Extinct or still out there? Disentangling influences on extinction and rediscovery helps to clarify the fate of species on the edge

Each year, two or three species that had been considered to be extinct are rediscovered. Uncertainty about whether or not a species is extinct is common, because rare and highly threatened species are difficult to detect. Biological traits such as body size and range size are expected to be associated with extinction. However, these traits, together with the intensity of search effort, might influence the probability of detection and extinction differently. This makes statistical analysis of extinction and rediscovery challenging. Here, we use a variant of survival analysis known as cure rate modelling to differentiate factors that influence rediscovery from those that influence extinction. We analyse a global data set of 99 mammals that have been categorized as extinct or possibly extinct. We estimate the probability that each of these mammals is still extant and thus estimate the proportion of missing (presumed extinct) mammals that are incorrectly assigned extinction. We find that body mass and population density are predictors of extinction, and body mass and search effort predict rediscovery. In mammals, extinction rate increases with body mass and population density, and these traits act synergistically to greatly elevate extinction rate in large species that also occurred in formerly dense populations. However, when they remain extant, larger‐bodied missing species are rediscovered sooner than smaller species. Greater search effort increases the probability of rediscovery in larger species of missing mammals, but has a minimal effect on small species, which take longer to be rediscovered, if extant. By separating the effects of species characteristics on extinction and detection, and using models with the assumption that a proportion of missing species will never be rediscovered, our new approach provides estimates of extinction probability in species with few observation records and scant ecological information.     

Road-Based Surveys for Estimating Wild Turkey Density in the Texas Rolling Plains

Abstract: Line-transect-based distance sampling has been used to estimate density of several wild bird species including wild turkeys (Meleagris gallopavo). We used inflatable turkey decoys during autumn (Aug-Nov) and winter (Dec-Mar) 2003-2005 at study sites in the Texas Rolling Plains, USA, to simulate Rio Grande wild turkey (M. g. intermedia) flocks. We evaluated detectability of flocks using logistic regression models. Our modeling effort suggested that distance to a flock and flock size played important roles in flock detectability. We also conducted surveys from roads for wild turkeys during November 2004-January 2006. The detection probability of decoy flocks was similar to wild turkey flocks during winter (decoy flock, 69.3 ± 6.2% [x̄ ± 95% CI]; wild turkey flock, 62.2 ± 18.3%) and autumn (decoy flock, 44.1 ± 5.1%; wild turkey flock, 44.7 ± 25.6%), which suggested that using decoys was appropriate for evaluating detectability of wild turkey flocks from roads. We conducted computer simulations to evaluate the performance of line-transect-based distance sampling and examined the power to detect trends in population change. Simulations suggested that population density may be underestimated by 12% during inter and 29% during autumn. Such bias occurred because of incomplete detectability of flocks near roads. Winter surveys tended to have less bias, lower relative variability, and greater power than did autumn surveys. During winter surveys, power was sufficient (≥0.80) to detect a 10-25% change in population density in 8-12 years using ≥100 16-km transects or ≥80 32-km transects. We concluded line-transect-based distance sampling from roads is an efficient, effective, and inexpensive technique for monitoring Rio Grande wild turkey populations across large scales.     

In Defense of Indices: The Case of Bird Surveys

Abstract: Indices to population size have come under increasing criticism in recent years, on the grounds that indices might not faithfully represent the entire population. Most criticisms involve surveys of birds, particularly those based on point counts, which is my focus here. A variety of quantitative methods have been developed to reduce the bias of point counts, such as distance sampling, multiple-observer surveys, and time-of-detection methods. I argue that these developments are valuable, in that they enhance understanding of the detection process, but that their practical application may well be limited, likely to intensive studies focusing on a small number of species. These quantitative methods are not generally applicable to extensive, multiple-species surveys. Although criticism of the thoughtless use of indices is welcome, their wholesale rejection is not.     

Rare but not lost: Endemic mountain lizard occupancy following megafire and grazing disturbances

Wildfire and grazing by invasive herbivores can influence habitat suitability for ground-dwelling fauna, such as reptiles. Australia has a large and diverse reptile fauna, with the Australian Alps bioregion in the southeast of the continent supporting a disproportionately high number of threatened species. In this bioregion, many species are threatened by fire, habitat loss or modification, and invasive species. The range of one such threatened endemic lizard, Cyclodomorphus praealtus (family Scincidae), was impacted by the 2019–20 megafires and is also subject to widespread grazing by invasive species. We investigated the relationship between C. praealtus site occupancy and fire and grazing. We completed 2045 surveys across 120 sites over 4 years, detecting the species at 43% of sites and increasing the species' known geographic range. Using single season detection occupancy models, we found C. praealtus occupancy was not associated with elevation, vegetation height or whether the site was burnt, but was positively associated with grazing activity. Our results indicate that C. praealtus has the capacity to persist following a single fire in some cases, and that habitats with high occupancy probabilities are subject to high grazing pressure. However, our results do not rule out more nuanced impacts associated with these disturbances, which affect a large proportion of C. praealtus' habitat. Our cumulative detection probability calculations revealed that considerable survey effort is often required to determine C. praealtus site occupancy. We therefore recommend that impact assessments assume species presence within areas of suitable habitat within the species' range. Our study improves our understanding of disturbance impacts on C. praealtus' occupancy, while demonstrating the need for sufficiently resourced impact assessments for cryptic and threatened species.     

Statistical approaches to account for false‐positive errors in environmental DNA samples

Environmental DNA (eDNA) sampling is prone to both false‐positive and false‐negative errors. We review statistical methods to account for such errors in the analysis of eDNA data and use simulations to compare the performance of different modelling approaches. Our simulations illustrate that even low false‐positive rates can produce biased estimates of occupancy and detectability. We further show that removing or classifying single PCR detections in an ad hoc manner under the suspicion that such records represent false positives, as sometimes advocated in the eDNA literature, also results in biased estimation of occupancy, detectability and false‐positive rates. We advocate alternative approaches to account for false‐positive errors that rely on prior information, or the collection of ancillary detection data at a subset of sites using a sampling method that is not prone to false‐positive errors. We illustrate the advantages of these approaches over ad hoc classifications of detections and provide practical advice and code for fitting these models in maximum likelihood and Bayesian frameworks. Given the severe bias induced by false‐negative and false‐positive errors, the methods presented here should be more routinely adopted in eDNA studies.     

Ecological traits modulate bird species responses to forest fragmentation in an Amazonian anthropogenic archipelago

Aim

We assessed patterns of avian species loss and the role of morpho‐ecological traits in explaining species vulnerability to forest fragmentation in an anthropogenic island system. We also contrasted observed and detectability‐corrected estimates of island occupancy, which are often used to infer species vulnerability.

Location

Tucuruí Hydroelectric Reservoir, eastern Brazilian Amazonia.

Methods

We surveyed forest birds within 36 islands (3.4–2,551.5 ha) after 22 years of post‐isolation history. We applied species–area relationships to assess differential patterns of species loss among three data sets: all species, forest specialists and habitat generalists. After controlling for phylogenetic non‐independence, we used observed and detectability‐corrected estimates of island occupancy separately to build competing models as a function of species traits. The magnitude of the difference between these estimates of island occupancy was contrasted against species detectability.

Results

The rate of species loss as a function of island area reduction was higher for forest specialists than for habitat generalists. Accounting for the area effect, forest fragmentation did not affect the overall number of species regardless of the data set. Only the interactive model including natural abundance, habitat breadth and geographic range size was strongly supported for both estimates of island occupancy. For 30 species with detection probabilities below 30%, detectability‐corrected estimates were at least tenfold higher than those observed. Conversely, differences between estimates were negligible or non‐existent for all 31 species with detection probabilities exceeding 45.5%.

Main conclusions

Predicted decay of avian species richness induced by forest loss is affected by the degree of habitat specialisation of the species under consideration, and may be unrelated to forest fragmentation per se. Natural abundance was the main predictor of species island occupancy, although habitat breadth and geographic range size also played a role. We caution against using occupancy models for low‐detectability species, because overestimates of island occupancy reduce the power of species‐level predictions of vulnerability.

     

Modelling variation in calling rates to develop a reliable monitoring method for the Australasian Bittern Botaurus poiciloptilus

Monitoring the abundance of cryptic species inevitably relies on the use of index methods. Unfortunately, detectability is often confounded by unidentified covariates. One such species is the critically endangered Australasian Bittern Botaurus poiciloptilus. Current monitoring relies upon the ability to count males based on the conspicuous breeding calls of males. However, as in many vocal species, calling rates vary spatially and temporally, making it necessary to account for this when using call counts to index abundance. We undertook 461 15‐min call counts of Australasian Bitterns, in a range of conditions, during two breeding seasons at Whangamarino wetland, New Zealand. We fitted a range of generalized linear mixed models to these data to determine which factors were the best predictors of calling rate per individual Bittern (CRPI), allowing us to make recommendations regarding the optimum time and conditions for monitoring. Bittern CRPI was predictable in terms of time of day, month, cloud cover, rainfall and certain moon parameters, but some spatial and temporal variation remained unexplained. Results showed that the best time to detect Australasian Bitterns was 1 h before sunrise, in September (austral spring), on a moonlit night with no cloud or rain. Such models are useful for identifying times and conditions when counts are the highest and least variable, and could be applied to any species or cue count monitoring method where detection depends on counting calling individuals. Results can be used to standardize index counts, or sensibly to adjust and compare counts from different times. Standardizing monitoring in this way can lead to the development of monitoring methods that have a greater power to show population changes across shorter time periods. Moreover, the use of modelling processes to estimate effect sizes creates potential for such methods to be applied in circumstances where monitoring conditions are rarely optimum and standardization creates logistical trade‐offs, something that is particularly common in studies of cryptic species.     

Effects of experimental playbacks on availability for detection during point counts

Point counts are the most commonly used technique for surveying passerines during the breeding season. Several methods for estimating probabilities of detection during point count surveys have been developed. These methods have focused primarily on accounting for the influence of environmental factors (e.g., weather and noise) on detectability, however, the probability that birds are available for detection (e.g., sings or moves) during point counts has received less attention. We used sequential point counts to determine the effect of playback of the mobbing calls of Black‐capped Chickadees (Poecile atricapillus) and the flight calls of Red‐tailed Hawks (Buteo jamaicensis) on availability for detection (e.g., singing or moving) during point‐count surveys. We conducted 180 point counts over a 2‐yr period in central – east central Minnesota to evaluate the possible effect of playbacks on observed density, overall species richness, minute of first detection, and distance of first detection. We also used removal models to quantify the magnitude of changes in detectability and direction of response to playbacks for 10 focal species. Playback of the mobbing calls of Black‐capped Chickadees increased observed density and decreased the average distance of detection and time of first detection, whereas playback of the flight calls of a Red‐tailed Hawk resulted in a decrease in observed density and species richness, and an increased time of first detection. Playback treatment was a covariate in all best performing models for the 10 species analyzed, but the magnitude and direction of response to playbacks were species specific. The importance of playback type in detectability models indicates that the calls of heterospecifics can influence species availability for detection. As such, researchers using playback methods should seek to quantify species‐specific responses in detection probability and consider how component detection probabilities could influence survey outcomes.