Identifying realistic recovery targets and conservation actions for tigers in a human-dominated landscape using spatially explicit densities of wild prey and their determinants

Conservation Prioritization in a Tiger Landscape: Is Umbrella Species Enough?

Conservation approaches in tiger landscapes have focused on single species and their habitat. Further, the limited extent of the existing protected area network in India lacks representativeness, habitat connectivity, and integration in the larger landscape. Our objective was to identify sites important for connected tiger habitat and biodiversity potential in the Greater Panna Landscape, central India. Further, we aimed to set targets at the landscape level for conservation and prioritize these sites within each district in the landscape as specific management/conservation zones. We used earth observation data to derive an index of biodiversity potential. Marxan was used to identify sites that met tiger and biodiversity conservation targets with minimum costs. We found that to protect 50% of the tiger habitat with connectivity, 20% of the landscape area must be conserved. To conserve 100% of high biodiversity potential, 50% moderate biodiversity potential, and 25% low biodiversity potential, 55% of the landscape area must be conserved. To represent both tiger habitat and biodiversity, 62% of the total landscape area requires conservation or restoration intervention. The prioritized zones can prove significant for hierarchical decision making, involving multiple stakeholders in the landscape, including other tiger range areas.

Extending density surface models to include multiple and double-observer survey data

Spatial models of density and abundance are widely used in both ecological research (e.g., to study habitat use) and wildlife management (e.g., for population monitoring and environmental impact assessment). Increasingly, modellers are tasked with integrating data from multiple sources, collected via different observation processes. Distance sampling is an efficient and widely used survey and analysis technique. Within this framework, observation processes are modelled via detection functions. We seek to take multiple data sources and fit them in a single spatial model. Density surface models (DSMs) are a two-stage approach: first accounting for detectability via distance sampling methods, then modelling distribution via a generalized additive model. However, current software and theory does not address the issue of multiple data sources. We extend the DSM approach to accommodate data from multiple surveys, collected via conventional distance sampling, double-observer distance sampling (used to account for incomplete detection at zero distance) and strip transects. Variance propagation ensures that uncertainty is correctly accounted for in final estimates of abundance. Methods described here are implemented in the dsm R package. We briefly analyse two datasets to illustrate these new developments. Our new methodology enables data from multiple distance sampling surveys of different types to be treated in a single spatial model, enabling more robust abundance estimation, potentially over wider geographical or temporal domains.

Spatially explicit density and its determinants for Asiatic lions in the Gir forests

Asiatic lions (Panthera leo persica) are an icon of conservation success, yet their status is inferred from total counts that cannot account for detection bias and double counts. With an effort of 4,797 km in 725 km² of western Gir Protected Area, India, we used polygon search based spatially explicit capture recapture framework to estimate lion density. Using vibrissae patterns and permanent body marks we identified 67 lions from 368 lion sightings. We conducted distance sampling on 35 transects with an effort of 101.5 km to estimate spatial prey density using generalized additive modeling (GAM). Subsequently, we modeled lion spatial density with prey, habitat characteristics, anthropogenic factors and distance to baiting sites. Lion density (>1-year-old lions) was estimated at 8.53 (SE 1.05) /100 km² with lionesses having smaller movement parameter (σ = 2.55 km; SE 0.12) compared to males (σ = 5.32 km; SE 0.33). Detection corrected sex ratio (female:male lions) was 1.14 (SE 0.02). Chital (Axis axis) was the most abundant ungulate with a density of 63.29 (SE 10.14) as determined by conventional distance sampling (CDS) and 58.17 (SE 22.17)/km² with density surface modeling (DSM), followed by sambar (Rusa unicolor) at 3.84 (SE 1.07) and 4.73 (SE 1.48)/km² estimated by CDS and DSM respectively. Spatial lion density was best explained by proximity to baiting sites and flat valley habitat but not as much by prey density. We demonstrate a scientifically robust approach to estimate lion abundance, that due to its spatial context, can be useful for management of habitat and human-lion interface. We recommend this method for lion population assessment across their range. High lion densities in western Gir were correlated with baiting. The management practice of attracting lions for tourism can perturb natural lion densities, disrupt behavior, lion social dynamics and have detrimental effects on local prey densities.

Effects of human-induced prey depletion on large carnivores in protected areas: Lessons from modeling tiger populations in stylized spatial scenarios

Prey depletion is a major threat to the conservation of large carnivore species globally. However, at the policy-relevant scale of protected areas, we know little about how the spatial distribution of prey depletion affects carnivore space use and population persistence. We developed a spatially explicit, agent-based model to investigate the effects of different human-induced prey depletion experiments on the globally endangered tiger (Panthera tigris) in isolated protected areas-a situation that prevails throughout the tiger's range. Specifically, we generated 120 experiments that varied the spatial extent and intensity of prey depletion across a stylized (circle) landscape (1,000 km2) and Nepal's Chitwan National Park (~1,239 km2). Experiments that created more spatially homogenous prey distributions (i.e., less prey removed per cell but over larger areas) resulted in larger tiger territories and smaller population sizes over time. Counterintuitively, we found that depleting prey along the edge of Chitwan National Park, while decreasing tiger numbers overall, also decreased female competition for those areas, leading to lower rates of female starvation. Overall our results suggest that subtle differences in the spatial distributions of prey densities created by various human activities, such as natural resource-use patterns, urban growth and infrastructure development, or conservation spatial zoning might have unintended, detrimental effects on carnivore populations. Our model is a useful planning tool as it incorporates information on animal behavioral ecology, resource spatial distribution, and the drivers of change to those resources, such as human activities.

Recovery planning towards doubling wild tiger Panthera tigris numbers: Detailing 18 recovery sites from across the range

With less than 3200 wild tigers in 2010, the heads of 13 tiger-range countries committed to doubling the global population of wild tigers by 2022. This goal represents the highest level of ambition and commitment required to turn the tide for tigers in the wild. Yet, ensuring efficient and targeted implementation of conservation actions alongside systematic monitoring of progress towards this goal requires that we set site-specific recovery targets and timelines that are ecologically realistic. In this study, we assess the recovery potential of 18 sites identified under WWF's Tigers Alive Initiative. We delineated recovery systems comprising a source, recovery site, and support region, which need to be managed synergistically to meet these targets. By using the best available data on tiger and prey numbers, and adapting existing species recovery frameworks, we show that these sites, which currently support 165 (118-277) tigers, have the potential to harbour 585 (454-739) individuals. This would constitute a 15% increase in the global population and represent over a three-fold increase within these specific sites, on an average. However, it may not be realistic to achieve this target by 2022, since tiger recovery in 15 of these 18 sites is contingent on the initial recovery of prey populations, which is a slow process. We conclude that while sustained conservation efforts can yield significant recoveries, it is critical that we commit our resources to achieving the biologically realistic targets for these sites even if the timelines are extended.

Spatial modelling of bilby (Macrotis lagotis) and rabbit (Oryctolagus cuniculus) pellets within a predator-proof enclosure

A traditional design-reliant estimate of abundance is calculated by multiplying a density estimate obtained from transects to reflect the size of the study area. This type of estimate tells nothing about the nature of a species’ distribution between the samples. In contrast, model-based inference can better estimate abundance by interpolating transect estimates over the study area with the aid of covariates. This study used density surface modelling (DSM) to predict spatial distribution of greater bilby (Macrotis lagotis) and rabbit (Oryctolagus cuniculus) pellets within a predator-proof enclosure at Currawinya National Park, south-west Queensland. Pellets and latrines were counted using distance sampling and plot sampling on 30 PPBio plots during 2012 and 2014. Pellets and latrines were not strongly associated with habitat features, reflecting the generalist nature of both species. Bilby pellets were found on 23 plots in 2012 and 5 plots in 2014. Rabbit pellets were present on 29 plots in 2012 and 16 plots during 2014. These substantial declines in pellet abundances coincided with invasion of the feral cat (Felis catus) into the enclosure. While DSM modelling can allow managers to make informed decisions about applying survey effort or management practices, it is not suited to all species or situations.

Losing time for the tiger Panthera tigris: delayed action puts a globally threatened species at risk of local extinction

Meeting global and regional environmental targets is challenging, given the multiplicity of stakeholders and their diverse and often competing policy agendas and objectives. Relatively few studies have sought to systematically analyse the progress, or lack thereof, of institutionally complex and diffuse projects. Here we analyse one such project, which aims to protect and restore a critical landscape corridor for tigers Panthera tigris in north-western India, using a temporal–analytic framework that integrates ecological information on species population status and spatial connectivity modelling with a systematic examination of the decision-making process. We find that even with adequate ecological knowledge the tiger population is on the verge of local extinction because of weak institutional support, poor adaptive planning and ineffective leadership in a complex political arena, which has led to delays in conservation action. From the outset the conservation agencies and NGOs that were the primary drivers of the project lacked awareness of the political idiosyncrasies of coordinating the actions of disparate agencies within the decision-making process. To secure better future environmental outcomes we recommend the adoption of an improved project appraisal methodology that explicitly encompasses an evaluation of organizational incentives, to determine political buy-in, including alignment with organizational objectives and funding availability.