Distance Sampling Data Research Articles

Recovery of globally threatened ungulate species in Huai Kha Khaeng Wildlife Sanctuary, Thailand

The forests of Southeast (SE) Asia support large ungulate species listed as globally threatened by the International Union for the Conservation of Nature’s Red List. In Thailand, Huai Kha Khaeng Wildlife Sanctuary (HKK) still contains the endangered banteng (Bos javanicus) and the vulnerable gaur (Bos gaurus) and sambar (Rusa unicolor). These species have been declining for decades in SE Asia. HKK is within the core area of the Western Forest Complex, one of the largest protected forest landscapes in SE Asia. This area is one of the few remaining that have the capacity to support species recovery for large ungulates and the tiger that depend on them. However, critical knowledge gaps remain on ungulate population metrics and trends of the species. In 2021, we revisited the locations surveyed in 2007–2008 and used distance sampling methodology to estimate density and abundance of banteng, gaur, and sambar. Using 32 transects in 2007, 2008 and 2021, we conducted replicated sample surveys with a total walk-efforts of 1,000 km per year and collected Distance-sampling data on target species. Banteng density increased two-fold from 0.91 ± 0.40–1.71 ± 0.79 individuals/km2, and sambar density doubled from 2.03 ± 0.51–3.99 ± 0.95 individuals/km2. In 2021, HKK supported 2,736 ± 1,264 banteng and 9,855 ± 2,347 sambar, the largest populations of these two globally threatened species in SE Asia. These results indicate evidence for recovery of large ungulate species in HKK and hope for the conservation of these globally threatened species. Our results illustrate the value of HKK as an important tiger recovery site in SE Asia, as these species are primary prey for tigers and underscore the value of ongoing efforts to eradicate poaching.

Abundance Determination of Large Mammals with Distance Sampling Perspective, the Case of Elephants of the Mole National Park (MNP) of Ghana

Distance sampling with line transect method has been applied by many researchers to monitor and observe varied animals and plants with the aim of determining the population density and or abundance of animals. The application of this method has not received the needed attention in Ghana, in particular to monitor, observe, and estimate the densities and abundance of animals and plants in the game reserves of the Mole National Park (MNP) is not without exception and the statistics of these are always reported based on guesses and without any scientific proof. This study has seen the application of line transect methodology in the MNP in which the abundance estimates are statistically determined with both the classical and Bayesian philosophies of statistical approaches. An alternative means of detectability estimation using the total probability concept has been established to enhance the probability of detection of a rare and elusive population of large mammals. In performing statistical investigations on rare and elusive population, it appears insufficient to model from the classical perspective, the use of PRIOR knowledge as seen in the Bayesian context cannot be underestimated. This study proposed that the concept of Total Probability with prior knowledge of animals and plants in line transect surveys must be well embraced, Periodic censuss must be conducted regularly to help in establishing the rate of extinction of units of interest in wildlife and Distance sampling data with line transect sampling methodology need not be analysed using only the classical reasoning. Attention must be given to the existence and availability of prior knowledge of the units under study.

Data fusion of distance sampling and capture-recapture data

Species distribution models (SDMs) are increasingly used in ecology, biogeography, and wildlife management to learn about the species–habitat relationships and abundance across space and time. Distance sampling (DS) and capture-recapture (CR) are two widely collected data types to learn about species–habitat relationships and abundance; still, they are seldomly used in SDMs due to the lack of spatial coverage. However, data fusion of the two data sources can increase spatial coverage, which can reduce parameter uncertainty and make predictions more accurate, and therefore, can be used with SDM. We developed a model-based approach for data fusion of DS and CR data. Our modeling approach accounts for two common missing data issues: 1) individuals that are missing not at random (MNAR) and 2) partially missing location information. Using a simulation experiment, we evaluated the performance of our modeling approach and compared it to existing approaches that use ad-hoc methods to account for missing data issues. Our results show that our approach provides unbiased parameter estimates with increased efficiency compared to the existing approaches. Finally, we demonstrated our approach using data collected for Grasshopper Sparrows (Ammodramus savannarum) in north-eastern Kansas, USA.

Modelling density surfaces of intraspecific classes using camera trap distance sampling

AbstractSpatially explicit densities of wildlife are important for understanding environmental drivers of populations, and density surfaces of intraspecific classes allow exploration of links between demographic ratios and environmental conditions. Although spatially explicit densities and class densities are valuable, conventional design‐based estimators remain prevalent when using camera‐trapping methods for unmarked populations.We developed a density surface model that utilized camera trap distance sampling data within a hierarchical generalized additive modelling framework. We estimated density surfaces of intraspecific classes of a common ungulate, white‐tailed deerOdocoileus virginianus, across three large management regions in Indiana, United States. We then extended simple statistical theory to test for differences in two ratios of density.Deer density was influenced by landscape fragmentation, wetlands and anthropogenic development. We documented class‐specific responses of density to availability of concealment cover, and found strong evidence that increased recruitment of young was tied to increased resource availability from anthropogenic agricultural land use. The coefficients of variation of the total density estimates within the three regions we surveyed were 0.11, 0.10 and 0.06.Synthesis and applications. Our strategy extends camera trap distance sampling and enables managers to use camera traps to better understand spatial predictors of density. Our density estimates were more precise than previous estimates from camera trap distance sampling. Population managers can use our methods to detect finer spatiotemporal changes in density or ratios of intraspecific‐class densities. Such changes in density can be linked to land use, or to management regimes on habitat and harvest limits of game species.

Seasonal Differences in the Encounter Rate of the Fat-Tailed Dwarf Lemur (Cheirogaleus medius) in the Transitional Forests of Northwest Madagascar: Implications for Reliable Population Density Assessment

Primate encounter rates often vary throughout the year due to seasonal differences in activity, ecology, and behaviour. One notably extreme behaviour is continuous hibernation. Although a rare adaptation in primates, the dwarf lemurs of Madagascar (genus Cheirogaleus) enter obligate hibernation each year during the dry season, after spending the wet season consuming high-energy foods. Whilst seasonal changes in activity in some Cheirogaleus populations are well-known, many species remain little-studied, and there is no specific information on their encounter rates, nor when they enter and emerge from hibernation. This uncertainty critically affects reliable calculation of population density estimates for these highly threatened lemurs. In this study, we assessed how encounter rates of the fat-tailed dwarf lemur (C. medius) vary seasonally in the transitional forests of the Sahamalaza-Iles Radama National Park, northwest Madagascar, during a 4-year period. We established a system of line transects (N = 60) throughout our study area, on which we conducted distance sampling of C. medius. We then used our distance sampling data to calculate encounter rate and population density data. We found encounter rates of C. medius to be significantly higher during the wet season compared with the dry season. Furthermore, encounter rates of C. medius were particularly low from May–August. These results provide some evidence to suggest the time period that C. medius hibernate in Sahamalaza-Iles Radama National Park. These findings underpin the importance of careful study design when surveying threatened species with seasonal differences in activity, such as those that hibernate. This study also demonstrates the importance of species-specific behavioural data for accurate population density assessment, which is required to inform conservation action.

Spatial density patterns of herbivore response to seasonal dynamics in the tropical deciduous forest of central India

AbstractResource dynamism in seasonal forests leads ungulates to differential habitat preference; hence, their distribution aligns with environmental covariates across spatial and temporal scales. Seasonal patterns of four species of ungulates, namely sambar, chital, nilgai, and wild pig, were investigated and identified as the environmental variables driving the density gradient across two seasons, summer and winter, in the tropical dry deciduous forest of Panna Tiger Reserve, central India. Distance sampling data were analyzed using density surface modeling for ungulates with a survey effort of 518 km in winter and 356 km in summer in a generalized additive modeling framework. We found that season significantly affected the spatial densities of all ungulates, with sambar, chital, and nilgai congregating in summer and wild pig in winter. All ungulates showed a clear seasonal shift to the valley in summer and preferred plateaus in winter. The spatially explicit map outputs draw attention to the seasonal hot spots for ungulates abundance and the species and season‐specific roles of environment variables in defining their distribution. These results provide a scientific basis for direct conservation efforts to the spatially prioritized habitats for cost‐effective management interventions.

Integrated spatial models foster complementarity between monitoring programmes in producing large‐scale bottlenose dolphin indicators

AbstractOver the last decades, large‐scale ecological projects have emerged that require collecting ecological data over broad spatial and temporal coverage. Yet, obtaining relevant information about large‐scale population dynamics from a single monitoring programme is challenging, and often several sources of data, possibly heterogeneous, need to be integrated. In this context, integrated models combine multiple data types into a single analysis to quantify the population dynamics of a targeted population. When working at large geographical scales, integrated spatial models have the potential to produce spatialized ecological estimates that would be difficult to obtain if data were analysed separately. In this study, we illustrate how spatial integrated modelling offers a relevant framework for conducting ecological inference at large scales. Focusing on the Mediterranean bottlenose dolphins (Tursiops truncatus), we combined 21,464 km of photo‐identification boat surveys collecting spatial capture–recapture data with 24,624 km of aerial line transect following a distance sampling protocol. We analysed spatial capture–recapture data together with distance sampling data to estimate the abundance and density of bottlenose dolphins. We compared the performances of the distance sampling model and the spatial capture–recapture model fitted independently to our integrated spatial model. The outputs of our spatial integrated models inform bottlenose dolphin ecological status in the French Mediterranean Sea and provide ecological indicators that are required for regional‐scale ecological assessments like the EU Marine Strategy Framework Directive. We argue that integrated spatial models are widely applicable and relevant to conservation research and biodiversity assessment at large spatial scales.

Estimating uncertainty in density surface models.

Providing uncertainty estimates for predictions derived from species distribution models is essential for management but there is little guidance on potential sources of uncertainty in predictions and how best to combine these. Here we show where uncertainty can arise in density surface models (a multi-stage spatial modelling approach for distance sampling data), focussing on cetacean density modelling. We propose an extensible, modular, hybrid analytical-simulation approach to encapsulate these sources. We provide example analyses of fin whales Balaenoptera physalus in the California Current Ecosystem.

Fisheries bycatch mitigation measures as an efficient tool for the conservation of seabird populations

Abstract The impact of industrial fisheries on marine biodiversity is conspicuous in large pelagic vertebrate's fisheries bycatch. In seabirds, this led to the decline of many populations since the 1980s following the rise of global fishing effort. Bycatch mitigation measures were implemented since the 2000s, but their effects on the concerned seabird populations remain poorly quantified and understood. We studied the effects of bycatch mitigation measures on the demography of the white‐chinned petrel, one of the most bycatch impacted seabirds whose populations suffered dramatic declines before the implementation of mitigation measures. To do so we (a) built multi‐event capture–recapture models to estimate the demographic parameters of a population from Possession Island (southern Indian Ocean) over 30 years, (b) assessed the effect of climate and fishery covariates on demographic parameters, (c) built a population matrix model to estimate stochastic growth rate according to the management in fisheries bycatch and (d) estimated changes in breeding population density using distance sampling data. The population declined from the 1980s to the mid‐2000s, while trawl and longline fisheries occurred with no bycatch mitigation measures. The negative effects of fishery bycatch through additive mortality and of rat predation on breeding success were likely the main drivers of this decline. Both modelled population growth rate and observed breeding densities showed an increase since the mid‐2000s. We explained this trend by the improvement in survival probability following implementation of fishery bycatch mitigation measures and in breeding success probability with the local control of the rat population and changes in sea ice conditions on foraging grounds. Synthesis and applications. We provide a holistic approach to assess the effects of management measures by analysing datasets from sampling methods commonly employed in seabird studies. Our conclusions should encourage the eradication of invasive predatory species in seabird breeding areas and the strengthening of bycatch mitigation measures for the vulnerable seabird species, especially in international waters, but also the development of such measures considering the other marine large pelagic species threatened by fisheries bycatch (sharks, rays, turtles and marine mammals) since it could be crucial to avoid populations' extinction.

Efficient effort allocation in line‐transect distance sampling of high‐density species: When to walk further, measure less‐often and gain precision

Abstract Line‐transect distance sampling is widely used to estimate population densities using distances of observed targets from transect lines to model detectability. When the target taxa are high density, the frequent measuring of distances may make the method seem impractical. We present a method that improves the efficiency of distance sampling when the target species occurs at high density. Only a proportion of targets are measured to model the detection function, and the time saved on the survey is then used to cover a longer total length of transect and accrue a larger ‘count only’ sample. This approach can improve the precision of the population density estimate when the cost of measuring the distance to a detected target is more than half the cost of walking to the next target. We find the optimal proportion of distances to measure that minimises the variance of the density estimate for a fixed survey budget. We quantify how much this optimised strategy increases the precision of the density estimate compared with conventional line‐transect distance sampling. We then use simulated distance sampling data to test our expressions, and illustrate circumstances under which the optimised approach would be beneficial using distance sampling data on high‐density plants. The simulations indicate that the optimised method delivers benefits in precision, but the magnitude of the benefit is lower than predicted from our expressions, which are based on an asymptotic approximation of the variance. We apply an adjustment to the predicted benefit equation to account for this difference, and show that, in all three plant case studies, the optimised approach could improve the precision gained from a distance sampling survey between 20% and 50%. This new approach could broaden the ecological contexts in which distance sampling is applied, to include estimation of densities of abundant taxa where plots are conventionally used. The method may have interesting applications for other survey types, including multispecies surveys or those using cues or signs that occur at high density.

Variance Propagation for Density Surface Models

Spatially explicit estimates of population density, together with appropriate estimates of uncertainty, are required in many management contexts. Density surface models (DSMs) are a two-stage approach for estimating spatially varying density from distance sampling data. First, detection probabilities—perhaps depending on covariates—are estimated based on details of individual encounters; next, local densities are estimated using a GAM, by fitting local encounter rates to location and/or spatially varying covariates while allowing for the estimated detectabilities. One criticism of DSMs has been that uncertainty from the two stages is not usually propagated correctly into the final variance estimates. We show how to reformulate a DSM so that the uncertainty in detection probability from the distance sampling stage (regardless of its complexity) is captured as an extra random effect in the GAM stage. In effect, we refit an approximation to the detection function model at the same time as fitting the spatial model. This allows straightforward computation of the overall variance via exactly the same software already needed to fit the GAM. A further extension allows for spatial variation in group size, which can be an important covariate for detectability as well as directly affecting abundance. We illustrate these models using point transect survey data of Island Scrub-Jays on Santa Cruz Island, CA, and harbour porpoise from the SCANS-II line transect survey of European waters. Supplementary materials accompanying this paper appear on-line.

Integrating Distance Sampling and Minimum Count Data: A Reply to Becker and Herreman

Integrating Distance Sampling and Minimum Count Data: A Reply to Becker and Herreman

Comments on Integrating Distance Sampling and Minimum Count Data (Schmidt et al. 2019)

Comments on Integrating Distance Sampling and Minimum Count Data (Schmidt et al. 2019)

Integrating distance sampling and presence-only data to estimate species abundance.

Integrated models combine multiple data types within a unified analysis to estimate species abundance and covariate effects. By sharing biological parameters, integrated models improve the accuracy and precision of estimates compared to separate analyses of individual data sets. We developed an integrated point process model to combine presence-only and distance sampling data for estimation of spatially explicit abundance patterns. Simulations across a range of parameter values demonstrate that our model can recover estimates of biological covariates, but parameter accuracy and precision varied with the quantity of each data type. We applied our model to a case study of black-backed jackals in the Masai Mara National Reserve, Kenya, to examine effects of spatially varying covariates on jackal abundance patterns. The model revealed that jackals were positively affected by anthropogenic disturbance on the landscape, with highest abundance estimated along the Reserve border near human activity. We found minimal effects of landscape cover, lion density, and distance to water source, suggesting that human use of the Reserve may be the biggest driver of jackal abundance patterns. Our integrated model expands the scope of ecological inference by taking advantage of widely available presence-only data, while simultaneously leveraging richer, but typically limited, distance sampling data.

Aerial surveys for Antarctic minke whales (Balaenoptera bonaerensis) reveal sea ice dependent distribution patterns.

This study investigates the distribution of Antarctic minke whales (AMW) in relation to sea ice concentration and variations therein. Information on AMW densities in the sea ice‐covered parts of the Southern Ocean is required to contextualize abundance estimates obtained from circumpolar shipboard surveys in open waters, suggesting a 30% decline in AMW abundance. Conventional line‐transect shipboard surveys for density estimation are impossible in ice‐covered regions, therefore we used icebreaker‐supported helicopter surveys to obtain information on AMW densities along gradients of 0%–100% of ice concentration. We conducted five helicopter surveys in the Southern Ocean, between 2006 and 2013. Distance sampling data, satellite‐derived sea‐ice data, and bathymetric parameters were used in generalized additive models (GAMs) to produce predictions on how the density of AMWs varied over space and time, and with environmental covariates. Ice concentration, distance to the ice edge and distance from the shelf break were found to describe the distribution of AMWs. Highest densities were predicted at the ice edge and through to medium ice concentrations. Medium densities were found up to 500 km into the ice edge in all concentrations of ice. Very low numbers of AMWs were found in the ice‐free waters of the West Antarctic Peninsula (WAP). A consistent relationship between AMW distribution and sea ice concentration weakens the support for the hypothesis that varying numbers of AMWs in ice‐covered waters were responsible for observed changes in estimated abundance. The potential decline in AMW abundance stresses the need for conservation measures and further studies into the AMW population status. Very low numbers of AMWs recorded in the ice‐free waters along the WAP support the hypothesis that this species is strongly dependent on sea ice and that forecasted sea ice changes have the potential of heavily impacting AMWs.

Inlabru: an R package for Bayesian spatial modelling from ecological survey data

Abstract Spatial processes are central to many ecological processes, but fitting models that incorporate spatial correlation to data from ecological surveys is computationally challenging. This is particularly true of point pattern data (in which the primary data are the locations at which target species are found), but also true of gridded data, and of georeferenced samples from continuous spatial fields. We describe here the R package inlabru that builds on the widely used RINLA package to provide easier access to Bayesian inference from spatial point process, spatial count, gridded, and georeferenced data, using integrated nested Laplace approximation (INLA, Rue et al., 2009). The package provides methods for fitting spatial density surfaces and estimating abundance, as well as for plotting and prediction. It accommodates data that are points, counts, georeferenced samples, or distance sampling data. This paper describes the main features of the package, illustrated by fitting models to the gorilla nest data contained in the package spatstat (Baddeley, & Turner, 2005), a line transect survey dataset contained in the package dsm (Miller, Rexstad, Burt, Bravington, & Hedley, 2018), and to a georeferenced sample from a simulated continuous spatial field.

Future Outlook of Building Indonesia National Hornbill Monitoring Protocol

Hornbills (Bucerotidae) are endangered bird species and current trade issues become major threats which surpass the information available on hornbill population. Differences in survey protocol from various regions in Indonesia may lead to acquisition of low quality data. Using Distance Sampling data from Sumatra and Sulawesi gathered from various projects, we examined different protocols for hornbill population estimation. The results demonstrated that most of hornbill species are distributed in the undisturbed forests of protected areas. The higher the efforts of the survey, the more observations are obtained. Density estimations were similar for Sumatran species except for the Rhinoceros Hornbill. Density estimation for Sulawesi hornbills was inconsistent, probably due to difference in site (mainland versus island). Line transect seemed more effective to achieve detections compared to point count suggest that detection at initial location may be violated in point count. These results suggested that understanding distance sampling is important in standardising protocols. Additionally, improvement on detections and hornbill identification is needed. Thus, training is crucial to ensure that a national hornbill monitoring protocol can be established and applied throughout Indonesia.

Model selection with overdispersed distance sampling data

AbstractDistance sampling (DS) is a widely used framework for estimating animal abundance.DSmodels assume that observations of distances to animals are independent. Non‐independent observations introduce overdispersion, causing model selection criteria such asAICorAICcto favour overly complex models, with adverse effects on accuracy and precision.We describe, and evaluate via simulation and with real data, estimators of an overdispersion factor (), and associated adjusted model selection criteria (QAIC) for use with overdispersedDSdata. In other contexts, a single value ofis calculated from the “global” model, that is the most highly parameterised model in the candidate set, and used to calculateQAICfor all models in the set; the resultingQAICvalues, and associated ΔQAICvalues andQAICweights, are comparable across the entire set. Candidate models of theDSdetection function include models with different general forms (e.g. half‐normal, hazard rate, uniform), so it may not be possible to identify a single global model. We therefore propose a two‐step model selection procedure by whichQAICis used to select among models with the same general form, and then a goodness‐of‐fit statistic is used to select among models with different forms. A drawback of this approach is thatQAICvalues are not comparable across all models in the candidate set.Relative toAIC,QAICand the two‐step model selection procedure avoided overfitting and improved the accuracy and precision of densities estimated from simulated data. When applied to six real datasets, adjusted criteria and procedures selected either the same model asAICor a model that yielded a more accurate density estimate in five cases, and a model that yielded a less accurate estimate in one case.ManyDSsurveys yield overdispersed data, including cue counting surveys of songbirds and cetaceans, surveys of social species including primates, and camera‐trapping surveys. Methods that adjust for overdispersion during the model selection stage ofDSanalyses therefore address a conspicuous gap in theDSanalytical framework as applied to species of conservation concern.

Hierarchical mark‐recapture distance sampling to estimate moose abundance

ABSTRACTEstimating the abundance of wide‐ranging wildlife, difficult under any circumstances, is particularly challenging when detection is low and affected by factors that also influence density and distribution. In northeastern Washington, moose (Alces alces) have evidently increased since the 1970s but spend most of their time under coniferous cover that makes detection from the air difficult. We used a Bayesian hierarchical approach to incorporate habitat use (in the form of availability as a function of canopy closure) into a detection model within a mark‐recapture distance sampling framework to estimate moose density. Our model of availability used a latent density surface employing habitat use data obtained from 17 adult female moose wearing global positioning system (GPS) collars. Distance sampling data, obtained from helicopter surveys in winters 2014, 2015, and 2016, consisted of double‐observer detections of 166 moose groups along 2,241 km of systematically placed line transects within 29 survey blocks selected using a stratified‐random design. We estimated moose density over the entire survey area as 0.49/km2 (95% credible interval = 0.33–0.67/km2). Extrapolated to the 10,513‐km2 survey area, we estimated 5,169 moose (95% credible interval = 3,510–7,034). Our methodology allowed us to adjust for availability bias and produce an estimate even where detection was difficult but required many hours of helicopter flights, acceptable weather conditions, and the availability of GPS collared‐moose. © 2018 The Wildlife Society.

Characterizing spatio-temporal variation in survival and recruitment with integrated population models

ABSTRACT Efforts to understand population dynamics and identify high-quality habitat require information about spatial variation in demographic parameters. However, estimating demographic parameters typically requires labor-intensive capture–recapture methods that are difficult to implement over large spatial extents. Spatially explicit integrated population models (IPMs) provide a solution by accommodating spatial capture–recapture (SCR) data collected at a small number of sites with survey data that may be collected over a much larger extent. We extended the spatial IPM framework to include a spatio-temporal point process model for recruitment, and we applied the model to 4 yr of SCR and distance-sampling data on Canada Warblers (Cardellina canadensis) near the southern extent of the species' breeding range in North Carolina, USA, where climate change is predicted to cause population declines and distributional shifts toward higher elevations. To characterize spatial variation in demographic parameters ...