Infrared camera trapping technology has been widely employed to monitor bird diversity. However, little is currently known about the extent to which this technique contributes to regional bird distribution and abundance, the nature of the sampling biases inherent in such monitoring, and the functional traits that give rise to these biases. We utilised monitoring data from 91 nature reserves across China and integrated avian taxonomy, functional traits, and phylogeny. Bayesian and random forest models were employed to investigate detection bias and identify the key traits influencing the detection of birds by camera trapping. The results show that: (1) camera trapping recorded approximately 42% of China’s forest bird species, with variations in detection frequency across different orders and families; specifically, the families Phasianidae, Muscicapidae, Corvidae and Leiothrichidae were detected more frequently than all other families; (2) the avian community detected by camera traps showed clear functional trait divergence, yet remained predominantly assembled by stochastic processes; (3) Bayesian and random forest models identified tarsometatarsal length and feeding behaviour were key predictors of the probability of being detected by the camera, and there was a significant interaction between the them. In particular, tarsometatarsal length (β = 0.04, 95% CI [0.02, 0.06]) and herbivory behaviour (β = 0.03, 95% CI [0.01, 0.04]) showed a positive effect, whereas vertebrate behaviour (β = −0.04, 95% CI [−0.07, −0.02]) had a negative effect, while other traits played only a minor role. These findings highlight the inherent detection biases associated with birds monitoring using camera trapping, providing a useful theoretical foundation and methodological guidance for targeted bias correction and the optimized application of camera trapping technology in bird diversity assessment.
Complex cognition is fundamental to animal survival and reproduction, and understanding its neural mechanisms remains a central goal in neuroscience. Birds exhibiting advanced cognitive abilities provide a unique model for such investigations, owing to their independent evolutionary history and markedly distinct brain architecture compared with mammals. Because complex cognitive functions rely on dynamic interactions among distributed brain-wide networks, functional magnetic resonance imaging (fMRI) has become a promising approach for investigating these processes. However, obtaining reliable fMRI data in a species for which fMRI methodologies have not yet been established generally requires the development of a species-specific and standardized analytical framework, owing to inherent neuroanatomical differences. Here, we developed the first open-source fMRI data analysis platform for parrots, using the Budgerigar (Melopsittacus undulatus) as a representative model species. Specifically, we constructed a high-resolution three-dimensional brain atlas and a standardized T2-weighted structural template based on MRI data acquired from 13 individuals. Building upon these resources, we developed an integrated fMRI analysis platform that automates key processing steps from raw data to spatially normalized datasets and generates atlas-based anatomical reports for statistical results. To validate the applicability of the platform, we employed a well-established visual stimulation paradigm and demonstrated that it can detect and anatomically localize expected task-related activation within the avian visual pathway using a standardized analytical framework. Together, these resources establish an integrated analysis platform and provide a technical foundation for rigorous and reproducible whole-brain functional imaging in parrots. By facilitating standardized data processing and anatomical interpretation, this platform provides an important methodological resource for avian neuroscience and supports future comparative studies on the neural basis and evolution of complex cognition.
Understanding how genomic differentiation accumulates during speciation is fundamental to elucidating the speciation process. Here, we compare whole-genome differentiation across six accentor species (Prunella) spanning different stages of speciation. Using species-specific PBS statistics, we identify genomic islands of elevated differentiation. These islands are wider in anciently divergent species than in recently divergent species, supporting their expansion over time via divergence hitchhiking. Across all six species, the islands consistently exhibit reduced nucleotide diversity, recombination rates, Tajima’s D, along with elevated codon sequence density and inbreeding. In recently divergent species, islands show reduced absolute divergence (DXY), whereas in anciently divergent species they show elevated DXY. This contrast indicates that initial island formation is driven by linked selection reducing ancestral diversity, while long-term maintenance and growth involve progressive accumulation of absolute divergence. Furthermore, the relative contributions of recombination and nucleotide diversity shift with divergence time: their interaction best explains differentiation in recently divergent species, whereas nucleotide diversity alone dominates in anciently divergent species. Thus, reduced nucleotide diversity, initially a consequence of linked selection and demographic decline such as inbreeding, later a driver of island expansion, becomes increasingly central as speciation proceeds. Together, these findings reveal that the interplay between nucleotide diversity and recombination landscape shapes expansion of genomic islands during speciation with gene flow.
Passive acoustic monitoring (PAM) for birds has become a vital approach in ecological monitoring due to its non-invasiveness and capability for long-term continuous observation. However, in practical regional monitoring applications, existing general bird sound classification models are constrained by habitat heterogeneity, variations in bird community composition, and the scarcity of locally labeled samples. Consequently, these models cannot be directly applied to specific regions while maintaining stable classification performance. To solve this problem, we propose a prompt-enhanced bird sound classification method tailored to localized few-shot scenarios. Based on the contrastive language-audio pretraining (CLAP) model, the proposed method improves the representation and discriminative capability of local bird sound acoustic features from two dimensions: textual semantic prompts and audio feature prompts. To evaluate the effectiveness of the proposed method, bird sound recordings were collected from Baiyun Mountain, Shimen National Forest Park, and Wanzuitou Wetland Park in Guangzhou to construct a local dataset. Comparative experiments were conducted between the proposed method and the prompt learning for audio language models (PALM) baseline, the fine-tuned BirdNET model, and the original BirdNET model. The experimental results demonstrate that under the few-shot setting, where only 16 clips of 3-s bird sound recordings per species were used for training, the proposed method achieved a Top-1 accuracy of 82.07% on the Guangzhou local dataset. It outperformed the fine-tuned BirdNET model, the PALM baseline, and the original BirdNET model by 2.32%, 6.26%, and 17.89% respectively. The findings verify that the few-shot modeling strategy integrating prompt learning and audio-text alignment can effectively boost regional bird sound classification performance. It can provide technical support for local bird diversity monitoring and acoustic surveys of rare bird species.
Avian haemosporidian parasites are widespread vector-borne pathogens that can influence host fitness, survival, and population dynamics. However, infection patterns and parasite lineage diversity remain poorly documented in many Asian bird populations, particularly within single host species sampled across contrasting geographic contexts. Here, we investigated haemosporidian infections in the Daurian Redstart (Phoenicurus auroreus), a widespread passerine in eastern China. Between 2022 and 2025, we sampled 508 individuals at four locations during breeding and non-breeding periods and screened for infections using nested PCR and mitochondrial cytochrome b sequencing. Overall infection prevalence was 30.31%, with Plasmodium as the dominant genus, followed by Leucocytozoon and Haemoproteus. We identified 44 parasite lineages, including 19 novel lineages, indicating substantial hidden diversity in this host species and region. Infection prevalence and parasite lineage composition differed significantly among locations, whereas no significant effects of host sex were detected. These results reveal substantial variation in haemosporidian infections among geographically and temporally distinct sampling contexts within a single host species, suggesting that local ecological and transmission conditions may contribute to host–parasite dynamics. Our study highlights the value of single-host, lineage-level surveys for uncovering hidden parasite diversity and documenting haemosporidian variation in understudied regions of Asia.
Antimicrobial resistance (AMR) is a critical global health threat, and migratory birds have emerged as unexpected yet key players in its ecology and dissemination. acting as both reservoirs and long-distance vectors. This review uniquely integrates evidence on the role of migratory birds in the global spread of “critical priority” antimicrobial resistance genes (ARGs), including blaCTX-M (ESBLs), carbapenemases (blaNDM, blaKPC), mcr-1 (colistin), and tet(X4) (tigecycline). We advance a three-step causal hypothesis comprising contaminated stopover acquisition, migration-assisted persistence, and shedding at distant receptor habitats. We further apply operational criteria to distinguish true vector species from sentinels and reservoirs: vectors are defined by longitudinal detection across multiple flyway nodes with evidence of viable shedding enroute; reservoirs by persistent colonization independent of migration; and sentinels by transient, locally acquired contamination. These criteria provide a framework for linking molecular resistance determinants to global migratory flyway connectivity. Unlike previous reviews that largely describe local or regional occurrences, we provide a structured comparative synthesis across all major flyways, revealing that scavenging bird guilds and proximity to anthropogenic contamination are the principal determinants of AMR prevalence, rather than flyway geography. Additionally, we highlight the role of mobile genetic elements (MGEs), e.g., plasmid replicon types, to assess whether MGEs in avian isolates mirror those in clinical settings; we observe substantial overlap for IncI2, IncX4, and IncFII replicons. Consequently, we identify the systematic comparison of avian versus clinical MGE profiles as a priority research gap, given their role in facilitating horizontal gene transfer and further amplifying resistance dissemination. By integrating published data from diverse geographic regions across Asia, Europe, the Americas, and Australia, this review advocates the urgent need for a three-tier, flyway-based surveillance framework specifying target species (Larus spp., Ardea spp., Anser spp.), sampling seasons (post-breeding migration), minimum sample sizes (n ≥ 50/site/season), priority resistance markers (blaCTX-M, mcr-1, blaNDM), and international data-sharing mechanisms through WHONET, EAAFP, and CMS. These advances move the field from descriptive catalogue towards a mechanistically grounded, operationally actionable One Health framework for managing AMR at the wildlife-environment interface.
Asian grasslands, including the vast Palearctic steppes and alpine meadows of the Qinghai-Tibetan Plateau (QTP), form a vital part of the global temperate grassland biome. These open landscapes support unique bird communities with many species specially adapted to harsh conditions. However, historical and ongoing land-use changes, climate shifts, and direct human activities threaten these birds, causing widespread population declines and biodiversity loss. This review synthesizes findings from 120 studies published over the past three decades, focusing on research published since 2010. It examines (1) spatial distribution, migration strategies, and evolutionary adaptations; (2) impacts of land-use practices (such as agricultural abandonment, intensification, and grazing) and climate change on populations, communities, and physiology; (3) the ecology, behavior, and conservation status of major groups including larks, raptors and bustards; (4) key threats like habitat loss, fragmentation, electrocution, hunting, and biotic homogenization; and (5) evidence-based conservation strategies, including sustainable land management, threat mitigation, and research priorities. We highlight the need for integrated, cross-scale, and multidisciplinary approaches to reconcile socio-economic development with biodiversity conservation, and to secure the long-term persistence of steppe bird assemblages under rapid environmental change.
Wildfires are major modifiers of Mediterranean forest ecosystems, with long-term impacts on biodiversity and ecological resilience. However, how fire-history contrasts and seasonal dynamics jointly relate to avian community structure remain insufficiently understood in Mediterranean pine forests. To address this gap, we assessed how avian communities vary across sites with different time since fire using passive acoustic monitoring (PAM) across four Pinus halepensis forests in northern Greece representing a post-fire chronosequence: unburned for at least 40 years and burned in 2001, 2009, and 2018. At each site, autonomous AudioMoth recorders captured 10-min soundscape samples every 30 min for ten consecutive days per season. Bird vocalizations within each sample were identified to species level using BirdNET (v2.4) followed by expert validation of a subset of detections to reduce misclassification. We used Non-metric Multidimensional Scaling (NMDS) to visualize how bird community composition varied among sites and seasons, and Bray–Curtis dissimilarities to describe differences in species composition. We detected 43 species and found that both seasonality and fire history significantly structured avian assemblages, with seasonal variation explaining the largest portion of community differences. Burned sites were dominated by generalist and shrubland species, while unburned forests supported mature-forest specialists. Indicator species analysis identified 13 taxa associated with different successional stages. Overall bird communities showed variation across sites with contrasting fire histories, reflecting potential changes in habitat structure across successional stages, although these patterns should be interpreted cautiously given the limited number of fire-history categories. Our findings highlight the importance of integrating fire history and seasonal dynamics in biodiversity monitoring, and support PAM as a scalable tool for evaluating ecological recovery in fire-affected Mediterranean landscapes.
Broadcasting recorded bird calls is a widely used technique in White-nest Swiftlet farming, yet the echolocation of House-farm Swiftlets remains poorly described. In this study, we analyzed 16 24-h audio-video recordings in the main nest area and the swiftlet entrance point of a commercial swiftlet house in Sitiawan, Perak, Malaysia, to characterize three aspects of the swiftlet echolocation: signal structure, behavioral context and diel rhythm. Based on 160 structurally clear double-clicks, we characterized a typical echolocation pattern consisting of two closely spaced clicks. Under the present recording and filtering settings, the measured frequency range was predominantly within the audible range, with median lower and upper frequency limits of approximately 1.2 and 8.6 kHz, respectively. The second click was slightly longer than the first and showed a higher maximum frequency and broader bandwidth, indicating a consistent within-pair asymmetry. Using infrared video footage, we assigned echolocation sequences to eight behavioral/movement-related contexts. Pulse repetition rate (PRR) of each category of echolocation sequence showed two distinct levels: (1) lower values (about 6‒8 pulses/s) when entering or exiting the nest, parent-chick interactions and performing parallel shift along the nest boards; and (2) higher values (about 11‒16 pulses/s) during short flights within the nest area, departures, returns and passages through the house entrance. Analysis of 30 min bins across the 16 sample days revealed a consistent bimodal diel pattern in the echolocation activity, peaking at dawn and dusk/early night and a pronounced trough from late morning to afternoon, closely matching previously reported peaks in parental visits and chick provisioning. Our results provide the first quantitative baseline for echolocation use by House-farm Swiftlets and highlight the potential of passive acoustic monitoring to track colony activity and inform acoustics-friendly design and management of these roosts.
Individual identification of wildlife is a cornerstone of behavioral and ecological research. In recent years, developments in deep learning for computer vision have broadened the applicability of image-based individual identification in animals. However, because high quality field images of birds are difficult to collect and curate, image-based individual identification in birds remains rare. In this study, we assessed the feasibility of individual identification in wild birds using field images of the Black-necked Crane (Grus nigricollis) as a focal species. A total of 45 individual cranes were photographed in the wild, yielding a dataset of 23,830 images. We processed these images using a three-stage workflow comprising object detection (YOLO11n), instance segmentation (Co-DINO), and individual classification (YOLO11s-cls). The model achieved a Top-1 accuracy of 90.5% on the original validation set, and this increased to 93.5% on a clarity-filtered validation set created by removing the 10% most blurred images from the original validation set. Finally, we evaluated its potential to identify new individuals. On the test set, when the model correctly identified 93.9% of images of new individuals, 18.8% of images of known individuals were falsely rejected. These results demonstrate the applicability of this approach for identifying individual Black-necked Cranes in the wild, and lay the groundwork by providing a methodological basis and practical reference for image-based identification in long-term field studies across large spatial scales.
Waterbird diversity and its landscape drivers along the YRB remain poorly understood across seasons and spatial scales. Waterbird observations from five synchronous surveys conducted between October 2022 and October 2023 were integrated into four seasonal datasets and combined with land-use data to examine diversity–landscape associations across spatial scales from 5 to 50 km. Waterbird diversity was significantly lower in winter than in spring and summer, and lower in the lower reaches compared with the middle and upper reaches. Relationships between landscape structure and waterbird diversity varied by season, scale, and region. Spring diversity was positively linked to PC3 derived from principal component analysis at 10 km (F = 9.273, p < 0.01, estimate = 0.096), implying a beneficial role of grassland–cropland heterogeneity. Summer effects indicated that grassland area, dominance, and impervious-surface aggregation may shape breeding-season diversity, with significant PC3 × region and PC2 × region interactions at 5 and 10 km, respectively (F = 3.951, p < 0.05; F = 3.040, p < 0.05). Autumn relationships showed a positive PC1 effect at 5 km (F = 12.814, p < 0.01, estimate = 0.184) and a significant region × PCA interaction at 50 km (F = 2.961, p < 0.01), suggesting that greater grassland extent and dominance may provide a favorable landscape context for migratory waterbirds. Winter at fine-scale showed that greater grassland availability provided a favorable local context, whereas more complex wetland and barren-land structures corresponded to lower diversity, as supported by contrasting principal component effects at 5 km (PC1: F = 9.143, p < 0.01, estimate = 0.169; PC2: F = 4.796, p < 0.05, estimate = −0.159; PC3: F = 6.266, p < 0.05, estimate = −0.068), whereas broad-scale gradients reflected wetland shape variability and regional landscape heterogeneity. These findings underscore the need for season-specific, multi-scale habitat management to conserve waterbirds across the YRB.
Urbanization drastically alters habitats, thereby affecting biodiversity. Reproduction is a key factor in population dynamics and can determine the success of colonization in urban areas. Here, our objective was to assess which factors determine the reproductive success of two birds associated with urban areas: Eared Dove (Zenaida auriculata, n = 300 nests monitored) and Ruddy Ground Dove (Columbina talpacoti, n = 204 nests monitored). The results demonstrate that, although the Ruddy Ground Dove and the Eared Dove share ecological and behavioral similarities, their daily survival rates (DSR) responds to various environmental predictors variables. For the Ruddy Ground Dove, the percentage of nest concealment was the best predictor of reproductive success, with more concealed nests having a higher probability of success, while for the Eared Dove, nest construction on anthropogenic substrates was the best predictor of success. The contrasting results for both species highlight that there is no single explanation for success of birds in urban areas. Understanding the nuances that govern the reproductive success of common species is fundamental to urban biodiversity management.
How their toothed jaws evolve into edentulous beaks remains a key puzzle in early birds. Although some studies using discrete data of dentition have rejected the hypotheses of long-term directional selection or weight-saving adaptations, a more detailed evolutionary pattern of dentition reduction remains poorly quantified. Here for the first time, we analyze four essential parameters including tooth number on the premaxilla, maxilla, and dentary, and average tooth crown size across a phylogeny of 21 key genera in Mesozoic birds. Our phylogenetic comparative analyses reveal that the evolution of tooth number is best described by the Brownian motion model, and this evolution is a random process driven primarily by time and phylogeny, with no evidence for a global trend towards tooth loss. However, at the same time, the evolution of tooth crown size shows a weak but detectable signal of directional selection (Ornstein-Uhlenbeck model) towards smaller teeth and independent of body size, which may reflect an evolutionary transition of jaw function or feeding behavior. Furthermore, we quantified that the complete tooth loss occurred independently in at least four lineages (e.g., Confuciusornis, Gobipteryx, Archaeorhynchus, Schizooura), and the key ancestral nodes are not edentulous. We conclude that the predominant mode of dentition evolution in Mesozoic birds is random. The localized and independent events of tooth loss are likely driven by lineage-specific ecological or developmental factors. Furthermore, the origin of a modern bird-like digestive system, rather than the strong selective pressure, should be the key biological trigger to the tooth loss.
Understanding why some insular bird species are especially vulnerable to local extinction under prolonged isolation is important for clarifying extinction processes and setting conservation priorities. However, the species traits linked to such vulnerability remain poorly understood in long-isolated island systems. Here, we examined how species traits influence local extinction vulnerability of terrestrial birds in the Zhoushan Archipelago, China, a land-bridge island system isolated from the mainland for approximately 7000–9000 years following Holocene sea-level rise. We surveyed birds on 40 islands and at 16 sites on the adjacent mainland. Using data on 10 ecological and life-history traits for 110 species, we applied phylogenetic generalized least squares models to test their associations with two vulnerability metrics: species restrictedness and species nestedness ranking. We further assessed island-level correlates of species occurrence using AICc-ranked generalized linear mixed models that included island area, isolation, habitat richness and human population density. Natural abundance on the adjacent mainland was the strongest predictor of both vulnerability metrics across breeding birds, wintering birds, resident birds and migratory birds. Species with lower natural abundance had more restricted island distributions, while other traits showed weaker and less consistent associations. Species occurrence probability increased with island area and habitat richness but showed no clear association with isolation or human population density. Conservation in this system should therefore prioritize species with low natural abundance and restricted island distributions, together with large, habitat-rich islands.
The arid plains of western India support significant populations of the Greater Flamingo (Phoenicopterus roseus) and the Lesser Flamingo (Phoeniconaias minor). “Flamingo City”, a seasonally flooded salt marsh (Rann), is the largest breeding site for these species in Asia. However, their movement ecology remains poorly understood, limiting effective conservation planning. We conducted the first GPS–GSM tracking study of flamingos in this region, tracking four individuals of each species (n = 8) from May 2019 to December 2023. All tracked flamingos remained primarily within the Kachchh region, with two instances of transboundary movement by Greater Flamingos to Afghanistan and Pakistan. Both species showed regular movements between Rann (during the breeding season) and inland and coastal wetlands (during the non-breeding season), while exhibiting strong site fidelity. A generalised additive mixed model revealed that daily movement distances were higher during the breeding season in both species, and were influenced by seasonal rainfall patterns. Kernel density estimates showed wider distributions during the breeding period, with Lesser Flamingos occupying a larger spatial extent. Species differed markedly in habitat use: Lesser Flamingos relied on saline habitats, while Greater Flamingos used freshwater and brackish wetlands. These findings suggest the presence of a resident breeding population of flamingos in western India. The study underscores the importance of conserving interconnected inland, coastal, and saline wetlands across the Kachchh landscape to maintain habitat connectivity for flamingos in this dynamic arid ecosystem.
Similar to many other animal taxa, bird species worldwide have experienced significant population declines in recent years. Public awareness plays a key role in helping to conserve these species. We measured public interest toward 527 North American bird species. We used the Google Trends tool to quantify temporal trends in Google searches for species names. Overall, a majority of these bird species (78%) have been increasingly searched on Google since 2004. This stands in stark contrast to the proportion of increasing Google search trends for worldwide mammal (20%) and amphibian species (8%) found in previous studies. Interestingly, several metrics of extinction risk, such as the proportion of population loss and IUCN Red List status were not significant predictors of increasing Google searches. We did find significant differences in public attention toward bird species living in different breeding biomes and between native and introduced species. We highlight some of the groups whose species are experiencing increases in Google searches, e.g., wetland birds and introduced species, as well as those of critical conservation concern that are receiving less public attention, e.g., grassland birds. Together, these results highlight increased levels of public interest in declining North American bird populations in comparison to other vertebrates and interesting differences between bird groups.
Urbanization alters avian communities through habitat modification, yet the mechanisms by which urban-induced microclimatic changes influence bird diversity remain poorly understood. In particular, the role of the urban heat island effect in mediating the impacts of urbanization and landscape fragmentation on bird richness has rarely been tested. The Yangtze River Delta Urban Agglomeration, a critical bottleneck along the East Asian–Australasian Flyway and one of the fastest-growing urban regions globally, was selected as the study area. Here, we leverage an extensive dataset of over 10,000 waterbird records (2007–2021) and multi-source remote sensing data to deconstruct the urbanization–urban heat island–waterbird cascade using Structural Equation Modeling. Results suggest that the urban heat island effect acts as a critical mediator linking the adverse impacts of urbanization level and landscape fragmentation on waterbird richness. Moreover, this mediation exhibits distinct spatiotemporal heterogeneity. Seasonally, UHI functions as a primary physiological stressor during the breeding season. Spatially, the driving mechanisms diverge along the urban–rural gradient: in rural peripheries, the urban heat island mediates the impacts of landscape fragmentation, whereas within urban cores, it also mediates the impacts of urbanization level. These findings indicate that traditional strategies focusing solely on habitat area are insufficient. Therefore, we advocate for a paradigm shift in conservation planning that prioritizes habitat climatic suitability, ensuring the persistence of waterbirds in an increasingly warming Anthropocene.
Nest defense is an essential behavior in birds that protects nests from predation. Defensive behavior often varies among individuals, yet how hemispheric lateralization, a phenomenon commonly existing in vertebrates and invertebrates, shapes this behavior in the wild remains poorly understood. In this study, we investigated the nest defense behavior in incubating female Japanese Tits (Parus minor) in a free-living wild population, with a focus on whether this behavior was associated with footedness, a behavioral indicator of hemispheric lateralization that was measured using their foot-use preference while grasping food items against a perch. Based on simulated nest intrusions under standardized procedures, we found that their active nest defense increased as the breeding season progressed, consistent with the hypothesis that the value of the clutch increases later in the breeding season when opportunities for re-nesting decline. Importantly, the results showed that as the distance from the nest to the nearest building increased, left-footed females were more likely to show active defense, whereas right-footed females showed a non-significant tendency toward reduced defense. Given the fitness consequences of nest defense, these findings provide field evidence for the adaptive significance of lateralization and advance an integrative view that connects potential neural specialization, environmental context, and parental strategies.
