Estimate Detection Probability Research Articles

The influence of sea ice on the detection of bowhead whale calls

Bowhead whales (Balaena mysticetus) face threats from diminishing sea ice and increasing anthropogenic activities in the Arctic. Passive acoustic monitoring is the most effective means for monitoring their distribution and population trends, based on the detection of their calls. Passive acoustic monitoring, however, is influenced by the sound propagation environment and ambient noise levels, which impact call detection probability. Modeling and simulations were used to estimate detection probability for bowhead whale frequency-modulated calls in the 80–180 Hz frequency band with and without sea ice cover and under various noise conditions. Sound transmission loss for bowhead calls is substantially greater during ice-covered conditions than during open-water conditions, making call detection ~ 3 times more likely in open-water. Estimates of daily acoustic detection probability were used to compensate acoustic detections for sound propagation and noise effects in two recording datasets in the northeast Chukchi Sea, on the outer shelf and continental slope, collected between 2012 and 2013. The compensated acoustic density suggests a decrease in whale presence with the retreat of sea ice at these recording sites. These results highlight the importance of accounting for effects of the environment on ambient noise and acoustic propagation when interpreting results of passive acoustic monitoring.

Comparative detection, density, and reproductive performance of Kirtland's warbler in jack and red pine

AbstractThe formerly endangered Kirtland's warbler (Setophaga kirtlandii) is among a growing number of conservation‐reliant species that depend on active management to avoid reverting to endangered status. Because the Kirtland's warbler is a habitat specialist of young, even‐aged jack pine (Pinus banksiana), managers of the recovery effort stressed creating new jack pine stands and monitoring numbers of singing males through an annual census using single visits to individual stands. Kirtland's warbler will occupy and breed in red pine (P. resinosa), but red pine has not been surveyed for Kirtland's warblers in the annual population census. Furthermore, the current monitoring approach cannot determine their species detection probability or individual detection probability, which is essential to evaluate both red pine use and the accuracy of the census. From 2016–2018 we estimated density and detection probabilities in jack pine and red pine stands through repeated visits to a limited number of stands rather than single visits to many stands. Estimates of species detection probability indicated that ≥1 male Kirtland's warbler would be detected on most sites when any were present, but individual detection probabilities were less and varied by stand type, indicating that single visits to sites would underestimate numbers and that accurate estimation of detection probability was important for estimation of density in different stand types. We offer quantitative estimates of detection probabilities for determination of Kirtland's warbler population size in jack pine versus red pine stands in the same areas and breeding seasons. Managers of Kirtland's warblers should incorporate detection probabilities into population surveys to achieve more accurate estimates of population size.

Methodology matters when estimating deer abundance: a global systematic review and recommendations for improvements

AbstractDeer (Cervidae) are key components of many ecosystems and estimating deer abundance or density is important to understanding these roles. Many field methods have been used to estimate deer abundance and density, but the factors determining where, when, and why a method was used, and its usefulness, have not been investigated. We systematically reviewed journal articles published during 2004–2018 to evaluate spatio‐temporal trends in study objectives, methodologies, and deer abundance and density estimates, and determine how they varied with biophysical and anthropogenic attributes. We also reviewed the precision and bias of deer abundance estimation methods. We found 3,870 deer abundance and density estimates. Most estimates (58%) were for white‐tailed deer (Odocoileus virginianus), red deer (Cervus elaphus), and roe deer (Capreolus capreolus). The 6 key methods used to estimate abundance and density were pedestrian sign (track or fecal) counts, pedestrian direct counts, vehicular direct counts, aerial direct counts, motion‐sensitive cameras, and harvest data. There were regional differences in the use of these methods, but a general pattern was a temporal shift from using harvest data, pedestrian direct counts, and aerial direct counts to using pedestrian sign counts and motion‐sensitive cameras. Only 32% of estimates were accompanied by a measure of precision. The most precise estimates were from vehicular spotlight counts and from capture–recapture analysis of images from motion‐sensitive cameras. For aerial direct counts, capture–recapture methods provided the most precise estimates. Bias was robustly assessed in only 16 studies. Most abundance estimates were negatively biased, but capture–recapture methods were the least biased. The usefulness of deer abundance and density estimates would be substantially improved by 1) reporting key methodological details, 2) robustly assessing bias, 3) reporting the precision of estimates, 4) using methods that increase and estimate detection probability, and 5) staying up to date on new methods. The automation of image analysis using machine learning should increase the accuracy and precision of abundance estimates from direct aerial counts (visible and thermal infrared, including from unmanned aerial vehicles [drones]) and motion‐sensitive cameras, and substantially reduce the time and cost burdens of manual image analysis.

Population densities and home range of the vulnerable Pyrenean brook newt in its core aquatic habitat

Abstract Accurate estimations of population size and space-use are critical issues, for assessing population trends and extinction risk. In this study, we applied spatially explicit capture-recapture (SECR) modelling customized for linear habitats to a three-year Capture-Mark-Recapture (CMR) monitoring of three populations of stream-dwelling Pyrenean brook newt (Calotriton asper), an endemic species of the Pyrenean mountain range. Our aim was to explore (1) spatial and temporal variation in population densities, home range size and individual detection probability during breeding season between populations and years, and (2) the influence of water temperature and flow on activity of newts. Estimates of population density range from 3044 to 4641 individuals/km of stream across the three years of the study. Linear home range along the stream was estimated to be 13.31 m from the home range centre on average over the three years. All detection probability estimates were similar, both between years and between sites, with a mean probability of 0.09, except in Fougax in 2020 where detection probability was 0.03. Activity of the Pyrenean brook newt was influenced by water temperature and flow, but these factors acted at different time scales. Overall, population densities are locally high in both populations but home range of the Pyrenean brook newt during breeding season is very small indicating a strong site attachment. The importance of water temperature and flow on activity emphasizes the sensitivity of the species to these factors that are forecasted to change in coming decades.

Accounting for uncertainty when estimating drivers of imperfect detection: An integrated approach illustrated with snorkel surveys for riverine fishes

Imperfect detection is a common issue affecting the accuracy of surveys that quantify animal abundance and distribution. To quantify detectability, counts are often calibrated to independent measures of abundance (e.g., via mark-recapture) but stochastic sampling variability in both data types is not typically accounted for. This practice may cause detectability to be quantified inaccurately and lead to overly confident predictions for out-of-sample applications. Our objective was to develop, apply, and simulation-test an integrated approach for quantifying detectability that better-accommodates uncertainty in the data. The method assumes mark-recapture and count surveys sample the same local abundance with error, allowing the construction of a joint likelihood function for both data sets. The model estimates coefficients that link detection probability to local covariates through a logit-linear model, which enables correcting counts for imperfect detection in locations where mark-recapture data are unavailable. We illustrate the application of the model with an empirical data set of over 100 paired snorkel count and mark-recapture electrofishing surveys for riverine salmonids in northeastern Oregon. Covariates that best explained heterogeneity in detectability included species, visibility, and channel unit type and depth, though substantial variability was attributed to site-level random effects. Estimated detection probability ranged from 0.02 to 0.92 among surveys and was higher for Chinook Salmon (Oncorhynchus tshawytscha) juveniles (mean: 0.38) than for steelhead/Rainbow Trout (O. mykiss; mean: 0.24). Simulation analyses revealed that our integrated model performed better (relative to a method that treated mark-recapture abundance estimates as known without error) with respect to (i) selection of covariates, (ii), credible interval coverage, (iii) accuracy of estimated random variability terms, and (iv) reduced sensitivity to violated mark-recapture assumptions surrounding behavioral effects. This model represents an improvement over simpler calibration methods, particularly for snorkel surveys, by applying more a rigorous statistical treatment of sources of variability while explicitly describing the mechanistic link between local conditions and detectability. The analytical methods we illustrate are general and could be broadly applied to quantify detectability in other biological surveys with paired abundance and count data.

A field experiment characterizing variable detection rates during plant surveys.

Surveys aimed at finding threatened and invasive species can be challenging due to individual rarity and low and variable individual detection rates. Detection rate in plant surveys typically varies due to differences among observers, among the individual plants being surveyed (targets), and across background environments. Interactions among these 3 components may occur but are rarely estimated due to limited replication and control during data collection. We conducted an experiment to investigate sources of variation in detection of 2 Pilosella species that are invasive and sparsely distributed in the Alpine National Park, Australia. These species are superficially similar in appearance to other yellow‐flowered plants occurring in this landscape. We controlled the presence and color of flowers on target Pilosella plants and controlled their placement in plots, which were selected for their variation in cover of non‐target yellow flowers and dominant vegetation type. Observers mimicked Pilosella surveys in the plots and reported 1 categorical and 4 quantitative indicators of their survey experience level. We applied survival analysis to detection data to model the influence of both controlled and uncontrolled variables on detection rate. Orange‐ and yellow‐flowering Pilosella in grass‐ and heath‐dominated vegetation were detected at a higher rate than nonflowering Pilosella. However, this detection gain diminished as the cover of other co‐occurring yellow‐flowering species increased. Recent experience with Pilosella surveys improved detection rate. Detection experiments are a direct and accessible means of understanding detection processes and interpreting survey data for threatened and invasive species. Our detection findings have been used for survey planning and can inform progress toward eradication. Interaction of target and background characteristics determined detection rate, which enhanced predictions in the Pilosella eradication program and demonstrated the difficulty of transferring detection findings into untested environments.

Finding rare species and estimating the probability that all occupied sites have been found

AbstractDetecting occupied sites of rare species, and estimating the probability that all occupied sites are known within a given area, are desired outcomes for many ecological or conservation projects. Examples include managing all occupied sites of a threatened species or eradicating an emerging invader. Occupied sites may remain undetected because (1) sites where the species potentially occurs had not been searched, and (2) the species could have been overlooked in the searched sites. For rare species, available data are typically scant, making it difficult to predict sites where the species probably occurs or to estimate detection probability in the searched sites. Using the critically endangered Rose's mountain toadlet (Capensibufo rosei), known from only two localities, we outline an iterative process aimed at estimating the probability that any unknown occupied sites remain and maximizing the chance of finding them. This includes fitting a species distribution model to guide sampling effort, testing model accuracy and sampling efficacy using the occurrence of more common proxy species, and estimating detection probability using sites of known presence. The final estimate of the probability that all occupied sites were found incorporates the uncertainties of uneven distribution, relative area searched, and detection probability. Our results show that very few occupied sites of C. rosei are likely to remain undetected. We also show that the probability of an undetected occupied site remaining will always be high for large unsearched areas of potential occurrence, but can be low for smaller areas intended for targeted management interventions. Our approach is especially useful for assessing uncertainty in species occurrences, planning the required search effort needed to reduce probability of unknown occurrence to desired levels, and identifying priority areas for further searches or management interventions.

What factors affect species richness and distribution dynamics within two Afromontane protected areas?

Context Biodiversity monitoring programs provide information on the status and trends in wildlife populations. These trends are unknown for most mammals within African montane forests, which harbour many endemic and threatened species. Camera traps are useful for assessing mammal populations, because they allow for the estimation of species richness, occupancy, and activity patterns. Aims We sought to explore the richness and distribution of small- to large-sized mammals by using occupancy models while accounting for imperfect detection in Volcanoes and Mgahinga Gorilla National Parks, in Rwanda and Uganda. Methods We used camera-trap data collected from 2014 to 2017 by the Tropical Ecology Assessment and Monitoring (TEAM) network and multi-season occupancy models with multispecies data to assess the dynamics of species richness and distribution in the Virunga Massif and the influence of site covariates on species detection probability, occupancy, colonisation and extinction. Key results We identified 17 species from 7047 trap-days, with most of them showing an uneven distributional pattern throughout the park. We found that average species richness per site increased from five to seven species in 2017. Average local colonisation was estimated at 0.13 (s.e. 0.014), but the probability of local extinction was 0.17 (s.e. 0.028) and negatively influenced by distance from the park boundary. Detection probability was highest for medium-sized species. For species distribution, we found that black-fronted duiker, Cephalophus nigrifrons, and bushbuck, Tragelaphus scriptus, declined in distribution but remained widespread in our study area, while all other species showed an increase in distribution over the study period. Conclusions Our approach allowed us to draw inferences on rare species, such as African golden cat, Caracal aurata, by estimating detection probability on the basis of shared covariate information with more common, widespread species. As such, we were able to estimate all occupancy parameters across the terrestrial mammal community. Implications The results of this study on the distribution of terrestrial mammal species can be used by park management to inform optimal ranger patrolling efforts for mitigating threats in areas of high species presence. Additionally, the results can highlight locations of potential human–wildlife conflicts on the basis of species commonly found along the park boundary.

Observer differences in individual identification of feral cats from camera trap images

Feral cats are a key threat to many Australian native fauna, with camera traps increasingly used to identify individuals for evaluation of management actions. However, observer bias and camera trap settings can affect individual identification rates. We compared feral cat individual identification by two observers with extremes of experience. Arrays of 39–50 camera traps were deployed continuously for 22 months at four sites in the Western Division of New South Wales. Where possible, feral cats were individually identified from phenotypic characteristics by an expert and naïve lay observer. We obtained 10 465 feral cat images, with 72 cats individually identified across the sites. The experienced observer attributed more feral cat events to a known individual compared with the lay observer (21.3 vs 12.9%, respectively). Forty three percent of cat images were similarly tagged by both observers. Daytime events yielded higher identification rates and match success (28.1 vs 19.5 and 17.9 vs 11.8% for day vs night events for the expert and lay observer, respectively). Lack of congruence between observers, combined with a small number of events where cats could be individually identified, and differences in identification accuracy over time and between sites, makes estimation of detection probabilities and errors difficult.

Accounting for the Lombard effect in estimating the probability of detection in passive acoustic surveys: Applications for single sensor mitigation and monitoring.

The detection range of calling animals is commonly described by the passive sonar equations. However, the sonar equations do not account for interactions between source and ambient sound level, i.e., the Lombard effect. This behavior has the potential to introduce non-linearities into the sonar equations and result in incorrectly predicted detection ranges. Here, we investigate the relationship between ambient sound and effective detection ranges for North Atlantic right whales (Eubalaena glacialis) in Cape Cod Bay, MA, USA using a sparse array of acoustic recorders. Generalized estimating equations were used to model the probability that a call was detected as a function of distance between the calling animal and the sensor and the ambient sound level. The model suggests a non-linear relationship between ambient sound levels and the probability of detecting a call. Comparing the non-linear model to the linearized version of the same model resulted in 12 to 25% increases in the effective detection range. We also found evidence of the Lombard effect suggesting that it is the most plausible cause for the non-linearity in the relationship. Finally, we suggest a simple modification to the sonar equation for estimating detection probability for single sensor monitoring applications.

A hierarchical dependent double‐observer method for estimating waterfowl breeding pairs abundance from helicopters

We applied a dependent double‐observer method for helicopter surveys and developed a hierarchical Bayesian model as a means to adjust counts of waterfowl for incomplete detection. We conducted our study using 52 plots in Labrador, Canada. A designated pair of primary observers reported counts and location of all waterfowl flocks that they detected to a pair of secondary observers, including details regarding the species, age and sex of observed birds. Secondary observers then reported any additional flocks observed by them but missed by the primary observers. The pairs of observers alternated between primary and secondary roles during the course of the survey, as well as position (front or back) within the helicopter. We used hierarchical Bayesian models to estimate detection probabilities of waterfowl flocks, as well as derive species‐specific detection‐corrected abundance and sex composition estimates of flocks. The hierarchical model output allowed us to derive estimates of indicated breeding pairs for each species in the survey area corrected for incomplete detection. Observers seated in the back of the helicopter had higher detection probabilities (0.89; 90% Bayesian credible intervals [BCI] = 0.82–0.95) than those in the front (0.74; 90% BCI = 0.66–0.83), and observer experience had a limited effect on detection. Total crew detection probabilities ranged between 0.99 (90% BCI = 0.97–1.00) and 0.97(90% BCI = 0.94–0.99), depending on the individual observers' position and role in the helicopter. Detection probabilities were higher for sea ducks and diving ducks and lower for dabbling ducks. Observers generally missed less than 5% of the total indicated pairs for all species. We recommend that detection in helicopter surveys be measured to control for observer turnover, observer experience and aircraft‐related differences in visibility.

Using historical dip net data to infer absence of flatwoods salamanders in stochastic environments.

Local extinction and undetected presence are two very different biological phenomena, but they can be challenging to differentiate. Stochastic environments hamper the development of standardized monitoring schemes for wildlife, and make it more challenging to plan and evaluate the success of conservation efforts. To avoid reintroductions of species at risk that could jeopardize extant populations, managers attempting translocation events require a higher level of confidence that a failure to confirm presence represents a true absence. For many pond breeding amphibians, monitoring of the breeding population occurs indirectly through larval surveys. Larval development and successful recruitment only occurs after a sequence of appropriate environmental conditions, thus it is possible for a breeding population of adults to exist at a site but for detectability of the species to be functionally zero. We investigate how annual variability in detection influences long-term monitoring efforts of Reticulated Flatwoods Salamanders (Ambystoma bishopi) breeding in 29 wetlands in Florida. Using 8 years of historic dip net data, we simulate plausible monitoring scenarios that incorporate environmental stochasticity into estimates of detection probability. We found that annual variation in environmental conditions precluded a high degree of certainty in predicting site status for low-intensity monitoring schemes. Uncertainty was partly alleviated by increasing survey effort, but even at the highest level of sampling intensity assessed, multiple years of monitoring are required to confidently determine presence/absence at a site. Combined with assessments of habitat quality and landscape connectivity, our results can be used to identify sites suitable for reintroduction efforts. Our methodologies can be generally applied to increase the effectiveness of surveys for diverse organisms for which annual variability in detectability is known.

Early detection monitoring for non-indigenous fishes; comparison of survey approaches during two species introductions in a Great Lakes port.

Assessing relative performance of different sampling methods used for early detection monitoring (EDM) is a critical step in understanding the likelihood of detecting new non-indigenous species (NIS) in an environment of interest. EDM performance metrics are typically based on the probability of detecting established NIS or rare indigenous species; however, detection probability estimates for these proxies may not accurately reflect survey effectiveness for newly introduced NIS. We used data from three different EDM survey approaches that varied by targeted life-stage (adult-juvenile versus ichthyoplankton), media (physical fish versus environmental DNA), and taxonomic method (morphology-based versus DNA-based taxonomy) to explore relative detection sensitivity for recently introduced white bass (Morone chrysops) and gizzard shad (Dorosoma cepedianum) in the Port of Duluth-Superior, a NIS introduction hot spot within the Laurentian Great Lakes. Detection efficiency, measured by the effort (number of samples) required to achieve 95% probability of detection, differed by EDM approach and species. Also, the relative sensitivity (detection rate) of each survey approach differed by species. For both species, detection in surveys using DNA-based taxonomy was generally as good or better than the adult-juvenile survey using morphology-based taxonomy. While both species appear to have been detected at early stages of invasion, white bass were likely present up to 5 years prior to initial detection, whereas gizzard shad may have been detected in the first year of introduction. We conclude that using complimentary sampling methods can help to balance the strengths and weaknesses of each approach and provide more reliable early detection of new invaders.

On the Estimation of POD and LOD of Qualitative Microbiological Assays from a Multi-Laboratory Validation Study.

Jarvis et al. in 2019 (J. AOAC Int. 102: 1617-1623) estimated the mean laboratory effect (µ), standard deviation of laboratory effects (σ), probability of detection (POD), and level of detection (LOD) from a multi-laboratory validation study of qualitative microbiological assays using a random intercept complementary log-log model. Their approach estimated σ based on a Laplace approximation to the likelihood function of the model, but estimated µ from a fixed effectmodel due to a limitation in the MS Excel spreadsheet which was used by the authors to develop a calculation tool. We compared the estimates of µ and σ from three approaches (the Laplace approximation that estimates µ and σ simultaneously from the random intercept model, adaptive Gauss-Hermite quadrature (AGHQ), and the method of Jarvis et al.) and introduced an R Shiny app to implement the AGHQ using the widely used "lme4" R package. We conducted a simulation study to compare the accuracy of the estimates of µ and σ from the three approaches and compared the estimates of µ, σ, LOD, etc. between the R Shiny app and the spreadsheet calculation tool developed by Jarvis et al. for an example dataset. Our simulation study shows that, while the three approaches produce similar estimates of σ, the AGHQ has generally the best performance for estimating µ (and hence mean POD and LOD). The differences in the estimates between the R Shiny app and the spreadsheet were demonstrated using the example dataset. The AGHQ is the best method for estimating µ, POD, and LOD. The user-friendly R Shiny app provides a better alternative to the spreadsheet.

Factors Influencing Colonization and Survival of Juvenile Blue Crabs Callinectes sapidus in Southeastern U.S. Tidal Creeks

Tidal creeks along the southeastern U.S. and Gulf of Mexico coastlines provide nursery habitats for commercially and ecologically important nekton, including juvenile blue crabs Callinectes sapidus, a valuable and heavily landed seafood species. Instream and watershed urbanization may influence the habitat value that tidal creeks provide to blue crabs. We investigated natural and anthropogenic factors influencing juvenile blue crab occupancy dynamics in eight first-order tidal creeks in coastal North Carolina (USA). An auto-logistic hierarchical multi-season (dynamic) occupancy model with separate ecological and observation sub-models was fitted to juvenile blue crab presence/absence data collected over replicate sampling visits in multiple seasons at three fixed trapping sites in each creek. Colonization and survival are the processes operating on occupancy that are estimated with this formulation of the model. Covariates considered in the ecological sub-model included watershed imperviousness, the percent of salt marsh in each creek’s high tide area, percent salt marsh edge, site-level water depth, and site-level salinity. Temperature, salinity, and dissolved oxygen were covariates considered in the observation sub-model. In the ecological sub-model, watershed imperviousness was a meaningful negative covariate and site-level salinity was a positive covariate of survival probability. Imperviousness and salinity were each marginally meaningful on colonization probability. Water temperature was a positive covariate of detection probability in the observation sub-model. Mean estimated detection probability across all sites and seasons of the study was 0.186. The results suggest that development in tidal creek watersheds will impact occupancy dynamics of juvenile blue crabs. This places an emphasis on minimizing losses of natural land cover classes in tidal creek watersheds to reduce the negative impacts to populations of this important species. Future research should explore the relationship between imperviousness and salinity fluctuations in tidal creeks to better understand how changing land cover influences water chemistry and ultimately the demographics of juvenile blue crabs.

Variation in herpetofauna detection probabilities: implications for study design

Population monitoring is fundamental for informing management decisions aimed at reducing the rapid rate of global biodiversity decline. Herpetofauna are experiencing declines worldwide and include species that are challenging to monitor. Raw counts and associated metrics such as richness indices are common for monitoring populations of herpetofauna; however, these methods are susceptible to bias as they fail to account for varying detection probabilities. Our goal was to develop a program for efficiently monitoring herpetofauna in southern Texas. Our objectives were to (1) estimate detection probabilities in an occupancy modeling framework using trap arrays for a diverse group of herpetofauna and (2) to evaluate the relative effectiveness of funnel traps, pitfall traps, and cover boards. We collected data with 36 arrays at 2 study sites in 2015 and 2016, for 2105 array-days resulting in 4839 detections of 51 species. We modeled occupancy for 21 species and found support for the hypothesis that detection probability varied over our sampling duration for 10 species and with rainfall for 10 species. For herpetofauna in our study, we found 14 and 12 species were most efficiently captured with funnel traps and pitfall traps, respectively, and no species were most efficiently captured with cover boards. Our results show that using methods that do not account for variations in detection probability are highly subject to bias unless the likelihood of false absences is minimized with exceptionally long capture durations. For monitoring herpetofauna in southern Texas, we recommend using arrays with funnel and pitfall traps and an analytical method such as occupancy modeling that accounts for variation in detection.

Estimating forest antelope population densities using distance sampling with camera traps

AbstractTraditional transect survey methods for forest antelopes often underestimate density for common species and do not provide sufficient data for rarer species. The use of camera trapping as a survey tool for medium and large terrestrial mammals has become increasingly common, especially in forest habitats. Here, we applied the distance sampling method to images generated from camera-trap surveys in Dja Faunal Reserve, Cameroon, and used an estimate of the proportion of time animals are active to correct for negative bias in the density estimates from the 24-hour camera-trap survey datasets. We also used multiple covariate distance sampling with body weight as a covariate to estimate detection probabilities and densities of rarer species. These methods provide an effective tool for monitoring the status of individual species or a community of forest antelope species, information urgently needed for conservation planning and action.

Brood success of northern bobwhite is biased by incomplete detectability during flush-counts

Flush-count surveys of game bird broods are a common method of deriving estimates of brood survival, but detection of chicks during surveys is < 1 due to factors such as vegetation obscurity, adult brooding behavior and variation by observer. Radio-telemetry is an alternative method for estimating survival that circumvents such factors and allow for higher detection. However, this practice is costly and labor-intensive and therefore not readily adopted. We sought to estimate detection probability of chicks during flush-counts as a function of vegetation height and adult brooding behavior. Secondly, we evaluated compromises in detection by comparing estimates of brood survival derived from flush-counts and radiotelemetry. Lastly, we compared counts between two observers to discern whether an additional observer could increase accuracy of counts.We radiomarked 247 northern bobwhite Colinus virginianus chicks at 10–12 days of age and conducted 46 flush-counts at 21 days of age. Vegetation height substantially decreased detection (β = –1.18; 95 CrI: –1.68 to –0.73); mean detection probability was 0.30 (95 CrI: 0.22–0.40). Observers failed to detect radio-marked chicks when adults exhibited running behavior (n = 16 chicks, n = 6 surveys), orphaning occurred (n = 11 chicks, n = 5 surveys) or brooding adults died or had transmitter failures (n = 4 chicks, n = 3 surveys). An additional observer did not affect counts with a mean difference of –0.6 chicks (95 CrI: –4.0 to 2.7) counted between observers. Chicks were not detected during 47% of surveys when ≥ 1 radio-marked individuals were known to be alive. Brood survival was 0.83 (95 CrI: 0.70–0.92) and 0.48 (95 CrI: 0.34–0.62) for radiotelemetry and naïve flush-counts, respectively. Because of low detectability of chicks during flush-counts, alternative methods should be considered by future researchers.

Application of Unmanned Aerial Vehicles and Thermal Imaging Cameras to Conduct Duck Brood Surveys

ABSTRACTBrood surveys are used to estimate productivity in ducks, but road‐side transects, aerial surveys, and double‐observer ground surveys have likely underestimated productivity. Duck broods are elusive and prefer wetlands with emergent vegetation where they hide at signs of disturbance, making it difficult to get accurate brood counts. Estimates of brood detection probabilities are typically below 50% and variable, which makes biological inferences about abundance tenuous. We conducted a study to evaluate the efficacy of using an unmanned aerial vehicle (UAV) equipped with a thermal imaging camera to survey duck broods in 2 study areas. In Manitoba we located 669 broods with the UAV, compared to 344 detected by double‐observer ground surveys. In Minnesota we detected 225 ducks broods with the UAV, whereas only 105 duck broods were detected by ground observers. Using a Huggins closed‐capture model in program MARK we estimated an average detection probability across both sites of 0.55 (SE = 0.02) with the UAV compared to 0.24 (SE = 0.02) for the ground crews. Although the UAV detected twice as many broods as the ground surveys, detection probability with the UAV was impacted by temperature, humidity, vegetation density, and the criteria we used to determine whether a brood could be classified as resighted. Nevertheless, using a UAV equipped with a thermal imaging camera effectively doubled the number of broods detected compared to traditional methods, and surveys were completed 3 times faster. With advancing drone and camera technology we believe UAV brood counts will become increasingly accurate and provide reliable measures of local duck productivity. © 2021 The Wildlife Society.

Using PIT-tags and portable antennas for quantification of fish movement and survival in streams under different environmental conditions.

While PIT-tag tracking using mobile antennas is being increasingly used to study fish movement and survival in streams, little is known about the limitations of the method, especially over longer periods of time and under different environmental settings. We used 6 years of data combining tagging, mobile antenna tracking and recaptures of Salmo trutta in multiple small streams in the Lake Lucerne drainage area in Switzerland to evaluate the relative importance of different environmental and intrinsic factors affecting the efficiency of the method. Our study system and experimental design allowed us to accurately verify the continuous presence and survival of recaptured fish in the stream after tracking, which meant that we could estimate detection probability with high confidence. The mean detection probability of tagged trout was 43%, but we found that fish length had a strong negative effect on detection probability, especially in males. Multivariate axes of stream environmental features did not predict efficiency but stream width alone was significantly positively correlated with efficiency. Additionally, stream temperature when tracking had a positive effect on fish detectability. Tag loss at recapture was globally rare (<8%) but common in large postspawn females (>30%). Based on the escape response of fish after detection, we could estimate the proportion of ghost tags, which reached a plateau of around 80% 2 years after tagging. We finally showed that our models of tag loss, fish detection and escape response are needed to interpret detection events. Our results highlight that individual variation in detection probability and tag loss is high and has to be considered for analysis.