An empirical demonstration of the effect of study design on density estimations

Abstract The simultaneous development of technology (e.g. camera traps) and statistical methods, particularly spatially capture–recapture (SCR), has improved monitoring of large mammals in recent years. SCR estimates are known to be sensitive to sampling design, yet existing recommendations about tr...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Muhammad Ali Nawaz, Barkat Ullah Khan, Amer Mahmood, Muhammad Younas, Jaffar ud Din, Chris Sutherland
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/7f8c4fcd1a754f6d959e88ce27e4705c
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:7f8c4fcd1a754f6d959e88ce27e4705c
record_format	dspace
spelling	oai:doaj.org-article:7f8c4fcd1a754f6d959e88ce27e4705c2021-12-02T17:14:29ZAn empirical demonstration of the effect of study design on density estimations10.1038/s41598-021-92361-22045-2322https://doaj.org/article/7f8c4fcd1a754f6d959e88ce27e4705c2021-06-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-92361-2https://doaj.org/toc/2045-2322Abstract The simultaneous development of technology (e.g. camera traps) and statistical methods, particularly spatially capture–recapture (SCR), has improved monitoring of large mammals in recent years. SCR estimates are known to be sensitive to sampling design, yet existing recommendations about trap spacing and coverage are often not achieved, particularly for sampling wide-ranging and rare species in landscapes that allow for limited accessibility. Consequently, most camera trap studies on large wide-ranging carnivores relies on convenience or judgmental sampling, and often yields compromised results. This study attempts to highlight the importance of carefully considered sampling design for large carnivores that, because of low densities and elusive behavior, are challenging to monitor. As a motivating example, we use two years of snow leopard camera trapping data from the same areas in the high mountains of Pakistan but with vastly different camera configurations, to demonstrate that estimates of density and space use are indeed sensitive to the trapping array. A compact design, one in which cameras were placed much closer together than generally recommended and therefore have lower spatial coverage, resulted in fewer individuals observed, but more recaptures, and estimates of density and space use were inconsistent with expectations for the region. In contrast, a diffuse design, one with larger spacing and spatial coverage and more consistent with general recommendations, detected more individuals, had fewer recaptures, but generated estimates of density and space use that were in line with expectations. Researchers often opt for compact camera configurations while monitoring wide-ranging and rare species, in an attempt to maximize the encounter probabilities. We empirically demonstrate the potential for biases when sampling a small area approximately the size of a single home range—this arises from exposing fewer individuals than deemed sufficient for estimation. The smaller trapping array may also underestimate density by significantly inflating $$\sigma$$ σ . On the other hand, larger trapping array with fewer detectors and poor design induces uncertainties in the estimates. We conclude that existing design recommendations have limited utility on practical grounds for devising feasible sampling designs for large ranging species, and more research on SCR designs is required that allows for integrating biological and habitat traits of large carnivores in sampling framework. We also suggest that caution should be exercised when there is a reliance on convenience sampling.Muhammad Ali NawazBarkat Ullah KhanAmer MahmoodMuhammad YounasJaffar ud DinChris SutherlandNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-9 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Muhammad Ali Nawaz Barkat Ullah Khan Amer Mahmood Muhammad Younas Jaffar ud Din Chris Sutherland An empirical demonstration of the effect of study design on density estimations
description	Abstract The simultaneous development of technology (e.g. camera traps) and statistical methods, particularly spatially capture–recapture (SCR), has improved monitoring of large mammals in recent years. SCR estimates are known to be sensitive to sampling design, yet existing recommendations about trap spacing and coverage are often not achieved, particularly for sampling wide-ranging and rare species in landscapes that allow for limited accessibility. Consequently, most camera trap studies on large wide-ranging carnivores relies on convenience or judgmental sampling, and often yields compromised results. This study attempts to highlight the importance of carefully considered sampling design for large carnivores that, because of low densities and elusive behavior, are challenging to monitor. As a motivating example, we use two years of snow leopard camera trapping data from the same areas in the high mountains of Pakistan but with vastly different camera configurations, to demonstrate that estimates of density and space use are indeed sensitive to the trapping array. A compact design, one in which cameras were placed much closer together than generally recommended and therefore have lower spatial coverage, resulted in fewer individuals observed, but more recaptures, and estimates of density and space use were inconsistent with expectations for the region. In contrast, a diffuse design, one with larger spacing and spatial coverage and more consistent with general recommendations, detected more individuals, had fewer recaptures, but generated estimates of density and space use that were in line with expectations. Researchers often opt for compact camera configurations while monitoring wide-ranging and rare species, in an attempt to maximize the encounter probabilities. We empirically demonstrate the potential for biases when sampling a small area approximately the size of a single home range—this arises from exposing fewer individuals than deemed sufficient for estimation. The smaller trapping array may also underestimate density by significantly inflating $$\sigma$$ σ . On the other hand, larger trapping array with fewer detectors and poor design induces uncertainties in the estimates. We conclude that existing design recommendations have limited utility on practical grounds for devising feasible sampling designs for large ranging species, and more research on SCR designs is required that allows for integrating biological and habitat traits of large carnivores in sampling framework. We also suggest that caution should be exercised when there is a reliance on convenience sampling.
format	article
author	Muhammad Ali Nawaz Barkat Ullah Khan Amer Mahmood Muhammad Younas Jaffar ud Din Chris Sutherland
author_facet	Muhammad Ali Nawaz Barkat Ullah Khan Amer Mahmood Muhammad Younas Jaffar ud Din Chris Sutherland
author_sort	Muhammad Ali Nawaz
title	An empirical demonstration of the effect of study design on density estimations
title_short	An empirical demonstration of the effect of study design on density estimations
title_full	An empirical demonstration of the effect of study design on density estimations
title_fullStr	An empirical demonstration of the effect of study design on density estimations
title_full_unstemmed	An empirical demonstration of the effect of study design on density estimations
title_sort	empirical demonstration of the effect of study design on density estimations
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/7f8c4fcd1a754f6d959e88ce27e4705c
work_keys_str_mv	AT muhammadalinawaz anempiricaldemonstrationoftheeffectofstudydesignondensityestimations AT barkatullahkhan anempiricaldemonstrationoftheeffectofstudydesignondensityestimations AT amermahmood anempiricaldemonstrationoftheeffectofstudydesignondensityestimations AT muhammadyounas anempiricaldemonstrationoftheeffectofstudydesignondensityestimations AT jaffaruddin anempiricaldemonstrationoftheeffectofstudydesignondensityestimations AT chrissutherland anempiricaldemonstrationoftheeffectofstudydesignondensityestimations AT muhammadalinawaz empiricaldemonstrationoftheeffectofstudydesignondensityestimations AT barkatullahkhan empiricaldemonstrationoftheeffectofstudydesignondensityestimations AT amermahmood empiricaldemonstrationoftheeffectofstudydesignondensityestimations AT muhammadyounas empiricaldemonstrationoftheeffectofstudydesignondensityestimations AT jaffaruddin empiricaldemonstrationoftheeffectofstudydesignondensityestimations AT chrissutherland empiricaldemonstrationoftheeffectofstudydesignondensityestimations
_version_	1718381327945826304

An empirical demonstration of the effect of study design on density estimations

Ejemplares similares