Analysis Of Capture-recapture Data Research Articles

Semiparametric empirical likelihood inference for abundance from one-inflated capture-recapture data.

Abundance estimation from capture-recapture data is of great importance in many disciplines. Analysis of capture-recapture data is often complicated by the existence of one-inflation and heterogeneity problems. Simultaneously taking these issues into account, existing abundance estimation methods are usually constructed on the basis of conditional likelihood under one-inflated zero-truncated count models. However, the resulting Horvitz-Thompson-type estimators may be unstable, and the resulting Wald-type confidence intervals may exhibit severe undercoverage. In this paper, we propose a semiparametric empirical likelihood (EL) approach to abundance estimation under one-inflated binomial and Poisson regression models. To facilitate the computation of the EL method, we develop an expectation-maximization algorithm. We also propose a new score test for the existence of one-inflation and prove its asymptotic normality. Our simulation studies indicate that compared with existing estimators, the proposed score test is more powerful and the maximum EL estimator has a smaller mean square error. The advantages of our approaches are further demonstrated by analyses of prinia data from Hong Kong and drug user data fromBangkok.

A proposal for a goodness-of-fit test to the Arnason-Schwarz multisite capture-recapture model.

In an analysis of capture-recapture data, the identification of a model that fits is a critical step. For the multisite (also called multistate) models used to analyze data gathered at several sites, no reliable test for assessing fit is currently available. We propose a test for the JMV model, a simple generalization of the Arnason-Schwarz (AS) model, in the form of interpretable contingency tables. For the AS model, we suggest complementing the test for the JMV model with a likelihood ratio test of AS vs. JMV. The examination of an example leads us to propose further a partitioning that emphasizes the role of the memory model of Brownie et al. (1993 Biometrics 49, 1173-1187) as a biologically more plausible alternative to the AS model.

Analysis of capture-recapture data with a Rasch-type model allowing for conditional dependence and multidimensionality.

In this article, we show that, if subjects are assumed to be homogeneous within a finite set of latent classes, the basic restrictions of the Rasch model (conditional independence and unidimensionality) can be relaxed in a flexible way by simply adding appropriate columns to a basic design matrix. When discrete covariates are available so that subjects may be classified into strata, we show how a joint modeling approach can achieve greater parsimony. Parameter estimates may be obtained by maximizing the conditional likelihood (given the total number of captures) with a combined use of the EM and Fisher scoring algorithms. We also discuss a technique for obtaining confidence intervals for the size of the population under study based on the profile likelihood.

DEMOGRAPHY OF A DECLINING POPULATION OF WARBLING VIREOS IN COASTAL CALIFORNIA

Abstract We investigated population trends, adult survival, and productivity indices of a breeding population of Warbling Vireos (Vireo gilvus) in north coastal California using capture-recapture data from 19 years of standardized mist-netting conducted at Point Reyes National Seashore. We found significant long-term declines in abundance of both breeding and migratory Warbling Vireos. In addition, we found a significant long-term decline of hatching-year birds captured during the breeding season. The number of new adults in a given year, defined as those birds not previously banded and assumed to be an index of adult recruitment, was significantly correlated with the capture rate of hatching-year birds in the previous year, suggesting a link between reproductive success and adult abundance and recruitment. Our analysis of capture-recapture data provided an estimate of 50.4% for adult annual survival, similar to estimates for other species in stable populations. We could detect no temporal trend in surviv...

A closure test for time-specific capture-recapture data

The assumption of demographic closure in the analysis of capture-recapture data under closed-population models is of fundamental importance. Yet, little progress has been made in the development of omnibus tests of the closure assumption. We present a closure test for time-specific data that, in principle, tests the null hypothesis of closed-population model Mt against the open-population Jolly-Seber model as a specific alternative. This test is chi-square, and can be decomposed into informative components that can be interpreted to determine the nature of closure violations. The test is most sensitive to permanent emigration and least sensitive to temporary emigration, and is of intermediate sensitivity to permanent or temporary immigration. This test is a versatile tool for testing the assumption of demographic closure in the analysis of capture-recapture data.

Competing events, mixtures of information and multistratum recapture models

A recent trend in the analysis of capture-recapture data, in the broad sense, is the development of survival models mixing different types of events such as live recapture and dead recoveries. These models are similar to those developed for analysing competing events and multistate life histories in human health studies, which are frequently based on Markov chains. The major difference is that, in the capture-recapture context, every event is subject to a detection probability in general lower than one. Multistratum capture-recapture models are then a natural tool. We show that models for information mixtures easily enter this framework, provided adequate, possibly non-observable, strata are defined. Similarities and differences between such models are then easy to ascertain and understand. A multistratum model for mixtures of live recaptures and dead recoveries is presented and an example of data from the literature is treated, using a prototype MATLAB program named M-SURGE. Further generalizations and expected developments are discussed.

Survival estimates for the northern spotted owl

An analysis of capture-recapture data is tailored to the biological characteristics of the female northern spotted owl (Strix occidentalis caurina) by conditioning female capture on capture of paired males. This procedure is relevant for animals such as the northern spotted owl where field studies indicate that capture of females generally is dependent on capture of their mates. Application of the method to data collected in Washington State suggests that survival rates for adult females have remained constant over the time span of our data, 1990-1997.

Estimator Selection for Closed-Population Capture: Recapture

For valid statistical inference, it is important to select an appropriate statistical model. In the analysis of capture-recapture data under the closed-population models of Otis et al. (1978), information theoretic and hypothesis testing approaches to model selection are not practical, because some of the models have likelihoods with nonidentifiable parameters. A further problem is that, for some of the Otis et al. models, multiple estimators exist but there is no objective basis for deciding which estimator to use for a particular dataset. In CAPTURE, a computer program for estimating parameters under the closed models of Otis et al., a linear discriminant classifier is used to select an appropriate model. This classifier frequently selects the incorrect generating model in simulation studies, and it provides no guidance on which estimator to use once a model has been selected. In this study, we develop new classifiers for selecting the best estimator (as opposed to the generating model) and evaluate their performance. In addition, we investigate an estimator averaging approach to estimation that is a modification of the model averaging approach described by Buckland et al. (1997). We found that, in general, the overall performance of the new classifiers was unimpressive. In contrast, the estimator averaging approach we investigated performed well.

Model Selection Strategy in the Analysis of Capture-Recapture Data

Analysis of capture-recapture data is critically dependent upon selection of a proper for inference. Model selection is particularly important in the analysis of multiple, interrelated data sets. This paper evaluates information theoretic approaches to selection of a parsimonious and compares them to the use of likelihood ratio tests using four a levels. The purpose of the evaluation is to compare selection strategies based on the quality of the inference, rather than on the degree to which differing selection strategies select the true model. A measure of squared bias and variance (termed RSS) is used as a basis for comparing different data-based selection strategies, assuming that a minimum RSS value is a reasonable target. In general, the information theoretic approaches consistently selected models with a smaller RSS than did the likelihood ratio testing approach. Two information theoretic criteria have a balance between underfitting and overfitting when compared to models where the average minimum RSS was known. Other findings are presented along with a discussion of the concept of a true model and dimension consistency in selection.

The structure of a natural population of the medicinal leech, Hirudo medicinalis, at Dungeness, Kent

SUMMARY 1. A hot branding method was used to mark medicinal leeches (Hirudo medicinalis L.) from five samples (four in 1984, one in 1985) at agravel pit at Dungeness in Kent. Analysis of capture-recapture data by the Fisher-Ford and Jolly methods gave similar results and indicated a population in excess of 10000 (0.112 leeches per m2) in 1984 and 1985. Successive estimates were each higher than the last suggesting a thriving community. 2. The population structure was skewed towards the lower weight classes. This may indicate good recruitment of hatchlings, as also indicated by the presence of cocoons, or the failure to capture heavier satiated or gravid leeches because of their relative unresponsiveness to the sampling method involving water disturbance, or their complete absence from the water during cocoon laying. Gravid leeches were always above 2 g in weight. Recaptured marked leeches revealed that they migrated throughout the site, indicating active foraging for hosts.

Erlangian models for capture-recapture data

Models for the analysis of capture-recapture data based on Erlangian distributions reflecting assumptions about the times at which individuals enter and leave the population are described. A procedure for fitting such models by the method of maximum likelihood is illustrated.

Recent Models for Mark-Recapture and Mark-Resighting Data

A topic of current interest in the analysis of capture-recapture data is the development of models that allow biologically meaningful constraints on survival and capture rates. Of particular interest are models in which survival is a function of environmental variables. Thus, in an important recent paper, Clobert and Lebreton (1985) described analysis of data from resighting of marked birds, where survival was modelled as a function of winter temperature. Clobert and Lebreton (1985) have, however, developed their methodology (including necessary computer routines) in the context of data from resighting samples only (thus assuming no losses on capture), whereas with appropriate interpretation of formulae their methods can be applied to recapture data that frequently include losses on capture. The purpose of this note is to indicate the modifications needed to make the Clobert-Lebreton methods applicable to recapture data. Also, to prevent future duplication of effort, relationships between models that have been developed separately for resighting data and recapture data are noted. Procedures in Clobert and Lebreton (1985) and Clobert et al. (1985) are based on the model for estimating time-specific survival and sighting rates from resighting data, proposed by Cormack (1964). The year after Cormack's model appeared, Jolly (1965) and Seber (1965) described what is now called the Jolly-Seber model for estimating survival rates (and population size) from recapture data, allowing for the possibility that physical recapture may result in loss of some animals. With a view to achieving parsimony and improving estimator precision, parallel efforts (including development of necessary computer routines) have since specialized the original Cormack and Jolly-Seber models to cases where survival and/or recapture (resighting) rates are assumed constant. Thus, Sandland and Kirkwood (1981) and Clobert et al. (1985) describe reduced-parameter versions of Cormack's (1964) model, and Jolly (1982) and Brownie, Hines, and Nichols (1986) describe analogous reduced-parameter versions of the Jolly-Seber model. Of the corresponding computer programs, the Clobert-Lebreton program (Clobert et al., 1985; Clobert and Lebreton, 1985) permits the greatest flexibility for modelling survival and capture rates. [For other somewhat parallel efforts, not immediately relevant to the discussion here, see e.g., Cormack (1981) and Crosbie and Manly (1985).] The Cormack and Jolly-Seber models differ in that new releases are viewed as uninformative constants in the former, and as random variables (providing information about population size) in the latter. The models yield equivalent formulae for obtaining estimates of survival and recapture rates, but lead to different interpretations of expectations in variance formulae (Seber, 1982; Brownie et al., 1986). When there are no losses on capture, the Cormack and Jolly-Seber models in fact produce identical point and interval estimates for survival and capture rates (Seber, 1982, pp. 214-215; Brownie and Robson, 1983). Reduced-parameter versions of the Cormack and Jolly-Seber models, which make the same assumptions about survival and capture rates, should yield numerically identical

Population Dynamics of Molting Pintail Drakes Banded in South-Central Saskatchewan

A total of 18,820 adult male pintails (Anas acuta) were trapped, banded, and released on Pel and Kutawagan marshes in south-central Saskatchewan in July 1955-58. Two hundred and fortyeight of these banded birds were recaptured on the same marshes during subsequent trapping operations, and 1,440 were recovered (shot or found dead) during legal hunting seasons, 1955-70. The size of the adult drake pintail population associated with Pel and Kutawagan marshes was estimated to have been approximately 313,600 in July 1956 and 232,800 in July 1957. Estimates of annual survival varied from approximately 60 percent to 85 percent with an average of 73 percent. Between July 1956 and July 1957, approximately 71,100 birds joined the population segment associated with Pel and Kutawagan marshes and 25,400 birds left the population segment for other areas. Indices to harvest pressure were low, varying from 2.26 in 1958 to 3.54 in 1956. Estimates of harvest rates varied from 5.6 percent to 8.8 percent, and kill rates varied from approximately 7.0 to 11.0 percent. The index to harvest pressure was highest in 1956, while survival during the 1956-57 period was the lowest, suggesting the possibility of an inverse relationship between harvest rate and total survival. J. WILDL. MANAGE. 38(2):266-274 Little quantitative information has been published concerning population dynamics of the pintail in North America. Early studies by Cartwright (1945), Munro (1944), and Van Den Akker and Wilson (1949) provided basic information on the pintail. Low (1949, 1957) summarized the geographic distribution of the pintail, and Lensink (1964) presented the results of the extensive banding program of flightless young. Recently, Henny (1973) discussed band recoveries of pintails in Russia and Asia and extended Smith's (1970) paper on the relationship between drought conditions in prairie Canada and northward migration in pintails. Pospahala et al. (Unpublished manuscript) have presented the results of extensive aerial surveys of breeding pintail populations during the 1955-71 period and updated the information presented annually in the Waterfowl Status Reports. The pintail is the third most abundant duck in North America, and since 1968 an average of 1.6 million have been harvested annually in the United States. This paper presents an analysis of a large banding experiment conducted on Pel and Kutawagan marshes in south-central Saskatchewan under the supervision of the junior author during July 1955-58. The banding was done to estimate basic population parameters and exploitation statistics of molting pintails in south-central Saskatchewan. The preliminary results of the experiment, including information on the distribution of band recoveries and homing aspects of molting pintails, were presented by Sterling (1966) for the 1955-63 period. Since then, additional recoveries have become available for study, and superior analytical methods have been developed for the analysis of capture-recapture data (Anderson 1972; Cormack 1968; Seber 1972, 1973). The senior author performed the analysis of data and led the report-writing phase of the investigation. We acknowledge J. B. Gollop, Canadian Wildlife Service, for his assistance during the course of the study and for the review of an early draft of the manuscript. A. D. Carothers, University of Edinburgh, revi wed the manuscript and made a number of suggestions. R. S. Miller, formerly at 266 J. Wildl. Manage. 38 (2):1974 This content downloaded from 157.55.39.192 on Wed, 24 Aug 2016 05:03:50 UTC All use subject to http://about.jstor.org/terms POPULATION DYNAMICS OF PINTAILS * Anderson and Sterling 267 the University of Saskatchewan, provided encouragement during part of the research. The field work was sponsored by Ducks Unlimited (Canada). J. R. Caldwell and G. H. Staines, both formerly of Ducks Unlimited (Canada), assisted with the field work. The banding and recovery data were processed and maintained by the U.S. Fish and Wildlife Service, and a computer summary of the recovery data was supplied by the Canadian Wildlife Service.

Electronic Data Processing of Capture-Recapture and Related Ecological Data

The disadvantages of ecological and wildlife management data processing and analysis by hand and with desk calculators are discussed along with the advantages of electronic data processing. A standard field form is suggested for capture-recapture observations which will facilitate the transfer of data to permanent punched-card storage. A time-cost comparison of manual and electronic processing of these data is presented which shows savings by electronic methods. Suggestions are made for additional use of electronic data processing in ecological studies. Limitations are discussed. Data processing for definitive studies of animals and their environments involves time-consuming, tedious operations of sorting, tabulation, and computation. This paper discusses the advantages of electronic data processing for ecological studies, presents an efficient generalized field record form for capture-recapture studies, and describes the use of a program for simultaneous analysis of capture-recapture data by three methods of population estimation. PROBLEMS OF DATA COLLECTION