Problem In Conservation Biology Research Articles

Modelling the distribution of badgers Meles meles: comparing predictions from field‐based and remotely derived habitat data

ABSTRACT Environmental heterogeneity is important in determining the distribution and abundance of organisms at various spatial scales. The ability to understand and predict distribution patterns is important for solving many management problems in conservation biology and wildlife epidemiology. The badger Meles meles is a highly adaptable, medium‐sized carnivore, distributed throughout temperate Eurasia, which shows a wide diversity of social and spatial organization. Within Britain, badgers are not only legally protected, but they also serve as a wildlife host for bovine tuberculosis Mycobacterium bovis. An evaluation of the role of badgers in the dynamics of this infection depends on understanding the responses of badgers to the environment at different spatial scales. The use of digital data to provide information on habitats for distribution models is becoming common. Digital data are increasingly accessible and are generally cheaper than field surveys. There has been little research, however, to compare the accuracy of models based on field‐derived and remotely derived data. In this paper, we make quantified comparisons between large‐scale presence/absence models for badgers in Britain, based on field‐surveyed habitat data and remotely derived digital data, comprising elevation, geology and soil. We developed four models: 1980s badger survey data using field‐based and digital data, and 1990s badger survey data using field‐based and digital data. We divided each of the four datasets into two subsets and used one subset for training (developing) the model and the other for testing it. All four training models had classification accuracies in excess of 69%. The models generated from digital data were slightly more accurate than those generated from field‐derived habitat data. The high classificatory ability of the digital‐based models suggests that the use of digital data may overcome many of the problems associated with field data in wildlife‐habitat modelling, such as cost and restricted geographical coverage, without any significant impact on model performance for some species. The more widespread use of digital data in wildlife‐habitat models should enhance their accuracy, repeatability and applicability and make them better‐suited as tools to aid policy‐ and decision‐making processes.

Does the invasive plant Elymus athericus modify fish diet in tidal salt marshes?

The invasion of Mont-Saint-Michel Bay salt marshes (France) by a grass species ( Elymus athericus) has led to important changes in vegetation cover, which is likely to modify the habitat for many invertebrates. Some of them constitute the main food items for several fish species, such as young sea bass ( Dicentrarchus labrax) and sand goby ( Pomatoschistus minutus), that feed in salt marsh creeks during high tides. As a result, fish nursery functions of salt marshes could be modified by the E. athericus invasion. In order to test this hypothesis, gut contents of the two most abundant fish species (sea bass and sand goby) were compared before and after E. athericus invasion in the same salt marsh creek and using the same methodology. The accessibility and availability of the main food item, the semi-terrestrial amphipod Orchestia gammarella, were estimated and compared between invaded (dominated by E. athericus) and original areas (dominated by Atriplex portulacoides). Gut content analysis showed a significantly greater percentage of fish leaving with empty guts from E. athericus areas than from A. portulacoides areas. The sea bass diet composition study showed a major shift in the relative importance of the main food items: before E. athericus invasion, diets were dominated by the semi-terrestrial species O. gammarella, whereas after the E. athericus invasion they were dominated by a marine mysid Neomysis integer. The same trend was found for sand gobies, with a shift of the main food item from O. gammarella before invasion to the polychaete Hediste diversicolor after invasion. These trophic changes may be explained by the lower accessibility and availability of O. gammarella in invaded communities than in natural ones. The E. athericus invasion, observed throughout northern Europe, is thus likely to disturb trophic function of natural salt marshes for fish. This preliminary study of the E. athericus invasion is also an illustration that invasive species are an urgent problem in conservation biology.

Complexity in Ecology and Conservation: Mathematical, Statistical, and Computational Challenges

Creative approaches at the interface of ecology, statistics, mathematics, informatics, and computational science are essential for improving our understanding of complex ecological systems. For example, new information technologies, including powerful computers, spatially embedded sensor networks, and Semantic Web tools, are emerging as potentially revolutionary tools for studying ecological phenomena. These technologies can play an important role in developing and testing detailed models that describe real-world systems at multiple scales. Key challenges include choosing the appropriate level of model complexity necessary for understanding biological patterns across space and time, and applying this understanding to solve problems in conservation biology and resource management. Meeting these challenges requires novel statistical and mathematical techniques for distinguishing among alternative ecological theories and hypotheses. Examples from a wide array of research areas in population biology and community ecology highlight the importance of fostering synergistic ties across disciplines for current and future research and application.

Assessment and Management of Invasive Alien Predators

Although invasive alien species have been identified as the second greatest threat to biodiversity after habitat loss, characterizing and quantifying their impacts on native species and habitats remains a fundamental problem in conservation biology. Here, I review the techniques that are currently used to assess the impact of invasive alien species on biodiversity, highlighting both their uses in invasive species ecology and their limitations in establishing a causal relationship. Adopting a hypothesis-driven experimental approach to impact assessment, and to eradication efforts through adaptive management, would benefit our ecological understanding of invasive species without delaying critical management action that could reduce the spread of invasive species populations.

Heuristic and optimal solutions for set‐covering problems in conservation biology

Area‐selection methods have recently gained prominence in conservation biology. A typical problem is to identify the minimum number of areas required to represent all species over some geographic region. Iterative heuristic methods have been developed by conservation scientists to solve these problems, although the solutions cannot be guaranteed to be optimal. Although optimal solutions can often be found, heuristics continue to be popular as they are perceived to be faster and more transparent as they are intuitively easy to understand. We used distributional data for 1921 bird species, 939 mammal species, 405 snake species, and 617 amphibian species compiled at the Zoological Museum, Univ. of Copenhagen for all 1° cells of mainland sub‐Saharan Africa to compare the quality of the solutions found using two heuristic methods (simple‐greedy algorithm and a progressive‐rarity algorithm) with optimal solutions. We found that the heuristic methods considered here often provide solutions as good as optimal solutions. Even in those cases where the optimal solutions were better the difference was relatively small, with the heuristics providing solutions requiring a 2–10% increase in area selected compared with the optimal solution, which importantly, represented an increase of <1% of the total area. Our study also suggests that the heuristic algorithms performed least well for datasets with few single cell endemics and taxa that tend to have larger range sizes. Despite the good quality of solutions using heuristics there was no time penalty associated with finding optimal solutions for the problems considered here, suggesting that the major obstacle to their use is making optimal methods accessible to conservation biologists. We encourage conservation biologists to work with operations researchers and so gain the benefit of their expertise and experience in solving these kinds of problems.

How to Excel in Conservation Biology

How to Excel in Conservation Biology

Place prioritization for biodiversity reserve network design: a comparison of the SITES and ResNet software packages for coverage and efficiency

Abstract. The place prioritization problem in conservation biology is that of establishing a sequentially prioritized list of places on the basis of biodiversity content. Such a list can then be used to select reserve networks that are designed to be fully representative of the biodiversity of an area as efficiently as possible (for instance, with minimum area or cost). The usual goal is the representation of all chosen biodiversity surrogates up to or beyond a required target, or to the greatest available extent. The purpose of this paper is to compare the respective performances of two place prioritization software packages, SITES and ResNet, on four datasets (distributions of termite genera in Namibia, breeding bird species in the Falkland Islands/Islas Malvinas, vertebrate species in Texas and flora and fauna species that are at risk in Québec), to determine their respective merits. The two software packages implement radically different algorithms: SITES is based on a simulated annealing procedure for finding (local) optima; ResNet uses an algorithm based on rarity and complementarity. This analysis indicates that the rarity‐complementarity based algorithm of ResNet surpasses the simulated annealing approach of SITES with respect to time and completeness. SITES, however, contains other features that are useful in conservation planning. Ways in which the two packages can be used together effectively are suggested.

The route to extinction in variable environments

Estimating the extinction risk of natural populations is not only an urgent problem in conservation biology but also involves some profound aspects of population dynamics. Apart from the obvious case of a continuous decrease in a population's carrying capacity, understanding the extinction process necessarily includes environmental and demographic stochasticity. Here, we build from first principles two stochastic, single‐population models that can account for various routes to extinction via demographic and environmental variability. The Ricker model of population dynamics generates extinctions from either low or high (around or above carrying capacity) population densities, primarily depending on the growth parameter r. Since extinctions from high densities seem ‘unnatural’, there is either something wrong with the model or with our intuition. Suitable data are scarce. Environmental variability has its strongest influence on extinction risk via per capita birth rates and is only marginally influencing that risk via per capita death rates if the growth parameter is high. The distribution of the environmental noise and the stochastic structure of the model have quantitative, but not qualitative effects on the estimates of extinction risks. We conclude that to determine the route to extinction and to estimate the extinction risk require a careful choice of both the deterministic component of the population model (e.g., under‐ or over‐compensation) and the structure of the demographic and environmental variabilities.

Science on the Bridge

Science on the Bridge

A Reliability Theoretical Quest for Keystones

Species are not equally important for the functioning of communities. This fact is widely recognized and the term keystone species has been introduced for the most important members of communities. In a recent paper, Power et al. called for methods suitable for a priori identification of keystone species, which is an urgent problem in conservation biology. Another problem is the scarcity of their quantitative description. In this paper, we propose an index for characterizing keystone species reflecting their role in food web flow networks. We present some calculations for characterizing the role of keystones in weighted webs and discuss some problems of food web dynamics. Our keystone index refers well to keystones only if trophic interactions are of prime importance in a community. We further analyse the relationship between keystone species and the reliability of network flow. Keystones make network flow less reliable Finally, we discuss some problems of stability and reliability in ecological systems.

The Role of Behavior in Recent Avian Extinctions and Endangerments

Abstract: Understanding patterns of differential extinction and predicting the relative risks of extinction among extant species are among the most important problems in conservation biology. Although recent studies reveal that behavior can be a critical component in many species’ extinctions or endangerments, current approaches to the problem of predicting extinction patterns largely ignore behavior. I reviewed how behavior can affect population persistence and then used recent avian extinctions and endangerments to illustrate behaviors relevant to extinction risk. Behaviors that affect population persistence can be grouped as aggregation, interspecific responses, dispersal, habitat selection, intraspecific behavior, and maladaptive behavior. Behavior that can affect extinction risk is not limited to birds; for example, in many taxonomic groups (vertebrate and invertebrate) there is evidence of socially facilitated reproduction in colonial species, Allee effects on reproductive success and survival, behavioral regulation of population size, and conspecific attraction to breeding sites. Incorporating specific behaviors into models predicting extinction probabilities and patterns should improve their predictions.

Using Statistical Probability to Increase Confidence of Inferring Species Extinction

One of the many daunting problems in conservation biology is determining when a species has gone extinct (Diamond 1987). A superficial evaluation of the problem might lead one to believe that this is a simple task-one either finds a species extant, or it is extinct. But species sometimes are not seen for many years and then are resighted (Atkinson et al. 1991). In the same vein, the literature regularly reports the discovery of new species, even within well-known taxa such as birds (Dinesen et al. 1994), demonstrating the difficulty of detecting species during surveys. The Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) and the World Conservation Union have arbitrarily set 50 years with no sightings as the cut-off for species extinction, even if no one has looked for the species (World Conservation Union 1982). Clearly, a more useful alternative for inferring extinction is needed. Extant species mnight not be seen during a survey because they are behaviorally or morphologically cryptic, because the area is difficult to survey, or because the species is unconmmon (Bibby et al. 1992). If species extinction is to be followed by habitat conversion (e.g., the habitat was previously protected only under laws such as the U.S. Endangered Species Act), increased certainty in inferring species extinction is critical. Solow (1993) presented two statistical methods for inferring species extinction based on repeated sampling of an area. His method assumes a Poisson distribution of species sightings prior to suspected extinction and uses either statistical inference or Bayesian methods to determine whether a recent lack of sightings is likely associated with extinction. I present a different method for assessing certainty of extinction using statistical probability. This method is based on statistical power (Cohen 1988) and addresses the question of how many visits without sighting a target species are required to be statistically confident of its absence.

Determining Minimum Habitat Requirements in Theory and Practice

A central problem in conservation biology is how to determine the amount of suitable habitat required for a population to persist. The critical minimum is what Lande (1987) terms the 'extinction threshold'. Lawton et al. (1994) and Nee (1994) have pointed out the fundamental equivalence of this quantity to the 'eradication threshold' used in epidemiology (Anderson and May 1991). The eradication threshold defines the maximum fraction of a susceptible population that an immunisation programme can afford to miss, while still succeeding in eradicating the disease. This has been shown to equal in magnitude the uninfected fraction of the susceptible population; in other words, it takes the same value as the unused amount of the disease's limiting resource (Anderson and May 1991). For a conservation biologist, the eradication threshold would define the smallest amount of suitable habitat or limiting resource that can sustain a population of animals. Lawton et al. (1994) give the example of a metapopulation consisting of local populations distributed among discrete patches. The metapopulation is at equilibrium when a local population colonises just one other patch before going extinct. Not all patches are occupied at equilibrium. This scenario parallels the epidemiological case of a human population of which some fraction are not immune to the disease and are either carrying it or susceptible to it. Some fraction, h, of the patches are habitable, and a fraction x* among these are unoccupied at equilibrium though susceptible to colonisation. By the simple device of presenting the equilibrium fraction of occupied patches as h x*, it follows that this reaches zero (extinction of the metapopulation) when h equals x*. In other words, the minimum fraction of patches required to avoid extinction, h, is simply x*, the unoccupied but susceptible fraction at equilibrium. Nee (1994) likewise demonstrates how the eradication threshold of a predator is equal in magnitude to the uneaten fraction of limiting prey. A population well buffered against extinction is thus one that occupies the majority of habitable patches; or it is one that maintains the equilibrium density of limiting prey well below what it would be in the absence of predation. Conversely, a population that is at risk of extinction in the event of a small reduction in habitable patches or prey is one whose members occupy few of the habitable patches at equilibrium, or make little impact on the density of their prey. Such situations can arise if the habitat is highly fragmented and dispersal incurs a substantial mortality risk. Equally a predator might make little impact on prey density if the prey have effective defences against predation (they are present in numbers, but difficult to catch). It is interesting to note that highly disturbed environments are likely to be inhabited by species that use the majority of suitable habitat, whereas historically undisturbed environments will support species that are less efficient in this respect. Knowledge of the unused fraction thus provides a way of indexing environmental disturbance. Most of the species-specific details that are commonly sought in conservation studies, such as migration rates between patches, foraging efficiency for limiting prey, birth rates, death rates, etc. are superfluous to this estimate of minimum habitat requirements (Nee 1994). This conclusion has obvious consequences for field research on rare species: the essential priority in conservation studies is to determine accurately what constitutes the limiting resource or the suitable habitat. Unfortunately, this is often the most difficult objective to realise in practice (see Caughley and Sinclair 1994 for examples). Several factors may combine to regulate population size (Sinclair 1989). Even if their combined influence points towards certain types of habitat as being more suitable than others, defining exactly what

Persistence of two species of gecko ( Oedura reticulata and Gehyra variegata) in remnant habitat

Identifying the characteristics that make a species vulnerable to extinction following habitat fragmentation is one of the most pressing problems in conservation biology. One suggestion is that habitat specialists are less able to cope with rapid changes to their habitat or to move through the modified landscape than habitat generalists and so will be more vulnerable to extinction. We examine this hypothesis by comparing the distribution of two species of gecko, Gehyra variegata and Oedura reticulata, in patches of remnant woodland. The former species is a habitat generalist relative to the latter and shows a markedly higher level of persistence (97% remnant occupancy vs 72%). Logistic regression modelling of the presence or absence of O. reticulata revealed a significant correlation between the number of smooth-barked eucalypts (both species preferred by O. reticulata) in the remnant and the presence of O. reticulata. This suggests that the probability of extinction for a given population is related to the amount of suitable habitat in the remnant and is a function of processes operating at the population level rather than on a regional basis. Pitfall trapping in this study and evidence from another long-term study suggest that the movement of O. reticulata between remnants is negligible. As a consequence, this species has been unable to form a metapopulation at equilibrium. In contrast, it is likely that G. variegata is maintaining its widespread distribution through a metapopulation structure. These results demonstrate the importance of the ability to form a metapopulation for a species to maintain persistence in recently and highly fragmented ecosystems.

DNA science and conservation

A wide array of DNA-based genetic techniques are now available for the study of many problems in conservation biology. Either directly or indirectly, such techniques are becoming increasingly available to scientists and managers alike. Although these technologies are generally known to conservationists, there is a need to clearly outline the principal characteristics of such genetic tools and to detail how they can most appropriately be used in the management of wildlife species. The essential characteristics of mitochondrial and chloroplast restriction fragment analyses are detailed, together with discussions of single locus nuclear restriction fragment length polymorphisms (RFLPs), multilocus DNA fingerprinting, microsatellite DNA, randomly amplified polymorphic DNAs (RAPDs), and DNA sequence variation. We also give relevant information about the development times for these techniques, their relative costs, and the quality of tissue required. In addition, we discuss which conservation problems are appropriate to each of these methods and give examples of their application and potential use in relation to New Zealand organisms. The following problems are considered: sex assignment; parentage and kinship; migration; species, population and strain identification; forensic applications; genetic effects of population bottlenecks; disease identification; feeding preferences; philopatry; pest control; and understanding population extinction. Finally, we suggest that both microsatellite and minisatellite DNA techniques have particular advantages over many other currently-available techniques and conclude that these two approaches are applicable to a wide range of the conservation problems.

The effects of habitat fragmentation on demography and on the loss of genetic variation in the red squirrel.

A major problem in conservation biology is the extent to which the loss of genetic variability in isolated populations reduces their chance of survival. We present data in which the loss of genetic diversity in small and isolated populations can be directly related to population dynamics. Genetic similarity in red squirrels is inversely correlated with population size. The loss of genetic variation and the lower population densities in isolated populations are both the result of reduced immigration. Our data suggest that population processes rather than genetic problems are the real threat to small squirrel populations.

Rapid human-induced evolution of insect–host associations

RAPID evolution of host association is now occurring independently in two populations of the host-specialist butterfly Euphydryas editha, each of which has recently incorporated a novel host species into its diet. The reasons for these episodes of rapid evolution lie in human land use practices: logging in one case and cattle ranching in the other. In contrast to other insects that have used tolerance of human activities to expand their ranges into disturbed habitats1–3, these rare butterflies have remained at their original sites and evolved adaptations to the changes occurring at those sites. At both sites, the proportion of insects preferring the novel host has increased, in one case clearly because of genetic changes in the insect population. This process is now starting to generate insects that refuse to accept their ancestral host, foreshadowing a new problem in conservation biology. By adapting genetically to human-induced changes in their habitat, the insects risk becoming dependent on continuation of the same practices. This is a serious risk, because human cultural evolution can be even faster than the rapid genetic adaptation that the insects can evidently achieve.

Times to extinction for small populations of large birds.

A major practical problem in conservation biology is to predict the survival times-"lifetimes"-for small populations under alternative proposed management regimes. Examples in the United States include the 'Alala (Hawaiian Crow; Corvus hawaiiensis) and Northern Spotted Owl (Strix occidentalis caurina). To guide such decisions, we analyze counts of all crow, owl, and hawk species in the most complete available data set: counts of bird breeding pairs on 14 European islands censused for 29-66 consecutive years. The data set yielded 129 records for analysis. We define the population ceiling as the highest number of breeding pairs observed from colonization to extinction, within a consecutive series of counts for a given species on a given island. The resulting distributions of population lifetimes as a function of population size prove to be highly skewed: most small populations disappear quickly, but a few last for a long time. Median (i.e., 50th percentile) lifetimes are calculated as only 1-5 yr for hawk, owl, and crow populations with ceilings of one or two breeding pairs. As expected if demographic accidents are the main cause of extinction for small populations, lifetimes rise by a factor of 3-4 for each additional pair up to three pairs. They rise more slowly thereafter. These observations suggest that lifetimes of the 'Alala (now reduced to about three pairs in the wild), and of populations of Northern Spotted Owl in the smallest forest fragments, will be short unless active management is implemented.

Environmental Variation and the Persistence of Small Populations

Understanding the factors that determine the continued survival of small populations is a central problem in conservation biology. The Acorn Woodpecker (Melanerpes formicivorus) naturally occurs in small, isolated populations throughout much of the American Southwest. In spite of this distributional pattern, the species is neither rare nor endangered. Thus it appears to have successfully "solved" the problems to the problems of habitat fragmentation. We used data from a 10-yr field study and simulation models to examine the effects of environmental stochasticity on population survival times. All simulated woodpecker populations went extinct within 49 yr, and the median survival time was only 16 yr. However, when immigration was allowed, persistence times greatly increased; with an immigrant rate of only five individuals per year, most populations lasted >1000 yr. The results of this and other analyses suggest that this population persists only because it is part of a larger "metapopulation," and because it is regularly rescued from extinction by immigration from other, independently varying, populations. This finding has important implications for the development of management strategies designed to preserve small populations that are faced with fragmented distributional patterns and high levels of environmental variation.

Conservation Genetics: Techniques and Fundamentals.

Conservation genetics utilizes the tools and concepts of genetics and applies them to problems in conservation biology. For example, molecular genetic techniques, such as protein electrophoresis, and analysis of mitochondrial DNA and highly variable nuclear genes (including DNA fingerprinting), have been important in documenting the extent and pattern of genetic variation in endangered species. We review these techniques and their advantages and disadvantages, and give examples of their application to endangered species. For captive animal populations, pedigree analysis has become the basic approach to evaluate breeding priority of particular individuals. Several pedigree analysis techniques are commonly used, but peeling and gene dropping give the most information. We compared these techniques and illustrate their value with applications to the Guam Rail, Przewalski's horse, and other endangered captive animals. The rationale for much conservation genetic interpretation is base in evolutionary genetics. We discuss the avoidance of inbreeding depression and the maintenance of genetic variation-both primary conservation genetic goals-from this perspective. In addition, we suggest aspects of these factors that deserve greater attention in their overall application to conservation planning. Finally, we briefly mention three evolutionary topics-the relationship of heterozygosity and fitness, population bottlenecks, and outbreeding depression-that have implications for conservation genetics. Although simple interpretation in these areas is appealing, we feel that because they are only generally understood and often quite controversial, their application to endangered-species management should be carefully evaluated and monitored.