Integrated spatial analysis for human–wildlife coexistence in the American West

Neil Carter ,MatthewAWilliamson , SophieGilbert, StacyALischka, Laura RPrugh, Joshua J Lawler, Alexander LMetcalf , Aerin L Jacob, Bray J Beltrán, Antonio JCastro, Abigail Sage andMorey Burnham 1 School for Environment and Sustainability, University ofMichigan, DanaBuilding, 440Church Street, AnnArbor,MI 48109,United States of America 2 Human-Environment Systems, Boise StateUniversity. 1910UniversityDr, Boise, ID 83725,United States of America 3 Department of Fish&Wildlife Sciences, University of Idaho, 875 PerimeterDrive,Moscow, ID 83844,United States of America 4 Department of Fish,Wildlife andConservation Biology, Colorado StateUniversity, 1474CampusDelivery, Fort Collins, CO80523, United States of America 5 School of Environmental and Forest Sciences, University ofWashington, Box 352100, Seattle,WA98195,United States of America 6 HumanDimensions Lab,WAFrankeCollege of Forestry andConservation, University ofMontana, 440CHCB, 32CampusDrive, Missoula,MT59812,United States of America 7 Yellowstone toYukonConservation Initiative, 200-1350RailwayAve., Canmore, AB, T1W1P6, Canada 8 Heart of the Rockies Initiative, 120Hickory St.Missoula,MT59801,United States of America 9 Centro Andaluz para la Evaluación y Seguimiento del CambioGlobal (CAESCG), Departamento de Biología Vegetal y Ecología, Universidad deAlmería, La Cañada de SanUrbano, E-04120, Almería, Spain 10 Idaho StateUniversity, Department of Biological Sciences, 921 South 8thAvenue, Pocatello, ID 83209,United States of America 11 Department of Sociology, SocialWork, andCriminology, Idaho StateUniversity, 921 South 8thAvenue, Pocatello, ID 83209,United States of America 12 Author towhomany correspondence should be addressed.


Introduction
The future of conservation and human-wildlife relationships in the American West is at a defining moment. The region consists of a mosaic of land-cover types, with large amounts of public land under varying degrees of protection, use, and ownership. This public land provides the foundation for high levels of connectivity and habitat for healthy populations of wildlife, including those with large resource requirements such as large and wide-ranging mammals (Barnes et al 2016). However, space for wildlife is under threat in the West. Energy development projects, urban and ex-urban sprawl, increasing road traffic and density, and amenity-driven human migration are dramatically changing the ecological landscape (Leu et al 2008). The social landscape is rapidly changing as well, with new residents bringing different worldviews, economic activities, and expectations regarding wildlife and their habitats (Teel and Manfredo 2010). Because maintaining and establishing landscape connectivity for wildlife in part depends on facilitating their movement across privately-owned lands that connect protected areas, balancing disparate human priorities with wildlife conservation across large landscapes in the American West requires novel approaches to conservation practice.
Inclusion of multi-level drivers of social processes and human behavior in spatial analysis and conservation planning represents a tremendous opportunity to improve outcomes for both wildlife and humans in shared landscapes (Lischka et al 2018). A growing body of work has demonstrated novel ways to spatially integrate social and ecological factors that can better inform decision making for human-wildlife coexistence under changing conditions (Bryan et al 2011, Behr et al 2017, Williamson et al 2018. Here, we build on that foundation to underscore the utility of integrating social factors into traditional spatial analysis to promote human-wildlife coexistence in the American West.

Conceptual framing for integrated spatial analysis
Contemporary conservation and land use plans integrate a substantial amount of information on landscapes' biophysical characteristics (e.g. land cover, topography, climatic conditions) and the (potentially pervasive) impacts of human actions (e.g. land use, environmental policy) to identify priority locations for conservation and management actions. This information is often generated OPEN ACCESS PUBLISHED 30 January 2020 Original content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence.
Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
using modeled interpolations of habitat suitability and structural or functional connectivity or other conservation metrics. Areas of biophysical importance are considered in conjunction with the monetary costs associated with developing and implementing a conservation plan. However, important social processes such as wealth distribution, institutional and governance structures, worldviews, and human attitudes-all drivers of human behaviors-may intervene to make monetary cost alone a poor proxy for the benefits and costs of coexistence (Carter and Linnell 2016). Failure to adequately consider these seldom used social dimensions will stymie implementation of plans or render them ineffective (figure 1).
We advocate an alternative framework wherein habitat quality, connectivity, or other conservation metrics are derived from attributes of both the biophysical and social landscapes (figure 1). Explicitly incorporating social factors into spatial analysis would allow practitioners to identify locations where coexistence strategies are both biologically critical and socially feasible. Moreover, a broader incorporation of the social factors that inhibit or promote conservation may help identify a more diverse suite of targeted interventions to achieve desired conservation outcomes.

Example Cases of integrated spatial analysis
Several key concepts, metrics, and data types in wildlife and human research are amenable to spatial integration (table 1). Below, we provide three example cases that highlight the value of integrating social dimensions into traditional wildlife-related metrics in the American West.
Integrating social dynamics into habitat assessments Measuring and mapping animal habitats are core activities of wildlife ecologists. Often spatial trends in human activities, such as urban sprawl, are included as predictors in habitat models via maps of land cover. However, social perceptions and changes in human institutions, attitudes, and behaviors associated with human demographic, cultural, or political change can also have strong effects on wildlife habitats. For example, rapid population growth in many areas of the American West is often associated with a decline in farming and ranching, and an increase in outdoor recreation (Hansen et al 2002). Such shifts not only alter habitat characteristics of landscapes (e.g. fragmenting riparian areas) but also the frequency of direct human-wildlife encounters. Furthermore, interactions between economic modernization (e.g. urbanization) and human demography have shifted worldviews toward wildlife (e.g. support for protection) in many parts of the world (Bruskotter et al 2017), including in the American West (Teel and Manfredo 2010), affecting how people perceive, value and behave toward wildlife (e.g. emphasizing nonconsumptive uses). Thus, changes in the characteristics of humans moving to or from an area may have a strong effect on local wildlife beyond physical changes to habitat (e.g. fragmentation from roads and recreational trails). The effects of changing social dimensions have not yet been sufficiently incorporated into spatial analyses and planning for wildlife conservation. Conceptual framework illustrating two different approaches to developing spatial coexistence strategies. Information on wildlife species attributes are often assessed in contemporary spatial conservation plans (top panel); however, human social factors, like attitudes and institutions, are rarely incorporated. Without these social factors, conservation actions might be less effective than intended, or even counterproductive in shared landscapes. In contrast, an approach based on a social-ecological systems foundation (bottom panel) would integrate social and wildlife spatial data to better identify coexistence opportunities that incorporate various costs and ensure planning success. Table 1. Summary of various concepts and measures in social-ecological science that are amenable to spatial integration for human-wildlife coexistence. We also indicate the degree to which we perceive these different data to be available or discoverable to researchers and practitioners. High = Publically available data covering large spatial extents; Moderate = Available on a project-by-project basis, over small spatial extents; Low = Not available, but possible to develop methods to collect. Incorporating human tolerance in connectivity surfaces In addition to habitat quality, ecologists often seek to map habitat connectivity, focusing on the factors that impede animal movements across landscapes, such as roads and inhospitable land-cover types. It is possible for wildlife to use human-occupied areas as habitat or movement corridors despite negative attitudes toward those animals, although these attitudes may impede efforts to restore wildlife populations, habitats or connectivity. Moreover, without spatial information on human tolerance, conservation actions may facilitate animal movement to ecological traps, where landscape features appear as suitable habitat yet human intolerance may lead to mortality. For example, human intolerance is a major impediment to reintroduction efforts of predator species, such as grizzly bears (Ursus arctos horribilis) or Mexican wolves (Canis lupus baileyi), where a large proportion of known mortalities are attributed to management removal or illegal retaliatory killing (USFWS 2016, USGS 2018). Indeed, high-quality biological habitat (e.g. floodplains, berry patches) is often also preferred by humans for development. Mapping human tolerance levels (attitude or behavior) and integrating them into existing analytical approaches for measuring connectivity will help identify priority areas for conservation that better account for the social dimension.
Evaluating spatial patterns of ecosystem services, disservices, and their recipients Researchers also seek to quantify and map ecosystem services provided by wildlife, such as ecotourism, crop pollination, or waste and pest removal ( instances of livestock depredation by wolves occurred in those same three states (Bradley et al 2015). In response, 326 partial packs and 48 full packs were killed (Bradley et al 2015). The spatial patterns of both wildlife-related ecosystem services and disservices, and their recipients, remain inadequately understood.

Opportunities for spatial data integration and analyses
There are various levels and methods of integrating human and wildlife data (table 2), each of which has its advantages, disadvantages, and outcomes for conservation planning. Below we highlight several promising methods.
Driving the big data revolution are remotelysensed and social media data, which open up new avenues for spatial integration at unprecedented scales and extents. The increased availability of worldwide high-resolution remote sensing products from a number of sources, such as the National Aeronautics and Space Administration (NASA)'s Landsat, moderate resolution imaging spectroradiometer (MODIS), Visible Infrared Imaging Radiometer Suite (VIIRS), the European Space Agency (ESA)'s Sentinel, and other ventures, enable inference and prediction of species distributions and their change (Jetz et al 2019). When combined with ancillary data, like wildlife population surveys, Earth Observation data enable the spatial characterization of an animal's realized niche, which might be constrained by human worldviews, attitudes, or behaviors. Social media platforms are also an increasingly important source of information for investigating human-nature interactions, including coupling location data with perceptions (e.g. ecosystem services), motivations, and behaviors. For example, recent work extracted data from users of different social media platforms to quantify and map their aesthetic and recreational values toward landscapes across European countries (Van Zanten et al 2016).
New analysis techniques, or those from other fields, offer promise for more robust integration of social dimensions into spatial analysis for wildlife conservation planning. Microtargeting, for example, allows conservationists to borrow tools from marketing and political sciences to understand conservation propensity at the individual level (Metcalf et al 2019). Coupled with increasing access to spatially explicit data on land ownership, these techniques could allow wildlife conservationists to identify prime corridor areas based on habitat quality and social receptivity. Geospatial tools, common in the field of human geography, can be used to spatially map and predict human tolerances toward wildlife (e.g. Struebig et al 2018) and integrate those surfaces into models of landscape resistance to animal movement (i.e. unidirectional relationship in table 2). Likewise, spatializing models of the policy processes (e.g. collaborative Table 2. Summary of existing ways of integrating social and ecological layers for human-wildlife coexistence. A non-exhaustive list of methods, considerations, and outcomes are described for each level of integration. Interactions between humans and wildlife are structured in time as well as space. For instance, conservation actions may alter the distribution of species (e.g. increased use of wildlife corridors may bring wildlife into areas where they were previously uncommon), which may alter human attitudes or tolerance towards those species going forward. To assess these changes, dynamic occupancy and spatial capturerecapture models can include various human activities as predictors of the probability that a species (re)occupies or vacates a portion of the landscape through time (Marescot et al 2019, table 2). Spatial econometric models can simulate how landowners respond to wildlife-related policies and measure the consequences of these decisions for wildlife conservation (Lewis et al 2011). Agent-based models also provide a means of incorporating the complexity of human decision making with the behavioral response of species (Van Schmidt et al 2019, table 2). Parameterizing these models could be based on telemetry and accelerometer data that measure an animal's behavioral response to the presence of different human activities (e.g. recreation, hazing), potentially augmented by other forms of wildlife or human data (e.g. remote camera traps, citizen science and social media data).

Conclusion
Although calls have been made in the past to integrate human and wildlife data in spatial analysis and conservation planning, conceptual and methodological hurdles persist. Here, we draw from multiple disciplines and work in various regions to provide suggestions for overcoming those hurdles, and highlight concrete examples of the utility of an integrated approach in shared landscapes, such as those that characterize the American West (Jones et al 2019). Mainstreaming integrated spatial analysis into coexistence strategies, however, will require developments in multiple areas, including: overcoming technical challenges of data awareness, processing, and access; establishing new spatial metrics of human social factors, like attitudes; quantifying spatial tradeoffs in human-wildlife interactions, such as in ecosystem (dis)services; protecting highly sensitive, spatial wildlife data (e.g. reproductive locations, high-use areas targeted by poachers) and human data (e.g. confidential information); and quantifying spatial feedbacks between humans and wildlife. Furthermore, as global change becomes ubiquitous and conservation needs and priorities fluctuate in space and time, integrated spatial analysis and conservation planning will need to become an iterative process, requiring increased use of forecasting, decision support frameworks, and involvement with multiple stakeholder groups. Progress in these areas is predicated on people recognizing the value of social-ecological analysis, investing in it, and innovating creative solutions to its constraints. Doing so will help advance the theory and practice of coexistence in globally pervasive shared landscapes.