Understanding distributions of invasive mammals in urban environments using remote cameras and citizen science

<p>Preserving biodiversity in urban environments is crucial not only for preventing local extinctions of native species, but also for educating the public about the importance of species conservation. Invasive mammalian species can have negative impacts for both people and biodiversity in urban environments. Understanding the factors influencing the distribution of these invasive species is crucial to comply with the ethical, ecological, and practical concerns associated with their management. Remote cameras are an increasingly popular tool for investigating the distribution and abundance of mammals. Yet few studies have used these cameras in urban environments. The time and effort required to classify remote camera data is the main constraint of this monitoring technique. To determine whether employing citizen science could facilitate the use of remote cameras in urban environments, I investigated the engagement, accuracy, and efficiency of volunteers (i.e., citizen scientists) in classifying animal images recorded by remote cameras in Wellington, New Zealand. Classifications from citizen scientists were in 84.2% agreement with classifications of expert ecologists. However, accuracy varied significantly among species and volunteers. Aggregating multiple classifications per image and highlighting animal movement in the images improved the accuracy of citizen scientists. Additionally, weighting their classifications based on previous accuracy, self-assessed confidence, and the species reported reduced the number of volunteer classifications required to achieve levels of accuracy comparable to that of experts. These results illustrate that citizen science allows for accurate and efficient classifications of remote camera data from urban areas. Using the classifications provided by citizen scientists, I then evaluated the suitability of remote cameras to monitor invasive mammals in urban environments. Based on data collected from forest and residential areas of Wellington, New Zealand, remote cameras detected significantly more European hedgehogs (Erinaceus europaeus) and rats (Rattus spp.) than tracking tunnels. Cameras, however, missed recording house mice (Mus musculus) on some occasions where tracking tunnels detected them, and vice-versa. Overall, my results demonstrate that remote cameras are a more efficient multi-species monitoring tool than tracking tunnels. Independent of habitat type, cats (Felis catus), hedgehogs, and mice were the species most frequently recorded. Data from remote cameras subsequently helped quantify differences in the occupancy rates of species between residential and forested areas furthering our ecological understanding of the distribution of invasive species in peopled landscapes. To identify the underlying processes influencing the distribution and abundances of invasive mammals found in urban patches of vegetation, I also used remote cameras to investigate the influence of habitat quality, management efforts, interspecific interactions and seasonality on the occupancy and relative abundance of invasive mammals in 47 patches of forest within Wellington. My results indicate that distance to forest edge influences positively on the relative abundance of rodents and negatively on the relative abundance of common brushtail possums (Trichosurus vulpecula), cats, European rabbits (Oryctolagus cuniculus), and hedgehogs. The cameras also revealed a positive interaction between the occupancy of ship rats (Rattus rattus) and the abundance of Norway rats (Rattus norvegicus), a positive influence of the nearby buildings on the occupancy of cats, and how management control reduces the occupancy of target species, particularly during spring. These results illustrate the importance of using season- and species-specific approaches to identify the most important factors influencing the distribution of invasive species in urban environments. Overall, my research highlights the benefits of engaging the public with scientific research, the advantages of using remote cameras to monitor mammals in urban environments and the importance of controlling invasive species at adequate spatial and temporal scales to ensure effective conservation management.</p>

Understanding distributions of invasive mammals in urban environments using remote cameras and citizen science

<p>Preserving biodiversity in urban environments is crucial not only for preventing local extinctions of native species, but also for educating the public about the importance of species conservation. Invasive mammalian species can have negative impacts for both people and biodiversity in urban environments. Understanding the factors influencing the distribution of these invasive species is crucial to comply with the ethical, ecological, and practical concerns associated with their management. Remote cameras are an increasingly popular tool for investigating the distribution and abundance of mammals. Yet few studies have used these cameras in urban environments. The time and effort required to classify remote camera data is the main constraint of this monitoring technique. To determine whether employing citizen science could facilitate the use of remote cameras in urban environments, I investigated the engagement, accuracy, and efficiency of volunteers (i.e., citizen scientists) in classifying animal images recorded by remote cameras in Wellington, New Zealand. Classifications from citizen scientists were in 84.2% agreement with classifications of expert ecologists. However, accuracy varied significantly among species and volunteers. Aggregating multiple classifications per image and highlighting animal movement in the images improved the accuracy of citizen scientists. Additionally, weighting their classifications based on previous accuracy, self-assessed confidence, and the species reported reduced the number of volunteer classifications required to achieve levels of accuracy comparable to that of experts. These results illustrate that citizen science allows for accurate and efficient classifications of remote camera data from urban areas. Using the classifications provided by citizen scientists, I then evaluated the suitability of remote cameras to monitor invasive mammals in urban environments. Based on data collected from forest and residential areas of Wellington, New Zealand, remote cameras detected significantly more European hedgehogs (Erinaceus europaeus) and rats (Rattus spp.) than tracking tunnels. Cameras, however, missed recording house mice (Mus musculus) on some occasions where tracking tunnels detected them, and vice-versa. Overall, my results demonstrate that remote cameras are a more efficient multi-species monitoring tool than tracking tunnels. Independent of habitat type, cats (Felis catus), hedgehogs, and mice were the species most frequently recorded. Data from remote cameras subsequently helped quantify differences in the occupancy rates of species between residential and forested areas furthering our ecological understanding of the distribution of invasive species in peopled landscapes. To identify the underlying processes influencing the distribution and abundances of invasive mammals found in urban patches of vegetation, I also used remote cameras to investigate the influence of habitat quality, management efforts, interspecific interactions and seasonality on the occupancy and relative abundance of invasive mammals in 47 patches of forest within Wellington. My results indicate that distance to forest edge influences positively on the relative abundance of rodents and negatively on the relative abundance of common brushtail possums (Trichosurus vulpecula), cats, European rabbits (Oryctolagus cuniculus), and hedgehogs. The cameras also revealed a positive interaction between the occupancy of ship rats (Rattus rattus) and the abundance of Norway rats (Rattus norvegicus), a positive influence of the nearby buildings on the occupancy of cats, and how management control reduces the occupancy of target species, particularly during spring. These results illustrate the importance of using season- and species-specific approaches to identify the most important factors influencing the distribution of invasive species in urban environments. Overall, my research highlights the benefits of engaging the public with scientific research, the advantages of using remote cameras to monitor mammals in urban environments and the importance of controlling invasive species at adequate spatial and temporal scales to ensure effective conservation management.</p>

A first assessment of the distribution and abundance of large pelagic species at Cocos Ridge seamounts (Eastern Tropical Pacific) using drifting pelagic baited remote cameras

Understanding the link between seamounts and large pelagic species (LPS) may provide important insights for the conservation of these species in open water ecosystems. The seamounts along the Cocos Ridge in the Eastern Tropical Pacific (ETP) ocean are thought to be ecologically important aggregation sites for LPS when moving between Cocos Island (Costa Rica) and Galapagos Islands (Ecuador). However, to date, research efforts to quantify the abundance and distribution patterns of LPS beyond the borders of these two oceanic Marine Protected Areas (MPAs) have been limited. This study used drifting-pelagic baited remote underwater video stations (BRUVS) to investigate the distribution and relative abundance of LPS at Cocos Ridge seamounts. Our drifting-pelagic BRUVS recorded a total of 21 species including elasmobranchs, small and large teleosts, dolphins and one sea turtle; of which four species are currently threatened. Depth of seamount summit was the most significant driver for LPS richness and abundance which were significantly higher at shallow seamounts (< 400 m) compared to deeper ones (> 400m). Distance to nearest MPA was also a significant predictor for LPS abundance, which increased at increasing distances from the nearest MPA. Our results suggest that the Cocos Ridge seamounts, specifically Paramount and West Cocos which had the highest LPS richness and abundance, are important aggregation sites for LPS in the ETP. However, further research is still needed to demonstrate a positive association between LPS and Cocos Ridge seamounts. Our findings showed that drifting pelagic BRUVS are an effective tool to survey LPS in fully pelagic ecosystems of the ETP. This study represents the first step towards the standardization of this technique throughout the region.

The 2019 Eruptive Activity at Stromboli Volcano: A Multidisciplinary Approach to Reveal Hidden Features of the “Unexpected” 3 July Paroxysm

In July and August 2019, Stromboli volcano underwent two dangerous paroxysms previously considered “unexpected” because of the absence of significant changes in usually monitored parameters. We applied a multidisciplinary approach to search for signals able to indicate the possibility of larger explosive activity and to devise a model to explain the observed variations. We analysed geodetic data, satellite thermal data, images from remote cameras and seismic data in a timespan crossing the eruptive period of 2019 to identify precursors of the two paroxysms on a medium-term time span (months) and to perform an in-depth analysis of the signals recorded on a short time scale (hours, minutes) before the paroxysm. We developed a model that explains the observations. We call the model “push and go” where the uppermost feeding system of Stromboli is made up of a lower section occupied by a low viscosity, low density magma that is largely composed of gases and a shallower section occupied by the accumulated melt. We hypothesize that the paroxysms are triggered when an overpressure in the lower section is built up; the explosion will occur at the very moment such overpressure overcomes the confining pressure of the highly viscous magma above it.

Variable strategies to solve risk–reward tradeoffs in carnivore communities

Mesopredator release theory suggests that dominant predators suppress subordinate carnivores and ultimately shape community dynamics, but the assumption that subordinate species are only negatively affected ignores the possibility of facilitation through scavenging. We examined the interplay within a carnivore community consisting of cougars, coyotes, black bears, and bobcats using contemporaneous Global Positioning System telemetry data from 51 individuals; diet analysis from 972 DNA-metabarcoded scats; and data from 128 physical investigations of cougar kill sites, 28 of which were monitored with remote cameras. Resource provisioning from competitively dominant cougars to coyotes through scavenging was so prolific as to be an overwhelming determinant of coyote behavior, space use, and resource acquisition. This was evident via the strong attraction of coyotes to cougar kill sites, frequent scavenging of cougar-killed prey, and coyote diets that nearly matched cougars in the magnitude of ungulate consumption. Yet coyotes were often killed by cougars and used space to minimize encounters, complicating the fitness benefits gained from scavenging. We estimated that 23% (95% CI: 8 to 55%) of the coyote population in our study area was killed by cougars annually, suggesting that coyote interactions with cougars are a complex behavioral game of risk and reward. In contrast, we found no indication that bobcat space use or diet was influenced by cougars. Black bears avoided cougars, but there was no evidence of attraction to cougar kill sites and much lower levels of ungulate consumption and carcass visitation than for coyotes. Interspecific interactions among carnivores are multifaceted, encompassing both suppression and facilitation.

A Systematic Deep Learning Based Overhead Tracking and Counting System Using RGB-D Remote Cameras

Automatic head tracking and counting using depth imagery has various practical applications in security, logistics, queue management, space utilization and visitor counting. However, no currently available system can clearly distinguish between a human head and other objects in order to track and count people accurately. For this reason, we propose a novel system that can track people by monitoring their heads and shoulders in complex environments and also count the number of people entering and exiting the scene. Our system is split into six phases; at first, preprocessing is done by converting videos of a scene into frames and removing the background from the video frames. Second, heads are detected using Hough Circular Gradient Transform, and shoulders are detected by HOG based symmetry methods. Third, three robust features, namely, fused joint HOG-LBP, Energy based Point clouds and Fused intra-inter trajectories are extracted. Fourth, the Apriori-Association is implemented to select the best features. Fifth, deep learning is used for accurate people tracking. Finally, heads are counted using Cross-line judgment. The system was tested on three benchmark datasets: the PCDS dataset, the MICC people counting dataset and the GOTPD dataset and counting accuracy of 98.40%, 98%, and 99% respectively was achieved. Our system obtained remarkable results.

American martens use vigilance and short-term avoidance to navigate a landscape of fear from fishers at artificial scavenging sites

Where two sympatric species compete for the same resource and one species is dominant, there is potential for the subordinate species to be affected through interference competition or energetic costs of avoiding predation. Fishers (Pekania pennanti) and American martens (Martes americana) often have high niche overlap, but fishers are considered dominant and potentially limiting to martens. We observed presence and vigilance of fishers and martens at winter carcass sites using remote cameras in Michigan, USA, to test the hypothesis that interference competition from fishers creates a landscape of fear for martens. Within winters, fishers co-occupied 78–88% of sites occupied by martens, and martens co-occupied 79–88% of sites occupied by fishers. Fishers displaced martens from carcasses during 21 of 6117 marten visits, while martens displaced fishers during 0 of 1359 fisher visits. Martens did not alter diel activity in response to fisher use of sites. Martens allocated 37% of time to vigilance compared to 23% for fishers, and martens increased vigilance up to 8% at sites previously visited by fishers. Fishers increased vigilance by up to 8% at sites previously visited by martens. Our results indicate that fishers were dominant over martens, and martens had greater baseline perception of risk than fishers. However, fishers appeared to be also affected as the dominant competitor by putting effort into scanning for martens. Both species appeared widespread and common in our study area, but there was no evidence that fishers spatially or temporally excluded martens from scavenging at carcasses other than occasional short-term displacement when a fisher was present. Instead, martens appeared to mitigate risk from fishers by using vigilance and short-term avoidance. Multiple short-term anti-predator behaviors within a landscape of fear may facilitate coexistence among carnivore species.

Dingoes dining with death

ABSTRACT Dingoes (Canis dingo) are known for hunting and killing animals to meet their energetic requirements, but like almost all predators they also scavenge animal remains. To improve our understanding of dingo scavenging ecology, we investigated the role of abiotic and biotic factors in shaping carcass utilisation by dingoes and further determined whether dingo scavenging influenced carcass persistence in the landscape. To do so, we monitored visitation and scavenging by dingoes using remote cameras positioned on 119 kangaroo carcasses in open and closed canopy habitats and in warm and cool seasons. The carcasses were monitored across multiple study sites, which incorporated forest, alpine and desert ecoregions in Australia. We found that season played an important role in shaping carcass utilisation by dingoes, as well as carcass persistence. Warmer seasons increased the rate of carcass discovery 6.3-fold in the Forest study site and 4.8-fold in the Alpine study site, and also increased the time dingoes spent feeding on carcasses in the Alpine study site. Further, across all study sites, carcasses persisted at least 4.7 times longer in cool compared with warm seasons. On the other hand, carcass utilisation by dingoes was not influenced by habitat, although carcasses were more likely to persist in open compared with closed canopy habitats in the Alpine study site. Finally, our study showed that dingo scavenging may contribute to substantial carcass removal in certain contexts. Indeed, decreased carcass persistence in the Forest study site was evident in the cool season, when dingo scavenging occurred during the first two weeks of monitoring. The variability in results highlights the complexity of patterns in dingo scavenging and, more broadly, of vertebrate scavenging. It emphasises the need to consider multiple abiotic and biotic factors to properly understand the functional roles of different scavenger species. Longer-term studies with additional seasonal replicates may also yield a more detailed picture of the role of dingoes as apex scavengers.

Accuracy and precision of citizen scientist animal counts from drone imagery

Repeated counts of animal abundance can reveal changes in local ecosystem health and inform conservation strategies. Unmanned aircraft systems (UAS), also known as drones, are commonly used to photograph animals in remote locations; however, counting animals in images is a laborious task. Crowd-sourcing can reduce the time required to conduct these censuses considerably, but must first be validated against expert counts to measure sources of error. Our objectives were to assess the accuracy and precision of citizen science counts and make recommendations for future citizen science projects. We uploaded drone imagery from Año Nuevo Island (California, USA) to a curated Zooniverse website that instructed citizen scientists to count seals and sea lions. Across 212 days, over 1,500 volunteers counted animals in 90,000 photographs. We quantified the error associated with several descriptive statistics to extract a single citizen science count per photograph from the 15 repeat counts and then compared the resulting citizen science counts to expert counts. Although proportional error was relatively low (9% for sea lions and 5% for seals during the breeding seasons) and improved with repeat sampling, the 12+ volunteers required to reduce error was prohibitively slow, taking on average 6 weeks to estimate animals from a single drone flight covering 25 acres, despite strong public outreach efforts. The single best algorithm was ‘Median without the lowest two values’, demonstrating that citizen scientists tended to under-estimate the number of animals present. Citizen scientists accurately counted adult seals, but accuracy was lower when sea lions were present during the summer and could be confused for seals. We underscore the importance of validation efforts and careful project design for researchers hoping to combine citizen science with imagery from drones, occupied aircraft, and/or remote cameras.