Abundance estimation of unmarked animals based on camera‐trap data

Conservation Biology, Journal Year: 2020, Volume and Issue: 35(1), P. 88 - 100

Published: April 16, 2020

The rapid improvement of camera traps in recent decades has revolutionized biodiversity monitoring. Despite clear applications conservation science, have seldom been used to model the abundance unmarked animal populations. We sought summarize challenges facing estimation animals, compile an overview existing analytical frameworks, and provide guidance for practitioners seeking a suitable method. When records multiple detections animal, one cannot determine whether images represent mobile individuals or single individual repeatedly entering viewshed. Furthermore, movement obfuscates definition sampling area and, as result, which estimate corresponds. Recognizing these challenges, we identified 6 approaches reviewed 927 camera-trap studies published from 2014 2019 assess use prevalence each Only about 5% any abundance-estimation methods identified. Most estimated local covariate relationships rather than predicting density over broader areas. Next, approach, compiled data requirements, assumptions, advantages, disadvantages help navigate landscape methods. appropriate method, should evaluate life history focal taxa, carefully define frame, consider what types collection are possible. challenge estimating populations persists; although exist, no method is optimal under all circumstances. As frameworks continue evolve animals becomes increasingly common, will become even more important informing decision-making.Estimación de la Abundancia Animales No Marcados con Base en Datos Cámaras Trampa Resumen La rápida mejoría las cámaras trampa décadas recientes ha revolucionado el monitoreo biodiversidad. A pesar su clara aplicación ciencias conservación, han sido utilizadas pocas veces para modelar abundancia poblaciones animales marcados. Buscamos resumir los retos que enfrenta estimación marcados, compilar una perspectiva general marcos analíticos trabajo existentes y proporcionar guía aquellos practicantes buscan un método adecuado. Cuando cámara registra múltiples detecciones se puede determinar si imágenes representan diferentes individuos movimiento o solo individuo entra repetidamente zona visión cámara. Sumado esto, ofusca definición del área muestreo y, como resultado, cual corresponde estimado abundancia. Después reconocer estos retos, identificamos seis estrategias analíticas revisamos estudios publicados entre evaluar uso prevalencia cada método. Solamente usó cualquiera métodos identificamos. mayoría estimaron relaciones covarianza lugar predecir densidad lo largo áreas más amplias. Después, estrategia analítica, recopilamos requerimientos datos, suposiciones, ventajas desventajas ayudar navegar paisaje busquen apropiado deberán historia vida taxón focal, definir cuidadosamente marco considerar cuáles tipos recolección datos son posibles. El reto estimar marcados persiste; aunque existan muchos métodos, hay único óptimo cumpla todas circunstancias. Mientras sigan evolucionando sea vez común, serán todavía importantes informar toma decisiones conservación.近几十年来红外相机陷阱技术的快速发展已经彻底改变了生物多样性监测的现状。尽管红外相机陷阱法在动物保护科学中有明确的应用, 但它很少被用来模拟无标记动物的种群数量。本研究旨在总结无标记动物的丰度估计所面临的挑战, 总结现有的分析框架并为寻求合适方法的实践者提供指导意见。当红外相机多次记录到无标记的动物时, 人们无法确定这些图像代表的是多个个体还是一个重复进入相机拍摄范围的个体。此外, 动物的运动导致不能清晰地划定采样区域, 因此也模糊了所对应区域的丰度估计。面对这些挑战, 我们确定了六种分析方法, 并综述了 年至 年发表的 项红外相机陷阱研究, 以评估每种方法的使用情况和流行程度。结果发现, 只有约 的研究使用了至少一种我们确定的丰度估计方法。这些研究大多是估计局部丰度或协变量关系, 而不是预测更大范围内的动物丰度或密度。接下来, 我们总结了每种分析方法的数据需求、假设、优点和缺点, 以帮助实践者了解丰度估计方法的总体情况。实践者在寻找合适的方法时, 应评估研究所关注类群的生活史, 谨慎地确定采样范围, 并考虑可能收集到的数据类型。无标记动物的种群数量估计仍面临挑战, 虽然已存在多种方法, 但没有一种方法对于所有红外相机陷阱数据都是最优的。随着分析框架的不断发展和对无标记动物数量估计变得越来越普遍, 红外相机陷阱法在为指导保护决策中也将更加重要。【翻译: 胡怡思; 审校: 聂永刚】.

---

A review of factors to consider when using camera traps to study animal behavior to inform wildlife ecology and conservation

Conservation Science and Practice, Journal Year: 2020, Volume and Issue: 2(8)

Published: June 19, 2020

Abstract Camera traps (CTs) are an increasingly popular method of studying animal behavior. However, the impact cameras on detected individuals—such as from mechanical noise, odor, and emitted light—has received relatively little attention. These impacts particularly important in behavioral studies conservation that seek to ascribe changes behavior relevant environmental factors. In this article, we discuss three sources bias using CTs: (a) disturbance caused by cameras; (b) variation animal‐detection parameters across camera models; (c) biased detection individuals age, sex, classes. We propose several recommendations aimed at mitigating responses CTs wildlife. Our offer a platform for development more rigorous robust CT technology and, if adopted, would result greater applied benefits management.

Language: Английский

---

Global camera trap synthesis highlights the importance of protected areas in maintaining mammal diversity

Conservation Letters, Journal Year: 2022, Volume and Issue: 15(2)

Published: Jan. 26, 2022

Abstract The establishment of protected areas (PAs) is a central strategy for global biodiversity conservation. While the role PAs in protecting habitat has been highlighted, their effectiveness at mammal communities remains unclear. We analyzed dataset from over 8671 camera traps 23 countries on four continents that detected 321 medium‐ to large‐bodied species. found strong positive correlation between taxonomic diversity and proportion surveyed area covered by scale ( β = 0.39, 95% confidence interval [CI] 0.19–0.60) Indomalaya 0.69, CI 0.19–1.2), as well functional PA coverage Nearctic 0.47, 0.09–0.85), after controlling human disturbances environmental variation. Functional was only weakly (and insignificantly) correlated with 0.22, −0.02–0.46), pointing need better understand response protection. Our study provides important evidence conserving terrestrial mammals emphasizes critical area‐based conservation post‐2020 framework.

Language: Английский

---

Density‐dependent space use affects interpretation of camera trap detection rates

Ecology and Evolution, Journal Year: 2019, Volume and Issue: 9(24), P. 14031 - 14041

Published: Nov. 22, 2019

Abstract Camera traps (CTs) are an increasingly popular tool for wildlife survey and monitoring. Estimating relative abundance in unmarked species is often done using detection rate as index of abundance, which assumes that has a positive linear relationship with true abundance. This assumption may be violated if movement behavior varies density, but the degree to density‐dependent across taxa unclear. The potential confounding population‐level indices by would depend on how regularly, what magnitude, home‐range size vary density. We conducted systematic review meta‐analysis quantify relationships between rate, size, terrestrial mammalian taxa. then simulated animal movements CT sampling test effect contrasting scenarios indices. Overall, were negatively correlated density positively one another. strength varied significantly populations. In simulations, rates related underestimated change, particularly slower moving small home ranges. situations where space use changes markedly we estimate up thirty percent change missed due movement, making trend estimation more difficult. common remains constant densities therefore wide range mammal species. When studying rates, researchers managers should explicitly consider such reflect both movement. Practitioners interpreting camera aware observed differences biased low Further information or methods do not assumptions density‐independent required make robust inferences population trends.

Language: Английский

---

Camera trap placement for evaluating species richness, abundance, and activity

Scientific Reports, Journal Year: 2021, Volume and Issue: 11(1)

Published: Nov. 29, 2021

Abstract Information from camera traps is used for inferences on species presence, richness, abundance, demography, and activity. Camera trap placement design likely to influence these parameter estimates. Herein we simultaneously generate compare estimates obtained (a) placed optimize large carnivore captures (b) random placement, infer accuracy biases Both setups recorded 25 when same number of trail cameras (n = 31) were compared. However, accumulation rate was faster with cameras. Relative abundance indices (RAI) surrogated estimated capture-mark-recapture distance sampling, while RAI biased higher carnivores Group size wild-ungulates both comparable. Random detected nocturnal activities wild ungulates in contrast mostly diurnal observed Our results show that setup give similar richness group size, but differ relative activity patterns. Therefore, made each designs the above parameters need be viewed within this context.

Language: Английский

---

Human disturbance compresses the spatiotemporal niche

Proceedings of the National Academy of Sciences, Journal Year: 2022, Volume and Issue: 119(52)

Published: Dec. 19, 2022

Human disturbance may fundamentally alter the way that species interact, a prospect remains poorly understood. We investigated whether anthropogenic landscape modification increases or decreases co-occurrence—a prerequisite for interactions—within wildlife communities. Using 4 y of data from >2,000 camera traps across human gradient in Wisconsin, USA, we considered 74 pairs (classifying as low, medium, high antagonism to account different interaction types) and used time between successive detections measure their co-occurrence probability define networks. Pairs averaged 6.1 [95% CI: 5.3, 6.8] d low-disturbance landscapes (e.g., national forests) but 4.1 [3.5, 4.7] high-disturbance landscapes, such those dominated by urbanization intensive agriculture. Co-occurrence networks showed higher connectance (i.e., larger proportion possible co-occurrences) greater proportions low-antagonism disturbed landscapes. Human-mediated abundance (possibly via resource subsidies) appeared more important than behavioral mechanisms changes daily activity timing) driving these patterns compressed The spatiotemporal compression co-occurrences likely strengthens interactions like competition, predation, infection unless can avoid each other at fine scales. Regardless, human-mediated with—and hence increased exposure to—predators competitors might elevate stress levels individual animals, with cascading effects populations, communities, ecosystems.

Language: Английский

---

Resource exploitation efficiency collapses the home range of an apex predator

Ecology, Journal Year: 2022, Volume and Issue: 103(5)

Published: Jan. 23, 2022

Abstract Optimizing energy acquisition and expenditure is a fundamental trade‐off for consumers, strikingly reflected in how mobile organisms use space. Several studies have established that home range size decreases as resource density increases, but the balance of costs benefits associated with exploiting given unclear. We evaluate ability consumers to exploit their resources through movement (termed “resource exploitation”) interacts influence size. then contrast two hypotheses exploitation influences across vast gradient productivity human‐created linear features (roads seismic lines) are known facilitate animal movements. Under Diffusion Facilitation Hypothesis, predicted lead more diffuse space larger ranges. Exploitation Efficiency increase foraging efficiency, resulting less being required meet energetic demands therefore smaller Using GPS telemetry data from 142 wolves ( Canis lupus ) distributed over than 500,000 km 2 , we found wolf was influenced by interaction between efficiency. Home decreased feature increased, supporting Hypothesis. However, effect on diminished productive areas, suggesting efficiency greater importance when low. These results suggest ranges will occur where both primary higher, thereby increasing regional density.

Language: Английский

---

Are we telling the same story? Comparing inferences made from camera trap and telemetry data for wildlife monitoring

Ecological Applications, Journal Year: 2022, Volume and Issue: 33(1)

Published: Sept. 15, 2022

Abstract Estimating habitat and spatial associations for wildlife is common across ecological studies it well known that individual traits can drive population dynamics vice versa. Thus, commonly assumed individual‐ population‐level data should represent the same underlying processes, but few have directly compared contemporaneous representing these different perspectives. We evaluated circumstances under which collected from Lagrangian (individual‐level) Eulerian (population‐level) perspectives could yield comparable inference to understand how scalable information population. used Global Positioning System (GPS) collar (Lagrangian) camera trap (Eulerian) seven species simultaneously in eastern Washington (2018–2020) compare inferences made survey fit respective streams resource selection functions (RSFs) occupancy models estimated habitat‐ space‐use patterns each species. Although previous considered whether generated information, ours first make this comparison multiple specifically ask two differed depending on focal found general agreement between predicted distributions most paired analyses, although specific relationships differed. hypothesize discrepancies arose due differences statistical power associated with GPS‐collar sampling, as mismatches data. Our research suggests individual‐based sampling methods capture coarse population‐wide a diversity of species, results differ when interpreting wildlife‐habitat relationships.

Language: Английский

---

Population assessment without individual identification using camera-traps: A comparison of four methods

Basic and Applied Ecology, Journal Year: 2022, Volume and Issue: 61, P. 68 - 81

Published: March 7, 2022

The use of camera traps to estimate population size when animals are not individually recognizable is gaining traction in the ecological literature, because its applicability conservation and management. We estimated synthetic with four trap sampling-based statistical models that do rely on individual recognition. Using a realistic model animal movement generate data, we compared random encounter model, staying time association time-to-event-model investigated impact violation assumptions estimates. While under ideal conditions these provide reliable estimates, movements were characterised by differences speed (due diverse behaviours such as locomotion, grazing resting) none provided both unbiased precise density results but tended overestimate size, while was less underestimate size. Lastly, unable results. found each tested very sensitive method used range field-of-view traps. Density estimates from also biases animals' speed. guidelines how get could be useful wildlife managers practitioners.

Language: Английский

---

Deer Behavior Affects Density Estimates With Camera Traps, but Is Outweighed by Spatial Variability

Frontiers in Ecology and Evolution, Journal Year: 2022, Volume and Issue: 10

Published: May 18, 2022

Density is a key trait of populations and an essential parameter in ecological research, wildlife conservation management. Several models have been developed to estimate population density based on camera trapping data, including the random encounter model (REM) trap distance sampling (CTDS). Both need account for variation animal behavior that depends, example, species sex animals along with temporally varying environmental factors. We examined whether estimates REM CTDS can be improved Europe’s most numerous deer species, by adjusting behavior-related parameters per accounting differences movement speeds between sexes, seasons, years. Our results showed bias through inadequate consideration was exceeded uncertainty estimates, which mainly influenced number independent observations locations. The neglection seasonal annual speed overestimated densities red autumn spring ca. 14%. This GPS telemetry-derived found problematic roe females summer when characterized small-scale displacements relative intervals fixes. In CTDS, foremost behavioral reactions traps (avoiding max. 19%), while species-specific delays photos had larger effect deer. general, applicability both would profit profoundly from improvements their precision reduction achieved exploiting available information data.

Language: Английский

---