Gaps and opportunities for the World Heritage Convention to contribute to global wilderness conservation

Wilderness areas are ecologically intact landscapes predominantly free of human uses, especially industrial‐scale activities that result in substantial biophysical disturbance. This definition does not exclude land and resource use by local communities who depend on such areas for subsistence and bio‐cultural connections. Wilderness areas are important for biodiversity conservation and sustain key ecological processes and ecosystem services that underpin planetary life‐support systems. Despite these widely recognized benefits and values of wilderness, they are insufficiently protected and are consequently being rapidly eroded. There are increasing calls for multilateral environmental agreements to make a greater and more systematic contribution to wilderness conservation before it is too late. We created a global map of remaining terrestrial wilderness following the established last‐of‐the‐wild method, which identifies the 10% of areas with the lowest human pressure within each of Earth's 62 biogeographic realms and identifies the 10 largest contiguous areas and all contiguous areas >10,000 km². We used our map to assess wilderness coverage by the World Heritage Convention and to identify gaps in coverage. We then identified large nationally designated protected areas with good wilderness coverage within these gaps. One‐quarter of natural and mixed (i.e., sites of both natural and cultural value) World Heritage Sites (WHS) contained wilderness (total of 545,307 km²), which is approximately 1.8% of the world's wilderness extent. Many WHS had excellent wilderness coverage, for example, the Okavango Delta in Botswana (11,914 km²) and the Central Suriname Nature Reserve (16,029 km²). However, 22 (35%) of the world's terrestrial biorealms had no wilderness representation within WHS. We identified 840 protected areas of >500 km² that were predominantly wilderness (>50% of their area) and represented 18 of the 22 missing biorealms. These areas offer a starting point for assessing the potential for the designation of new WHSs that could help increase wilderness representation on the World Heritage list. We urge the World Heritage Convention to ensure that the ecological integrity and outstanding universal value of existing WHS with wilderness values are preserved.     

Individual and collective decision-making during nest site selection by the ant Leptothorax albipennis

Social insect colonies possess remarkable abilities to select the best among several courses of action. In populous societies with highly efficient recruitment behaviour, decision-making is distributed across many individuals, each acting on limited local information with appropriate decision rules. To investigate the degree to which small societies with less efficient recruitment can also employ distributed decision-making, we studied nest site selection in Leptothorax albipennis. Colonies were found to make sophisticated choices, taking into account not only the intrinsic qualities of each site, but also its value relative to the available options. In choices between two sites, individual ants were able to visit both sites, compare them and choose the better one. However, most ants encountered only one site in the course of an emigration. These poorly informed ants also contributed to the colony's decision, because their probability of initiating recruitment to a site depended on its quality. This led to shorter latencies between discovery and recruitment to a superior site, and so created greater amplification via positive feedback of the population at the better site. In short, these small colonies make use of a distributed mechanism of information processing, but also take advantage of direct decision-making by well-informed individuals. The latter feature may in part stem from the limitations of their social structure, but may also reflect the stringent demand for unanimous decisions by house-hunting colonies of any size.     

Recreational Portage Trails as Corridors Facilitating Non-Native Plant Invasions of the Boundary Waters Canoe Area Wilderness (U.S.A.)

Abstract:  Wilderness areas are protected and valued in part for recreation; recreational use, however, can negatively impact these areas. In particular, recreational use can facilitate transport of non-native propagules and create open sites for establishment of non-native species. We examined the role of recreational portage trails in the introduction and establishment of non-native plants into the Boundary Waters Canoe Area Wilderness of northern Minnesota (U.S.A.). On 20 portages, we sampled non-native plant richness and cover at four distances (0, 10, 25, and 50 m) from trails. Non-native richness and cover were not related to distance from wilderness entry point. Non-native richness and cover were, however, negatively related to distance from trails. All six non-native species we observed were either directly on or within 1 m of trails. These results suggest that recreational trails act as corridors facilitating invasions of non-native plants into wilderness areas. It remains unclear, however, whether these effects are caused by dispersal of propagules, creation of bare ground, or changes in the native plant community.     

RIVPACS models for predicting the expected macroinvertebrate fauna and assessing the ecological quality of rivers

The European Union Water Framework Directive recognises the need for and value of biological monitoring. This paper reviews the modelling approach known as River Invertebrate Prediction and Classification System (RIVPACS for assessing the ecological quality of river sites using macroinvertebrate sampling. The RIVPACS philosophy is to develop statistical relationships between the fauna and the environmental characteristics of a large set of high quality reference sites which can be used to predict the macroinvertebrate fauna to be expected at any site in the absence of pollution or other environmental stress. The observed fauna at new test sites can then be compared with their site-specific expected fauna to derive indices of ecological quality. All methodological decisions in any such model development have implications for the reliability, precision and robustness of any resulting indices for assessing the ecological quality and ecological grade (‘status’) of individual river stretches. The choice of reference sites and environmental predictor variables, the site classification and discrimination methods, the estimation of the expected fauna, and indices for comparing the agreement, or lack of it, between the observed and expected fauna, are all discussed. The indices are assessed on the reference sites and on a separate test set of 340 sites, which are subject to a wide range of types and degrees of impairment.     

Assessing the Conservation Value of New Zealand's Offshore Islands

New Zealand has over 500 offshore islands. Many consist of relatively pristine wilderness and contain a disproportionately large amount of New Zealand's biological wealth. Increased demand for recreation has increased the pressure to develop these areas. Although the New Zealand public has shown widespread support for the preservation and safeguarding of both wilderness areas and endangered species, economic values attached to the benefits of preservation are not known. Using a telephone survey, we found evidence that the public attaches significant value to the conservation of Little Barrier Island. The results show that these values are influenced by different levels and sources of prior information.     

Forecasting the Risk of Brown Tree Snake Dispersal from Guam: a Mixed Transport-Establishment Model

Abstract:  The brown tree snake ( Boiga irregularis ) is a devastating invader that has ecologically and economically affected Guam and is poised to disperse further. Interdiction efforts are being conducted on Guam and some of the potential receiving sites, but no tools exist for evaluating the potential for snake incursion; thus, the amount of effort that should be invested in protecting particular sites is unknown. We devised a model that predicts the relative risk of establishment of the brown tree snake (BTS) at a given site. To calculate overall risk, we incorporated in the model information on the likelihood of an organism entering the transportation system, avoiding detection, surviving to arrive at another location, and establishing at the receiving end. On the basis of documented rates of snake arrival at receiving sites, the model produced realistic predictions of invasion risk. Model outputs can thus be used to prioritize interdiction efforts to focus on especially vulnerable receiving locations. We provide examples of the utility of the model in evaluating the impacts of changes in transportation parameters. Finally, the model can be used to evaluate the impacts that BTS establishment at an additional site and that creation of a new source of snakes would have. The use of qualitative inputs allows the model to be adapted by substituting data on other invasive species or transportation systems.     

Mapping opportunities and challenges for rewilding in Europe

Farmland abandonment takes place across the world due to socio‐economic and ecological drivers. In Europe agricultural and environmental policies aim to prevent abandonment and halt ecological succession. Ecological rewilding has been recently proposed as an alternative strategy. We developed a framework to assess opportunities for rewilding across different dimensions of wilderness in Europe. We mapped artificial light, human accessibility based on transport infrastructure, proportion of harvested primary productivity (i.e., ecosystem productivity appropriated by humans through agriculture or forestry), and deviation from potential natural vegetation in areas projected to be abandoned by 2040. At the continental level, the levels of artificial light were low and the deviation from potential natural vegetation was high in areas of abandonment. The relative importance of wilderness metrics differed regionally and was strongly connected to local environmental and socio‐economic contexts. Large areas of projected abandonment were often located in or around Natura 2000 sites. Based on these results, we argue that management should be tailored to restore the aspects of wilderness that are lacking in each region. There are many remaining challenges regarding biodiversity in Europe, but megafauna species are already recovering. To further potentiate large‐scale rewilding, Natura 2000 management would need to incorporate rewilding approaches. Our framework can be applied to assessing rewilding opportunities and challenges in other world regions, and our results could guide redirection of subsidies to manage social‐ecological systems.     

Wilderness forms and their implications for global environmental policy and conservation

With the intention of securing industry-free land and seascapes, protecting wilderness entered international policy as a formal target for the first time in the zero draft of the Post-2020 Global Biodiversity Framework under the Convention on Biological Diversity. Given this increased prominence in international policy, it is timely to consider the extent to which the construct of wilderness supports global conservation objectives. We evaluated the construct by overlaying recently updated cumulative human pressure maps that offer a global-scale delineation of industry-free land as wilderness with maps of carbon stock, species richness, and ground travel time from urban centers. Wilderness areas took variable forms in relation to carbon stock, species richness, and proximity to urban centers, where 10% of wilderness areas represented high carbon and species richness, 20% low carbon and species richness, and 3% high levels of remoteness (>48 h), carbon, and species richness. Approximately 35% of all remaining wilderness in 2013 was accessible in <24 h of travel time from urban centers. Although the construct of wilderness can be used to secure benefits in specific contexts, its application in conservation must account for contextual and social implications. The diverse characterization of wilderness under a global environmental conservation lens shows that a nuanced framing and application of the construct is needed to improve understanding, communication, and retention of its variable forms as industry-free places.     

The Links between Protected Areas, Faiths, and Sacred Natural Sites

Most people follow and are influenced by some kind of spiritual faith. We examined two ways in which religious faiths can in turn influence biodiversity conservation in protected areas. First, biodiversity conservation is influenced through the direct and often effective protection afforded to wild species in sacred natural sites and in seminatural habitats around religious buildings. Sacred natural sites are almost certainly the world's oldest form of habitat protection. Although some sacred natural sites exist inside official protected areas, many thousands more form a largely unrecognized "shadow" conservation network in many countries throughout the world, which can be more stringently protected than state-run reserves. Second, faiths have a profound impact on attitudes to protection of the natural world through their philosophy, teachings, investment choices, approaches to land they control, and religious-based management systems. We considered the interactions between faiths and protected areas with respect to all 11 mainstream faiths and to a number of local belief systems. The close links between faiths and habitat protection offer major conservation opportunities, but also pose challenges. Bringing a sacred natural site into a national protected-area system can increase protection for the site, but may compromise some of its spiritual values or even its conservation values. Most protected-area managers are not trained to manage natural sites for religious purposes, but many sacred natural sites are under threat from cultural changes and habitat degradation. Decisions about whether or not to make a sacred natural site an "official" protected area therefore need to be made on a case-by-case basis. Such sites can play an important role in conservation inside and outside official protected areas. More information about the conservation value of sacred lands is needed as is more informed experience in integrating these into wider conservation strategies. In addition, many protected-area staff need training in how to manage sensitive issues relating to faiths where important faith sites occur in protected areas.     

Importance of Roadless Areas in Biodiversity Conservation in Forested Ecosystems: Case Study of the Klamath-Siskiyou Ecoregion of the United States

Abstract: Although many roadless areas on federal lands have been mapped in the United States since the 1970s, there has been little specific research on how and to what degree roadless areas contribute to biodiversity conservation. We examined the ecological attributes of mapped roadless areas for the Klamath-Siskiyou ecoregion of northwestern California and southwestern Oregon (U.S.A.). Attributes examined include special elements (such as natural heritage, serpentine geology, late-seral forests, Port Orford cedar [ Chamaecyparis lawsoniana ] ), and key watersheds; elevation and habitat representation; and overall landscape connectivity. We compared designated wilderness to roadless areas, giving special attention to the relative importance of small roadless areas (405–2024 ha). We mapped nearly 500 roadless areas of ≥405 ha. Roadless areas occupied more than twice the land area of wilderness (approximately 27% of the entire ecoregion) and contained approximately 36% of the known occurrences of heritage elements, 37% of the mapped serpentine habitats, 36% of the remaining late-seral forests, 60% of Port Orford cedar strongholds, and 42% of key watersheds for aquatic biodiversity. In addition, roadless areas were composed of significant amounts of low- and mid-elevation sites and a substantial number of the 214 mapped physical-biological habitat types with strong complementarity with designated wilderness. Fragmentation analyses showed that roadless areas contributed to regional connectivity in important ways. Also, small roadless areas were an important component of the roadless-areas conservation assessment. For the Klamath-Siskiyou ecoregion, roadless areas and designated wilderness provide an important foundation upon which to develop a comprehensive regional conservation strategy.     

Effectiveness of multiple release sites in reintroduction of Persian fallow deer

Releasing animals in more than one location may increase or decrease the probability of success of a reintroduction project, yet the question of how many release sites to use has received little attention. We used empirical data from the reintroduction program of the Persian fallow deer (Dama mesopotamica) (Galilee region in northern Israel) in an individual-based spatially explicit simulation model to assess the effects of releasing deer from multiple sites. We examined whether multiple release sites increase reintroduction success, and if so, whether the optimal number of sites for a given scenario can be determined and whether the outcome differs if animals are released alternately (i.e., the location of the release alternates yearly between sites) or consecutively (i.e., one release site is used for several years, then another is used, and so forth). We selected 8 potential release sites in addition to the original site and simulated the release of 180 individuals at a rate of 10 individuals per year in different combinations of the original site and 1-4 additional sites. In our model, releasing animals into the wild at multiple sites produced higher population growth and greater spatial expansion than releasing animals at only one site and a consecutive-release approach was superior to an alternate-release approach. We suggest that through the use of simulation modeling that is based on empirical data from previous releases, managers can make better-informed decisions regarding the use of multiple release sites and greatly improve the probability of reintroduction success.     

Improving multi-site benefit functions via Bayesian model averaging: A new approach to benefit transfer

A benefit function transfer obtains estimates of willingness-to-pay (WTP) for the evaluation of a given policy at a site by combining existing information from different study sites. This has the advantage that more efficient estimates are obtained, but it relies on the assumption that the heterogeneity between sites is appropriately captured in the benefit transfer model. A more expensive alternative to estimate WTP is to analyze only data from the policy site in question while ignoring information from other sites. We make use of the fact that these two choices can be viewed as a model selection problem and extend the set of models to allow for the hypothesis that the benefit function is only applicable to a subset of sites. We show how Bayesian model averaging (BMA) techniques can be used to optimally combine information from all models.The Bayesian algorithm searches for the set of sites that can form the basis for estimating a benefit function and reveals whether such information can be transferred to new sites for which only a small data set is available. We illustrate the method with a sample of 42 forests from U.K. and Ireland. We find that BMA benefit function transfer produces reliable estimates and can increase about 8 times the information content of a small sample when the forest is ‘poolable’.     

Optimal design of compact and connected nature reserves for multiple species

When designing a conservation reserve system for multiple species, spatial attributes of the reserves must be taken into account at species level. The existing optimal reserve design literature considers either one spatial attribute or when multiple attributes are considered the analysis is restricted only to one species. We built a linear integer programing model that incorporates compactness and connectivity of the landscape reserved for multiple species. The model identifies multiple reserves that each serve a subset of target species with a specified coverage probability threshold to ensure the species' long‐term survival in the reserve, and each target species is covered (protected) with another probability threshold at the reserve system level. We modeled compactness by minimizing the total distance between selected sites and central sites, and we modeled connectivity of a selected site to its designated central site by selecting at least one of its adjacent sites that has a nearer distance to the central site. We considered structural distance and functional distances that incorporated site quality between sites. We tested the model using randomly generated data on 2 species, one ground species that required structural connectivity and the other an avian species that required functional connectivity. We applied the model to 10 bird species listed as endangered by the state of Illinois (U.S.A.). Spatial coherence and selection cost of the reserves differed substantially depending on the weights assigned to these 2 criteria. The model can be used to design a reserve system for multiple species, especially species whose habitats are far apart in which case multiple disjunct but compact and connected reserves are advantageous. The model can be modified to increase or decrease the distance between reserves to reduce or promote population connectivity.     

The new approach to wilderness preservation through benefit-cost analysis

Over the last 15 years, economists of Resources for the Future led by John Krutilla have been developing a new approach to wilderness preservation. This paper offers a concise exposition of their approach and adds a few steps. The basic ingredient of the new approach is a perception of declining net benefits of wilderness development and rising net benefits of wilderness preservation over time. These alter the entire shape of the benefit-cost analysis of potential development. The new steps involve reversibility, postponability, and recognition of multiple interest rates.     

Effects of Subsistence Hunting on Vertebrate Community Structure in Amazonian Forests

Abstract: Subsistence hunting affects vast tracts of tropical wilderness that otherwise remain structurally unaltered, yet distinguishing hunted from nonhunted tropical forests presents a difficult problem because this diffuse form of resource extraction leaves few visible signs of its occurrence. I used a standardized series of line-transect censuses conducted over a 10-year period to examine the effects of subsistence game harvest on the structure of vertebrate communities in 25 Amazonian forest sites subjected to varying levels of hunting pressure. Crude vertebrate biomass, which was highly correlated with hunting pressure, gradually declined from nearly 1200 kg km⁻² at nonhunted sites to less than 200 kg km⁻² at heavily hunted sites. Hunting had a negative effect on the total biomass and relative abundance of vertebrate species in different size classes at these forest sites, but it did not affect their overall density. In particular, persistent hunting markedly reduced the density of large-bodied game species (>5 kg), which contributed a large proportion of the overall community biomass at nonhunted sites (65–78%) and lightly hunted sites (55–71%). Nutrient-rich floodplain forests contained a consistently greater game biomass than nutrient-poor unflooded forests, once I controlled for the effects of hunting pressure. Conservative estimates of game yields indicate that as many as 23.5 million game vertebrates, equivalent to 89,224 tons of bushmeat with a market value of US$190.7 million, are consumed each year by the rural population of Brazilian Amazonia, which illustrates the enormous socioeconomic value of game resources in the region. My cross-site comparison documents the staggering effect of subsistence hunters on tropical forest vertebrate communities and highlights the importance of considering forest types and forest productivity in game management programs.     

Ecological Costs of Livestock Grazing in Western North America

Livestock grazing is the most widespread land management practice in western North America. Seventy percent of the western United States is grazed, including wilderness areas, wildlife refuges, national forests, and even some national parks. The ecological costs of this nearly ubiquitous form of land use can be dramatic. Examples of such costs include loss of biodiversity; lowering of population densities for a wide variety of taxa; disruption of ecosystem functions, including nutrient cycling and succession; change in community organization; and change in the physical characteristics of both terrestrial and aquatic habitats. Because livestock congregate in riparian ecosystems, which are among the biologically richest habitats in arid and semiarid regions, the ecological costs of grazing are magnified in these sites. Range science has traditionally been laden with economic assumptions favoring resource use. Conservation biologists are encouraged to contribute to the ongoing social and scientific dialogue on grazing issues.     

Colony choice in birds: models based on temporally invariant site quality

We propose two stochastic models to explain how birds choose colonies. In the resource choice model, birds settle at each site at a rate proportional to the total resources the site contains. In the reduced resource choice model, a smaller cohort of birds enters sites at a rate determined by the total resources at each, and the remaining individuals enter sites at a rate that is linearly proportional to the total number of birds already nesting at each site. Thus, a fraction of birds chooses sites based on the resources present, and the remainder are attracted to a site by the presence of other birds. Colony site quality is assumed not to vary between years. Both models result, on average, in an ideal free distribution of colony sizes if the birds' settlement rate is linearly related to the resources in a site, if resources are distributed equally among individuals within sites, and if individuals with equal resources have equal fitness. We applied these models to long-term data on colony sizes and site usage of cliff swallows in south-western Nebraska. A test of the resource choice model suggested that the swallow population as a whole did not choose sites based strictly on site quality or the total resources contained at each site. However, a test of the reduced resource choice model suggested that a smaller fraction of the individuals in each colony may have based their choice of site on local resource availability, with the remaining birds aggregating at those sites based on the number of birds already settled there. Tests of these models may provide insight into how individuals choose colony sites and why colonies vary in size.     

A simulation model for the management of low density recreational areas

In order to take account of the congestion externalities associated with increased use intensity during a given period at a low density recreational resource, it is necessary to determine the effects of alternative use levels on the expected quality of each individual's experience. This paper describes a simulation model for wilderness recreation designed to provide such measures. In order to illustrate the model it is applied to the Spanish Peaks Areas in Montana.     

Dynamic reserve design with the union-find algorithm

When reserve networks are established over time, there is a risk that sites will be developed in areas planned for future reservation, reducing the effectiveness of reserves. We developed a dynamic reserve design model that maximizes the expected number of species conserved, taking account of the risk of future habitat loss and fragmentation. The model makes use of the union-find algorithm, which is an efficient method for maintaining a list of connected regions in a graph as nodes and edges are inserted. A simple extension of the algorithm allows us to efficiently determine, for each species, when a sequence of site selections results in a reserve in which the species can persist. The extension also allows us to determine when a sequence of deforestation events results in the species becoming non-viable. The dynamic reserve design model is much more effective than commonly used heuristics, particularly when multiple connected sites are required for species persistence. The model also is able to solve much larger problems with greater effectiveness than the only previous dynamic reserve design model that considered site connectivity relationships. The union-find algorithm has much scope for addressing ecological management problems in which dynamic connectivity needs to be considered.     

An integrated approach for predicting fates of reintroductions with demographic data from multiple populations

We devised a novel approach to model reintroduced populations whereby demographic data collected from multiple sites are integrated into a Bayesian hierarchical model. Integrating data from multiple reintroductions allows more precise population-growth projections to be made, especially for populations for which data are sparse, and allows projections that account for random site-to-site variation to be made before new reintroductions are attempted. We used data from reintroductions of the North Island Robin (Petroica longipes), an endemic New Zealand passerine, to 10 sites where non-native mammalian predators are controlled. A comparison of candidate models that we based on deviance information criterion showed that rat-tracking rate (an index of rat density) was a useful predictor of robin fecundity and adult female survival, that landscape connectivity and a binary measure of whether sites were on a peninsula were useful predictors of apparent juvenile survival (probably due to differential dispersal away from reintroduction sites), and that there was unexplained random variation among sites in all demographic rates. We used the two best supported models to estimate the finite rate of increase (λ) for populations at each of the 10 sites, and for a proposed reintroduction site, under different levels of rat control. Only three of the reintroduction sites had λ distributions completely >1 for either model. At two sites, λ was expected to be >1 if rat-tracking rates were <5%. At the other five reintroduction sites, λ was predicted to be close to 1, and it was unclear whether growth was expected. Predictions of λ for the proposed reintroduction site were less precise than for other sites because distributions incorporated the full range of site-to-site random variation in vital rates. Our methods can be applied to any species for which postrelease data on demographic rates are available and potentially can be extended to model multiple species simultaneously.