Explore the vast range of titles available.

The scaly-naped pigeon (Patagioenas squamosa) is threatened by hunting in the Caribbean. At present, the pigeon is abundant in Puerto Rico, but overharvesting is a major concern; therefore, the development of a sustainable harvest strategy is a management priority. The management objective of the harvest strategy is to maximize hunting opportunity while keeping the population above an abundance threshold (NT) of 260,000 pigeons. To facilitate operational development and implementation of the harvest strategy, we conducted point-transect distance sampling to estimate population size, and mail and telephone hunter surveys to estimate total harvest. We used monitoring data and a Bayesian state-space model to estimate population and harvest management parameters, and predicted changes in population size as a function of expected total harvest. Population size averaged 262,899 pigeons (SE = 122,087) in April-June 1986-2012, and total harvest averaged 40,760 pigeons (SD = 43,405) in September-November 1986-2011. Intrinsic rate of growth was 0.442 (SD = 0.142), carrying capacity was 524,900 pigeons (SD = 119,200), maximum sustainable harvest rate was 0.221 (SD = 0.071) with a total harvest of 57,988 pigeons (SD = 14,640), and equilibrium population size was 262,500 pigeons (SD = 59,620). Because the population recently recovered from deforestation and has been affected by hurricanes, a conservative harvest level was prescribed at 50,311 pigeons (SD = 24,939). However, harvest in 2008-2011 was 2.1 times larger than the maximum sustainable yield. Assuming 122,905 pigeons were harvested per year in 2012 and 2013, we predicted a population size of 216,000 pigeons (SD = 90,770) in 2014, suggesting that restrictive regulations may need to be prescribed to meet the management objective. Our monitoring and modeling framework is an important first step in the development and implementation of a sustainable harvest strategy for the scaly-naped pigeon, and the approach can be applied to the management of other columbids in Puerto Rico.

Many wildlife monitoring programmes collect annual data on population abundance. The resulting abundance estimates fluctuate over time partly because of true population change and partly because of observation error. These two components of variation can be separated by fitting the estimates to a population dynamics model within a Bayesian state-space modelling framework. By constraining the population trajectory to be biologically realistic, more precise estimates can be obtained. Independent biological knowledge can be incorporated through choice of model structure and by specifying informative prior distributions on demographic parameters. We illustrate the approach using a 31-year point transect study of the Hawai’i ’ākepa (Loxops coccineus). We fitted five models, each making different assumptions about how population change, recruitment and/or adult survival varied over time. Overall, the ’ākepa geometric mean growth rate was 1.02, indicating an increasing population over the 31-year time series, although there were periods of slow decline potentially associated with low recruitment and more rapid recovery associated with pulses of high recruitment. Abundance estimates derived from the population models were substantially more precise than the ‘raw’ point transect estimates: 95% credible interval (CrI) was on average 51.7% (s.d. = 14.1%) narrower.

Birds are often obligate to specific habitats which can result in study areas with complex boundaries due to sudden changes in vegetation or other features. This can result in study areas with concave arcs or that include holes of unsuitable habitat such as lakes or agricultural fields. Spatial models used to produce species’ distribution and density estimates need to respect such boundaries to make informed decisions for species conservation and management. The soap film smoother is one model for complex study regions which controls the boundary behaviour, ensuring realistic values at the edges of the region. We apply the soap film smoother to account for boundary effects and compare it with thin plate regression spline (TPRS) smooth and design-based conventional distance sampling methods to produce abundance estimates from point-transect distance sampling collected data on Hawai‘i ‘Ākepa Loxops coccineus in the Hakalau Forest Unit of the Big Island National Wildlife Refuge Complex, Hawai‘i Island, USA. The soap film smoother predicted zero or near zero densities in the northern part of the domain and two hotspots (in the southern and central parts of the domain). Along the boundary the soap film model predicted relatively high densities where ‘Ākepa occur in the adjacent forest and near zero elsewhere. The design-based and soap film abundance estimates were nearly identical. The width of the soap film confidence interval was 16.5% and 0.8% wider than the width of the TPRS smooth and design-based confidence intervals, respectively. The peaks in predicted densities along the boundary indicates leakage by the TPRS smooth. We provide a discussion of the statistical methods, biological findings and management implications of applying soap film smoothers to estimate forest bird population status.

Insular species, particularly birds, experience high levels of speciation and endemism. Similarly, island birds experience extreme levels of extinction. Based on a 2012 taxonomic assessment, historically there were four reed-warbler species in the Mariana Islands, the Guam Reed-warbler Acrocephalus luscinia (Guam), the Nightingale Reed-warbler Acrocephalus hiwae (Saipan and Alamagan), the Aguijuan Reed-warbler A. nijoi (Aguiguan or Aguijuan), and the Pagan Reed-warbler A. yamashinae (Pagan). Between 2008 and 2010 we surveyed for three of these species on Alamagan, Aguiguan, and Pagan. Our results indicate that reed-warblers are extinct on Aguiguan, likely extinct on Pagan, and only the Nightingale Reed-warbler on Alamagan and Saipan remains. We estimated the global population at between 1,019 and 6,356 birds (95% CI; mean estimate 3,688), which has declined by more than 1,000 birds since the first quantitative surveys were conducted in 1982, i.e. a 24% decline in 28 years. Camp et al. (2009) describe the status of the Nightingale Reed-warbler on Saipan, which has also declined. We estimated the Alamagan population to be between 428 and 1,762 birds in 2010 (mean estimate 946). Thus, the Alamagan population is ~25 % of the global population, and it has declined slightly since 2000. This decline was not significant but is concerning, especially given a similar decline on Saipan. Restoration and protection of tall-stature native and secondary forest could benefit the Alamagan population, as would similar conservation on Saipan that includes wetland habitat. After suitable restoration of forest and wetland habitats on Aguiguan, Guam and Pagan, individuals from Alamagan and Saipan could serve as founder populations. Careful consideration of the extent and habitat preference of individuals translocated to Tinian, where an unknown reed-warbler species previously occurred, is warranted.

We used distance sampling to assess density and the detectability of male black grouse (Tetrao tetrix) and rock ptarmigan (Lagopus muta) in 2 protected areas of the Italian Alps. Our sampling effort was not sufficient to provide reliable inference for monitoring projects. Therefore, we used our field results to structure a simulation study to compare the performance of plot-based and distance sampling estimators of density. These 2 methods have different assumptions: plot sampling assumes a perfect detection of animals within the surveyed plots, whereas distance sampling assumes a decrease in detectability as the distance between observer and animal increases. The density and the detectability conditions adopted in the simulation were designed to be similar to those observed in our 2 studies, whereas the spatial patterns presumed for the simulated populations described a wide range of possible scenarios. Sampling points were allocated according to both random and stratified distributions. Simulation results showed that plot sampling underestimated density with bias invariably greater than 30% and confidence intervals with coverage lower than the nominal level of 95%. Conversely, distance sampling estimators provided bias levels invariably smaller than those obtained using plot sampling and bootstrap confidence intervals with empirical coverage near to or greater than 95%. Based on our simulations, the distance sampling estimator was superior to the plot sampling estimator for these grouse species on our study areas in terms of precision and accuracy and the stratified allocation was superior to the random allocation. However, distance sampling would be very costly to implement. Based on our simulations, 4-5 points per km² would be necessary to achieve reliable estimates of density and density changes. If distance sampling cannot be completed at this intensity, other sampling methods should be adopted.

1. Distance sampling is a widely used technique for estimating the size or density of biological populations. Many distance sampling designs and most analyses use the software Distance. 2. We briefly review distance sampling and its assumptions, outline the history, structure and capabilities of Distance, and provide hints on its use. 3. Good survey design is a crucial prerequisite for obtaining reliable results. Distance has a survey design engine, with a built-in geographic information system, that allows properties of different proposed designs to be examined via simulation, and survey plans to be generated. 4. A first step in analysis of distance sampling data is modelling the probability of detection. Distance contains three increasingly sophisticated analysis engines for this: conventional distance sampling, which models detection probability as a function of distance from the transect and assumes all objects at zero distance are detected; multiple-covariate distance sampling, which allows covariates in addition to distance; and mark-recapture distance sampling, which relaxes the assumption of certain detection at zero distance. 5. All three engines allow estimation of density or abundance, stratified if required, with associated measures of precision calculated either analytically or via the bootstrap. 6. Advanced analysis topics covered include the use of multipliers to allow analysis of indirect surveys (such as dung or nest surveys), the density surface modelling analysis engine for spatial and habitat modelling, and information about accessing the analysis engines directly from other software. 7.Synthesis and applications. Distance sampling is a key method for producing abundance and density estimates in challenging field conditions. The theory underlying the methods continues to expand to cope with realistic estimation situations. In step with theoretical developments, state-of-the-art software that implements these methods is described that makes the methods accessible to practising ecologists.

Imperfect detection during animal surveys biases estimates of abundance and can lead to improper conclusions regarding distribution and population trends. Farnsworth et al. (2005) developed a combined distance-sampling and time-removal model for point-transect surveys that addresses both availability (the probability that an animal is available for detection; e.g., that a bird sings) and perceptibility (the probability that an observer detects an animal, given that it is available for detection). We developed a hierarchical extension of the combined model that provides an integrated analysis framework for a collection of survey points at which both distance from the observer and time of initial detection are recorded. Implemented in a Bayesian framework, this extension facilitates evaluating covariates on abundance and detection probability, incorporating excess zero counts (i.e. zero-inflation), accounting for spatial autocorrelation, and estimating population density. Species-specific characteristics, such as behavioral displays and territorial dispersion, may lead to different patterns of availability and perceptibility, which may, in turn, influence the performance of such hierarchical models. Therefore, we first test our proposed model using simulated data under different scenarios of availability and perceptibility. We then illustrate its performance with empirical point-transect data for a songbird that consistently produces loud, frequent, primarily auditory signals, the Golden-crowned Sparrow (Zonotrichia atricapilla); and for 2 ptarmigan species (Lagopus spp.) that produce more intermittent, subtle, and primarily visual cues. Data were collected by multiple observers along point transects across a broad landscape in southwest Alaska, so we evaluated point-level covariates on perceptibility (observer and habitat), availability (date within season and time of day), and abundance (habitat, elevation, and slope), and included a nested point-within-transect and park-level effect. Our results suggest that this model can provide insight into the detection process during avian surveys and reduce bias in estimates of relative abundance but is best applied to surveys of species with greater availability (e.g., breeding songbirds).

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.

The dominant source of variance in line transect sampling is usually the encounter rate variance. Systematic survey designs are often used to reduce the true variability among different realizations of the design, but estimating the variance is difficult and estimators typically approximate the variance by treating the design as a simple random sample of lines. We explore the properties of different encounter rate variance estimators under random and systematic designs. We show that a design-based variance estimator improves upon the model-based estimator of Buckland et al. (2001, Introduction to Distance Sampling. Oxford: Oxford University Press, p. 79) when transects are positioned at random. However, if populations exhibit strong spatial trends, both estimators can have substantial positive bias under systematic designs. We show that poststratification is effective in reducing this bias.

Conventional distance sampling adopts a mixed approach, using model-based methods for the detection process, and design-based methods to estimate animal abundance in the study region, given estimated probabilities of detection. In recent years, there has been increasing interest in fully model-based methods. Model-based methods are less robust for estimating animal abundance than conventional methods, but offer several advantages: they allow the analyst to explore how animal density varies by habitat or topography; abundance can be estimated for any sub-region of interest; they provide tools for analysing data from designed distance sampling experiments, to assess treatment effects. We develop a common framework for model-based distance sampling, and show how the various model-based methods that have been proposed fit within this framework.