Aerial Surveys Research Articles

Environmental, behavioral, and design-related factors affect accuracy and precision of beluga abundance estimates from aerial surveys

Abundance estimation of wildlife populations is frequently derived from systematic survey data. Accuracy and precision of estimates, however, depend on the number of replicate surveys, and on adjustments made for animals unavailable to (availability bias), or available but undetected (perception bias) by observers. This study offers a comprehensive analysis of the relative influence of methodological, environmental and behavioral factors on availability bias estimates from photographic and visual aerial surveys of a small cetacean with a highly clumped distribution, the beluga (Delphinapterus leucas). It also estimates the effect of the number of surveys on accuracy and precision of abundance estimates, using 28 replicate visual surveys flown within a 16—29 day window depending on survey year. Availability bias was estimated using detailed dive data from 27 beluga from the St. Lawrence Estuary, Canada, and applied to systematic visual and photographic aerial surveys of this population, flown using various survey platforms. Dive and surface interval durations varied among individuals, and averaged (weighted) 176.6 s (weighted s.e. = 12.6 s) and 51.6 s (weighted s.e. = 4.5 s), respectively. Dive time and instantaneous availability, but not surface time, were affected by local turbidity, seafloor depth, whale behavior (i.e., whether beluga were likely in transit or not), and latent processes that were habitat-specific. Overall, adjustments of availability for these effects remained minor compared to effects from survey design (photographic or visual) and type of platform, and observer search patterns. For instance, mean availability varied from 0.33—0.38 among photographic surveys depending on sightings distribution across the study area, but exceeded 0.40 for all visual surveys. Availability also varied considerably depending on whether observers searched within 0-90° (0.42—0.60) or 170° (0.70—0.80). Simulation-based power analysis indicates a large benefit associated with conducting more than 1 or 2 survey reps, but a declining benefit of conducting > 5—10 survey reps. An increase in sample size from 2, to 5, and 10 reps decreased the CV from 30, to 19 and 13%, respectively, and increased the probability of the abundance estimate being within 15% of true abundance from 0.42, to 0.59 and 0.69 in species like beluga.

The harbour porpoise Phocoena phocoena in the Belgian part of the North Sea: trends in abundance and distribution

The harbour porpoise Phocoena phocoena is currently common in the Belgian part of the North Sea (BPNS), after decades of virtual absence. This article describes the results of aerial surveys to assess its distribution and density in Belgian waters, in concert with the first results of tests with passive acoustic monitoring devices (Porpoise Detectors - PoDs), and a basic analysis of strandings. The strandings data over four decades clearly demonstrate an increase of harbour porpoises in the BPNS, with only few stranded animals between 1970 and 1997 (0 to 6 per year), increasing numbers between 1998 and 2004 (8 to 40 per year), and even higher numbers between 2005 and 2009 (62 to 94 per year). The combined results of aerial surveys, strandings monitoring and tests with passive acoustic monitoring reveal a seasonal pattern, with harbour porpoises being abundant from February to April, and more scarce from May to January. Average densities in 2008 and 2009, as estimated by aerial monitoring covering most of the BPNS (with the exclusion of the nearshore 5 km strip), ranged from 0.05 animals per km², or in total less than 200 animals in an area equivalent to the BPNS in August 2009, to 1.01 animals per km², or in total almost 3,700 animals in an area equivalent to the BPNS during April 2008. In the first quarter of the year porpoises occur throughout the BPNS, including territorial waters (12 mile zone), whereas they are restricted to more offshore and northerly waters later in the year. Erratic invasions in the BPNS however blur general seasonal spatio-temporal patterns, which complicates our understanding of spatial distribution and migration.

Smart Cities and The UAVs Applications. A Review Of The Ongoing Research in The UAVs Unit

Nowadays, drones are one of the most important and valuable sources for acquiring aerial imagery and data related to inspection, monitoring, mapping, smart city organization, disaster monitoring and prevention, and 3D modeling. Drones are a low-cost alternative to aerial and space photography because they provide users with images with high spatial resolution down to centimeters. These images enable the production of accurate 3D models such as Digital Surface and Terrain Models (DTM/DSM), contour lines, vector information, etc. This research provides a brief overview of the studies conducted in the Unmanned Aircraft Unit - Remote Sensing Center - University of Mosul. The programs, sequence of operations, and treatments that were followed on the drone images taken by these aircraft were reviewed to prepare them for use in remote sensing and aerial survey operations. This flows towards the use of this technology in the development of smart cities. Specialized GIS software was also used to analyze and process the aerial images that were taken using the DJI Phantom 4 aircraft to study several areas in the city of Mosul, where several sorties were conducted at different periods and times of time.

Quantifying spatiotemporal variation of nearshore forage fish schools with aerial surveys in Prince William Sound, Alaska

AbstractObjectiveChanges in abundance and distribution of schooling forage fish, such as the Pacific Sand Lance Ammodytes hexapterus and Pacific Herring Clupea pallasii, can be difficult to document using traditional boat‐based methods, especially in the shallow, nearshore habitats frequented by these species. In contrast, nearshore fish schools are easily observed and quantified from aircraft when light and sea conditions are favorable. We used aerial shoreline surveys to assess interannual variability in the distribution and abundance of schooling forage fish in Prince William Sound, Alaska, during the summers of 2010 and 2012–2022.MethodsDuring the surveys, aerial observers classified fish schools by their size, species, and (in some cases) age‐class. All observations were georeferenced along the flight path, converted to estimated surface area (m2) based on school diameter, and standardized by effort (shoreline kilometers surveyed).ResultPacific Herring were widely distributed, and school densities varied annually; there were several spikes in school density of up to 54.38 m2/km interspersed among years of lower average densities (7.73–25.57 m2/km). In contrast, Pacific Sand Lance were usually limited in their distribution to a few predictable locations. School density in these consistent areas varied across years, from a high of 50.98 m2/km in 2010 to a low of 0.15 m2/km in 2017. We validated 88 schools during aerial surveys conducted in 2014–2016 and 2019–2022, of which 76 (86%) were correctly identified to species.ConclusionHere, we provide indices of Pacific Herring and Pacific Sand Lance school density over time in shallow nearshore coastal areas of Prince William Sound, Alaska. These indices were generated from aerial surveys, which offer an effective alternative to boat‐based surveys for tracking forage fish schools when they occur in shallow and nearshore coastal habitats.

Uncovering extensive populations of the ‘threatened’ Black Crowned Cranes in Gambella, Ethiopia

A reliable assessment of the population size of threatened species is crucial for making informed decisions about their conservation and management. However, there is a lack of comprehensive knowledge on the population of the eastern Black Crowned Crane (Balearica pavonina ceciliae), which is globally recognized as a threatened taxon experiencing a declining population trend. Here, we investigated the density, abundance and factors that influence flock size in the wetlands of the Gambella region in Ethiopia. We used line transects to collect data during dry and wet seasons. Population density and abundance were estimated using DISTANCE software, whereas factors influencing crane flock size were evaluated using a generalized linear model. We found that crane density ranged from 444 to 965 birds per sq km in the dry season and from 62 to 105 individuals per sq km in the wet season. We estimated that the population size within the designated survey areas as 7634 during the dry season and 866 during the wet season. Furthermore, for the broader survey region that encompasses permanent wetlands, the estimated population ranged from 9138 to 19,849 individuals in the dry season and from 1272 to 2166 individuals in the wet season. We found that grazing intensity and proximity to woodlands and rivers had significant positive impacts on crane flock size, whereas an increase in wetland water depth and grass height remarkably reduced flock size. As a result, wetland habitat degradation-related problems are not currently a threat to cranes in this region, but are likely to increase in the future due to large-scale agricultural expansion. The crane population size we estimated in Gambella is high, three to four times higher than the estimate from the entirety of Ethiopia. Our research underscores the global significance of Gambella wetlands as an important habitat for cranes. However, to obtain an accurate population estimate in this region, it is imperative to include seasonal wetlands, savannah grasslands, farmlands and the extended wetlands of South Sudan, which are hydrologically connected to the wetlands of Gambella region. To achieve this, comprehensive surveys and estimations should incorporate various techniques, such as aerial surveys, satellite telemetry, and high-resolution satellite imagery.

Lepidopteran scales in lake sediments as a reliable proxy for spruce budworm outbreak events in the boreal forest of Eastern Canada

Characterizing disturbance regimes over long time scales is paramount for describing and identifying their variability. The most important biotic disturbance in the eastern Canadian boreal forest is the defoliation caused by the eastern spruce budworm, a moth of the insect order Lepidoptera. Lepidopteran scales have recently been used to reconstruct spruce budworm population fluctuations throughout the Holocene. However, this novel proxy has yet to be compared to an independent proxy. This study aimed to determine whether lepidopteran scales found in the surface sediments of boreal lakes tracked large spruce budworm populations, that is, outbreaks, using yearly aerial surveys (1967–present) of spruce budworm defoliation as an independent proxy. Scales were extracted (1 cm resolution) from the top 20 cm of 210Pb-dated sediment cores recovered from nine lakes. To identify significant abundance peaks of scales in the time series, we removed background noise using a modified version of CharAnalysis. A 100-year smoothing window width combined with a 60th percentile threshold yielded the highest true positive and true negative occurrences, and the lowest false positive and false negative occurrences, with values of 0.69 and 0.70 for Cohen’s Kappa and Matthews correlation coefficient, respectively. Our findings demonstrate that lepidopteran scales are a suitable proxy for identifying spruce budworm outbreaks in the sediment record enabling the reconstruction of budworm and other lepidopteran species outbreak dynamics at millennial timescales.

A negative trend in abundance and an exceeded mortality limit call for conservation action for the Vulnerable Belt Sea harbour porpoise population

The management and conservation of biodiversity relies on information on both the abundance of species and the potential impact of threats. Globally, one of the largest threats towards marine biodiversity is bycatch in fisheries. Under the Marine Strategy Framework Directive (MSFD), EU Member States are required to assess the status of species, such as the harbour porpoise (Phocoena phocoena), in relation to their abundance and mortality due to bycatch every six years. The Vulnerable (HELCOM) Belt Sea population of harbour porpoise has been surveyed to determine its abundance six times using dedicated aerial or ship-based line-transect distance sampling surveys. Here, we estimated the first trend in population abundance over an 18 year period (2005-2022). Using the most recent abundance estimate, we computed a mortality limit applying the modified Potential Biological Removal (mPBR) method based on the regionally agreed conservation objective to restore or maintain 80% of carrying capacity over 100 years with an 80% probability. Over the past 18 years there has been a strong negative trend (-2.7% p.a.; 95% CI: -4.1%; + 1.3%) in abundance, with a 90.5% probability. The mortality limit was estimated to be 24 animals, which the current bycatch estimates (~900 porpoises/year from the commercial Danish and Swedish set net fishery fleets, with no data from Germany and other fishery types) exceed by far. The frequency and quality of data available on abundance for this population are higher than those available for the majority of marine species. Given the observed population decline and likely unsustainable levels of bycatch, the results presented here provide a strong basis to make informed, evidence-based management decisions for action for this population. Such action is needed urgently, before the dire situation of other porpoise species and populations around the globe is repeated.

Temporal changes in habitat use by dugongs in a spatially restricted coral reef environment

Context The dynamic habitat use by dugongs (Dugong dugon) in small coral reef lagoon systems spatially limited by tides is not well understood and has hampered adaptive management. Aims We investigated how dugongs locally used a high conservation value coral-reef lagoon system during different seasons and tides to support local management. Methods We conducted local-scale aerial surveys to document the seasonal and tidal changes in the distribution and number of dugongs over seagrass and non-seagrass coral reef habitats in a small lagoon in New Caledonia. The surveys were conducted fortnightly over 18 months in different seasons and at different tides. Temperature loggers and existing local footage of dugong herding behaviour were used to study the habitat use and behaviour of the animals in the area. Key results More dugongs were sighted in Cap Goulvain during the cool season than during the warm season. As tides restricted access to the intertidal seagrass meadows and during the cool season, more dugongs were sighted outside the lagoon on the fore reef shelf outside the lagoon. Dugongs were resting in large aggregations during their use of this non-seagrass habitat. Conclusions Our study emphasises the importance of non-seagrass habitats for dugongs in spatially restricted coral reef environments as well as the importance of considering outer lagoon habitats as key dugong management areas. This study also adds evidence of behavioural thermoregulation in dugongs. Implications The development of dynamic management strategies is essential, especially where different habitats should be protected at different times of the year.

Erosion Monitoring in Benggang Based on Control-Free Images and Nap-of-the-Object Photogrammetry Techniques

Erosion Monitoring in Benggang Based on Control-Free Images and Nap-of-the-Object Photogrammetry Techniques

Trans-lesion synthesis and mismatch repair pathway crosstalk defines chemoresistance and hypermutation mechanisms in glioblastoma

Almost all Glioblastoma (GBM) are either intrinsically resistant to the chemotherapeutical drug temozolomide (TMZ) or acquire therapy-induced mutations that cause chemoresistance and recurrence. The genome maintenance mechanisms responsible for GBM chemoresistance and hypermutation are unknown. We show that the E3 ubiquitin ligase RAD18 (a proximal regulator of TLS) is activated in a Mismatch repair (MMR)-dependent manner in TMZ-treated GBM cells, promoting post-replicative gap-filling and survival. An unbiased CRISPR screen provides an aerial map of RAD18-interacting DNA damage response (DDR) pathways deployed by GBM to tolerate TMZ genotoxicity. Analysis of mutation signatures from TMZ-treated GBM reveals a role for RAD18 in error-free bypass of O6mG (the most toxic TMZ-induced lesion), and error-prone bypass of other TMZ-induced lesions. Our analyses of recurrent GBM patient samples establishes a correlation between low RAD18 expression and hypermutation. Taken together we define molecular underpinnings for the hallmark tumorigenic phenotypes of TMZ-treated GBM.

First recorded ice entrapment of a beluga whale (Delphinapterus leucas) in east Greenland

Beluga whales are rare along the coast of east Greenland and the closest recognized stock occurs around Svalbard. Here we report on an ice entrapment of an adult beluga whale (Delphinapterus leucas) in north-east Greenland. The whale was observed entrapped in the fast ice on 21 April 2023 in Loch Fyne (73°54’N, 21°51’W) during a visual aerial survey for polar bears (Ursus maritimus). The whale was located >100 km from open water (i.e., pack ice) and appeared in poor body condition. A literature review back to the early 1900s failed to produce any other records of beluga whale ice entrapments in east Greenland.

Optimization strategy for flexible barrier structures: investigation and back analysis of a rockfall disaster case in southwestern China

Abstract. Field investigations and back analyses were conducted on a rockfall hazard. The flexible barrier protection system constructed along the roadside was damaged by the rockfall impact and lost its mitigation ability. Vital physical characteristics such as rockfall trajectory and kinetic energy were presumed based on the data from the aerial survey and the slope digital model. A numerical model, including slope, rockfalls, and flexible barrier, was created and thus the impacting process was reproduced. It demonstrates that the impact kinetic energy of the rockfall is only around 40 % of its design protection energy. The improper connections of members are the leading causes of damage, which prevent the flexible barrier from producing significant deformation and reduce its capacity to absorb impact force. The damage can be avoided by changing the connections of the members to improve the ability of the nets and ropes to slide and deform. The calculation results indicate that the impact resistance of the optimized model is 3 times better than the actual project. The findings can be used as a guide when designing a flexible protection system that performs better.

Motion‐sensitive cameras track population abundance changes in a boreal mammal community in southwestern Yukon, Canada

AbstractMotion‐sensitive cameras are commonly used to monitor wildlife occupancy rates; however, few studies have assessed whether data from cameras are correlated with density estimates obtained from more traditional labor‐intensive methods such as those based on capture‐mark‐recapture. We used data from a boreal forest community to test whether camera data were correlated with densities estimated from independent monitoring methods. We placed 72 covert cameras in the forest around Lhù'ààn Mân' (Kluane Lake), Yukon, Canada, for 7 years and tracked changes in population densities by camera hit rates. We independently estimated population densities of snowshoe hares (Lepus americanus) and red squirrels (Tamiasciurus hudsonicus) using capture‐mark‐recapture via live trapping, and Canada lynx (Lynx canadensis), coyotes (Canis latrans), and moose (Alces americanus) by snow track transects. Density estimates obtained from conventional aerial surveys were also periodically available for moose. Except for red squirrels, camera hit rates were highly correlated with population density estimates obtained by traditional methods, including across a large range of estimated densities corresponding to cyclic population dynamics in several species. Accordingly, we infer that motion‐sensitive cameras could supplement or replace traditional methods for monitoring key species in boreal forest food webs. Using cameras to monitor population change has several advantages; they require less effort in the field, are non‐invasive compared to live‐trapping, include multiple species at the same time, and rely less on weather than either aerial surveys or snow track transects. Tracking changes across the vast boreal forest is becoming increasingly necessary because of climate and landscape change and our data validate the use of motion‐sensitive cameras to provide a useful quantitative method for state‐of‐the‐environment reporting.

Evaluation of branch sampling, ocular assessments, and aerial surveys for estimating spruce budworm defoliation

We compared three methods for estimating current-year spruce budworm ( Choristoneura fumiferana (Clem.)) defoliation from 2014 to 2021 using a network of 99 permanent sample plots in central Gaspé Peninsula, Québec. Percent current-year defoliation was measured by assessing shoots from mid-crown branches, ocular ratings of all individual trees using binoculars, and provincial government aerial surveys. Ocular survey defoliation differed from branch sample defoliation in 5–6 out of 7 years, consistently underestimating defoliation, across the full range of defoliation severity observed. Nested mixed-effect models for fir–spruce combined, balsam fir, white spruce, and black spruce ocular survey defoliation bias resulted in marginal R2 of 0.40, 0.47, 0.82, and 0.86, respectively. Current defoliation severity and its interaction with previous year defoliation and weather conditions significantly affected ocular survey bias. Correspondence of aerial survey estimates and mean plot defoliation occurred in only 43% of all plot-years and ranged from 14%–58% in individual years. Differences between aerial survey defoliation and plot values mainly resulted from assigning an adjacent class (e.g., light <30% assigned as moderate 31%–70% defoliation) or misplaced defoliation polygon boundaries, suggesting that assignment of aerial survey defoliation to plots or specific ground areas needs ground truth sampling.

Unmanned air/ground vehicle survey following a radiological dispersal event

AbstractThis paper presents the method and results of surveying a dispersed radioactive field using an unmanned air vehicle (UAV) and an unmanned ground vehicle (UGV). A 35 GBq of La material was distributed in a specific geometric l‐polygon pattern measuring a 120 m 20 m longitudinal rectangle and an 80 m 10 m transverse rectangle. Two methods were used to determine the amount and distribution of the lanthanum over the polygon including 20 plywood coupons distributed over the area and a UGV equipped with a Kromek GR1® driving over the area. The aerial survey was conducted using an unmanned aerial vehicle UAV carrying the Kromek GR1® while flying a traditional grid pattern and circular pattern at different elevations and speeds over the area. The data collected by the UAV were further postprocessed using N‐Visage, a 3D radiation modeling method developed by Createc, to create a model of the ground activity. This model was compared with the reference data collected on the ground by the UGV and was found to be in agreement with 11%.

Evaluating the Use of a Thermal Sensor to Detect Small Ground-Nesting Birds in Semi-Arid Environments during Winter

Aerial wildlife surveys with fixed-wing airplanes and helicopters are used more often than on-the-ground field surveys to cover areas that are both extensive and often inaccessible. Drones with high-resolution thermal sensors are being widely accepted as research tools to aid in monitoring wildlife species and their habitats. Therefore, our goal was to assess the feasibility of detecting northern bobwhite quail (Colinus virginianus, hereafter ‘bobwhite’) using drones with a high-resolution thermal sensor. Our objectives were (1) to identify the altitudes at which bobwhites can be detected and (2) compare the two most used color palettes to detect species (black-hot and isotherm). We achieved this goal by performing drone flights at different altitudes over caged tame bobwhites and capturing still images and video recordings at altitudes from 18 to 42 m. We did not observe or detect any obvious signs of distress, movement, or fluttering of bobwhites inside cages caused by the noise or presence of the drone during data acquisition. We observed the highest counts of individual bobwhites with the black-hot thermal palette at 18 m (92%; x¯ = 47 bobwhites; SE = 0.41) and at 24 m (81%; x¯ = 41 bobwhites; SE = 0.89). The isotherm thermal palette had lower count proportions. The use of video to count quail was not feasible due to the low resolution of the video and the species size. Flying drones with high-resolution thermal sensors provided reliable imagery to detect roosting bobwhite individuals in South Texas during the winter.

Mottled ducks, feral mallards, and their hybrids in Florida

AbstractThe nonmigratory and endemic Florida mottled duck (Anas fulvigula fulvigula) is facing conservation threats from the combined effects of urbanization and introgressive hybridization with feral mallards (Anas platyrhynchos) and mallard x mottled duck hybrids. In the past, the status of the Florida mottled duck population was assessed during annual aerial surveys and most brown ducks (mottled ducks, mallards, and hybrids of them) detected during the survey would have been mottled ducks. But the release of domesticated mallards for aesthetic purposes has led to increases in the prevalence of mallards‐hybrids (mallards or mallard x mottled duck hybrids) throughout peninsular Florida, USA, and because it is impossible to differentiate among mottled ducks, female mallards, and hybrids during aerial surveys, helicopter surveys were halted in 2009 until state researchers could conduct a range‐wide study to determine what proportion of brown ducks are mottled ducks versus mallards‐hybrids. We used plumage keys and high‐resolution photography to categorize brown ducks from 557 wetland grid points as either mottled ducks or mallards‐hybrids. Of the 5,179 brown ducks categorized, 40.1% were mottled ducks and 59.9% were mallards‐hybrids. We used logistic regression analysis to model the interactive effect of a site's latitude and level of urbanization (urban gradient value within a 2‐km buffer) to generate a predictive raster surface (1‐km resolution) of the study area with values corresponding to the probability that a brown duck observed within a cell is a pure mottled duck. Predicted values will be used as correction factors when estimating final mottled duck population abundance from brown‐duck survey data. Additionally, the predictive raster surface will be used to identify wetlands where mottled ducks remain predominant so that these sites can be targeted for preservation. Overall, mallards‐hybrids outnumbered mottled ducks throughout most of peninsular Florida, especially in more urbanized regions, and their current prevalence rate presents a serious conservation threat, via hybridization, to extant mottled duck populations.

Review of Satellite Remote Sensing and Unoccupied Aircraft Systems for Counting Wildlife on Land

Although many medium-to-large terrestrial vertebrates are still counted by ground or aerial surveys, remote-sensing technologies and image analysis have developed rapidly in recent decades, offering improved accuracy and repeatability, lower costs, speed, expanded spatial coverage and increased potential for public involvement. This review provides an introduction for wildlife biologists and managers relatively new to the field on how to implement remote-sensing techniques (satellite and unoccupied aircraft systems) for counting large vertebrates on land, including marine predators that return to land to breed, haul out or roost, to encourage wider application of these technological solutions. We outline the entire process, including the selection of the most appropriate technology, indicative costs, procedures for image acquisition and processing, observer training and annotation, automation, and citizen science campaigns. The review considers both the potential and the challenges associated with different approaches to remote surveys of vertebrates and outlines promising avenues for future research and method development.

The ACCOBAMS survey initiative: the first synoptic assessment of cetacean abundance in the Mediterranean Sea through aerial surveys

The “ACCOBAMS Survey Initiative” (ASI) is a pilot programme aimed at establishing an integrated and coordinated monitoring system for cetaceans across the Agreement on the Conservation of Cetaceans of the Black Sea, Mediterranean Sea and contiguous Atlantic (hereafter “ACCOBAMS”) area. Conducted in coordination with Mediterranean coastal countries, it supports the implementation of European and regional policies, in particular the EU Marine Strategy Framework Directive and the Ecosystem Approach process. In summer 2018, a synoptic survey was conducted across the Mediterranean Sea and contiguous Atlantic area, combining visual monitoring from aircrafts with visual and passive acoustic monitoring from vessels. Species density and abundance were estimated through design-based approach in a line-transect sampling framework. Based on data arising from the aerial survey only, uncorrected design-based abundance was obtained for striped (N=426,744, CV=0.13), common (N=65,359, CV=0.4), bottlenose (N=63,333, CV=0.17), and Risso´s dolphins (N=26,006, CV=0.3), Cuvier’s beaked whales (N=2,929, CV=0.4) and long-finned pilot whales (N=5,540 CV=0.4). A merged category of either striped or common dolphins resulted in 212,828 individuals (CV=0.26). Fin whales abundance of 1,749 animals (CV=0.3) was corrected for both availability and perception biases and resulted in 3,282 (CV=0.31). The ASI survey offers an overall picture of the distribution and abundance of cetaceans throughout the Mediterranean basin, providing robust estimates to be considered as a baseline for future regional systematic monitoring programmes. The ASI survey is the first step towards establishing a long-term monitoring program across the entire ACCOBAMS area, and, as such, it sets the basis for further future basin-wide monitoring efforts using systematic, shared, coordinated and comparable methods. The information gathered will further enhance knowledge on cetacean status, facilitating the development of informed conservation and mitigation measures, as well as supporting the implementation of international obligations. Furthermore, the outcomes of this survey will support both place- and threat-based conservation efforts in the ACCOBAMS area, through the identification of Important Marine Mammal Areas and Cetacean Critical Habitats. Here the results of the ASI survey are presented and discussed alongside proposed management and conservation actions aimed at ensuring the persistence of cetacean populations in the region.

Detection Probability and Bias in Machine-Learning-Based Unoccupied Aerial System Non-Breeding Waterfowl Surveys

Unoccupied aerial systems (UASs) may provide cheaper, safer, and more accurate and precise alternatives to traditional waterfowl survey techniques while also reducing disturbance to waterfowl. We evaluated availability and perception bias based on machine-learning-based non-breeding waterfowl count estimates derived from aerial imagery collected using a DJI Mavic Pro 2 on Missouri Department of Conservation intensively managed wetland Conservation Areas. UASs imagery was collected using a proprietary software for automated flight path planning in a back-and-forth transect flight pattern at ground sampling distances (GSDs) of 0.38–2.29 cm/pixel (15–90 m in altitude). The waterfowl in the images were labeled by trained labelers and simultaneously analyzed using a modified YOLONAS image object detection algorithm developed to detect waterfowl in aerial images. We used three generalized linear mixed models with Bernoulli distributions to model availability and perception (correct detection and false-positive) detection probabilities. The variation in waterfowl availability was best explained by the interaction of vegetation cover type, sky condition, and GSD, with more complex and taller vegetation cover types reducing availability at lower GSDs. The probability of the algorithm correctly detecting available birds showed no pattern in terms of vegetation cover type, GSD, or sky condition; however, the probability of the algorithm generating incorrect false-positive detections was best explained by vegetation cover types with features similar in size and shape to the birds. We used a modified Horvitz–Thompson estimator to account for availability and perception biases (including false positives), resulting in a corrected count error of 5.59 percent. Our results indicate that vegetation cover type, sky condition, and GSD influence the availability and detection of waterfowl in UAS surveys; however, using well-trained algorithms may produce accurate counts per image under a variety of conditions.