Bee species richness through time in an urbanizing landscape of the southeastern United States.

Compared to non-urban environments, cities host ecological communities with altered taxonomic diversity and functional trait composition. However, we know little about how these urban changes take shape over time. Using historical bee (Apoidea: Anthophila) museum specimens supplemented with online repositories and researcher collections, we investigated whether bee species richness tracked urban and human population growth over the past 118 years. We also determined which species were no longer collected, whether those species shared certain traits, and if collector behavior changed over time. We focused on Wake County, North Carolina, United States where human population size has increased over 16 times over the last century along with the urban area within its largest city, Raleigh, which has increased over four times. We estimated bee species richness with occupancy models, and rarefaction and extrapolation curves to account for imperfect detection and sample coverage. To determine if bee traits correlated with when species were collected, we compiled information on native status, nesting habits, diet breadth, and sociality. We used non-metric multidimensional scaling to determine if individual collectors contributed different bee assemblages over time. In total, there were 328 species collected in Wake County. We found that although bee species richness varied, there was no clear trend in bee species richness over time. However, recent collections (since 2003) were missing 195 species, and there was a shift in trait composition, particularly lost species were below-ground nesters. The top collectors in the dataset differed in how often they collected bee species, but this was not consistent between historic and contemporary time periods; some contemporary collectors grouped closer together than others, potentially due to focusing on urban habitats. Use of historical collections and complimentary analyses can fill knowledge gaps to help understand temporal patterns of species richness in taxonomic groups that may not have planned long-term data.

Bird species responses to rangeland management in relation to their traits: Rio de la Plata Grasslands as a case study.

Areas used for livestock production and dominated by native grasses represent a unique opportunity to reconcile biodiversity conservation and livestock production. However, limited knowledge of individual species' responses to rangeland management restricts our capacity to design grazing practices that favor endangered species and other priority birds. In this work, we applied Hierarchical Modelling of Species Communities (HMSC) to study individual species responses, as well as the influence of traits on such responses, to variables related to rangeland management using birds of the Rio de la Plata Grasslands as a case study. Based on presence-absence data collected in 454 paddocks across 46 ranches we inferred the response of 69 species considering imperfect detection. This degree of detail fills a major gap in rangeland management, as species-level responses can be used to achieve targeted conservation goals other than maximizing richness or abundance. We found that artificial pastures had an overall negative impact on many bird species, whereas the presence of tussocks had a positive effect, including all threatened species. Grassland specialists were in general sensitive to grass height and tended to respond positively to tussocks but negatively to tree cover. Controlling grass height via adjustments in stocking rate can be a useful tool to favor grassland specialists. To favor a wide range of bird species in ranches, a mosaic of short and tall native grasslands with patches of tussocks and trees is desirable. We also found that species-specific responses were modulated by their traits: small-sized birds responded positively to tussocks and tree cover while large species responded negatively to increasing grass height. Ground foragers preferred short grass while birds that scarcely use this stratum were not affected by grass height. Results on the influence of traits on bird responses are an important novelty in relation to previous work in rangelands and potentially increase our predicting capacity and model transferability across grassland regions.

The context dependency of fish-habitat associations in separated karst ecoregions.

Fish populations may be isolated via natural conditions in geographically separated ecoregions. Although reconnecting these populations is not a management goal, we need to understand how these populations persist across landscapes to develop meaningful conservation actions, particularly for species occupying sensitive karst ecosystems. Our study objective was to determine the physicochemical factors related to the occurrence of four spring-associated fishes. Arbuckle Uplift and Ozark Highlands ecoregions, USA. We used a hierarchical approach to identify habitat relationships at multiple spatial scales. We collected detection data using snorkeling and seining. We examined the physicochemical relationships related to the detection and occurrence of four spring-associated fishes using occupancy modeling in a Bayesian framework. We found physicochemical relationships that differed and were similar between ecoregions for several fishes. For three species, we found different water temperature relationships between ecoregions. Smallmouth bass were ubiquitous in their use of drainage areas in the Ozark Highlands but only associated with the lower network of the Arbuckle Uplift. There were several mirrored relationships between ecoregions, including an interaction between residual pool depth and water temperature, where sites with deeper pools were more likely to be occupied during warmer water temperatures. There were single-species occurrence relationships with percent vegetation and percent agriculture. Lastly, snorkeling was a more efficient sampling method compared to seining for all fishes. Our results indicate stream temperature mitigation may be possible via the maintenance of key channel morphologies, and we identify shared stressors between ecoregions. Channel mitigation to maintain reaches with deeper pools may be an important strategy for maintaining thermal refugia, particularly when considering climate change. Identifying the mechanistic underpinning of other multiscale ecological relationships would be helpful to discern if some of the different ecoregion relationships represent warning signals or interactions with unmeasured biotic or abiotic factors.

Challenges of copro-parasitological surveys in wild Iberian ibex (Capra pyrenaica) populations addressed through a combination of molecular and statistical tools.

Copro-parasitological surveys in wildlife face challenges due to the secretive nature of many species and the unknown performance of the diagnostic tests employed. To overcome these issues, we used a combination of hierarchical models (site-occupancy and N-mixture models) applied to copro-parasitological data obtained from fecal samples assigned to the host species by molecular methods in the Iberian ibex in north-western Iberian Peninsula. The aims were to compare the performance of four diagnostic tests (Mini-FLOTAC, McMaster, Willis flotation, and natural sedimentation) and to use this methodological approach (molecular analysis with hierarchical models) to better estimate positivity proportion and shedding intensity in a wild ibex population. Pooled fecal samples were collected, and those confirmed by molecular analyses to be the host species in question were included in the study. Hierarchical models confirmed different performances of each diagnostic test, with Mini-FLOTAC showing higher sensitivity for eimeriid coccidia, Willis flotation (for proportion positive) and McMaster (for shedding intensity) in gastrointestinal Strongylida, and equal performance of MiniFlotac/Willis flotation (for proportion positive) and MiniFlotac/McMaster (for shedding intensity) in Moniezia spp. This study employed a combination of molecular and statistical methods that improved the estimates of prevalence and shedding intensity and allowed us to compare the performance of four diagnostic tests while assessing the effect of covariates. Such improvements are critical to enhancing inference in non-invasive wildlife copro-parasitological studies.

Spatiotemporal distributions of mammals occurring in an agro-prairie ecosystem.

Anthropogenic activities since the European colonization of the North American Great Plains have drastically altered landscape composition and configuration, subsequently affecting native biodiversity. These contemporary human-modified landscapes may affect mammal species' distributions, diel activity patterns, habitat use, and interspecific interactions, though a better understanding of these effects on mammals occurring in remaining prairie landscapes is needed. To fill this gap, we surveyed 381 randomly selected sites in 2018, 2019, and 2020 using motion-sensing camera traps across the western part of the US state of Kansas (7,160,077 ha). Sites were separated by ≥2 km ( x ¯ $$ \overline{x} $$ = 8.16 km, SD = 3.61), and cameras were secured to a metal post 40 cm above ground and randomly oriented toward the north or south. We placed an olfactory attractant (mixture of skunk essence and petroleum jelly) on a wooden stake 3 m in front of each camera. Cameras were in place at each site for 28 consecutive days for each year. We manually identified all mammal species detected at each site, collating these data into a database that included taxonomic information for 14 families of mammals (Antilocapridae, Bovidae, Canidae, Cervidae, Cricetidae, Dasypodidae, Didelphidae, Erethizontidae, Felidae, Heteromyidae, Leporidae, Mephitidae, Mustelidae, Procyonidae, Sciuridae, and Muridae) comprising 28 total species. We recorded 31,178 mammal photographs (nonindependent events) over 27,954 camera trap nights during 2018 (n = 10,351), 2019 (n = 9478), and 2020 (n = 8125). Additionally, we included the time and date of each photocapture. Moreover, we gathered survey-specific data useful for modeling species-specific detection along with site-level habitat composition data taken at each site each year. These data will be useful for examining habitat use, species distributions, diel activity patterns, and spatiotemporal interactions between species and across guilds of mammals occurring in a rapidly changing agro-prairie ecosystem. There are no copyright restrictions, but we ask researchers to cite this paper when using these data for publication.

Fast Bayesian inference for large occupancy datasets.

In recent years, the study of species' occurrence has benefited from the increased availability of large-scale citizen-science data. While abundance data from standardized monitoring schemes are biased toward well-studied taxa and locations, opportunistic data are available for many taxonomic groups, from a large number of locations and across long timescales. Hence, these data provide opportunities to measure species' changes in occurrence, particularly through the use of occupancy models, which account for imperfect detection. These opportunistic datasets can be substantially large, numbering hundreds of thousands of sites, and hence present a challenge from a computational perspective, especially within a Bayesian framework. In this paper, we develop a unifying framework for Bayesian inference in occupancy models that account for both spatial and temporal autocorrelation. We make use of the Pólya-Gamma scheme, which allows for fast inference, and incorporate spatio-temporal random effects using Gaussian processes (GPs), for which we consider two efficient approximations: subset of regressors and nearest neighbor GPs. We apply our model to data on two UK butterfly species, one common and widespread and one rare, using records from the Butterflies for the New Millennium database, producing occupancy indices spanning 45 years. Our framework can be applied to a wide range of taxa, providing measures of variation in species' occurrence, which are used to assess biodiversity change.

Model selection in occupancy models: Inference versus prediction.

Occupancy models are a vital tool for ecologists studying the patterns and drivers of species occurrence, but their use often involves selecting among models with different sets of occupancy and detection covariates. The information-theoretic approach, which employs information criteria such as Akaike's information criterion (AIC) is arguably the most popular approach for model selection in ecology and is often used for selecting occupancy models. However, the information-theoretic approach risks selecting models that produce inaccurate parameter estimates due to a phenomenon called collider bias, a type of confounding that can arise when adding explanatory variables to a model. Using simulations, we investigated the consequences of collider bias (using an illustrative example called M-bias) in the occupancy and detection processes of an occupancy model, and explored the implications for model selection using AIC and a common alternative, the Schwarz criterion (or Bayesian information criterion, BIC). We found that when M-bias was present in the occupancy process, AIC and BIC selected models that inaccurately estimated the effect of the focal occupancy covariate, while simultaneously producing more accurate predictions of the site-level occupancy probability than other models in the candidate set. In contrast, M-bias in the detection process did not impact the focal estimate; all models made accurate inferences, while the site-level predictions of the AIC/BIC-best model were slightly more accurate. Our results show that information criteria can be used to select occupancy covariates if the sole purpose of the model is prediction, but must be treated with more caution if the purpose is to understand how environmental variables affect occupancy. By contrast, detection covariates can usually be selected using information criteria regardless of the model's purpose. These findings illustrate the importance of distinguishing between the tasks of parameter inference and prediction in ecological modeling. Furthermore, our results underline concerns about the use of information criteria to compare different biological hypotheses in observational studies.

Size and degree of protection of native forest remnants drive the local occupancy of an endangered neotropical primate.

Although the species-area relationship is well known, it may interact with and be augmented or cancelled out by other factors, such as local human disturbance. We used data on site occupancy of the Endangered blonde capuchin monkey (Sapajus flavius) based primarily on a standardized program of local interviews to model the influence of past human disturbance on the occurrence of this species across remaining forest patches of northeastern Brazil within the Atlantic Forest and Caatinga biomes. To do so, we assessed environmental covariates that best represent the history of human impacts. We then used single-species occupancy models to assess site occupancy, while controlling for detection error during sampling. Surprisingly, we obtained a higher occupancy rate in the more arid Caatinga remnants than in the more mesic Atlantic Forest. Habitat patch size, history of site protection, and annual precipitation were the best predictors of local occupancy. Historical human disturbance, including subsistence hunting, has exerted considerable impact on the modern distribution of the blonde capuchin, whose geographic range largely spans a region historically lacking any wildlife protection. Matrix vegetation structure across the Caatinga, which so far has averted large-scale mechanized agriculture, also creates a benign landscape that likely benefits contemporary capuchin occupancy. Local extinctions of this endangered primate will most likely continue unabated unless a ban on hunting in remaining Atlantic Forest and Caatinga fragments can be enforced.

Environmental DNA detects Spawning Habitat of an ephemeral migrant fish (Anadromous Rainbow Smelt: Osmerus mordax).

BACKGROUND: Anadromous rainbow smelt (Osmerus mordax) have experienced a large range reduction in recent decades and the status of remnant spawning populations is poorly known in Maine, where these fish have significant ecological, cultural, and commercial relevance. Defining the remnant range of anadromous smelt is more difficult than for many declining fish species because adults are only ephemerally present while spawning in small coastal streams at night during spring runoff periods when traditional assessments can be unreliable or even hazardous. We hypothesized that eDNA might facilitate improved survey efforts to define smelt spawning habitat, but that detection could also face challenges from adult eDNA quickly flushing out of these small stream systems. We combined daytime eDNA sampling with nighttime fyke netting to ascertain a potential window of eDNA detection before conducting eDNA surveys in four streams of varying abundance. Hierarchical occupancy modeling was in turn employed to estimate eDNA encounter probabilities relative to numbers of sampling events (date), samples within events, and qPCR replicates within samples. RESULTS: Results from the combined eDNA and fyke net study indicated eDNA was detectable over an extended period, culminating approximately 8-13 days following peak spawning, suggesting developing smelt larvae might be the primary source of eDNA. Subsequently, smelt eDNA was readily detected in eDNA surveys of four streams, particularly following remediation of PCR inhibitors. Hierarchical occupancy modeling confirmed our surveys had high empirical detection for most sites, and that future surveys employing at least three sampling events, three samples per event, and six qPCR replicates can afford greater than 90% combined detection capability in low abundance systems. CONCLUSIONS: These results demonstrate that relatively modest eDNA sampling effort has high capacity to detect this ephemerally present species of concern at low to moderate abundances. As such, smelt eDNA detection could improve range mapping by providing longer survey windows, safer sampling conditions, and lower field effort in low density systems, than afforded by existing visual and netting approaches.

Assessing the Effects of Landscape Change on the Occupancy Dynamics of the Greater White-Toothed Shrew Crocidura russula.

Land-use change is the main driver of biodiversity loss in the Mediterranean basin. New socio-economic conditions produced a rewilding process so that cultural landscapes are being invaded by more natural habitats. We analyze the effects of landscape change on the demography and the spatial distribution of Crocidura russula in six protected areas of the western Mediterranean basin. The study was conducted in the period 2008-2020 on 19 live trapping plots representing the three main natural habitats of the area (scrubland, pinewood, and holm oak woodland). We used a multiscale approach to ensure that the scale of response matched landscape structure (from plot to landscape) using either vegetation profiles (LiDAR) and land use data obtained from years 2007 and 2017. Statistical models (multiple-season single-species occupancy models) showed that C. russula populations were strongly associated to habitat features at the plot level. These models were used to predict occupancy at sampling units for the whole study area (850 km2), showing contrasting trends that shifted at relatively small spatial scales (expansions and retractions of species ranges). Parks showing extreme scrubland encroachment (-8% of area) and afforestation (+6%) significantly reduced habitat suitability for shrews and reductions in occupancy (-5%). Results would indicate faster changes in the spatial distribution of the target species than previously expected on the basis of climate change, driven by fast landscape changes.

Spatial ecology of crested porcupine in a metropolitan landscape.

Human settlements, including cities, may provide wildlife with new ecological niches, in terms of habitat types and food availability, thus requiring plasticity for adaptation. The crested porcupine Hystrix cristata is a habitat-generalist, large-sized rodent, also recorded in some suburban areas, but no information is available on its habitat use in metropolitan landscapes. Here, we assessed the land-use factors influencing the presence of crested porcupines in a metropolitan area of Central Italy. We collected data on the occurrence of crested porcupines from the metropolitan area of Rome, following an observer-oriented approach to record occurrences and retreive pseudo-absences. We then related the presence/absence of H. cristata to landscape composition. Occupancy models showed that cultivations and scrubland were positively related to porcupine presence, most likely as they provide food resources and shelter sites, respectively. Although the crested porcupine has been confirmed as a "generalist" species in terms of habitat selection, a strong preference for areas limiting the risk of being killed and providing enough food and shelter was observed. We therefore suggest that the crested porcupine may adapt to deeply modified landscapes such as large cities by selecting specific favourable land-use types.

Climate change winners and losers among North American bumblebees.

Mounting evidence suggests that climate change, agricultural intensification and disease are impacting bumblebee health and contributing to species' declines. Identifying how these factors impact insect communities at large spatial and temporal scales is difficult, partly because species may respond in different ways. Further, the necessary data must span large spatial and temporal scales, which usually means they comprise aggregated, presence-only records collected using numerous methods (e.g. diversity surveys, educational collections, citizen-science projects, standardized ecological surveys). Here, we use occupancy models, which explicitly correct for biases in the species observation process, to quantify the effect of changes in temperature, precipitation and floral resources on bumblebee site occupancy over the past 12 decades in North America. We find no evidence of genus-wide declines in site occupancy, but do find that occupancy is strongly related to temperature, and is only weakly related to precipitation or floral resources. We also find that more species are likely to be climate change 'losers' than 'winners' and that this effect is primarily associated with changing temperature. Importantly, all trends were highly species-specific, highlighting that genus or community-wide measures may not reflect diverse species-specific patterns that are critical in guiding allocation of conservation resources.

Is scat marking a reliable tool for otter census and surveys at the landscape scale?

Biological significance of scat marking by otters has been a controversial subject among scientists. Using multiyear (2014-2017) data of otter spraint counts in South Korea, this study aimed to test whether the observed pattern of spraint presence/absence is driven by detection error and if/how scat counts can be a proxy for otter abundance at the landscape scale. To test the first hypothesis, spraint presence/absence was analyzed through occupancy models, which relied on environmental variables related to otter detectability and presence. Spraint count models were used to test the second hypothesis against resource-related covariates in combination with landscape, anthropogenic, and climate variables through machine learning algorithms (MLAs). The detection probability has specifically decreased in areas characterized by high rainfall and human population densities, whereas the probability has increased near food-rich sites, characterized by high marking frequencies. The temporal trends of spraint count predictions were in line with changes in the diversity of fish communities in 2014-2017 instead of fish biomass, suggesting that the availability of feeding resources is higher where fish communities are more diverse. Because diverse fish communities can attract otters, fish diversity conservation is critical for preserving this mammal's populations. This fine scale four-year monitoring has contributed to the disentanglement of the role of spraint presence/absence and spraint counts in detectability and population trends. This will assist in identifying key resource areas and planning strategies to promote otter conservation and dispersal dynamics.

Scale dependence of coral reef oases and their environmental correlates.

Identifying relatively intact areas within ecosystems and determining the conditions favoring their existence is necessary for effective management in the context of widespread environmental degradation. In this study, we used 3766 surveys of randomly selected sites in the United States and U.S. Territories to identify the correlates of sites categorized as "oases" (defined as sites with relatively high total coral cover). We used occupancy models to evaluate the influence of 10 environmental predictors on the probability that an area (21.2-km2 cell) would harbor coral oases defined at four spatial extents: cross-basin, basin, region, and subregion. Across all four spatial extents, oases were more likely to occur in habitats with high light attenuation. The influence of the other environmental predictors on the probability of oasis occurrence were less consistent and varied with the scale of observation. Oases were most likely in areas of low human population density, but this effect was evident only at the cross-basin and subregional extents. At the regional and subregional extents oases were more likely where sea-surface temperature was more variable, whereas at the larger spatial extents the opposite was true. By identifying the correlates of oasis occurrence, the model can inform the prioritization of reef areas for management. Areas with biophysical conditions that confer corals with physiological resilience, as well as limited human impacts, likely support coral reef oases across spatial extents. Our approach is widely applicable to the development of conservation strategies to protect biodiversity and ecosystems in an era of magnified human disturbance.

Bird occupancy in intensively managed agroecosystems under large-scale organic and conventional farming in Argentina: A multi-species approach.

Several studies in European and North American agroecosystems conclude that organic farming benefits birds compared to conventional farming. Nevertheless, there are some biases toward these geographic regions and farm size. Argentinian agroecosystems are particularly homogeneous with large arable fields and sparse uncultivated field margins (i.e. large-scale homogenous cropping systems). In Argentina only 0.55% of the total farmland is under organic farming. Thus, our aims were to assess differences in bird occupancy between organic versus conventional farming regimes, and whether bird occupancy varied in relation to annual crop proportion in both farming regimes in central Argentina agroecosystems. We surveyed 156 points in farms under conventional and 154 in organic farming regimes during two bird-breeding seasons. We used multi-species occupancy models with a Bayesian approach to estimate bird occupancy. We observed that the type of farming regime (organic in relation to conventional) had a weak effect on avian occupancy, varying by species and groups. Probability of occupancy was higher for a few insectivorous and omnivorous species but lower for carnivores in organic farms in relation to conventional ones. The proportion of annual crops was positively correlated with occupancy of an insectivore aerial forager, some insectivore foliage gleaners, a granivore, and some omnivorous species in organic farms, but not conventional farms. This work contributes to reducing geographic and small-scale heterogeneous cropping system biases in the avian agroecological literature. Our results, together with future studies needed to assess landscape configuration and composition, and resource availability for birds in each farming regime, will allow the evaluation of organic farming as a tool for the conservation of bird species in large-scale homogeneous cropping systems in temperate regions.

Using single visits into integrated occupancy models to make the most of existing monitoring programs.

A major challenge in statistical ecology consists of integrating knowledge from different data sets to produce robust ecological indicators. To estimate species distribution, occupancy models are a flexible framework that can accommodate several data sets obtained from different sampling methods. However, repeating visits at sampling sites is a prerequisite for using standard occupancy models. Occupancy models were recently developed to analyze detection/non-detection data collected during a single visit. To date, single-visit occupancy models have never been used to integrate several different data sets. Here, we showcase an approach that combines two data sets into an integrated single-visit occupancy model. As a case study, we estimated the distribution of common bottlenose dolphin (Tursiops truncatus) over the northwestern Mediterranean Sea by combining 24,624 km of aerial surveys and 21,464 km of at-sea monitoring. We compared the outputs of single- vs. repeated-visit occupancy models into integrated occupancy models. Integrated models allowed a better sampling coverage of the targeted population, which provided a better precision for occupancy estimates than occupancy models using data sets in isolation. Overall, single- and repeated-visit integrated occupancy models produced similar inference about the distribution of bottlenose dolphins. We suggest that single-visit occupancy models open promising perspectives for the use of existing ecological data sets.

Limited influence of hunting on the activity patterns and habitat use of Pampas fox (Lycalopex gymnocercus) in agroecosystems of central Argentina.

The Pampas fox (Lycalopex gymnocercus) has suffered from hunting and human persecution for decades, both for fur trade or due to conflicts with livestock. However, studies assessing the effects of hunting pressure on this canid population ecology are lacking. In this work, we assessed the influence of several hunting-related variables on the daily activity patterns and habitat use of the Pampas fox. In private farms of central Argentina, we performed two camera-trapping surveys: one during the non-hunting season and other during the hunting season, distinguishing between sites with or without access by hunters. The habitat use of Pampas foxes did not vary between seasons, but their detectability was lower during the hunting season in habitats that allow humans to see foxes easily (i.e., habitats with high visibility). Pampas foxes selected dusk and night-time, increasing their activity at dusk hours on sites with hunting and showed differences in activity patterns related with the level of visibility of the habitat and to season. Hunting pressure may interact with anthropogenic habitat modifications and create ecological traps for the Pampas fox in agroecosystems.

What is the effect of poaching activity on wildlife species?

Poaching is a pervasive threat to wildlife, yet quantifying the direct effect of poaching on wildlife is rarely possible because both wildlife and threat data are infrequently collected concurrently. In this study, we used poaching data collected through the Management Information System (MIST) and wildlife camera trap data collected by the Tropical Ecology Assessment and Monitoring (TEAM) network from 2014 to 2017 in Volcanoes National Park, Rwanda. We implemented co-occurrence multi-season occupancy models that accounted for imperfect detection to investigate the effect of poaching on initial occupancy, colonization, and extinction of five mammal species. Specifically, we focused on two species of conservation concern (mountain gorilla [Gorilla beringei beringei] and golden monkey [Cercopithecus mitis kandti]), and three species targeted by poachers (black-fronted duiker [Cephalophus nigrifrons], bushbuck [Tragelaphus scriptus], and African buffalo [Syncerus caffer]). We found that the probability of local extinction was highest in sites with poaching activity for golden monkey and bushbuck. In addition, the probability of initial occupancy for golden monkey was highest in sites without poaching activity. We only found weak evidence of effects of poaching on parameters governing the occupancy dynamics of the other species. All species showed evidence of poaching presence affecting the probability of detection of the wildlife species. This is the first study to our knowledge to combine direct threat observations from ranger-based monitoring data with camera trap wildlife observations to quantify the effect of poaching on wildlife. Given the widespread collection of ranger-based monitoring and camera trap data, our approach is broadly applicable to numerous protected areas and has the potential to significantly improve conservation management. Specifically, the relationship between poaching activity and wildlife population dynamics can be combined with information on the relationship between ranger patrols and poaching activity to develop models useful for making wise decisions about ranger patrol deployment.

Can dynamic occupancy models improve predictions of species' range dynamics? A test using Swiss birds.

Predictions of species' current and future ranges are needed to effectively manage species under environmental change. Species ranges are typically estimated using correlative species distribution models (SDMs), which have been criticized for their static nature. In contrast, dynamic occupancy models (DOMs) explicitily describe temporal changes in species' occupancy via colonization and local extinction probabilities, estimated from time series of occurrence data. Yet, tests of whether these models improve predictive accuracy under current or future conditions are rare. Using a long-term data set on 69 Swiss birds, we tested whether DOMs improve the predictions of distribution changes over time compared to SDMs. We evaluated the accuracy of spatial predictions and their ability to detect population trends. We also explored how predictions differed when we accounted for imperfect detection and parameterized models using calibration data sets of different time series lengths. All model types had high spatial predictive performance when assessed across all sites (mean AUC > 0.8), with flexible machine learning SDM algorithms outperforming parametric static and DOMs. However, none of the models performed well at identifying sites where range changes are likely to occur. In terms of estimating population trends, DOMs performed best, particularly for species with strong population changes and when fit with sufficient data, while static SDMs performed very poorly. Overall, our study highlights the importance of considering what aspects of performance matter most when selecting a modelling method for a particular application and the need for further research to improve model utility. While DOMs show promise for capturing range dynamics and inferring population trends when fitted with sufficient data, computational constraints on variable selection and model fitting can lead to reduced spatial accuracy of predictions, an area warranting more attention.

A multi-state occupancy model to non-invasively monitor visible signs of wildlife health with camera traps that accounts for image quality.

Camera traps are an increasingly popular tool to monitor wildlife distributions. However, traditional analytical approaches to camera trap data are difficult to apply to visible wildlife characteristics in single images, such as infection status. Several parasites produce visible signs of infection that could be sampled via camera traps. Sarcoptic mange Sarcoptes scabiei is an ideal disease to study using cameras because it results in visible hair loss and affects a broad host range. Here, we developed a multi-state occupancy model to estimate the occurrence of mange in coyotes Canis latrans across an urban gradient. This model incorporates a secondary detection function for apparent by-image infection status to provide detection-corrected estimates of mange occurrence. We analysed a multi-year camera trap dataset in Chicago, Illinois, United States, to test whether the apparent occurrence of sarcoptic mange in coyotes Canis latrans increases with urbanization or varies through time. We documented visible signs consistent with current or recovering mange infection and variables we hypothesized would improve mange detection: The proportion of the coyote in frame, image blur and whether the image was in colour. We were more likely to detect coyotes with mange in images that were less blurry, in colour, and if a greater proportion of the coyote was visible. Mangy coyote occupancy was significantly higher in urban developed areas with low housing density and higher canopy cover whereas coyote occupancy, mangy or otherwise, decreased with urbanization. By incorporating image quality into our by-image detection function, we provide a robust method to non-invasively survey visible aspects of wildlife health with camera traps. Apparently mangy coyotes were associated with low-density forested neighbourhoods, which may offer vegetated areas while containing sources of anthropogenic resources. This association may contribute to human-wildlife conflict and reinforces posited relationships between infection risk and habitat use. More generally, our model could provide detection-corrected occupancy estimates of visible characteristics that vary by image such as body condition or injuries.