Integrating acoustic telemetry into mark–recapture models to improve the precision of apparent survival and abundance estimates (original) (raw)
Related papers
TheRcapturePackage: Loglinear Models for Capture-Recapture inR
Journal of Statistical Software, 2007
This article introduces Rcapture, an R package for capture-recapture experiments. The data for analysis consists of the frequencies of the observable capture histories over the t capture occasions of the experiment. A capture history is a vector of zeros and ones where one stands for a capture and zero for a miss. Rcapture can fit three types of models. With a closed population model, the goal of the analysis is to estimate the size N of the population which is assumed to be constant throughout the experiment. The estimator depends on the way in which the capture probabilities of the animals vary. Rcapture features several models for these capture probabilities that lead to different estimators for N. In an open population model, immigration and death occur between sampling periods. The estimation of survival rates is of primary interest. Rcapture can fit the basic Cormack-Jolly-Seber and Jolly-Seber model to such data. The third type of models fitted by Rcapture are robust design models. It features two levels of sampling; closed population models apply within primary periods and an open population model applies between periods. Most models in Rcapture have a loglinear form; they are fitted by carrying out a Poisson regression with the R function glm. Estimates of the demographic parameters of interest are derived from the loglinear parameter estimates; their variances are obtained by linearization. The novel feature of this package is the provision of several new options for modeling capture probabilities heterogeneity between animals in both closed population models and the primary periods of a robust design. It also implements many of the techniques developed by R. M. Cormack for open population models.
Fisheries Research, 2019
Spatiotemporal variability in fishing patterns and species distributions from ocean climatology confound analysis of pelagic longline CPUE. A generalized computer program, LLSIM, was developed to simulate such data to test methods used to quantify abundance trends. The method employs a Monte-Carlo algorithm with a probability of capture computed for each hook on each set based on overlaps of the species-and hook-depth distributions. The method was tested using characteristics of the longline gears and fishing locations of the US pelagic longline fleet and blue marlin latitude-longitude-depth distributions predicted using a species distribution model fitted to 1986-2012 monthly oceanographic data. Catch data were simulated for two hypothetical trends in total abundance. The simulator was capable of performing complex but controlled simulation experiments. Analyses demonstrated the advantage of comparing estimates from alternative standardizations to known true values that are not possible with real data.
We introduce a new method that uses generalized linear mixed models to infer the depth distribution of pelagic fishes. It uses existing data from research surveys and observers on commercial vessels to estimate changes in catchability when longline fishing gear is lengthened to access deeper water. We infer the depth distribution of catchability for 37 fish species that are caught on pelagic longlines in the Pacific Ocean. We show how the estimates of catchability can be used to correct abundance indices for variations in longline depth. Our method facilitates the inclusion of data from early surveys in the time series of commercial catch rates used to estimate abundance. It also resolves inconsistencies in the time series caused by a rapid switch to deep longlining in the 1970s. The catchability distribution does not always match depth preferences derived from tracking studies. Therefore, depth preferences from tracking studies should not be used to correct abundance indices without additional information on feeding behavior. Résumé : Nous présentons une nouvelle méthode qui utilise des modèles linéaires généralisés mixtes pour estimer la répartition des poissons pélagiques en fonction de la profondeur. La méthode exploite les données existantes d'inventaires scientifiques et d'observations faites sur les navires commerciaux afin d'estimer les changements de capturabilité qui se produisent lorsqu'on allonge les palangres pour pêcher en eau plus profonde. Nous estimons la répartition de la capturabilité en fonction de la profondeur chez 37 espèces de poissons récoltés à la palangre pélagique dans le Pacifique. Nous démontrons comment les estimations de capturabilité peuvent servir à corriger les indices d'abondance en fonction des variations de la profondeur des palangres. Notre méthode facilite l'inclusion de données d'inventaires plus anciens dans la série chronologique de taux de capture commerciaux utilisée pour estimer l'abondance. Elle permet aussi de résoudre les irrégularités dans la série chronologique causées par un passage rapide à la pêche à la palangre en profondeur durant les années 1970. La répartition de la capturabilité ne correspond pas toujours aux préférences de profondeur déterminées par les études qui traquent les poissons; il ne faut donc pas utiliser les préférences de profondeurs obtenues de ces études pour corriger les indices d'abondance s'il n'existe pas de rensei-gnements supplémentaires sur le comportement alimentaire.
Accurate assessments of population parameters, such as survival and abundance, are critical for effective wildlife conservation. In order for wildlife managers to undertake long-term monitoring of populations, the data collection must be as costeffective as possible. Two demographic modelling techniques commonly used are markrecapture and mark-resight. Mark-resight can be used in conjunction with biotelemetry methods and offers a more cost effective alternative to the traditional mark-recapture models. However, there has been no empirical comparison of the demographic parameters obtained from the two modelling techniques. This study used photographs of natural markings to individually identify wobbegong sharks (Orectolobus maculatus) sighted during underwater surveys over a 2 year period, during eight distinct sampling periods, and analysed with Pollock's robust design mark-recapture models. These estimates were then Communicated by Simon Ingram.
Fisheries Research, 2010
We estimate recent (1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005) trends in relative abundance for Northwest Atlantic oceanic and large coastal sharks, using generalized linear mixed models to standardize catch rates of eight species groups as recorded by U.S. pelagic longline fishery observers. Models suggest precipitous (76%) declines in hammerhead (Sphyrna species) and large coastal (dusky, night, and silky shark, genus Carcharhinus) species, and moderate declines (53%) in blue and oceanic whitetip sharks over this period. In contrast, mako and thresher sharks appear to have stabilized, and the tiger shark population appears to be increasing. A comparison of nominal shark catch rates from this fleet's observer and logbook data (to evaluate the veracity of trends previously estimated from the latter) showed a high degree of concordance for each species group, both in individual sub-areas and overall. Models of these two datasets for the common time period (1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000) show that compared to the observer data the logbook data indicate greater declines for some species, but lesser declines for others. Signs of recovery for some shark species are encouraging, but must also be set in the context of the significant declines that occurred in previous decades.
Cormack–Jolly–Seber models: time and age perspectives
Stochastic Environmental Research and Risk Assessment, 2020
Survival is a key demographic characteristic in many areas including both human demography and population ecology. However, it is often the case that data collection protocols are different in these areas, resulting in different models and methods of analysis. This paper is motivated for the different emphasis given to the elicitation of the temporal scale (and consequently, on the origin time) in ecological and medical survival studies. Specifically, in medical studies, the origin time is often determined in advance with individuals followed over a period of time at regular (or irregular) intervals, thus focusing on time within study (or age to a given reference point). However, in ecological capture-recapture studies, the capture occasions are typically fixed in advance, with an imperfect detection process observing individuals at these times. Moreover, the temporal scale is often primarily specified at the capture occasion level. In this work we focus on an ecological capture-recapture study related to guillemots and compare and contrast two different temporal scales: (i) calendar (or capture occasion); and (ii) age (or time within study), in terms of the way the data may be represented and in relation to the ecological Cormack-Jolly-Seber-type model. The different temporal scales provides insights into the different underlying structures, which can then be combined into a joint (calendar and age) dependence model.
Canadian Journal of Fisheries and Aquatic Sciences, 2013
One key issue for standardizing catch per unit effort (CPUE) of bycatch species is how to model observations of zero catch per fishing operation. Typically, the fraction of zero catches is high, and catch counts may be overdispersed. In this study, we develop a model selection and multimodel inference approach to standardize CPUE in a case study of oceanic whitetip shark (Carcharhinus longimanus) bycatch in the Hawaii-based pelagic longline fishery. Alternative hypotheses for shark catch per longline set were characterized by the variance to mean ratio of the count distribution. Zero-inflated and non-inflated Poisson, negative binomial, and delta-gamma models were fit to fishery observer data using stepwise variable selection. Alternative hypotheses were compared using multimodel inference. Results from the best-fitting zero-inflated negative binomial model showed that standardized CPUE of oceanic whitetip sharks decreased by about 90% during 1995-2010 because of increased zero catch sets and decreased CPUE on sets with positive catch. Our model selection approach provides an objective way to address the question of how to treat zero catches when analyzing bycatch CPUE.
Estimation of g(0) in line-transect surveys of cetaceans
2004
Estimating g(0)-the probability that an object that is on the line is detected, is crucial for any study on abundance and distribution using standard line-transect methods. In cetaceans, g(0) is usually <1, since whales and dolphins are submerged most of the time and are therefore unavailable for visual detection. Furthermore, species such as minke whales or harbour porpoises are inconspicuous and can easily be missed by observers. It is difficult and challenging to estimate g(0) precisely. Most approaches involve the analysis of data obtained from independent platforms. There are also new approaches, for example combinations of acoustic and visual surveys, which might provide exciting prospects for future studies. The aim of the workshop was to exchange ideas and share experiences about how to estimate g(0) in line-transect surveys of cetaceans. Our intention was that those who have worked with the issue could see what other teams are doing, while those who are planning new surveys would be able to receive inspiration and ideas. The workshop took place on Sunday, 28 th March 2004, at the Vildmarkshotellet of the Kolmårdens Djurpark, Kolmården, Sweden prior to the start of the 18 th Annual Conference of the European Cetacean Society. Around 80 persons from 14 countries attended the workshop (see list of participants at the end of this volume).