ednaoccupancy: An r package for multiscale occupancy modelling of environmental DNA data

Molecular Ecology Resources, Journal Year: 2017, Volume and Issue: 18(2), P. 368 - 380

Published: Nov. 9, 2017

In this article, we describe ednaoccupancy, an r package for fitting Bayesian, multiscale occupancy models. These models are appropriate surveys that include three nested levels of sampling: primary sample units within a study area, secondary collected from each unit and replicates unit. This design is commonly used in environmental DNA (eDNA). ednaoccupancy allows users to specify fit with or without covariates, estimate posterior summaries occurrence detection probabilities, compare different using Bayesian model-selection criteria. We illustrate these features by analysing two published data sets: eDNA fungal pathogen amphibians endangered fish species.

Language: Английский

---

A comment on priors for Bayesian occupancy models

PLoS ONE, Journal Year: 2018, Volume and Issue: 13(2), P. e0192819 - e0192819

Published: Feb. 26, 2018

Understanding patterns of species occurrence and the processes underlying these is fundamental to study ecology. One more commonly used approaches investigate occupancy modeling, which can account for imperfect detection a during surveys. In recent years, there has been proliferation Bayesian modeling in ecology, includes fitting models. The framework appealing ecologists many reasons, including ability incorporate prior information through specification distributions on parameters. While almost exclusively intend choose priors so that they are "uninformative" or "vague", such easily be unintentionally highly informative. Here we report how "vague" normally distributed (i.e., Gaussian) coefficients models influence parameter estimation. Using both simulated data empirical examples, illustrate this issue likely compromises inference about species-habitat relationships. extent informative depends set, researchers should conduct sensitivity analyses ensure intended inference, employ less (e.g., logistic t distributions). We provide suggestions addressing studies, an online tool exploring under different contexts.

Language: Английский

---

The use of Bayesian priors in Ecology: The good, the bad and the not great

Methods in Ecology and Evolution, Journal Year: 2020, Volume and Issue: 11(8), P. 882 - 889

Published: May 18, 2020

Abstract Bayesian data analysis (BDA) is a powerful tool for making inference from ecological data, but its full potential has yet to be realized. Despite generally positive trajectory in research surrounding model development and assessment, far too little attention been given prior specification. Default priors, sub‐class of non‐informative distributions that are often chosen without critical thought or evaluation, commonly used practice. We believe the fear being ‘subjective’ prevented many researchers using any information their analyses despite fact defending choice (informative not) promotes good statistical In this commentary, we provide an overview how BDA currently random sample articles, discuss implications if current bad practices continue, highlight sub‐fields where knowledge about system improved promoted through careful justified use informative priors. hope inspire renewed discussion priors Ecology with particular paid specification justification. also emphasize all result subjective choice, should discussed way.

Language: Английский

---

Integrated community models: A framework combining multispecies data sources to estimate the status, trends and dynamics of biodiversity

Journal of Animal Ecology, Journal Year: 2023, Volume and Issue: 92(12), P. 2248 - 2262

Published: Oct. 25, 2023

Abstract Data deficiencies among rare or cryptic species preclude assessment of community‐level processes using many existing approaches, limiting our understanding the trends and stressors for large numbers species. Yet evaluating dynamics whole communities, not just common charismatic species, is critical to responses biodiversity ongoing environmental pressures. A recent surge in both public science government‐funded data collection efforts has led a wealth data. However, these programmes use wide range sampling protocols (from unstructured, opportunistic observations wildlife well‐structured, design‐based programmes) record information at variety spatiotemporal scales. As result, available vary substantially quantity content, which must be carefully reconciled meaningful ecological analysis. Hierarchical modelling, including single‐species integrated models hierarchical community models, improved ability assess predict processes. Here, we highlight emerging ‘integrated modelling’ framework that combines integration modelling improve inferences on species‐ dynamics. We illustrate with series worked examples. Our three case studies demonstrate how can used extend geographic scope when distributions richness patterns; discern population over time; estimate demographic rates growth communities sympatric implemented examples multiple software methods through R platform via packages formula‐based interfaces development custom code JAGS, NIMBLE Stan. Integrated provide an exciting approach model biological observational types sources simultaneously, thus accounting uncertainty error within unified framework. By leveraging combined benefits produce valuable about as well dynamics, allowing holistic evaluation effects global change biodiversity.

Language: Английский

---

Genetic rescue often leads to higher fitness as a result of increased heterozygosity across animal taxa

Molecular Ecology, Journal Year: 2024, Volume and Issue: 33(19)

Published: Sept. 16, 2024

Abstract Biodiversity loss has reached critical levels partly due to anthropogenic habitat and degradation. These landscape changes are damaging as they can fragment species distributions into small, isolated populations, resulting in limited gene flow, population declines reduced adaptive potential. Genetic rescue, the translocation of individuals increase genetic diversity ultimately fitness, produced promising results for fragmented populations but remains underutilized a lack long‐term data monitoring. To promote better understanding rescue its potential risks benefits over short‐term, we reviewed analysed published attempts identify whether increases following translocation, if this change is associated with increased fitness. Our review identified 19 studies that provided fitness from before after translocation; majority these were on mammals, included experimental, natural conservation‐motivated translocations. Using Bayesian meta‐analytical approach, found average, post translocations, although there some exceptions trend. Overall, was positive predictor cases relationship extended three generations post‐rescue. suggest single have lasting benefits, support another tool facilitate conservation success. Given number data, echo need monitoring post‐translocation understand also limit long‐term.

Language: Английский

---

Evidential statistics as a statistical modern synthesis to support 21st century science

Population Ecology, Journal Year: 2015, Volume and Issue: 58(1), P. 9 - 29

Published: Dec. 24, 2015

Abstract During the 20th century, population ecology and science in general relied on two very different statistical paradigms to solve its inferential problems: error statistics (also referred as classical frequentist statistics) Bayesian statistics. A great deal of good was done using these tools, but both schools suffer from technical philosophical difficulties. At turning 21st century (Royall Statistical evidence: a likelihood paradigm. Chapman & Hall, London, 1997 ; Lele The nature scientific statistical, empirical considerations. University Chicago Press, Chicago, pp 191–216, 2004a ), evidential emerged seriously contending Drawing refining elements statistics, likelihoodism, information criteria, robust methods, is modern synthesis that smoothly incorporates model identification, uncertainty, comparison, parameter estimation, pre‐data control error, post‐data strength evidence into single coherent framework. We argue currently most effective paradigm support science. Despite power paradigm, we think there no substitute for learning how clarify arguments with arguments. In this paper sketch relate conceptual bases also discuss number misconceptions about have hindered practitioners, well some real problems solved by

Language: Английский

---

An integrated data model to estimate spatiotemporal occupancy, abundance, and colonization dynamics

Ecology, Journal Year: 2016, Volume and Issue: 98(2), P. 328 - 336

Published: Nov. 5, 2016

Ecological invasions and colonizations occur dynamically through space time. Estimating the distribution abundance of colonizing species is critical for efficient management or conservation. We describe a statistical framework simultaneously estimating spatiotemporal occupancy dynamics species. Our method accounts several issues that are common when modeling ecological data including multiple levels detection probability, sources, computational limitations making fine-scale inference over large domain. apply model to estimate colonization sea otters (Enhydra lutris) in Glacier Bay, southeastern Alaska.

Language: Английский

---

An efficient extension of N‐mixture models for multi‐species abundance estimation

Methods in Ecology and Evolution, Journal Year: 2017, Volume and Issue: 9(2), P. 340 - 353

Published: July 24, 2017

In this study we propose an extension of the N-mixture family models that targets improvement statistical properties rare species abundance estimators when sample sizes are low, yet typical for tropical studies. The proposed method harnesses information from other in ecological community to correct each species' estimator. We provide guidance determine size required estimate accurately attempting single species.We evaluate methods using assumption 50 m radius plots and perform simulations comprising a broad range sizes, true abundances detectability values complex data generating process. model is achieved by assuming detection probabilities drawn at random beta distribution multi-species fashion. This hierarchical avoids having specify probability parameter per targeted community. Parameter estimation done via Maximum Likelihood.We compared our approach with previously models, which show biased densities less than seven individuals 100 hectares. here outperforms traditional Multi-species allowing organisms lower controlling bias estimation.We illustrate how methodology can be used suggest organisms, these either rare, common or abundant. When interest full communities, approaches, particular methodology, as practical solution organism rapid inventory datasets. inferences Likelihood also group according their detectabilities.

Language: Английский

---

Ensuring identifiability in hierarchical mixed effects Bayesian models

Ecological Applications, Journal Year: 2020, Volume and Issue: 30(7)

Published: May 4, 2020

Abstract Ecologists are increasingly familiar with Bayesian statistical modeling and its associated Markov chain Monte Carlo (MCMC) methodology to infer about or discover interesting effects in data. The complexity of ecological data often suggests implementation (statistical) models a commensurately rich structure effects, including crossed nested (i.e., hierarchical multi‐level) structures fixed and/or random effects. Yet, our experience that most ecologists not subtle but important problems arise such their popular software. Of foremost consideration for us is the notion effect identifiability, which generally concerns how well data, models, approaches inform about, i.e., identify, quantities interest. In this paper, we focus on pitfalls potentially misinform subsequent inference, despite otherwise informative models. We illustrate aforementioned issues using regressions synthetic show diagnose identifiability remediate these model reparameterization computational coding practices software, JAGS, OpenBUGS, Stan. also solutions can be extended more complex involving multiple groups nested, crossed, additive, multiplicative Finally, provide example code (JAGS/OpenBUGS Stan) practitioners modify use own applications.

Language: Английский

---

Estimation and analysis of insect population dynamics parameters via physiologically based models and hybrid genetic algorithm MCMC methods

Ecological Informatics, Journal Year: 2023, Volume and Issue: 77, P. 102232 - 102232

Published: July 26, 2023

Decision support systems are gaining importance in several fields of agriculture, forest, and ecological management. Their predictive potential, entrusted to mathematical models, is fundamental set up opportune strategies control pests adversities that may occur seriously compromise the natural equilibria. Among others, population dynamics one crucial challenges field. Despite scientific community recent years providing valuable models faithfully represent terrestrial arthropods populations, such as insects, main concerns still represented by parameter estimation. Parameters, fact, characterise species their estimation often dedicated laboratory experiments require specific equipment highly qualified personnel. In this study we propose a novel method estimate model parameters directly from field data, where experimental activities less expensive time consuming. combination least squares methods via genetic algorithms preliminary evaluate best values Markov Chain Monte Carlo approach obtain distribution. The algorithm has been tested special case Drosophila suzukii, quantify part an almost validated two steps: i) first pseudo-validation using perturbed numerical solutions, ii) validation real data. results highlighted potentialities estimating opened perspectives for further improvements both computational point view.

Language: Английский

---