasing of distributional records from collections and literature sources, at least in lesser-known taxa such as invertebrates. We discuss possible reasons for our findings, which may help shape GBIF strategies for providing more informative data.

Sufficient data to describe spatial distributions of rare and threatened populations are typically difficult to obtain. For example, there are minimal modern offshore sightings of the endangered southern right whale, limiting our knowledge of foraging grounds and habitat use patterns. Using historical exploitation data of southern right whales (SRW), we aim to better understand their seasonal offshore distribution patterns in relation to broad-scale oceanography, and to predict their exposure to shipping traffic and response to global climate change.

Australasian region between 130° W and 100° E, and 30° S and 55° S.

We model 19th century whaling data with boosted regression trees to determine functional responses of whale distribution relative to environmental factors. Habitat suitability maps are generated and we validate these predictions with independent historical and recent sightings. We identify areas of increased risk of ship-strike by integrating predicted whale distribution maps with shipping traffic patterns. We implement predicted ocean temperatures for the 2090–2100 decade in our models to predict changes in whale distribution due to climate change.

Temperature in the upper 200 m, distance from the subtropical front, mixed layer depth, chlorophyll concentration and distance from ridges are the most consistent and influential predictors of whale distribution. Validation tests of predicted distributions determined generally high predictive capacity. We identify two areas of increased risk of vessel strikes and predict substantial shifts in habitat suitability and availability due to climate change.

Our results represent the first quantitative description of the offshore foraging habitat of SRW. Conservation applications include identifying areas and causes of threats to SRW, generating effective mitigation strategies, and directing population monitoring and research efforts. Our study demonstrates the benefits of incorporating unconventional datasets such as historical exploitation data into species distribution models to inform management and help combat biodiversity loss.

During this century rapid climate change will have a profound effect on global biodiversity, and species survival will be contingent on their ability to adapt or disperse. Species distribution models are a popular tool for gauging how the distribution of favourable climate may change over space and time. Evaluating the level of dispersal between the current distribution and potential future distribution of a species is a key to predicting their survival, but rarely estimated. Here we applied species distribution models and a genetic estimate of dispersal to quantitatively assess dispersal to new distributions in the timescale imposed by future climate change.

We sampled 635 adult Gehyra variegata (2n = 40a/38b) throughout central and eastern Australia, encompassing much of the recorded distribution for this gecko. We genotyped all individuals at 16 microsatellite loci, from which we estimated mean annual dispersal distance using Wright's neighbourhood size. Species distribution modelling predicted the current and future distribution of the species, and we used annual dispersal distances to evaluate whether the species could keep up with shifts in the range of their favourable climatic conditions.

Our estimates of mean dispersal showed that 17–41% of the current G. variegata (2n = 40a/38b) distribution was unlikely to contribute to their future distribution given the timescale imposed by future global climate change.

Our approach can make further use of molecular and occurrence record datasets to answer whether a species has the capacity to reach future areas of favourable climate and the extent to which the current distribution will contribute to this process.

Two alternative genetic patterns are typical in recently established populations. One is reduced genetic diversity but significant population structuring compared with original populations. The other is the persistence of genetic polymorphisms and the lack of differentiation in frontier populations. This study aims to test for these patterns by examining population genetics of the Light-vented Bulbul (Pycnonotus sinensis), an Oriental bird that has been undergoing rapid range expansion.

Eastern China.

Molecular analyses were conducted on mitochondrial DNA and microsatellite datasets from 256 individuals. Intraspecific phylogeny was reconstructed by Bayesian inference and network analysis. Analysis of molecular variance (AMOVA) and Bayesian clustering were applied to determine population structure. Genetic diversity was tested to determine whether there were significant differences between frontier and source populations. Population expansions were tested in mtDNA and microsatellites. Gene flow and recent migrants were estimated by Bayesian methods.

Both high- and low-nucleotide diversities were recorded in frontier populations, and differences in genetic diversity between frontier and source populations were not statistically significant. Population differentiation was recorded in some source but not in frontier populations. Population size expansion was detected both in frontier and in source populations, and for the whole dataset, commencing before the Last Glacial Maximum.

Genetic diversity can be maintained in expanding populations of the Light-vented Bulbul, and genetic differentiation can be avoided, where substantial genetic exchanges are occurring. This study also discusses the potential effects of evolutionary properties such as historical population growth on recently recorded range expansion.

In the Northern Hemisphere, bird migration from the tropic to the temperate zone in spring is thought to proceed at a rate determined in large part by local phenology. In contrast, little is understood about where birds go or the factors that determine why they move or where they stop during the post-breeding period.

Study sites were in Oregon, Nebraska and Vermont, and location data we collected extend south to Argentina.

We deployed light-level geolocators on individual Bobolinks from three populations across the breeding range and compare their southbound movement phenology to austral greening as indicated by the Normalized Difference Vegetation Index.

Bobolinks from all breeding populations synchronously arrived and remained for up to several weeks in two sequential, small non-breeding areas that were separated by thousands of kilometres, before staging for pre-alternate moult. Similar to the migration patterns of birds to northern breeding areas, movements into the Southern Hemisphere corresponded to increasing primary productivity.

Our findings suggest that the Bobolink's southbound migration is broadly constrained by resource availability, and its non-breeding distribution has been shaped by the seasonal phenology of grasslands in both time and space. This is the first documentation of individual birds from across a continental breeding range exhibiting phenological matching during their post-breeding southward migration. Known conservation threats overlap temporally and spatially with large concentrations of Bobolinks, and should be closely examined. We emphasize the need to consider how individuals move and interact with their environment throughout their annual cycle and over hemispheric scales.

To build and assess pathway-level non-indigenous species (NIS) establishment curves generated using a propagule pressure (PP) proxy and historical establishment data.

North America

Our analysis examines the utility and behaviour of pathway-level NIS establishment curves that relate species-level PP to establishment probability. Using theoretical and empirical methods, we examine the behaviour of pathway-level establishment models when species are heterogeneous in their ability to establish. Next, we examine the implications of using PP proxy and historical establishment data to parameterize these models. Finally, we test the model by building an establishment curve for aquarium fish establishments in the United States using import data as a proxy for PP.

First, we show theoretically how species' heterogeneity and the use of a proxy metric for PP affect model parameterization and the interpretation of the establishment curve. Second, we demonstrate that import data are relatively consistent across space and time for aquarium fish species. Finally, we demonstrate how basic import-level data can improve our ability to predict which species are at risk of establishment using aquarium fish introductions to the United States as a case study.

Pathway-level analyses generated using species-level PP information can provide a snapshot of establishment probability for use in risk analyses without in-depth knowledge of species' abiotic and biotic interactions. Proxy data for PP can be a good metric for such analyses, and valid predictions can be expected when the PP data are relatively consistent across the time period for which establishments are recorded.

Introduced alien species are frequently implicated in ecosystem disruption and biodiversity loss, but some ecologists have recently argued that efforts to manage ecosystems should be refocused on known problematic species without regard to whether such species are native or alien. This argument rests on the premise that native and alien species in general do not differ in their impacts. Although there are numerous cases that suggest alien predators and herbivores can sometimes cause severe declines or even local extinctions of native species, very few studies have compared the impacts of native and alien consumers on native populations.

World-wide.

We have conducted a meta-analysis on a global dataset to compare the effects of native and alien predators and herbivores on native populations occupying a broad range of terrestrial and aquatic environments.

The distribution of positive, negative and neutral effects on native prey abundance differed significantly by consumer origin, with alien consumers associated with more negative and fewer positive effects than expected, opposite the finding for native consumers. The effect size of alien consumers was 2.4 times greater than that of native consumers and did not differ between predators and herbivores. The impact of alien consumers did not differ significantly in aquatic (lakes, rivers, oceans) versus terrestrial (continental, island) habitats. Similarly, there was no significant interaction between consumer origin and location, as consumers had similar effects in insular (freshwater, island) and open (continental, marine) systems – contrary to the notion that alien species impacts are mainly problematic for island biota.

We hypothesize that the ecological naïveté of native biota facilitates their enhanced suppression by alien predators and herbivores relative to native enemies. Our results counter the assertion that the biogeographical origin of species has no bearing on their ecological impact.

The critically endangered Cross River gorilla is a patchily distributed taxon for which habitat selection has been modelled only at coarse spatial scales, using remotely sensed landscape data and large-scale species distribution maps. These coarse-scale models fail to explain why Cross River gorillas (CRG) display a highly fragmented distribution within what appears to be a large, continuous area of suitable habitat. This study aimed to refine our understanding of CRG habitat use to inform conservation planning both for the subspecies and for other fragmented species of conservation concern.

Cross River gorillas occur only in a discontinuous distribution in the southern portion of the Cameroon-Nigeria border region, an area that represents one of Africa's biodiversity hotspots. This study was carried out in the Northern Mone-Mt. Oko region, part of the Mone/Mbulu forest system located in the Manyu division of the South-west Province of Cameroon.

We used resource selection functions to understand habitat use by CRG at multiple scales. Specifically, we employed generalized additive models at the scale of the annual subpopulation range and conditional logistic regression at the scale of individual movements.

Cross River gorillas habitat selection is highly scale dependent. Localized measures of habitat quality strongly influenced selection at the subpopulation or landscape scale, while human activity and food availability were the best predictors of selection at finer scales.

Understanding why CRG do not occur in seemingly suitable habitat is crucial for designating critical habitat both within and between CRG subpopulations. Our results indicate that conservation planning to maintain critical habitat and connectivity among CRG populations will require an integrative, multi-scale planning approach incorporating large-scale landscape characteristics, human use patterns and CRG food availability.

Determination of taxonomic diversity changes, and patterns and drivers of homogenization/differentiation processes in fish (meta)communities at spatial watershed and temporal decadal scales.

The 333-km-long lowland Pilica River, with a solitary large dam reservoir constructed in the early 1970s, central Poland.

Fish data sets obtained along the river upstream and downstream of the reservoir in the 1960s and 2000s are compared in terms of several community properties. The true gamma diversity of order one (¹D_γ), decomposed into true alpha (¹D_α) and true beta (¹D_β) components, is calculated for the taxonomic classification of the sets. Pairwise community distances against the fluvial distances of each course and period are also assessed using the Bray–Curtis measure (BC) applied to occurrence and by ¹D_β, applied to abundance.

Based on ¹D_β, the fish fauna became homogenized taxonomically by 33.0% upstream, while it differentiated by 8.6% downstream. ¹D_α and ¹D_γ decreased considerably in both courses. The changes in diversity were due to the range expansions of the native species. Despite the exposure to alien invasive species, the Pilica was not invaded by any of them between the surveys, although one species probably established itself in the lower course.

Decreases in ¹D_α and ¹D_γ in both Pilica courses and in ¹D_β upstream were not sharp enough to allow the invasions of alien species to which the Pilica was exposed. Consequently, either other drivers than river fragmentation, introductions, climate warming and improvement in water quality may be essential for such invasions, or longer time periods than those presently considered are necessary for them to occur. An individual species' occurrence and abundance analysis suggests that the homogenization and differentiation resulted from variously combined impacts of several natural and human-mediated drivers. The river fragmentation exerted both course-specific and species-specific impacts.

Many species exhibit a time-lag between habitat loss and its extinction, resulting in extinction debt. Although extinction debt is considered a widespread phenomenon, differences in methodological approaches can affect its detection. We aim to contribute to this methodological debate by exploring whether extinction debt is either a phenomenon common to all patches or idiosyncratic to the patch and landscape attributes of a given patch. We also aim to determine whether the scale dependency of species richness might help to explain extinction debt.

Southern Catalonia (NE Iberian Peninsula).

We studied the effects of habitat loss on plant species richness (total, specialists and generalists) in stable (habitat loss < 40% since 1956) and regressive (habitat loss more than 40% since 1956) patches of Mediterranean grasslands at both quadrat and patch scales using general linear models.

We detected extinction debt at patch scale but only in regressive patches. The magnitude of extinction debt was not constant but was related to the percentage of patch area reduction. Contrastingly, regressive patches presented fewer species than stable patches at quadrat scale.

Quadrat scale extinctions in regressive patches lead to rarefaction, but not immediate extinction, of some species at patch scale and created an extinction debt. Species loss at quadrat scale constitutes an early warning indicator of the effects of habitat loss on biodiversity, while delayed extinctions offer an opportunity for conservation initiatives.

Modern extirpations within the Carnivora have generally followed the human footprint. The contagion hypothesis predicts that range contractions should occur along gradients in human activity, leaving relict populations in remote areas at range edges. We evaluated this hypothesis for cougars (Puma concolor), a widely distributed and heavily exploited North American carnivore.

Colorado Plateau and Great Basin ecoregions within Utah, USA.

We examined the spatial distribution of anthropogenic cougar mortality (n = 4217) using indices of remoteness and habitat quality within a GIS/multiple-regression analytical framework. To identify areas of disproportionately high or low exploitation rates, we used break-points from the literature and local field studies. We defined de facto refugia as watersheds with mean annual harvest rates ≤ 24% of the predicted population, whereas ecological traps were those watersheds that exceeded this value.

Cougar harvest rates were greater in the core and lower along the periphery of their statewide geographic range. The largest refugia were overrepresented in arid ecoregions with low human population densities, whereas ecological traps were concentrated in areas of low remoteness. Ecological traps were within mean cougar dispersal distances from refugia, highlighting the potential for source-sink dynamics. Patterns of anthropogenic cougar mortality generally followed the predictions of the contagion hypothesis, being spatially correlated with human access in high-quality habitats.

Low-quality habitats on the range margins are likely to harbour carnivore populations in the event of widespread human-caused declines, and therefore may have greater conservation value than has previously been assumed. Resource managers may consider using the distribution of de facto refugia and ecological traps within a source-sink context to develop conservation strategies for cougars and other wide-ranging, low-density carnivores with high dispersal tendencies.

We use data based on species checklists and geographical range maps for 3096 amphibian, 4100 bird and 1878 terrestrial mammal species to explore possible discrepancies in estimates of species richness for protected areas. We predict substantial differences between checklist and range map richness for smaller areas, where higher errors of commission from range maps are likely to occur. We also explore how discrepancies between species inventories and range maps may affect assessments of reserve prioritization and representation.

Continental America and associated islands.

We obtained species numbers based on checklists for 619 protected areas from a variety of sources, including management plans, technical reports, environmental assessments, official websites, biodiversity databases and conservation agencies, and evaluated how these correlate with species richness estimates from geographical range map databases, also testing for an effect of reserve size on range map omission and commission errors.

Species richness estimates obtained from checklists and range maps showed a significant positive relationship for all three groups of vertebrates, although for most protected areas, estimates of richness from range overlap maps tended to be overestimates, especially for amphibians and mammals in species-rich regions. Protected area size explained little of the discrepancies between checklist and range map richness. Species representation in protected areas based on checklists was in general lower than representation based on range maps.

Our results suggest that range maps, although far from perfect, have the advantage of reducing geographical biases and filling gaps that exist in point locality data at least in species-rich protected areas, while most available checklists in these regions were extremely variable in quality and availability, lacking of basic information on survey design, reliability and completeness. However, range maps will likely overestimate species representation in protected areas and should thus be treated with caution in this context.

To investigate the use of bioclimatic envelope models for predicting distributions of species that have experienced severe human-induced geographical range contractions. Bioclimatic envelope model predictions of current and future distributions were contrasted with those from models that used biotic indicators of suitable habitat as predictors rather than climate.

Temperate grassy woodlands of South Australia.

We modelled the distribution of two native grassland plant species, key habitat indicators of the endangered and geographically restricted pygmy bluetongue lizard (Tiliqua adelaidensis), using climate and landscape variables with aggregated boosted regression trees. We forecast annual changes in the plant species distributions from 2000 to 2100 under a no-climate-policy ‘Reference’ scenario (high global greenhouse gas emissions) and a climate stabilization ‘Policy’ scenario. We compared current and future predicted distributions of the lizard estimated directly using bioclimatic envelope models with those derived indirectly from climate-driven changes in habitat suitability of the grassland plant species with which the lizard has a strong association (termed plant-habitat models).

Both coupled plant-habitat models and bioclimatic envelope models described the current distribution of the pygmy bluetongue lizard almost equally well; however, future projections of changes in the species range were markedly more pessimistic (i.e. greater range contraction) for bioclimatic envelope models. Further, bioclimatic envelope models that included interactions among variables projected rapid increases in area of occupancy that are unlikely to be attainable given dispersal constraints, but no such increases were projected from plant-habitat models.

Capturing species–environment relationships for threatened and range-restricted species using surrogate biotic variables that represent resource requirements of the focal species – which themselves respond to environmental variation and are in stable equilibrium – allows more confident and ecologically realistic forecasts of potential range changes for species most susceptible to climate change.

Selecting informative variables is crucial for species distribution modelling and ecological studies in general. Proxies quantifying water accumulation may have suitable properties because hydromorphy partly determines plant and animal communities. Topographic wetness index (TWI) was developed to locate wetlands but has largely been ignored from ecological studies despite the value of these areas for biodiversity and the ecosystem services they provide. We assessed here the ability of TWI to predict the occurrence of grassland passerines and tested different settings to determine which was the best predictor for our dataset.

Floodplain meadows in the Loire valley, France, Western Europe.

We recorded the occurrence of four grassland passerines on 64 transects in large hay meadow patches. We computed four TWIs based on digital elevation models (DEMs). TWIs compute water accumulation as a function of slope and catchment. We tested two DEM resolutions (50 m and 250 m) and four TWI algorithms to identify which combination yielded the best model fits to our dataset.

Results depended on the predictor settings and the species considered. TWI predicted the occurrence of the Whinchat, the most specialized species, and the combined occurrence of the others three passerines. One TWI algorithm (SWI) yielded the poorest fit, and we could not determine the best algorithm among the others three. The coarser DEM resolution (250 m pixel size) also yielded better fitting model than the finer resolution (50 m).

Topographic wetness index appears as an informative predictor of species occurrence, at least for the Whinchat, and a useful proxy to detect suitable areas for floodplain grassland birds. This family of indices may improve our ability to model the habitat of wetlands species. However, DEM resolution and algorithm should be selected with caution as they may impact the predictive potential of the proxy.

Use of local-scale non-native plant species (NNS) distribution models has the potential to decrease survey effort and improve population prioritization for management. We developed and evaluated data collection methods and minimum sampling requirements to inform local-scale models of NNS distribution. We also evaluated overall model predictive performance for 16 species at two sites and determined how classes of variables contributed to model performance and suggest invasion drivers.

Wyoming and Idaho, USA

A simulation study was used to test the efficiency of different sampling methods to predict imposed species distributions. Empirical distribution models of species occurrence data from two environmentally disparate sites were cross-validated at increasing sample sizes, and the asymptotic maximum predictive performance and relative contribution of classes of variables were determined for 16 NNS.

Transect sampling was the most efficient method for maximizing model performance after accounting for logistics. Minimum sample sizes to reach model maximum predictive performance were similar for the simulation (< 0.5% of study area) and empirical studies (mean of 0.13% using transects). Maximum predictive performance tended to be greater at the site with steeper environmental gradients, and topo-climatic/biotic variables were most important to model improvement.

Local-scale SDMs can be useful to NNS managers. Using transect methodology, enough data can be collected (ca. 0.13% of the management area) to fit models within logistical/budgetary constraints. These models are most predictive for well-established species as opposed to new invaders and in areas with steeper environmental gradients. Finally, topo-climatic/biotic predictors are the most important variables for predicting more established species, but disturbance and dispersal limitation should be considered and quantified to ensure variables associated with dominant processes are included in the SDM.

Modelling the distribution of rare and invasive species often occurs in situations where reliable absences for evaluating model performance are unavailable. However, predictions at randomly located sites, or ‘background’ sites, can stand in for true absences. The maximum value of the area under the receiver operator characteristic curve, AUC, calculated with background sites is believed to be 1 − a/2, where a is the typically unknown prevalence of the species on the landscape. Using a simple example of a species' range, I show how AUC can achieve values > 1 − a/2 when test presences do not represent each inhabited region of a species__ range in proportion to its area. Values of AUC that surpass 1 − a/2 are associated with higher model predictions in areas overrepresented in the test data set, even if they are less environmentally suitable than other regions the species occupies. Pursuit of high AUC values can encourage inclusion of spurious predictors in the final model if they help to differentiate areas with disproportionate representation in the test data. Choices made during modelling to increase AUC calculated with background sites on the assumption that higher scores connote more accurate models can decrease actual accuracy when test presences disproportionately represent inhabited areas.

Urbanization as a major global trend profoundly changes biodiversity patterns, and homogenization of urban biota due to expanding exotic species and declining native species is of increasing concern. Previous studies on this topic have mostly taken place at large scales that include high habitat heterogeneity. Here, we aimed at disentangling the effects of urbanization and plant invasion on species composition through the analysis of similarity patterns of urban plant assemblages at the community scale where species interact.

Berlin, Germany.

We analysed how different levels of urbanization, specific components of the urban matrix and the dominance of a native (Betula pendula) versus an exotic tree species (Robinia pseudoacacia) affect alpha and beta diversity of urban woodland understorey vegetation in sixty-eight 100-m² plots.

Exotic dominance reduced alpha diversity, but not beta diversity of the total species pool. Comparing beta diversity among different species groups revealed significant but divergent effects of exotic dominance, habitat connectivity and levels of urbanization in native and non-native species assemblages. In particular, urbanity proved to homogenize the native species pool, whereas the beta diversity of the non-native species pool showed a more pronounced response to exotic dominance.

Our data provide evidence that both the urban context and the dominance of exotic species can modify homogenization processes at the community level. These novel insights into the mechanisms of biotic homogenization of urban floras may contribute to mitigating the effects of urbanization on biodiversity.

To provide a procedure for measuring the rarity of both invertebrate species and assemblages of species from multiple scales without the need for fine-resolution datasets over broad areas.

The western Palearctic (WP) and western France.

On the basis of different datasets from different geographical extents, we applied a multiscale rarity weight to species occurrence from multiple scales. Multiscale rarity weights were then averaged at an assemblage level in a multiscale index of relative rarity (I_RR). These rarity weights were calculated using a flexible, scale-dependent method that ensures equitable contributions of each scale to the final index. We provided a simple two-scale example of the application, on spiders of western France, for which we obtained occurrence information from a regional-extent dataset (regional scale) and a western Palearctic-extent dataset (WP scale). Thus, we showed the necessity of a two-scale approach by successively analysing species occurrence, multiscale rarity weights of species and multiscale indices of species assemblage. Finally, we presented a case study within a nature reserve.

Species occurrences are not predictable from one scale to another, and rarity indices of assemblages are poorly congruent among scales, which supports the necessity of a two-scale approach. Multiscale rarity weights accurately showed information on species rarity from both scales. Multiscale indices of assemblages were congruent but with additional information over each one-scale index.

The novelty of the multiscale method developed here is to accurately combine different datasets of varying extents and resolutions to provide multiscale rarity weights for species and indices for assemblages. Given the increasing availability of datasets for invertebrate taxa, this method represents a significant improvement for rarity and conservation studies on invertebrates.

We examine variation in woody plant species richness and endemism within tropical dry forest on oceanic islands and determine what climatic and biogeographic metrics best explain native species richness and endemism across archipelagos, islands and plots.

Oceanic islands in the Pacific.

Stand-level sampling (0.1 ha) at 35 different dry forest sites across 16 islands, and five archipelagos (New Caledonia, Fiji, Marquesas, Marianas and Hawaii). Descriptors of native species richness and endemism were calculated at the plot, island and archipelago level. Biogeographic and climate metrics at the archipelago, island and plot level were drawn from the literature and computer databases. The effects of biogeographic and climate metrics were investigated using linear mixed-effects models.

Dry forests of New Caledonia and Fiji had the highest native species richness, while New Caledonia and Hawaii had the highest endemism. Native species richness and endemism within tropical dry forests on oceanic islands are primarily influenced by biogeographic metrics, especially isolation of the archipelago, and not climatic metrics. Most variance in native species richness and endemism (60% and 64%) is at the archipelago level compared with the island (8%, 16%) and plot (32%, 15%) level. At the island level, species richness in tropical dry forest is affected by precipitation, while island area significantly affects endemism. The area of forest fragments is an important predictor of native species richness and endemism in plots.

Although dry forests in the Pacific have been exceptionally deforested and degraded, high native species richness and endemism remains in a number of forest fragments. Biogeographic metrics explain most of the variance in native species richness and endemism across scales, while climatic metrics are important at the island level. First-order assessments of native richness and endemism at the archipelago, island and stand-level are possible for forest types on oceanic archipelagos.

Cryptic species in the Neofusicoccum parvum–N. ribis species complex have only recently been described, invalidating previous interpretations on host and geographical distribution. This study aimed to characterize the diversity and distribution of these species and to understand the patterns of host association, likely origins and their patterns of spread.

Australia, Brazil, Cameroon, Chile, China, Colombia, Ethiopia, France, Greece, India, Indonesia, Iran, Italy, Japan, Kenya, Mexico, New Zealand, Panama, Portugal, Puerto Rico, South Africa, South Korea, Spain, Swaziland, Taiwan, Thailand, Uganda, United States of America, Uruguay, Zambia and Zimbabwe.

Using the unique polymorphisms that separate species within the complex, we evaluated sequence search results available in public and in our own databases. In addition, the global distribution of diversity of N. parvum was analysed using seven microsatellite markers.

Neofusicoccum parvum is found in 90 hosts across six continents and 29 countries. Neofusicoccum kwambonambiense is found on four continents, six countries and on 14 hosts; N. occulatum is found on four continents, four countries and on 11 hosts; N. umdonicola is found on two continents, countries and hosts; N. cordaticola is found on three continents, countries and hosts; N. batangarum is found on two continents, three countries and three hosts; and N. ribis is found on one host in one country. Population genetic analysis of the global N. parvum population reflects admixture and repeat introductions.

This study illustrates the unfettered and frequent movement of latent pathogens across international borders. Amongst the species in the N. parvum–N. ribis complex, N. parvum is the most widespread and has been reported on the majority of the hosts studied. The current dispersal of N. parvum and its sister species is probably due to repeated introductions of plant material into new growing areas, with Eucalyptus and Vitis vinifera being two prominent candidates for material transfer.

Assessing long-term (1992–2009) trends of littoral invertebrate and phytoplankton metacommunities in boreal lakes with emphasis on separating the nestedness and turnover components of beta diversity. Deriving implications for regional biodiversity conservation and management, based on a data-intensive approach with high ecological realism.

Sweden (Northern Europe).

A recently published method was used to partition beta diversity into species turnover and nestedness components. Regression analyses were used to test for monotonic temporal change of these diversity fractions through time. Associations between the temporal diversity patterns of taxonomic groups and environmental variables were studied using correlation analyses.

Turnover of both metacommunities increased monotonically over the study period, while nestedness decreased. In invertebrates, these changes correlated mainly with regional changes in acidity, while phytoplankton responded more to changing water clarity. Nestedness and turnover patterns were inversely correlated in both groups, but neither turnover nor nestedness was correlated between invertebrates and phytoplankton. Nestedness of both groups explained a lower percentage of the partitioned variance compared with turnover.

Results suggest that all lakes contribute more equally to regional diversity over time and are, as a result, all potential targets of management actions. Not only is a regional conservation strategy logistically difficult, it is also a financially expensive expectation.

Snakes are more vulnerable to extinction than many other taxa. Additionally, their secretive behaviour makes it difficult to acquire the baseline ecological knowledge required to reliably evaluate extinction risks. Consequently, the conservation status of snakes has only been assessed for small populations; reliable methods for large-scale evaluation remain to be tested. In this study, we explored how habitat-suitability models (HSMs) could be used to provide relevant information to help assess extinction risks and formulate appropriate conservation strategies for the Orsini's viper (Vipera ursinii), a rare, endangered snake species.

Provence-Alpes-Côte d'Azur region in south-eastern France (c. 30,000 km²).

We developed a high-resolution HSM (50 × 50 m) using a large sample of species presence data and nine climatic and land cover predictors. We used this model to predict the potential distribution of the Orsini's viper as well as to investigate the main environmental drivers explaining this distribution. We also assessed the geographical barriers between local populations and tested whether forest cutting would reduce fragmentation.

The occurrence of the Orsini's viper was strongly correlated with the annual cumulative temperature and with vegetation cover type. The total extent of suitable habitat covered 2.98% of the study area and was highly fragmented into 1417 distinct areas. Among these areas of suitable habitat, 21 were confirmed to have the species. These represented 22,134 ha and a potential carrying capacity of 168,000 individuals.

Our HSM was consistent with the past assessment of the distribution of the Orsini's viper. Our HSM represents a sound benchmark for the distribution of the species and can provide a powerful tool to help with the search of new populations, the identification of areas for habitat restoration, the test conservation strategies and effects of climate change. We found that forest cutting may lead to reconnect close isolated areas of suitable habitat.

To offer a test of expert knowledge about rarity of twenty Amazon forest bird species following an approach that equates rarity with low site occupancy and formally accounts for imperfect species detection. We define ten pairs of closely related species, each pair with one hypothetically common and one hypothetically rare species. Our null hypothesis is that members of each pair have similar occupancy, with hypothesized differences due to detection errors alone.

A 1000-ha plot of primary rainforest in the central Brazilian Amazon.

We visited each of 55 sampling sites multiple times per season for three field seasons and estimated the probability of site occupancy by each species following a maximum likelihood state-space approach that also estimates the probability that a species is present yet undetected at a site. To maximize detection and account for its variation, we employed three different sampling techniques while systematically training and testing observer's ability to recognize species.

Occupancy estimates agree with expert predictions in all but two species pairs and show no evidence of clear temporal variation in occupancy between sampling seasons. Detection probability had a positive relation with observer ability, a strong relation to time of day across species, and a strong relation with the use of playback for some species. Detection with point counts and with autonomous recorders varied between species pairs.

We reject the null hypothesis of equal occupancy within pairs, concluding that expert knowledge on species rarity is useful and worth eliciting. Our results replace qualitative ratings of rarity with statistical estimates of occupancy, establishing a reliable baseline for future comparisons. Besides illustrating the relevance of expert knowledge, this application to Amazonian birds illustrates a flexible approach that can be used for testing knowledge about rarity for a variety of species groups and spatial scales.

The indirect consequences of biotic homogenization, the process of a gradual increase in the similarity of regional biotas driven by the combined effects of species invasions and extinctions, are still poorly understood. In this study, we aimed to assess the ability of a native affiliate species to maintain its host resources under the condition of biotic homogenization of host communities.

Central (Vltava River Basin, Czech Republic) and western (Douro River Basin, Portugal) Europe.

We tested the ability of non-native species to serve as an alternative partner in local host–affiliate relationships. We used a European freshwater mussel, Anodonta anatina, which is considered to be a host generalist of native fish species, and compared the compatibility of its glochidia with native versus non-native fishes in two distinct European regions. Subsequently, we projected the obtained host compatibility data into the recent progress of biotic homogenization and estimated the degree of host dilution.

We found significant differences in the ability of A. anatina glochidia to parasitize the native and non-native fish species in both the central and peripheral parts of the mussel's distribution range. As a result, the increasing presence of non-native species within fish communities across Europe likely significantly decreases the availability of the mussel's host. Biotic homogenization of host communities may interfere with general life history traits (host specificity) of their local affiliate species.

This study demonstrates that the mixing of regional biotas may lead to an excessive loss of host availability even for host generalists, such as the freshwater mussel A. anatina, with potentially broad consequences for their population dynamics. Conservation strategies of endangered affiliate species need to incorporate the biogeographical context of host–affiliate relationships and particularly the consequences of biotic homogenization.

Both anthropogenic habitat disturbance and the breadth of habitat use by alien species have been found to facilitate invasion into novel environments, and these factors have been hypothesized to be important within coccinellid communities specifically. In this study, we address two questions: (1) Do alien species benefit more than native species from human-disturbed habitats? (2) Are alien species more generalized in their habitat use than natives within the invaded range or can their abundance patterns be explained by specialization on the most common habitats?

Chile.

We quantified the use of different habitat types by native and alien coccinellid beetles by sampling individuals in nine habitat types that spanned a gradient in disturbance intensity, and represented the dominant natural, semi-natural and agricultural habitats in the landscape.

Our results provide strong support for the hypotheses that more-disturbed habitats are differentially invaded. Both the proportion of alien individuals and the proportion of alien species increased significantly with increasing disturbance intensity. In contrast, we found no evidence that alien species were more generalized in their habitat use than native species; in fact, the trend was in the opposite direction. The abundance of specialized alien coccinellid species was not correlated with the area of the habitat types in the landscape.

The results suggest that successfully established alien coccinellid species may be ‘disturbance specialists’ that thrive within human-modified habitats. Therefore, less-disturbed agroecosystems are desirable to promote the regional conservation of native species within increasingly human-dominated landscapes.

As the global climate is changing rapidly, there is a need to make conservation decisions to facilitate species' persistence under climate change. Models employed to make predictions regarding the impacts of climate change on species' distributions, and ultimately persistence, typically assume that interannual variability in environmental conditions is independent between years. However, the colour of environmental noise has been shown to affect extinction risk in populations occupying spatially static environments, and should therefore affect persistence during climate change. This study aims to investigate the importance of noise colour for extinction risk during climate-induced range shifts.

We use a spatially explicit coupled map lattice with a latitudinal gradient in climatic suitability, together with time series of environmental noise, to simulate periods of directional climate change and investigate the effects of noise colour on extinction risk and range size.

Extinction risk increases with reddening of the environmental noise, and this effect is particularly pronounced over short time frames when climate change is rapid.

Given that management decisions are typically made over such short time frames, and the rapid rates of climate change currently being experienced, we highlight the importance of incorporating realistic time series of environmental noise into models used for conservation planning under climate change.

To highlight where in Madagascar the phylogenetically and spatially rare lemur species at risk of extinction are concentrated.

Madagascar.

Phylogenetic endemism (PE) is a combined measure for apportioning a phylogenetic tree across a landscape used to identify those geographical locations that contain spatially rare phylogenetic diversity (Rosauer et al., 2009). We present a simple extension (imperilled phylogenetic endemism) that scales this phylogenetic diversity by the probability of its loss to extinction. We apply these measures to a composite phylogeny of all confirmed Malagasy lemuriform species using International Union for Conservation of Nature (IUCN) extent of occurrence and threat status data.

We find that, because nonimperilled species are scattered about the lemuriform tree, interior branches of the tree are still quite secure: this also means that areas of high phylogenetic endemism for Madagascar lemuriformes are often the same areas as those of high imperilled PE (IPE), as both are heavily weighted by branches nearer the tips. However, although the North of Madagascar holds the largest amount of spatially rare evolutionary history using both PE and IPE, there are additional pockets of imperilled history in the south and west.

Correlations of endemism and threat status with phylogenetic isolation are modest across lemurs and so are not substitutable conservation values. They might best be integrated on the landscape using IPE. As illustrated here, IPE successfully highlights areas containing species which are at once threatened with extinction and that are phylogenetically and spatially rare.

Biological atlases are a globally widespread and effective means for documenting the distribution of numerous taxa and have been used to study many macroecological relationships. A common assumption when analysing atlas data is that species are detected perfectly (p = 1). This assumption is likely incorrect, but the application of methods to account for heterogeneous detectability (p < 1) has been difficult to implement. We provide an application of current methods of occupancy estimation and modelling to account for imperfect detection in the analysis of atlas data.

New York, USA.

We employed multiseason occupancy models substituting spatial replicates for temporally repeated surveys to examine changes in distribution of the Canada Warbler (Cardellina canadensis) using breeding bird atlases from 1980–1985 and 2000–2005. We compared estimates from models accounting for p < 1 versus those assuming p = 1 in assessing statewide patterns of occupancy, colonization and extinction.

We found forest cover, observer effort, information on previous detections and the sampling year were important predictors of detection. Environmental predictors of statewide occupancy dynamics were similar among models accounting for p < 1 versus those assuming p = 1. Despite these similarities, site-level estimates of occupancy from the model accounting for imperfect detection indicated 14% and 19% higher site occupancy in the 1980–1985 and 2000–2005 Atlases, respectively. In addition, relative to the model accounting for p < 1, the model assuming perfect detectability underestimated persistence and overestimated extinction between atlases. The model accounting for p < 1 had 0.3% and 7.9% higher accuracy in predicting occupancy in the 1980–1985 and 2000–2005 Atlases, respectively, than the models assuming p = 1.

Occupancy modelling and estimation can be successfully applied to broad-scale surveys, such as atlases, that do not explicitly implement repeated visits to a survey block. Occupancy modelling allows for a more rigorous analysis of atlas data for exploring species–environmental relationships and modelling species distributions while accounting for imperfect detection.

Understanding patterns in a species' occupancy and abundance across multiple scales is important for management purposes, particularly for protecting threatened species. Here, we develop a new quantitative, multiscale model of occupancy and abundance that characterizes seven types of rarity and one of commonness in high-diversity communities of staghorn coral. We examine how rare species influence coral community structure and explore spatial variability in underlying patterns of community structure in the context of optimizing the data needed to protect biodiversity.

North-west Pacific Ocean.

We present categorical abundance data for 87 species of staghorn corals occurring within 100 sites across five reefs in the north-west Pacific Ocean. We develop a new model that combines measures of global distribution, local distribution and local abundance to describe eight mutually exclusive occupancy–abundance patterns, which can be used to prioritize regional species conservation. Traditional and new analytical approaches are compared to explore how rare species influence multidimensional space and community structure.

We show that five types of occupancy–abundance relationships exist in staghorn coral assemblages, including four patterns of rarity. The overwhelming majority of species (73%) are rare according to local abundance and/or distribution criteria. Occupancy–abundance patterns are spatially variable in staghorn coral communities, and no single underlying distribution fits all assemblages. Our findings suggest that 54 species are at risk of regional extinction, 30 of which are also classified as ‘vulnerable’ by the IUCN.

Our model demonstrates that multiple occupancy–abundance patterns exist in staghorn coral assemblages. We conclude that 66% of the pool of staghorn coral fauna in the north-west Pacific is at risk of regional extinction. At the locations and scales examined here, occupancy–abundance patterns and the underlying distributions of coral communities are spatially variable, suggesting that it may not be appropriate to apply unified ecological theory to communities with a large proportion of threatened species because this may jeopardize biodiversity conservation.

Habitat loss and fragmentation can influence the genetic structure of biological populations. Large terrestrial predators can often avoid genetic subdivision due to fragmentation because they have high rates of dispersal-mediated gene flow. Leopards (Panthera pardus) are found in a variety of habitats and are the most widely distributed of the large and middle-sized felids. We investigated the genetic diversity and population substructure of leopards (P.p fusca) in a fragmented meta-population comprised of four populations from five protected areas and inter-connecting corridors spread over an area of 45,000 km² in central India.

Kanha, Pench, Satpura and Melghat Tiger Reserves, and interconnecting corridors in the Satpura-Maikal Landscape of Central India.

We collected faecal samples and used genetic methods to identify individuals, estimate the genetic variation and evaluate the patterns of genetic substructuring within this meta-population of leopards.

We identified 217 individual leopards using a panel of seven microsatellite loci. Leopards showed high levels of genetic diversity in all sampled populations. Spatial and non-spatial Bayesian analysis revealed at least two admixed genetic populations with indications of ongoing genetic subdivision. Genetic differentiation between populations was not explained by geographic distance. We identified nine individuals as migrants, most of which were assigned to reserves connected by corridors.

Our study demonstrates that the leopard, an adaptable and vagile species, can become genetically differentiated with increased habitat fragmentation. Contrary to our hypothesis of panmixia, our results indicate that although leopards in this landscape are admixed, there is genetic substructuring at both the landscape and the fine-scale level. We conclude that this is due to habitat fragmentation and corridors are of immense value in maintaining genetic connectivity in this landscape.

We explored lineage diversification within desert-dwelling fauna. Our goals were (1) to determine whether phylogenetic lineages and population expansions were consistent with younger Pleistocene climate fluctuation hypotheses or much older events predicted by pre-Pleistocene vicariance hypotheses, (2) to assess concordance in spatial patterns of genetic divergence and diversity among species and (3) to identify regional evolutionary hotspots of divergence and diversity and assess their conservation status.

Mojave, Colorado, and Sonoran Deserts, USA.

We analysed previously published gene sequence data for twelve species. We used Bayesian gene tree methods to estimate lineages and divergence times. Within each lineage, we tested for population expansion and age of expansion using coalescent approaches. We mapped interpopulation genetic divergence and intra-population genetic diversity in a GIS to identify hotspots of highest genetic divergence and diversity and to assess whether protected lands overlapped with evolutionary hotspots.

In seven of the 12 species, lineage divergence substantially predated the Pleistocene. Historical population expansion was found in eight species, but expansion events postdated the Last Glacial Maximum (LGM) in only four. For all species assessed, six hotspots of high genetic divergence and diversity were concentrated in the Colorado Desert, along the Colorado River and in the Mojave/Sonoran ecotone. At least some proportion of the land within each recovered hotspot was categorized as protected, yet four of the six also overlapped with major areas of human development.

Most of the species studied here diversified into distinct Mojave and Sonoran lineages prior to the LGM – supporting older diversification hypotheses. Several evolutionary hotspots were recovered but are not strategically paired with areas of protected land. Long-term preservation of species-level biodiversity would entail selecting areas for protection in Mojave and Sonoran Deserts to retain divergent genetic diversity and ensure connectedness across environmental gradients.

The aim of the present study was to describe the diversity patterns along elevational gradients that are crucial for conservation management and for understanding diversification processes in montane environments. This article analyses the distributional range and elevation patterns of endemic and non-endemic (NE) vascular plants living on a unique set of Neotropical table mountains (tepuis). To investigate the potential causes of the high degree of diversity and endemism we tested the role of elevation, area and the mid-domain effect (MDE). We also aimed to discuss the origin of the current tepuian biota.

The Guayana Highlands (northern South America). Emphasis is placed on the mountaintops above 1500 m elevation, which form the highly biodiverse Pantepui biogeographical province.

We examined the distribution patterns of vascular plant species richness in relation to elevation, area, and the MDE using generalized additive models. We used Range Model for the MDE.

We found that regional endemics richness show a hump-shaped curve in relation to elevation. Single-tepui endemics (STE) increase with elevation, whereas NE and total species richness decrease. Area and MDE influence this pattern for Pantepui and STE, but not for NE. We also observed that the spatial distribution of endemic richness displays a left-skewed distribution pattern due to the dominance of STE.

Our results demonstrate that a combination of elevation, area and MDE provide a basic explanation for the diversity of vascular plants in Pantepui. In addition, the present study indicates that maxima of STE are located at the highest altitudes, where the possibility of biotic connection (via migration) and gene flux has been minimal, even during glacial phases when most migration pathways amongst the tepui mountains were open. We also suggest that climatic filtering due to the extreme conditions atop the tepuis and low dispersal capacity stand out as the main drivers of the decline in NE species richness with elevation.

Across the tropics, large-bodied mammal species are threatened by rapid and widespread forest habitat conversion by either commercial logging or agricultural expansion. How such species use these habitats is an important area of research for guiding their future management. The tropical forest-dwelling sun bear, Helarctos malayanus, is the least known of the eight bear species. Consequently, the IUCN/SSC Bear Specialist Group ranks research on this species as a top priority. This study aims to investigate landscape variables that influence sun bear habitat use in forests under varying levels of degradation and protection.

A 20,998 km² Sumatra forest landscape covering Kerinci Seblat National Park (KSNP), Batang Hari Protection Forest (BHPF) and neighbouring logging and agricultural concessions.

An occupancy-based sampling technique using detection/non-detection data with 10 landscape covariates was applied in six study areas that operated a total of 125 camera traps. The potential differences between habitat use (ψ) of sun bears were first modelled with broad-scale covariates of study area, land-use types and forest type. Sun bear habitat use was then investigated with the finer-scale landscape features associated within these areas.

From 10,935 trap nights, sun bears were recorded at altitudes ranging from 365 to 1791 m. At a broad-scale, habitat use increased with protection status, being highest in KSNP (0.688 ± 0.092, ± SE) and BHPF (0.621 ± 0.110) compared to production (0.418 ± 0.121) and convertible (0.286 ± 0.122) forests. Within these areas, sun bears showed a preference for forest that was further from public roads and villages and at a lower elevation.

The habitat suitability model identified several high-quality habitat patches outside of the priority conservation areas for immediate protection. Consequently, conservation management strategies should emphasize the importance of high conservation value forests and prohibit further conversion of threatened lowland forests.

Conservation studies often investigate the biodiversity of one taxonomic group with the expectation that it reflects biodiversity of other taxa as well. However, previous studies have found that biodiversity patterns are often only weakly correlated across ecologically very different taxa. Using three arthropod taxa that share the same habitat, utilize similar resources and are sampled with identical technique, we investigate the applicability of two levels of biodiversity indication: (1) prediction of biodiversity patterns, and (2) inference of environment–biodiversity relationships. The second aspect is of high relevance to applied conservation management yet mostly neglected, at least in terrestrial systems, when discussing the indicator concept.

Northern Switzerland

We used a very large data set of pitfall trap samples from different habitats for three taxonomic groups (Carabidae, Staphylinidae and Araneae). We quantified biodiversity by different metrics of local diversity (species richness, effective number of species) and of pairwise faunal dissimilarities (Sørensen, Bray–Curtis). We investigated the congruence of (1) biodiversity patterns by cross-taxa regressions, and (2) environmental models of biodiversity by comparing fitted coefficients, and resulting extrapolations across the research region.

We found positive yet not very strong correlations in biodiversity patterns, while environmental models differed considerably between taxa as well as between diversity metrics.

Inferences of environment–biodiversity relationships can differ between taxonomic groups even if biodiversity patterns alone show significant correlation. This may be either because species indeed respond differently to environmental variation or because of misspecifications inherent in ecological modelling. Both possibilities suggest a need for caution in selecting and applying biodiversity indicators. Furthermore, the choice of diversity metric can strongly affect results, and therefore, decisions about which metric to use in any given situation need to be made carefully.

Risk assessments in applied scientific disciplines have evolved somewhat in isolation, adopting conventions, assumptions and tools from other disciplines almost haphazardly. This editorial provides background for the articles in this special issue, which sample six broad themes in risk assessment in conservation biology and presenting new innovations and applications.

Global.

The articles in the special issue address themes related to species distribution modelling, population viability analysis, threatened species management, biosecurity, uncertainty analysis, cost–benefit analysis and foresight. We sought articles that address new and emerging topics in each of these areas.

The articles identify new and potentially useful innovations in a variety of areas relevant to conservation biology. Collectively, they paint a picture of risk assessment as an important element in supporting transparent, rational decisions and effective policy.

Policy makers and conservation managers aspire to set evidence-based priorities, and technical specialists aim to have their methods used in decision-making. Scientists will succeed if, as the articles in this issue exemplify, they develop a sound understanding of the context of the decisions in which their tools are to be used and shape them accordingly.

Decision-making for conservation management often involves evaluating risks in the face of environmental uncertainty. Models support decision-making by (1) synthesizing available knowledge in a systematic, rational and transparent way and (2) providing a platform for exploring and resolving uncertainty about the consequences of management decisions. Despite their benefits, models are still not used in many conservation decision-making contexts. In this article, we provide evidence of common objections to the use of models in environmental decision-making. In response, we present a series of practical solutions for modellers to help improve the effectiveness and relevance of their work in conservation decision-making.

Global review.

We reviewed scientific and grey literature for evidence of common objections to the use of models in conservation decision-making. We present a set of practical solutions based on theory, empirical evidence and best-practice examples to help modellers substantively address these objections.

We recommend using a structured decision-making framework to guide good modelling practice in decision-making and highlight a variety of modelling techniques that can be used to support the process. We emphasize the importance of participatory decision-making to improve the knowledge-base and social acceptance of decisions and to facilitate better conservation outcomes. Improving communication and building trust are key to successfully engaging participants, and we suggest some practical solutions to help modellers develop these skills.

If implemented, we believe these practical solutions could help broaden the use of models, forging deeper and more appropriate linkages between science and management for the improvement of conservation decision-making.

Rare aquatic species are a substantial component of biodiversity, and their conservation is a major objective of many management plans. However, they are difficult to assess, and their optimal habitats are often poorly known. Methods to effectively predict the likely locations of suitable rare aquatic species habitats are needed. We combine two modelling approaches to predict occurrence and general abundance of several rare fish species.

Allegheny watershed of western New York State (USA)

Our method used two empirical neural network modelling approaches (species specific and assemblage based) to predict stream-by-stream occurrence and general abundance of rare darters, based on broad-scale habitat conditions. Species-specific models were developed for longhead darter (Percina macrocephala), spotted darter (Etheostoma maculatum) and variegate darter (Etheostoma variatum) in the Allegheny drainage. An additional model predicted the type of rare darter-containing assemblage expected in each stream reach. Predictions from both models were then combined inclusively and exclusively and compared with additional independent data.

Example rare darter predictions demonstrate the method's effectiveness. Models performed well (R² ≥ 0.79), identified where suitable darter habitat was most likely to occur, and predictions matched well to those of collection sites. Additional independent data showed that the most conservative (exclusive) model slightly underestimated the distributions of these rare darters or predictions were displaced by one stream reach, suggesting that new darter habitat types were detected in the later collections.

Broad-scale habitat variables can be used to effectively identify rare species' habitats. Combining species-specific and assemblage-based models enhances our ability to make use of the sparse data on rare species and to identify habitat units most likely and least likely to support those species. This hybrid approach may assist managers with the prioritization of habitats to be examined or conserved for rare species.

To quantify the consequences of major threats to biodiversity, such as climate and land-use change, it is important to use explicit measures of species persistence, such as extinction risk. The extinction risk of metapopulations can be approximated through simple models, providing a regional snapshot of the extinction probability of a species. We evaluated the extinction risk of three species under different climate change scenarios in three different regions of the Mexican cloud forest, a highly fragmented habitat that is particularly vulnerable to climate change.

Cloud forests in Mexico.

Using Maxent, we estimated the potential distribution of cloud forest for three different time horizons (2030, 2050 and 2080) and their overlap with protected areas. Then, we calculated the extinction risk of three contrasting vertebrate species for two scenarios: (1) climate change only (all suitable areas of cloud forest through time) and (2) climate and land-use change (only suitable areas within a currently protected area), using an explicit patch-occupancy approximation model and calculating the joint probability of all populations becoming extinct when the number of remaining patches was less than five.

Our results show that the extent of environmentally suitable areas for cloud forest in Mexico will sharply decline in the next 70 years. We discovered that if all habitat outside protected areas is transformed, then only species with small area requirements are likely to persist. With habitat loss through climate change only, high dispersal rates are sufficient for persistence, but this requires protection of all remaining cloud forest areas.

Even if high dispersal rates mitigate the extinction risk of species due to climate change, the synergistic impacts of changing climate and land use further threaten the persistence of species with higher area requirements. Our approach for assessing the impacts of threats on biodiversity is particularly useful when there is little time or data for detailed population viability analyses.

Most risk assessments and decisions in conservation are based on surrogate approaches, where a group of species or environmental indicators are selected as proxies for other aspects of biodiversity. In the focal species approach, a suite of species is selected based on life history characteristics, such as dispersal limitation and area requirements. Testing the validity of the focal species concept has proved difficult, due to a lack of theory justifying the underlying framework, explicit objectives and measures of success. We sought to understand the conditions under which the focal species concept has merit for conservation decisions.

Our model system comprised 10 vertebrate species in 39 patches of native forest embedded in pine plantation in New South Wales, Australia.

We selected three focal species based on ecological traits. We used a multiple-species reserve selection method that minimizes the expected loss of species, by estimating the risk of extinction with a metapopulation model. We found optimal reserve solutions for multiple species, including all 10 species, the three focal species, for all possible combinations of three species, and for each species individually.

Our case study suggests that the focal species approach can work: the reserve system that minimized the expected loss of the focal species also minimized the expected species loss in the larger set of 10 species. How well the solution would perform for other species and given landscape dynamics remains unknown.

The focal species approach may have merit as a conservation short cut if placed within a quantitative decision-making framework, where the aspects of biodiversity for which the focal species act as proxies are explicitly defined, and success is determined by whether the use of the proxy results in the same decision. Our methods provide a framework for testing other surrogate approaches used in conservation decision-making and risk assessment.

Species distribution models (SDMs) coupled with metapopulation dynamics models can integrate multiple threats and population-level processes that influence species distributions. However, multiple sources of uncertainties could lead to substantial differences in model outputs and jeopardize risk assessments. We evaluate uncertainties in coupled species distribution–metapopulation dynamics models and focus on two often underappreciated sources of uncertainty: the choice of general circulation model (GCM) and demographic parameter uncertainty of the metapopulation model. We rank the risks associated with potential climate changes and habitat loss on projected range margin dynamics of the Hooded Warbler (Setophaga citrina).

Breeding range of the Hooded Warbler, North America.

Using SDMs, we quantified variability in projected future distributions using four GCMs and a consensus model at the biogeographic scale and assessed the propagation of uncertainty through to metapopulation viability projections. We applied a global sensitivity analysis to the coupled species distribution–metapopulation models to rank the influence of choice of GCM, parameter uncertainty and simulated effects of habitat loss on metapopulation viability, thereby addressing error propagation through the whole modelling process.

The Hooded Warbler range was consistently projected to shift north: choice of GCMs influenced the magnitude of change, and variability was spatially structured. Variability in the choice of GCMs propagated through to metapopulation viability at the northern range boundary. Although viability measures were sensitive to the GCM used, measures of direct habitat loss were more influential. Despite the high ranking of vital rates in the global sensitivity analysis, direct habitat loss had a larger negative influence on extinction risk than potential future climate changes.

This work underscores the importance of a global sensitivity analysis framework applied to coupled models to disentangle the relative influence of uncertainties on projections. The use of multiple GCMs enabled the exploration of a range of possible outcomes relative to the consensus GCM, helping to inform risk estimates. Ranking uncertainties informs the prioritization of management actions for species affected by dynamic anthropogenic threats over multiple spatial scales.

Population viability analysis (PVA) is used to quantify the risks faced by species under alternative management regimes. Bayesian PVAs allow uncertainty in the parameters of the underlying population model to be easily propagated through to the predictions. We developed a Bayesian stochastic patch occupancy model (SPOM) and used this model to assess the viability of a metapopulation of the growling grass frog (Litoria raniformis) under different urbanization scenarios.

Melbourne, Victoria, Australia.

We fitted a Bayesian model that accounted for imperfect detection to a multiseason occupancy dataset for L. raniformis collected across northern Melbourne. The probability of extinction was modelled as a function of effective wetland area, aquatic vegetation cover and connectivity, using logistic regression. The probability of colonization was modelled as a function of connectivity alone. We then simulated the dynamics of a metapopulation of L. raniformis subject to differing levels of urbanization and compensatory wetland creation. Uncertainty was propagated by conducting simulations for 5000 estimates of the parameters of the models for extinction and colonization.

There was considerable uncertainty in both the probability of quasi-extinction and the minimum number of occupied wetlands under most urbanization scenarios. Uncertainty around the change in quasi-extinction risk and minimum metapopulation size increased with increasing habitat loss. For our focal metapopulation, the analysis revealed that significant investment in new wetlands may be required to offset the impacts of urbanization.

Bayesian approaches to PVA allow parametric uncertainty to be propagated and considered in management decisions. They also provide means of identifying parameters that represent critical uncertainties, and, through the use of informative priors, can easily assimilate new data to reduce parametric uncertainty. These advantages, and the ready availability of software to run Bayesian analyses, will ensure that Bayesian approaches are used increasingly for PVAs.

Geographic mapping of risks is a useful analytical step in ecological risk assessments and in particular, in analyses aimed to estimate risks associated with introductions of invasive organisms. In this paper, we approach invasive species risk mapping as a portfolio allocation problem and apply techniques from decision theory to build an invasion risk map that combines risk and uncertainty in a single map product.

Canada.

We divide the study area into a set of spatial domains and treat each domain as an individual ‘portfolio’ with a unique distribution of the expected impacts of invasion. The risk of invasion is then mapped by finding nested ‘efficient’ portfolio sets that identify the geographic areas exhibiting the worst combinations of the estimated risk of invasion and the uncertainty in that estimate. For Canadian municipalities, we apply the approach to quantify the risk that a given location will receive invasive forest pests with commercial freight transported via the North American road network. We compare risk allocation techniques that employ the concepts of nested mean-variance (M-V) frontiers and second-degree stochastic dominance.

While both methods based on M-V and the stochastic dominance principles identified similar areas of highest risk, they differed in how they demarcated moderate-risk areas. Furthermore, they address uncertainty in different ways, treating it as a risk premium (in the case of nested M-V frontiers) or producing risk-averse delineations (in the case of stochastic dominance).

The portfolio-based approach offers a viable strategy for dealing with the typically wide variability in risk estimates caused by a lack of knowledge about a new invader. The methodology also provides a tractable way of incorporating decision-making preferences into the final risk estimates and thus better aligns risk assessments with particular decision-making scenarios about the organism of concern.

Decision-making in weed management involves consideration of limited budgets, long time horizons, conflicting priorities, and as a result, trade-offs. Economics provides tools that allow these issues to be addressed and is thus integral to management of the risks posed by weeds. One of the critical issues in weed risk management during the early stages of an invasion concerns feasibility of eradication. We briefly review how economics may be used in weed risk management, concentrating on this management strategy.

Australia.

A range of innovative studies that investigate aspects of weed risk management are reviewed. We show how these could be applied to newly invading weeds, focussing on methods for investigating eradication feasibility. In particular, eradication feasibility is analysed in terms of cost and duration of an eradication programme, using a simulation model based on field-derived parameter values for chromolaena, Chromolaena odorata.

The duration of an eradication programme can be reduced by investing in progressively higher amounts of search effort per hectare, but increasing search area will become relatively more expensive as search effort increases. When variation in survey and control success is taken into account, increasing search effort also reduces uncertainty around the required duration of the eradication programme.

Economics is integral to the management of the risks posed by weeds. Decision analysis, based on economic principles, is now commonly used to tackle key issues that confront weed managers. For eradication feasibility, duration and cost of a weed eradication programme are critical components; the dimensions of both factors can usefully be estimated through simulation.

Biosecurity responses to incursions aim to achieve pest- or disease-free status as quickly as possible. One of the critical initial response activities involves tracing known movements (trace events) to and from an infected or infested property (IP) that could spread the pest or pathogen. During an incursion response, managers allocate surveillance resources to follow up trace events in order of priority. Prioritizing trace events is difficult and typically subjective. We present a simulation model where several dispersal mechanisms spread a pest between areas. We use model outputs to test different search strategies, using citrus canker (caused by Xanthomonas citri) as a case study. Model scenarios are based on an outbreak of citrus canker that occurred in Queensland in 2004.

Australia.

Model parameters were extracted from published scientific reports and elicited from experts. We used model outputs to assess three search strategies to determine how best to monitor citrus canker spread. Parameters governing disease detectability and host susceptibility were varied in a sensitivity analysis.

In all simulation scenarios, the ‘adaptive radius’ rule performed best, whereby a circular search area was placed around the IP where the disease outbreak was first detected, with a radius proportional to the estimated number of months the property was infected. Importantly, none of the search rules tested detected all IPs without searching all areas with susceptible hosts in the region.

We identify a simple rule of thumb for searching during a citrus canker outbreak that is robust to uncertainty. We cannot generalize the results of this study for tracing other pests or pathogens. The model has created a framework that may be used to explore other contexts and disease dynamics, leading perhaps to more general rules for disease outbreak management.

Effective decisions for managing invasive species depend on feedback about the progress of eradication efforts. Panetta & Lawes () developed the eradograph, an intuitive graphical tool that summarizes the temporal trajectories of delimitation and extirpation to support decision-making. We correct and extend the tool, which was affected by incompatibilities in the units used to measure these features that made the axes impossible to interpret biologically.