Quantifying the bias in density estimated from distance sampling and camera trapping of unmarked individuals (original) (raw)
Related papers
On bias, precision and accuracy in wildlife aerial surveys
Wildlife Research, 2008
Bias, precision and accuracy have been studied extensively in wildlife population estimation including aerial surveys. A review of the literature shows that the concepts of bias and precision are used broadly consistently. Aerial survey data from known populations of feral pig carcases and white-tailed deer show that few density estimates are unbiased and precise. Research is needed, however, to clarify how much bias and how much precision are enough for the various types of wildlife management activities. Accuracy is used in two closely related but different ways. One set of definitions of accuracy relates to deviations from the true value (bias) and the second set relates to squared deviations from the true value (bias and precision). The implications are that authors are encouraged to clearly state which definition of accuracy they use, or focus solely on bias and precision.
Application of the Random Encounter Model in citizen science projects to monitor animal densities
Abundance and density are vital metrics for assessing a species' conservation status and for developing effective management strategies. Remote-sensing cameras are being used increasingly as part of citizen science projects to monitor wildlife, but current methodologies to monitor densities pose challenges when animals are not individually recognizable. We investigated the use of camera traps and the Random Encounter Model (REM) for estimating the density of West European hedgehogs (Erinaceus europaeus) within a citizen science framework. We evaluated the use of a simplified version of the REM in terms of the parameters' estimation (averaged vs. survey-specific) and assessed its potential application as part of a large-scale, long-term citizen science project. We compared averaged REM estimates to those obtained via spatial capture-recapture (SCR) using data from nocturnal spotlight surveys. There was a high degree of concordance in REM-derived density estimates from averaged parameters versus those derived from survey-specific parameters. Averaged REM density estimates were also comparable to those produced by SCR at eight out of nine sites; hedgehog density was 7.5 times higher in urban (32.3 km À2) versus rural (4.3 km 2) sites. Power analyses indicated that the averaged REM approach would be able to detect a 25% change in hedgehog density in both habitats with >90% power. Furthermore, despite the high start-up costs associated with the REM method, it would be cost-effective in the long term. The averaged REM approach is a promising solution to the challenge of large-scale and long-term species monitoring. We suggest including the REM as part of a citizen science monitoring project, where participants collect data and researchers verify and implement the required analysis. Introduction Information about animal abundance and density, and how these are affected by biotic and/or abiotic factors, are important when developing management strategies and allocating conservation efforts (Fryxell et al. 2014). However, the range of methods available for estimating animal density is substantial (Williams et al. 2002), such that it can be a challenge to decide which method is best for specific species in different contexts. Ideally, the chosen method should be the one best suited to answering the research question, but factors such as accuracy,
Assessing precision and requirements of three methods to estimate roe deer density
PLOS ONE
Roe deer (Capreolus capreolus) is the most abundant cervid in Europe and, as such, has a considerable impact over several human activities. Accurate roe deer population size estimates are useful to ensure their proper management. We tested 3 methods for estimating roe deer abundance (drive counts, pellet-group counts, and camera trapping) during two consecutive years (2012 and 2013) in the Apennines (Italy) in order to assess their precision and applicability. During the study period, population density estimates were: drive counts 21.89±12.74 roe deer/km 2 and pellet-group counts 18.74±2.31 roe deer/km 2 in 2012; drive counts 19.32±11.12 roe deer/km 2 and camera trapping 29.05±7.48 roe deer/km 2 in 2013. Precision of the density estimates differed widely among the 3 methods, with coefficients of variation ranging from 12% (pellet-group counts) to 58% (drive counts). Drive counts represented the most demanding method on account of the higher number of operators involved. Pellet-group counts yielded the most precise results and required a smaller number of operators, though the sampling effort was considerable. When compared to the other two methods, camera trapping resulted in an intermediate level of precision and required the lowest sampling effort. We also discussed field protocols of each method, considering that volunteers, rather than technicians, will more likely be appointed for these tasks in the near future. For this reason, we strongly suggest that for each method managers of population density monitoring projects take into account ease of use as well as the quality of the results obtained and the resources required.
EFSA Supporting Publications
Hunting statistics can be suitable to determine wild boar density estimates if a calibration with an accepted rigorous method is performed. Here, densities calculated from drive counts during collective drive hunting activities are compared against density values calculated by camera trapping using the random encounter method. For this purpose, we selected 10 study sites in Spain, from North to South representing a diversity of habitats, management and hunting traditions without artificial feeding, plus one study site in Czech Republic where artificial feeding was practiced. Density values estimated from both drive counts and camera trapping were strongly positively correlated (R 2 =0.84 and 0.87 for linear and non-linear models, respectively) and showed a good agreement. Drive counts data might be therefore used as a density estimate to calibrate models for estimating density in large areas and potentially, to compare densities among areas. For these purposes, there is still the need to harmonise hunting data collection across Europe to make them usable at a large scale. Our results need to be confirmed across a wider number of European populations to provide valid geographical wild boar density predictions across Europe.
Large scale wildlife monitoring studies: statistical methods for design and analysis
Environmetrics, 2002
Techniques for estimation of absolute abundance of wildlife populations have received a lot of attention in recent years. The statistical research has been focused on intensive small-scale studies. Recently, however, wildlife biologists have desired to study populations of animals at very large scales for monitoring purposes. Population indices are widely used in these extensive monitoring programs because they are inexpensive compared to estimates of absolute abundance. A crucial underlying assumption is that the population index (C) is directly proportional to the population density (D). The proportionality constant, , is simply the probability of`detection' for animals in the survey. As spatial and temporal comparisons of indices are crucial, it is necessary to also assume that the probability of detection is constant over space and time. Biologists intuitively recognize this when they design rigid protocols for the studies where the indices are collected. Unfortunately, however, in many ®eld studies the assumption is clearly invalid. We believe that the estimation of detection probability should be built into the monitoring design through a double sampling approach. A large sample of points provides an abundance index, and a smaller sub-sample of the same points is used to estimate detection probability. There is an important need for statistical research on the design and analysis of these complex studies. Some basic concepts based on actual avian, amphibian, and ®sh monitoring studies are presented in this article.
Wildlife Research, 2020
Context. Precise and accurate estimates of animal numbers are often essential for population and epidemiological models, as well as for guidance for population management and conservation. This is particularly true for threatened species in landscapes facing multiple threats. Estimates can be derived by different methods, but the question remains as to whether these estimates are comparable. Aims. We compared three methods to estimate population numbers, namely, distance sampling, mark-recapture analysis, and home-range overlap analysis, for a population of the iconic threatened species, the koala (Phascolarctos cinereus). This population occupies a heavily fragmented forest and woodland habitat on the Liverpool Plains, northwestern New South Wales, Australia, on a mosaic of agricultural and mining lands. Key results. All three methods produced similar estimates, with overlapping confidence intervals. Distance sampling required less expertise and time and had less impact on animals, but also had less precision; however, future estimates using the method could be improved by increasing both the number and expertise of the observers. Conclusions. When less intrusive methods are preferred, or fewer specialised practitioners are available, we recommend distance sampling to obtain reliable estimates of koala numbers. Although its precision is lower with a low number of sightings, it does produce estimates of numbers similar to those from the other methods. However, combining multiple methods can be useful when other material (genetic, health and demographic) is also needed, or when decisions based on estimates are for high-profile threatened species requiring greater confidence. We recommend that all estimates of population numbers, and their precision or variation, be recorded and reported so that future studies can use them as prior information, increasing the precision of future surveys through Bayesian analyses.
Biodiversity and Conservation, 2012
Ecologists and conservation biologists seem increasingly attracted to sophisticated modelling approaches, sometimes at the expense of attention to data quality and appropriateness of fieldwork design. This dissociation may lead to a loss of perspective promoting biological unrealities as conclusions, which may be used in conservation applications. We illustrate this concern by focusing on recent attempts to estimate population size of breeding birds at large scales without any explicit testing of the reliability of the predictions through comparison with direct counts. Disconnection of analysts from ''nature'' can lead to cases of biological unrealities such as that used here to illustrate such trends. To counter this risk, we encourage investment in well-rounded scientists or more collaborative, multi-disciplinary teams capable of integrating sophisticated analyses with in-depth knowledge of the natural history of their study subjects.
How many are there? The use and misuse of continental-scale wildlife abundance estimates
Wildlife Research
The number of individuals in a wildlife population is often estimated and the estimates used for wildlife management. The scientific basis of published continental-scale estimates of individuals in Australia of feral cats and feral pigs is reviewed and contrasted with estimation of red kangaroo abundance and the usage of the estimates. We reviewed all papers on feral cats, feral pigs and red kangaroos found in a Web of Science search and in Australian Wildlife Research and Wildlife Research, and related Australian and overseas scientific and 'grey' literature. The estimated number of feral cats in Australia has often been repeated without rigorous evaluation of the origin of the estimate. We propose an origin. The number of feral pigs in Australia was estimated and since then has sometimes been quoted correctly and sometimes misquoted. In contrast, red kangaroo numbers in Australia have been estimated by more rigorous methods and the relevant literature demonstrates active refining and reviewing of estimation procedures and management usage. We propose four criteria for acceptable use of wildlife abundance estimates in wildlife management. The criteria are: use of appropriate statistical or mathematical analysis; precision estimated; original source cited; and age (current or out-of-date) of an estimate evaluated. The criteria are then used here to assess the strength of evidence of the abundance estimates and each has at least one deficiency (being out-of-date). We do know feral cats, feral pigs and red kangaroos occur in Australia but we do not know currently how many feral cats or feral pigs are in Australia. Our knowledge of red kangaroo abundance is stronger at the state than the continental scale, and is also out-of-date at the continental scale. We recommend greater consideration be given to whether abundance estimates at the continental scale are needed and to their use, and not misuse, in wildlife management.