Birds of paradise are bright and colorful birds found in the rainforests of Papua New Guinea,eastern Indonesia and northern Australia.There are 45 known species,and a new study has found that 37 of them can glow using...Birds of paradise are bright and colorful birds found in the rainforests of Papua New Guinea,eastern Indonesia and northern Australia.There are 45 known species,and a new study has found that 37 of them can glow using biofluorescence(生物荧光).This is when a living thing absorbs light and gives it off again in a different color.展开更多
Wetland degradation is an escalating global challenge with profound impacts on animal diversity,particularly during successional processes.Birds,as highly mobile and environmentally sensitive organisms,serve as effect...Wetland degradation is an escalating global challenge with profound impacts on animal diversity,particularly during successional processes.Birds,as highly mobile and environmentally sensitive organisms,serve as effective indicators of ecological change.While previous studies have primarily focused on local community structures and species diversity during a specific season,there is a need to extend the research timeframe and explore broader spatial variations.Additionally,expanding from simple species diversity indices to more multidimensional diversity indices would provide a more comprehensive understanding of wetland health and resilience.To address these gaps,we investigated the effects of wetland degradation on bird diversity across taxonomic,phylogenetic,and functional dimensions in the Zoige Wetland,a plateau meadow wetland biodiversity hotspot.Surveys were conducted during both breeding(summer)and overwintering(winter)seasons across 20 transects in 5 sampling areas,representing 4 degradation levels(pristine,low,medium,and high).Our study recorded a total of 106 bird species from 32 families and 14 orders,revealing distinct seasonal patterns in bird community composition and diversity.Biodiversity indices were significantly higher in pristine and low-degraded wetlands,particularly benefiting waterfowl(Anseriformes,Ciconiiformes)and wading birds(Charadriiformes)in winter,when these areas provided superior food resources and habitat conditions.In contrast,medium and highly degraded wetlands supported increased numbers of terrestrial birds(Passeriformes)and raptors(Accipitriformes,Falconiformes).Seasonal differences in taxonomic,phylogenetic,and functional diversity indices highlighted the contrasting ecological roles of wetlands during breeding and overwintering periods.Furthermore,indicator species analysis revealed key species associated with specific degradation levels and seasons,providing valuable insights into wetland health.This study underscores the importance of spatiotemporal dynamics in understanding avian responses to wetland degradation.By linking seasonal patterns of bird diversity to habitat conditions,our findings contribute to conservation efforts and provide a framework for assessing wetland degradation and its ecological impacts.展开更多
Passive acoustic monitoring(PAM)technology is increasingly becoming one of the mainstream methods for bird monitoring.However,detecting bird audio within complex natural acoustic environments using PAM devices remains...Passive acoustic monitoring(PAM)technology is increasingly becoming one of the mainstream methods for bird monitoring.However,detecting bird audio within complex natural acoustic environments using PAM devices remains a significant challenge.To enhance the accuracy(ACC)of bird audio detection(BAD)and reduce both false negatives and false positives,this study proposes a BAD method based on a Dual-Feature Enhancement Fusion Model(DFEFM).This method incorporates per-channel energy normalization(PCEN)to suppress noise in the input audio and utilizes mel-frequency cepstral coefficients(MFCC)and frequency correlation matrices(FCM)as input features.It achieves deep feature-level fusion of MFCC and FCM on the channel dimension through two independent multi-layer convolutional network branches,and further integrates Spatial and Channel Synergistic Attention(SCSA)and Multi-Head Attention(MHA)modules to enhance the fusion effect of the aforementioned two deep features.Experimental results on the DCASE2018 BAD dataset show that our proposed method achieved an ACC of 91.4%and an AUC value of 0.963,with false negative and false positive rates of 11.36%and 7.40%,respectively,surpassing existing methods.The method also demonstrated detection ACC above 92%and AUC values above 0.987 on datasets from three sites of different natural scenes in Beijing.Testing on the NVIDIA Jetson Nano indicated that the method achieved an ACC of 89.48%when processing an average of 10 s of audio,with a response time of only 0.557 s,showing excellent processing efficiency.This study provides an effective method for filtering non-bird vocalization audio in bird vocalization monitoring devices,which helps to save edge storage and information transmission costs,and has significant application value for wild bird monitoring and ecological research.展开更多
This paper addresses the shortcomings of the Sparrow and Eagle Optimization Algorithm (SBOA) in terms of convergence accuracy, convergence speed, and susceptibility to local optima. To this end, an improved Sparrow an...This paper addresses the shortcomings of the Sparrow and Eagle Optimization Algorithm (SBOA) in terms of convergence accuracy, convergence speed, and susceptibility to local optima. To this end, an improved Sparrow and Eagle Optimization Algorithm (HS-SBOA) is proposed. Initially, the algorithm employs Iterative Mapping to generate an initial sparrow and eagle population, enhancing the diversity of the population during the global search phase. Subsequently, an adaptive weighting strategy is introduced during the exploration phase of the algorithm to achieve a balance between exploration and exploitation. Finally, to avoid the algorithm falling into local optima, a Cauchy mutation operation is applied to the current best individual. To validate the performance of the HS-SBOA algorithm, it was applied to the CEC2021 benchmark function set and three practical engineering problems, and compared with other optimization algorithms such as the Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), and Whale Optimization Algorithm (WOA) to test the effectiveness of the improved algorithm. The simulation experimental results show that the HS-SBOA algorithm demonstrates significant advantages in terms of convergence speed and accuracy, thereby validating the effectiveness of its improved strategies.展开更多
One of the main objectives of artificial intelligence lies in the simulation of the behavior of living organisms;emotions are a fundamental part of life, and they cannot be left aside when simulating behavior. In this...One of the main objectives of artificial intelligence lies in the simulation of the behavior of living organisms;emotions are a fundamental part of life, and they cannot be left aside when simulating behavior. In this research, software is developed that simulates the behavior of birds with different characteristics. The latter interacts by considering different stimuli from the environment (external), and the internal state of the subject (objectives). To achieve this, a model of birds in the role of prey and predators is developed that focuses on the study of the interaction between these organisms that exhibit specific behaviors in their environment. This project is a seminal work that aims to represent the emotions of birds, and the latter caused by stimuli from a dynamic environment.展开更多
Habitat fragmentation poses a significant threat to bird communities, especially those in open and semi-open ecosystems such as steppes. This study investigates how steppe birds adapt to and utilize fragmented habitat...Habitat fragmentation poses a significant threat to bird communities, especially those in open and semi-open ecosystems such as steppes. This study investigates how steppe birds adapt to and utilize fragmented habitats by combining niche modeling with ecological trait analysis. We conducted standardized point surveys to examine the habitat preferences of 32 bird species in Inner Mongolia, China, and quantified their habitat niche parameters using the Outlying Mean Index (OMI). Our results reveal distinct habitat preferences among species, with some thriving in intact environments while others are better adapted to fragmented areas. Grassland species showed high specialization along the fragmentation gradient, while others exhibited adaptability to varying levels of fragmentation. Using a Generalized Additive Model (GAM), we identified three key traits influencing habitat occupancy: hand-wing index, body mass, and range size. Specifically, species with medium hand-wing indices, moderate body mass, and larger range sizes were more likely to occupy heavily fragmented habitats. These findings provide empirical evidence on how habitat fragmentation affects bird species in steppe ecosystems. The study highlights the importance of functional traits in understanding avian responses to habitat fragmentation and offers a foundation for developing effective conservation strategies to preserve biodiversity in fragmented landscapes.展开更多
The factors affecting the behavior of non-specialized nectar-feeding passerines have received little attention in the literature on plant-pollinator interactions. Puya chilensis (Bromeliaceae) has sterile branch apice...The factors affecting the behavior of non-specialized nectar-feeding passerines have received little attention in the literature on plant-pollinator interactions. Puya chilensis (Bromeliaceae) has sterile branch apices that project outward from the inflorescence. In this study, we evaluate the functional role of sterile apices as support systems for bird foraging behavior. We recorded bird visitation and flower probing in the presence and absence of sterile branches during the spring seasons of 2021 and 2024. The results revealed that experimental plants with excised branches received fewer bird visits and flower probings than control plants, indicating that sterile branches play an important role in the nectar-feeding behavior of passerine birds in P. chilensis.展开更多
The publisher regrets that the Appendix A.Supplementary data was not updated as per author and editor’s request.The publisher would like to apologise for any inconvenience caused.
Bird species classification is not only a challenging topic in artificial intelligence but also a domain closely related to environmental protection and ecological research.Additionally,performing edge computing on lo...Bird species classification is not only a challenging topic in artificial intelligence but also a domain closely related to environmental protection and ecological research.Additionally,performing edge computing on low-level devices using small neural networks can be an important research direction.In this paper,we use the EfficientNetV2B0 model for bird species classification,applying transfer learning on a dataset of 525 bird species.We also employ the BiRefNet model to remove backgrounds from images in the training set.The generated background-removed images are mixed with the original training set as a form of data augmentation.We aim for these background-removed images to help the model focus on key features,and by combining data augmentation with transfer learning,we trained a highly accurate and efficient bird species classification model.The training process is divided into a transfer learning stage and a fine-tuning stage.In the transfer learning stage,only the newly added custom layers are trained;while in the fine-tuning stage,all pre-trained layers except for the batch normalization layers are fine-tuned.According to the experimental results,the proposed model not only has an advantage in size compared to other models but also outperforms them in various metrics.The training results show that the proposed model achieved an accuracy of 99.54%and a precision of 99.62%,demonstrating that it achieves both lightweight design and high accuracy.To confirm the credibility of the results,we use heatmaps to interpret the model.The heatmaps show that our model can clearly highlight the image feature area.In addition,we also perform the 10-fold cross-validation on the model to verify its credibility.Finally,this paper proposes a model with low training cost and high accuracy,making it suitable for deployment on edge computing devices to provide lighter and more convenient services.展开更多
Estimating bird abundance is key to assess threats,and to prioritize conservation actions.However,few studies focus on this topic,particularly in developing countries,which may hamper conservation efficiency.We used d...Estimating bird abundance is key to assess threats,and to prioritize conservation actions.However,few studies focus on this topic,particularly in developing countries,which may hamper conservation efficiency.We used data collected from Guangdong Province,China,between 2000 and 2020 to estimate bird abundance using MaxEnt modelling.In total,258 bird species were included,with an average density of 1485.2±489.3 ind./km^(2)(range:242.9-4502.2 ind./km^(2)).The highest density occurred in the Pearl River Estuary and on the Leizhou Peninsula.For forest birds,203 species were included with an average density of 1236.2±424.5 ind./km^(2)(143.7-2373.1 ind./km^(2)),and highest densities were found in the Pearl River Estuary and North River regions.For the 55 species of waterbirds,the average density was 249.0±351.8 ind./km^(2)(0.3-2336.1 ind./km^(2)).The total number of birds in Guangdong was estimated to be 2.58×108 ind.(2.24-3.06×10^(8)),with a total number of forest birds estimated to be 2.15×10^(8)ind.(1.90-2.49×10^(8)).The most abundant forest species(>107 individuals)were:Huet's Fulvetta(Alcippe hueti)with 2.84×10^(7)ind.,(range:2.73-2.95×10~7),Light-vented Bulbul(Pycnonotus sinensis)with 1.13×10^(7)ind.(1.07-1.20×10^(7)),Swinhoe's White-eye(Zosterops simplex)with 1.13×10^(7)ind.(1.09-1.17×10^(7)),and Red-whiskered Bulbul(Pycnonotus jocosus)with 1.01×10^(7)ind.(9.66-10.47×10^(6)),The total number of waterbirds in Guangdong was estimated to be 4.37×10^(7)ind.(3.38-5.75×10^(7)).The most abundant waterbirds(>106 individuals)were Black-headed Gull(Chroicocephalus ridibundus)with 6.35×10^(6)ind.(5.48-7.36×10^(6)),Pied Avocet(Recurvirostra avosetta)with 5.56×10^(6)ind.(3.75-8.24×10^(6)),and Little Egret(Egretta garzetta)with 5.01×10^(6)ind.(4.19-6.00×10^(6)).The densities and abundances of the 41 species listed as threatened in IUCN(higher than NT)or Chinese National Protected lists(higher thanⅡ)were evaluated,of which the population sizes of nine species were estimated for the first time in Guangdong.Moreover,all 41 species'average densities significantly declined from 2012 to 2017 in Nanling National Natural Reserve,Guangdong.展开更多
The Natural Forest Protection Project(NFPP),initiated by the Chinese government in 2000,is a crucial ecological construction project that has played a significant role in forest restoration in China.Forests in the Qin...The Natural Forest Protection Project(NFPP),initiated by the Chinese government in 2000,is a crucial ecological construction project that has played a significant role in forest restoration in China.Forests in the Qinghai-Tibet Plateau(QTP)serve as important habitats for many rare and endemic birds.Understanding the conservation efficiency of NFPP implementation on these birds holds significant practical significance.In this study,we utilized land use change matrices to analyze the forest changes in the QTP before and after NFPP implementation,predicted the potential spatial distribution of 16 nationally protected birds using Species Distribution Models(SDMs),and compared the impacts of this project on bird habitats under different carbon emission scenarios.Mann-Whitney U tests were employed to analyze the adaptation of different birds to forest changes during NFPP implementation.Our results showed that NFPP protected 172,398 km^(2) of primary forests and added 6379 km^(2) of secondary forests in our study area.The potential spatial distribution and sympatric species richness of the 16 protected birds slightly increased after NFPP implementation under different climate change scenarios,and NFPP implementation contributed to improving the potential spatial distribution of birds.Compared to newly established secondary forests,protected primary forests exhibited enhanced conservation for forest birds(Z-value>0 for six bird species,P<0.1),while being less suitable for non-forest birds(significantly unsuitable for three non-forest bird species,Z-value<0,P<0.05;non-significantly unsuitable for four non-forest bird species,Zvalue<0,P>0.1).This indicates that the protection of primary forests during NFPP implementation benefits forest bird conservation while the addition of secondary forests is beneficial to non-forest birds.To enhance the role of NFPP in avian conservation in the QTP,it is suggested to increase the landscape heterogeneity of forest,particularly in newly established secondary forests.展开更多
Mimetic seeds attract birds to disperse seeds mainly by mimicking fleshy fruits or arillate seeds,however,they provide little nutritive reward for bird dispersers.The key characteristics of mimetic seeds are conspicuo...Mimetic seeds attract birds to disperse seeds mainly by mimicking fleshy fruits or arillate seeds,however,they provide little nutritive reward for bird dispersers.The key characteristics of mimetic seeds are conspicuous seed color,hard seed coat,certain toxic secondary metabolites,and perhaps smooth waxy layer.In this review,we discuss the global distribution of mimetic seeds,the interaction of mimetic seeds with bird dispersers,and secondary metabolites that underlie key characteristics of mimetic seeds.Mimetic-seed species mainly occur in the tropics,with large numbers distributed along coastal areas.The interaction between mimetic-seed species and bird dispersers can be antagonistic,mutualistic,or both.These interactions are generally established by conspicuous visual cues and hard tactile cues from mimetic seeds.The formation and variation of key characteristics of mimetic seeds may contribute to the metabolism of several kind of secondary compounds.Here,we also discuss mimetic-seed dispersal in the context of an evolutionary game,and propose several aspects of mimetic-seed dispersal that remain unstudied.While this review is based on preliminary findings and does not account for other potential influencing factors such as climate,it is expected to contribute to an improved understanding of mimetic-seed dispersal.展开更多
The transformation of natural habitats into human-modified landscapes has far-reaching consequences for species distribution and abundance.As species adapt to these changing environments,shifts in distribution pattern...The transformation of natural habitats into human-modified landscapes has far-reaching consequences for species distribution and abundance.As species adapt to these changing environments,shifts in distribution patterns,niche dynamics,and interspecies interactions may occur,impacting biodiversity at multiple levels and potentially leading to ecosystem imbalances.This study aims to assess the impact of variations in vegetation composition and human disturbance on the distribution of sympatric breeding birds and to determine the extent of niche overlap or differentiation among these species.We conducted field surveys and collected data on bird distribution,vegetation composition,and level of human disturbance in eastern Inner Mongolian grasslands.We focused on the six most frequently co-occurring breeding birds,representing a mix of sparrows,larks,and corvids.Generalized Additive Models revealed varying responses of species occurrence along habitat gradients.Species like the Eurasian Skylark(Alauda arvensis),Mongolian Lark(Melanocorypha mongolica),and Asian Shorttoed Lark(Calandrella cheleensis),increased in larger and more connected habitats,while others,like the Tree Sparrow(Passer montanus),Eurasian Magpie(Pica pica),and Barn Swallow(Hirundo rustica),adapted to more fragmented habitats.Niche analysis indicated habitat generalists tended to occupy larger niches than grassland specialists.Substantial niche overlap was also found among the six co-occurring bird species.Conservation efforts should consider the specific needs of specialist species and strive to maintain or restore critical grassland habitats.Additionally,promoting sustainable agricultural practices that balance the needs of birds and human activities can contribute to the coexistence of generalist and specialist bird species in modified landscapes.展开更多
Animal behavioral studies are often combined with research concerning cognitive abilities.Larger brains usually mean more complex neural networks and advanced cognitive functions.By measuring the brain size of differe...Animal behavioral studies are often combined with research concerning cognitive abilities.Larger brains usually mean more complex neural networks and advanced cognitive functions.By measuring the brain size of different individual animals,we can explore differences in behavioral complexity between populations or species.However,obtaining accurate measurements of brain size is challenging both in field and laboratory environments,especially for rare and endangered species.Therefore,there is an urgent need to develop reliable methods for performing cranial brain mass.This study tests which external structures of the avian head can most accurately predict brain size.We selected five bird species from four orders,categorized external head measures into three types of parameters(direct,calculated and composite measurements),and analyzed these in relation to brain mass.The results showed that while head size can partially explain brain mass,the parameters of head height x head width were the most accurate predictors of brain mass in birds(90.4%).In addition,the positive correlation between endocranial volume and brain mass once again confirmed that avian endocranial volume can,to a certain extent,serve as a valid proxy for brain mass.Our study demonstrates that in the future we can more conveniently perform non-invasive measurements to better understand the relationship between bird brain size and behavior,ecology,and evolution.展开更多
Increasing human activity is altering the struc-ture of forests,which affects the composition of communi-ties,including birds.However,little is known about the key forest structure variables that determine the richnes...Increasing human activity is altering the struc-ture of forests,which affects the composition of communi-ties,including birds.However,little is known about the key forest structure variables that determine the richness of bird communities in European temperate oak forests.We,there-fore,aimed to identify key variables in these habitats that could contribute to the design of management strategies for forest conservation by surveying 11 oak-dominated forest sites throughout the mid-mountain range of Hungary at 86 survey points to reveal the role of different compositional and structural variables for forest stands that influence the breeding bird assemblages in the forests at the functional group and individual species levels.Based on decision tree modelling,our results showed that the density of trees larger than 30 cm DBH was an overall important variable,indi-cating that large-diameter trees were essential to provide diverse bird communities.The total abundance of birds,the foliage-gleaners,primary and secondary cavity nest-ers,residents,and five specific bird species were related to the density of high trunk diameter trees.The abundance of shrub nesters was negatively influenced by a high density of trees over 10 cm DBH.The density of the shrub layer positively affected total bird abundance and the abundance of foliage gleaners,secondary cavity nesters and residents.Analysis of the co-dominant tree species showed that the presence of linden,beech,and hornbeam was important in influencing the abundance of various bird species,e.g.,Eur-asian Treecreeper(Certhia familiaris),Marsh Tit(Poecile palustris)and Wood Warbler(Phylloscopus sibilatrix).Our results indicated that large trees,high tree diversity,and dense shrub layer were essential for forest bird communities and are critical targets for protection to maintain diverse and abundant bird communities in oak-dominated forest habitats.展开更多
The migratory nature of avian species is well known not only to researchers but also to the general public,becoming engrained in cultural traditions and even children's fairy tales.However,our understanding of the...The migratory nature of avian species is well known not only to researchers but also to the general public,becoming engrained in cultural traditions and even children's fairy tales.However,our understanding of these charismatic behaviors made great strides in the 1990s with the advent of small,light-weighted satellite transmitters capable of longterm tracking(Argos,2016).The emergence of this new technology made it possible to track a broader range of species at higher resolution than ever before.In turn,this data enabled detailed understanding of individual avian behavior and habitats,including transboundary migration routes.展开更多
Nestedness is one of the important patterns in island biogeography,community ecology and conservation biology.However,most previous nestedness studies focus on the taxonomic dimension while neglecting the functional a...Nestedness is one of the important patterns in island biogeography,community ecology and conservation biology.However,most previous nestedness studies focus on the taxonomic dimension while neglecting the functional and phylogenetic processes in generating nestedness.Moreover,few studies have examined the seasonal change of the nestedness and underlying processes.Here,we examined the seasonal nestedness of bird assemblages in taxonomic,functional,and phylogenetic dimensions,and determined the underlying processes of nestedness patterns in the Zhoushan Archipelago,China.We surveyed the occurrence, abundance,and habitats of birds on 40 islands.We calculated taxonomic,functional,and phylogenetic nestedness using WNODF and treeNODF.We determined the processes underlying nestedness by relating nestedness ranks to island characteristics and species traits.The WNODF analyses showed that bird assemblages in winter and summer were both significantly nested.The habitat-by-site matrix was also significantly nested.The nestedness of bird assemblages was significantly correlated with island area,habitat diversity,habitat specificity,minimum area requirement,habitat specificity and hand-wing index(HWI) of birds in both seasons.While the distance to the nearest mainland(DTM) exerted the influence on nestedness in summer,the distance to the nearest larger island(DTNL)affected nestedness only in winter.However,the nestedness of bird assemblages was not caused by passive sampling or human disturbance.The results of treeNODF analyses illustrated that bird assemblages were functionally and phylogenetically nested in summer and winter,but the exact mechanisms were somewhat different in these two seasons.Overall,our results supported the habitat nestedness hypothesis,selective extinction hypothesis,and selective colonization hypothesis in both seasons.From a conservation viewpoint,we should protect islands with large area and diverse habitats,islands close to the mainland,and species with large area requirement and high habitat specificity to prevent local extinction.展开更多
Bird monitoring and protection are essential for maintaining biodiversity,and fine-grained bird classification has become a key focus in this field.Audio-visual modalities provide critical cues for this task,but robus...Bird monitoring and protection are essential for maintaining biodiversity,and fine-grained bird classification has become a key focus in this field.Audio-visual modalities provide critical cues for this task,but robust feature extraction and efficient fusion remain major challenges.We introduce a multi-stage fine-grained audiovisual fusion network(MSFG-AVFNet) for fine-grained bird species classification,which addresses these challenges through two key components:(1) the audiovisual feature extraction module,which adopts a multi-stage finetuning strategy to provide high-quality unimodal features,laying a solid foundation for modality fusion;(2) the audiovisual feature fusion module,which combines a max pooling aggregation strategy with a novel audiovisual loss function to achieve effective and robust feature fusion.Experiments were conducted on the self-built AVB81and the publicly available SSW60 datasets,which contain data from 81 and 60 bird species,respectively.Comprehensive experiments demonstrate that our approach achieves notable performance gains,outperforming existing state-of-the-art methods.These results highlight its effectiveness in leveraging audiovisual modalities for fine-grained bird classification and its potential to support ecological monitoring and biodiversity research.展开更多
Ensuring food security for a rapidly growing global population amidst resource limitations and climate change is a major challenge.Agroforestry an ecologically sustainable land-use system that integrates trees,crops,a...Ensuring food security for a rapidly growing global population amidst resource limitations and climate change is a major challenge.Agroforestry an ecologically sustainable land-use system that integrates trees,crops,and sometimes livestock offers significant promise by enhancing biodiversity,ecosystem services,and agricultural productivity.A central concern in such systems is pest management,which traditionally relies on chemical pesticides.However,their excessive use has led to environmental degradation,pest resistance,and health hazards.This review explores the potential of insectivorous birds as natural pest control agents in agroforestry systems.It focuses on how habitat features,vegetation complexity,and species-specific behaviors influence bird-mediated biological control.Insectivorous birds manage pest populations through direct predation,targeting a range of insect pests including caterpillars,beetles,and grasshoppers.Their foraging activity helps maintain pest populations below the economic threshold.Vegetation strata comprising ground cover,shrubs,understory,and canopy offer diverse foraging niches and nesting habitats that enhance bird diversity and functional roles.Pest control efficiency is closely linked to seasonality,resource availability,and habitat structure.Differentiating between beneficial(predatory)and pestiferous birds is essential to maximize ecosystem services and minimize crop losses or damage to beneficial insects.Conservation of beneficial bird species,informed vegetation planning,and regular monitoring are vital to strengthening multitrophic interactions and achieving sustainable pest control.Future research should focus on bird behavior,predator-prey interactions,and habitat management to optimize bird-friendly pest regulation strategies in agroforestry landscapes.展开更多
Bird vocalizations are pivotal for ecological monitoring,providing insights into biodiversity and ecosystem health.Traditional recognition methods often neglect phase information,resulting in incomplete feature repres...Bird vocalizations are pivotal for ecological monitoring,providing insights into biodiversity and ecosystem health.Traditional recognition methods often neglect phase information,resulting in incomplete feature representation.In this paper,we introduce a novel approach to bird vocalization recognition(BVR)that integrates both amplitude and phase information,leading to enhanced species identification.We propose MHARes Net,a deep learning(DL)model that employs residual blocks and a multi-head attention mechanism to capture salient features from logarithmic power(POW),Instantaneous Frequency(IF),and Group Delay(GD)extracted from bird vocalizations.Experiments on three bird vocalization datasets demonstrate our method's superior performance,achieving accuracy rates of 94%,98.9%,and 87.1%respectively.These results indicate that our approach provides a more effective representation of bird vocalizations,outperforming existing methods.This integration of phase information in BVR is innovative and significantly advances the field of automatic bird monitoring technology,offering valuable tools for ecological research and conservation efforts.展开更多
文摘Birds of paradise are bright and colorful birds found in the rainforests of Papua New Guinea,eastern Indonesia and northern Australia.There are 45 known species,and a new study has found that 37 of them can glow using biofluorescence(生物荧光).This is when a living thing absorbs light and gives it off again in a different color.
基金supported by the Southwest Minzu University Research Startup Funds (No.16011221038,RQD2022021)Double World-Class Project (No.CX2023010)。
文摘Wetland degradation is an escalating global challenge with profound impacts on animal diversity,particularly during successional processes.Birds,as highly mobile and environmentally sensitive organisms,serve as effective indicators of ecological change.While previous studies have primarily focused on local community structures and species diversity during a specific season,there is a need to extend the research timeframe and explore broader spatial variations.Additionally,expanding from simple species diversity indices to more multidimensional diversity indices would provide a more comprehensive understanding of wetland health and resilience.To address these gaps,we investigated the effects of wetland degradation on bird diversity across taxonomic,phylogenetic,and functional dimensions in the Zoige Wetland,a plateau meadow wetland biodiversity hotspot.Surveys were conducted during both breeding(summer)and overwintering(winter)seasons across 20 transects in 5 sampling areas,representing 4 degradation levels(pristine,low,medium,and high).Our study recorded a total of 106 bird species from 32 families and 14 orders,revealing distinct seasonal patterns in bird community composition and diversity.Biodiversity indices were significantly higher in pristine and low-degraded wetlands,particularly benefiting waterfowl(Anseriformes,Ciconiiformes)and wading birds(Charadriiformes)in winter,when these areas provided superior food resources and habitat conditions.In contrast,medium and highly degraded wetlands supported increased numbers of terrestrial birds(Passeriformes)and raptors(Accipitriformes,Falconiformes).Seasonal differences in taxonomic,phylogenetic,and functional diversity indices highlighted the contrasting ecological roles of wetlands during breeding and overwintering periods.Furthermore,indicator species analysis revealed key species associated with specific degradation levels and seasons,providing valuable insights into wetland health.This study underscores the importance of spatiotemporal dynamics in understanding avian responses to wetland degradation.By linking seasonal patterns of bird diversity to habitat conditions,our findings contribute to conservation efforts and provide a framework for assessing wetland degradation and its ecological impacts.
基金supported by the Beijing Natural Science Foundation(5252014)the National Natural Science Foundation of China(62303063)。
文摘Passive acoustic monitoring(PAM)technology is increasingly becoming one of the mainstream methods for bird monitoring.However,detecting bird audio within complex natural acoustic environments using PAM devices remains a significant challenge.To enhance the accuracy(ACC)of bird audio detection(BAD)and reduce both false negatives and false positives,this study proposes a BAD method based on a Dual-Feature Enhancement Fusion Model(DFEFM).This method incorporates per-channel energy normalization(PCEN)to suppress noise in the input audio and utilizes mel-frequency cepstral coefficients(MFCC)and frequency correlation matrices(FCM)as input features.It achieves deep feature-level fusion of MFCC and FCM on the channel dimension through two independent multi-layer convolutional network branches,and further integrates Spatial and Channel Synergistic Attention(SCSA)and Multi-Head Attention(MHA)modules to enhance the fusion effect of the aforementioned two deep features.Experimental results on the DCASE2018 BAD dataset show that our proposed method achieved an ACC of 91.4%and an AUC value of 0.963,with false negative and false positive rates of 11.36%and 7.40%,respectively,surpassing existing methods.The method also demonstrated detection ACC above 92%and AUC values above 0.987 on datasets from three sites of different natural scenes in Beijing.Testing on the NVIDIA Jetson Nano indicated that the method achieved an ACC of 89.48%when processing an average of 10 s of audio,with a response time of only 0.557 s,showing excellent processing efficiency.This study provides an effective method for filtering non-bird vocalization audio in bird vocalization monitoring devices,which helps to save edge storage and information transmission costs,and has significant application value for wild bird monitoring and ecological research.
文摘This paper addresses the shortcomings of the Sparrow and Eagle Optimization Algorithm (SBOA) in terms of convergence accuracy, convergence speed, and susceptibility to local optima. To this end, an improved Sparrow and Eagle Optimization Algorithm (HS-SBOA) is proposed. Initially, the algorithm employs Iterative Mapping to generate an initial sparrow and eagle population, enhancing the diversity of the population during the global search phase. Subsequently, an adaptive weighting strategy is introduced during the exploration phase of the algorithm to achieve a balance between exploration and exploitation. Finally, to avoid the algorithm falling into local optima, a Cauchy mutation operation is applied to the current best individual. To validate the performance of the HS-SBOA algorithm, it was applied to the CEC2021 benchmark function set and three practical engineering problems, and compared with other optimization algorithms such as the Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), and Whale Optimization Algorithm (WOA) to test the effectiveness of the improved algorithm. The simulation experimental results show that the HS-SBOA algorithm demonstrates significant advantages in terms of convergence speed and accuracy, thereby validating the effectiveness of its improved strategies.
文摘One of the main objectives of artificial intelligence lies in the simulation of the behavior of living organisms;emotions are a fundamental part of life, and they cannot be left aside when simulating behavior. In this research, software is developed that simulates the behavior of birds with different characteristics. The latter interacts by considering different stimuli from the environment (external), and the internal state of the subject (objectives). To achieve this, a model of birds in the role of prey and predators is developed that focuses on the study of the interaction between these organisms that exhibit specific behaviors in their environment. This project is a seminal work that aims to represent the emotions of birds, and the latter caused by stimuli from a dynamic environment.
基金supported by the National Natural Science Foundation of China(No.32201304)the Fundamental Research Funds for the Central Universities(No.2412022QD026).
文摘Habitat fragmentation poses a significant threat to bird communities, especially those in open and semi-open ecosystems such as steppes. This study investigates how steppe birds adapt to and utilize fragmented habitats by combining niche modeling with ecological trait analysis. We conducted standardized point surveys to examine the habitat preferences of 32 bird species in Inner Mongolia, China, and quantified their habitat niche parameters using the Outlying Mean Index (OMI). Our results reveal distinct habitat preferences among species, with some thriving in intact environments while others are better adapted to fragmented areas. Grassland species showed high specialization along the fragmentation gradient, while others exhibited adaptability to varying levels of fragmentation. Using a Generalized Additive Model (GAM), we identified three key traits influencing habitat occupancy: hand-wing index, body mass, and range size. Specifically, species with medium hand-wing indices, moderate body mass, and larger range sizes were more likely to occupy heavily fragmented habitats. These findings provide empirical evidence on how habitat fragmentation affects bird species in steppe ecosystems. The study highlights the importance of functional traits in understanding avian responses to habitat fragmentation and offers a foundation for developing effective conservation strategies to preserve biodiversity in fragmented landscapes.
基金supported by grants FONDECYT 1180850 and 1231757 to RM.
文摘The factors affecting the behavior of non-specialized nectar-feeding passerines have received little attention in the literature on plant-pollinator interactions. Puya chilensis (Bromeliaceae) has sterile branch apices that project outward from the inflorescence. In this study, we evaluate the functional role of sterile apices as support systems for bird foraging behavior. We recorded bird visitation and flower probing in the presence and absence of sterile branches during the spring seasons of 2021 and 2024. The results revealed that experimental plants with excised branches received fewer bird visits and flower probings than control plants, indicating that sterile branches play an important role in the nectar-feeding behavior of passerine birds in P. chilensis.
文摘The publisher regrets that the Appendix A.Supplementary data was not updated as per author and editor’s request.The publisher would like to apologise for any inconvenience caused.
文摘Bird species classification is not only a challenging topic in artificial intelligence but also a domain closely related to environmental protection and ecological research.Additionally,performing edge computing on low-level devices using small neural networks can be an important research direction.In this paper,we use the EfficientNetV2B0 model for bird species classification,applying transfer learning on a dataset of 525 bird species.We also employ the BiRefNet model to remove backgrounds from images in the training set.The generated background-removed images are mixed with the original training set as a form of data augmentation.We aim for these background-removed images to help the model focus on key features,and by combining data augmentation with transfer learning,we trained a highly accurate and efficient bird species classification model.The training process is divided into a transfer learning stage and a fine-tuning stage.In the transfer learning stage,only the newly added custom layers are trained;while in the fine-tuning stage,all pre-trained layers except for the batch normalization layers are fine-tuned.According to the experimental results,the proposed model not only has an advantage in size compared to other models but also outperforms them in various metrics.The training results show that the proposed model achieved an accuracy of 99.54%and a precision of 99.62%,demonstrating that it achieves both lightweight design and high accuracy.To confirm the credibility of the results,we use heatmaps to interpret the model.The heatmaps show that our model can clearly highlight the image feature area.In addition,we also perform the 10-fold cross-validation on the model to verify its credibility.Finally,this paper proposes a model with low training cost and high accuracy,making it suitable for deployment on edge computing devices to provide lighter and more convenient services.
基金supported by the National Natural Science Foundation of China(31961123003,32001098,31672265)the DFGP project of fauna of Guangdong(202115)the GDAS Special Project of Science and Technology Development(2022GDASZH-2022010106)。
文摘Estimating bird abundance is key to assess threats,and to prioritize conservation actions.However,few studies focus on this topic,particularly in developing countries,which may hamper conservation efficiency.We used data collected from Guangdong Province,China,between 2000 and 2020 to estimate bird abundance using MaxEnt modelling.In total,258 bird species were included,with an average density of 1485.2±489.3 ind./km^(2)(range:242.9-4502.2 ind./km^(2)).The highest density occurred in the Pearl River Estuary and on the Leizhou Peninsula.For forest birds,203 species were included with an average density of 1236.2±424.5 ind./km^(2)(143.7-2373.1 ind./km^(2)),and highest densities were found in the Pearl River Estuary and North River regions.For the 55 species of waterbirds,the average density was 249.0±351.8 ind./km^(2)(0.3-2336.1 ind./km^(2)).The total number of birds in Guangdong was estimated to be 2.58×108 ind.(2.24-3.06×10^(8)),with a total number of forest birds estimated to be 2.15×10^(8)ind.(1.90-2.49×10^(8)).The most abundant forest species(>107 individuals)were:Huet's Fulvetta(Alcippe hueti)with 2.84×10^(7)ind.,(range:2.73-2.95×10~7),Light-vented Bulbul(Pycnonotus sinensis)with 1.13×10^(7)ind.(1.07-1.20×10^(7)),Swinhoe's White-eye(Zosterops simplex)with 1.13×10^(7)ind.(1.09-1.17×10^(7)),and Red-whiskered Bulbul(Pycnonotus jocosus)with 1.01×10^(7)ind.(9.66-10.47×10^(6)),The total number of waterbirds in Guangdong was estimated to be 4.37×10^(7)ind.(3.38-5.75×10^(7)).The most abundant waterbirds(>106 individuals)were Black-headed Gull(Chroicocephalus ridibundus)with 6.35×10^(6)ind.(5.48-7.36×10^(6)),Pied Avocet(Recurvirostra avosetta)with 5.56×10^(6)ind.(3.75-8.24×10^(6)),and Little Egret(Egretta garzetta)with 5.01×10^(6)ind.(4.19-6.00×10^(6)).The densities and abundances of the 41 species listed as threatened in IUCN(higher than NT)or Chinese National Protected lists(higher thanⅡ)were evaluated,of which the population sizes of nine species were estimated for the first time in Guangdong.Moreover,all 41 species'average densities significantly declined from 2012 to 2017 in Nanling National Natural Reserve,Guangdong.
基金funded by Central Fiscal Forestry and Grassland Ecological Protection and Restoration Fund in 2022(grant number:HYGJ22069P(2022zfcg03469)-HT01)National Natural Science Foundation of China(grant number:3152010390332070452)。
文摘The Natural Forest Protection Project(NFPP),initiated by the Chinese government in 2000,is a crucial ecological construction project that has played a significant role in forest restoration in China.Forests in the Qinghai-Tibet Plateau(QTP)serve as important habitats for many rare and endemic birds.Understanding the conservation efficiency of NFPP implementation on these birds holds significant practical significance.In this study,we utilized land use change matrices to analyze the forest changes in the QTP before and after NFPP implementation,predicted the potential spatial distribution of 16 nationally protected birds using Species Distribution Models(SDMs),and compared the impacts of this project on bird habitats under different carbon emission scenarios.Mann-Whitney U tests were employed to analyze the adaptation of different birds to forest changes during NFPP implementation.Our results showed that NFPP protected 172,398 km^(2) of primary forests and added 6379 km^(2) of secondary forests in our study area.The potential spatial distribution and sympatric species richness of the 16 protected birds slightly increased after NFPP implementation under different climate change scenarios,and NFPP implementation contributed to improving the potential spatial distribution of birds.Compared to newly established secondary forests,protected primary forests exhibited enhanced conservation for forest birds(Z-value>0 for six bird species,P<0.1),while being less suitable for non-forest birds(significantly unsuitable for three non-forest bird species,Z-value<0,P<0.05;non-significantly unsuitable for four non-forest bird species,Zvalue<0,P>0.1).This indicates that the protection of primary forests during NFPP implementation benefits forest bird conservation while the addition of secondary forests is beneficial to non-forest birds.To enhance the role of NFPP in avian conservation in the QTP,it is suggested to increase the landscape heterogeneity of forest,particularly in newly established secondary forests.
基金supported by the Yunnan Ten Thousand Talents Plan Young&Elite Talents Project(YNWR-QNBJ-2018-017)the National Natural Science Foundation of China(32371564)+2 种基金the Key Project of Basic Research of Yunnan Province,China(202101AS070035202301AS070001)to G.ChenYunnan Provincial Science and Technology Talent and Platform Plan(202305AM070005).
文摘Mimetic seeds attract birds to disperse seeds mainly by mimicking fleshy fruits or arillate seeds,however,they provide little nutritive reward for bird dispersers.The key characteristics of mimetic seeds are conspicuous seed color,hard seed coat,certain toxic secondary metabolites,and perhaps smooth waxy layer.In this review,we discuss the global distribution of mimetic seeds,the interaction of mimetic seeds with bird dispersers,and secondary metabolites that underlie key characteristics of mimetic seeds.Mimetic-seed species mainly occur in the tropics,with large numbers distributed along coastal areas.The interaction between mimetic-seed species and bird dispersers can be antagonistic,mutualistic,or both.These interactions are generally established by conspicuous visual cues and hard tactile cues from mimetic seeds.The formation and variation of key characteristics of mimetic seeds may contribute to the metabolism of several kind of secondary compounds.Here,we also discuss mimetic-seed dispersal in the context of an evolutionary game,and propose several aspects of mimetic-seed dispersal that remain unstudied.While this review is based on preliminary findings and does not account for other potential influencing factors such as climate,it is expected to contribute to an improved understanding of mimetic-seed dispersal.
基金supported by the National Natural Science Foundation of China (No.32201304)the Fundamental Research Funds for the Central Universities (No.2412022QD026)。
文摘The transformation of natural habitats into human-modified landscapes has far-reaching consequences for species distribution and abundance.As species adapt to these changing environments,shifts in distribution patterns,niche dynamics,and interspecies interactions may occur,impacting biodiversity at multiple levels and potentially leading to ecosystem imbalances.This study aims to assess the impact of variations in vegetation composition and human disturbance on the distribution of sympatric breeding birds and to determine the extent of niche overlap or differentiation among these species.We conducted field surveys and collected data on bird distribution,vegetation composition,and level of human disturbance in eastern Inner Mongolian grasslands.We focused on the six most frequently co-occurring breeding birds,representing a mix of sparrows,larks,and corvids.Generalized Additive Models revealed varying responses of species occurrence along habitat gradients.Species like the Eurasian Skylark(Alauda arvensis),Mongolian Lark(Melanocorypha mongolica),and Asian Shorttoed Lark(Calandrella cheleensis),increased in larger and more connected habitats,while others,like the Tree Sparrow(Passer montanus),Eurasian Magpie(Pica pica),and Barn Swallow(Hirundo rustica),adapted to more fragmented habitats.Niche analysis indicated habitat generalists tended to occupy larger niches than grassland specialists.Substantial niche overlap was also found among the six co-occurring bird species.Conservation efforts should consider the specific needs of specialist species and strive to maintain or restore critical grassland habitats.Additionally,promoting sustainable agricultural practices that balance the needs of birds and human activities can contribute to the coexistence of generalist and specialist bird species in modified landscapes.
基金supported by the National Key R&D Program of China(2023YFF1304600)。
文摘Animal behavioral studies are often combined with research concerning cognitive abilities.Larger brains usually mean more complex neural networks and advanced cognitive functions.By measuring the brain size of different individual animals,we can explore differences in behavioral complexity between populations or species.However,obtaining accurate measurements of brain size is challenging both in field and laboratory environments,especially for rare and endangered species.Therefore,there is an urgent need to develop reliable methods for performing cranial brain mass.This study tests which external structures of the avian head can most accurately predict brain size.We selected five bird species from four orders,categorized external head measures into three types of parameters(direct,calculated and composite measurements),and analyzed these in relation to brain mass.The results showed that while head size can partially explain brain mass,the parameters of head height x head width were the most accurate predictors of brain mass in birds(90.4%).In addition,the positive correlation between endocranial volume and brain mass once again confirmed that avian endocranial volume can,to a certain extent,serve as a valid proxy for brain mass.Our study demonstrates that in the future we can more conveniently perform non-invasive measurements to better understand the relationship between bird brain size and behavior,ecology,and evolution.
基金supported part ia l l y by LIFE4Oak Forests Project LIFE16NAT/IT/000245)the RRF 2.3.121202200008 projectthe MERLiN project funded under the European Commission H2020 Programme(101036337 MERLiN H2020 LC GD 2020)。
文摘Increasing human activity is altering the struc-ture of forests,which affects the composition of communi-ties,including birds.However,little is known about the key forest structure variables that determine the richness of bird communities in European temperate oak forests.We,there-fore,aimed to identify key variables in these habitats that could contribute to the design of management strategies for forest conservation by surveying 11 oak-dominated forest sites throughout the mid-mountain range of Hungary at 86 survey points to reveal the role of different compositional and structural variables for forest stands that influence the breeding bird assemblages in the forests at the functional group and individual species levels.Based on decision tree modelling,our results showed that the density of trees larger than 30 cm DBH was an overall important variable,indi-cating that large-diameter trees were essential to provide diverse bird communities.The total abundance of birds,the foliage-gleaners,primary and secondary cavity nest-ers,residents,and five specific bird species were related to the density of high trunk diameter trees.The abundance of shrub nesters was negatively influenced by a high density of trees over 10 cm DBH.The density of the shrub layer positively affected total bird abundance and the abundance of foliage gleaners,secondary cavity nesters and residents.Analysis of the co-dominant tree species showed that the presence of linden,beech,and hornbeam was important in influencing the abundance of various bird species,e.g.,Eur-asian Treecreeper(Certhia familiaris),Marsh Tit(Poecile palustris)and Wood Warbler(Phylloscopus sibilatrix).Our results indicated that large trees,high tree diversity,and dense shrub layer were essential for forest bird communities and are critical targets for protection to maintain diverse and abundant bird communities in oak-dominated forest habitats.
文摘The migratory nature of avian species is well known not only to researchers but also to the general public,becoming engrained in cultural traditions and even children's fairy tales.However,our understanding of these charismatic behaviors made great strides in the 1990s with the advent of small,light-weighted satellite transmitters capable of longterm tracking(Argos,2016).The emergence of this new technology made it possible to track a broader range of species at higher resolution than ever before.In turn,this data enabled detailed understanding of individual avian behavior and habitats,including transboundary migration routes.
基金supported by the National Natural Science Foundation of China (32271734)。
文摘Nestedness is one of the important patterns in island biogeography,community ecology and conservation biology.However,most previous nestedness studies focus on the taxonomic dimension while neglecting the functional and phylogenetic processes in generating nestedness.Moreover,few studies have examined the seasonal change of the nestedness and underlying processes.Here,we examined the seasonal nestedness of bird assemblages in taxonomic,functional,and phylogenetic dimensions,and determined the underlying processes of nestedness patterns in the Zhoushan Archipelago,China.We surveyed the occurrence, abundance,and habitats of birds on 40 islands.We calculated taxonomic,functional,and phylogenetic nestedness using WNODF and treeNODF.We determined the processes underlying nestedness by relating nestedness ranks to island characteristics and species traits.The WNODF analyses showed that bird assemblages in winter and summer were both significantly nested.The habitat-by-site matrix was also significantly nested.The nestedness of bird assemblages was significantly correlated with island area,habitat diversity,habitat specificity,minimum area requirement,habitat specificity and hand-wing index(HWI) of birds in both seasons.While the distance to the nearest mainland(DTM) exerted the influence on nestedness in summer,the distance to the nearest larger island(DTNL)affected nestedness only in winter.However,the nestedness of bird assemblages was not caused by passive sampling or human disturbance.The results of treeNODF analyses illustrated that bird assemblages were functionally and phylogenetically nested in summer and winter,but the exact mechanisms were somewhat different in these two seasons.Overall,our results supported the habitat nestedness hypothesis,selective extinction hypothesis,and selective colonization hypothesis in both seasons.From a conservation viewpoint,we should protect islands with large area and diverse habitats,islands close to the mainland,and species with large area requirement and high habitat specificity to prevent local extinction.
基金supported by the Beijing Natural Science Foundation(No.5252014)the Open Fund of The Key Laboratory of Urban Ecological Environment Simulation and Protection,Ministry of Ecology and Environment of the People's Republic of China (No.UEESP-202502)the National Natural Science Foundation of China (No.62303063&32371874)。
文摘Bird monitoring and protection are essential for maintaining biodiversity,and fine-grained bird classification has become a key focus in this field.Audio-visual modalities provide critical cues for this task,but robust feature extraction and efficient fusion remain major challenges.We introduce a multi-stage fine-grained audiovisual fusion network(MSFG-AVFNet) for fine-grained bird species classification,which addresses these challenges through two key components:(1) the audiovisual feature extraction module,which adopts a multi-stage finetuning strategy to provide high-quality unimodal features,laying a solid foundation for modality fusion;(2) the audiovisual feature fusion module,which combines a max pooling aggregation strategy with a novel audiovisual loss function to achieve effective and robust feature fusion.Experiments were conducted on the self-built AVB81and the publicly available SSW60 datasets,which contain data from 81 and 60 bird species,respectively.Comprehensive experiments demonstrate that our approach achieves notable performance gains,outperforming existing state-of-the-art methods.These results highlight its effectiveness in leveraging audiovisual modalities for fine-grained bird classification and its potential to support ecological monitoring and biodiversity research.
文摘Ensuring food security for a rapidly growing global population amidst resource limitations and climate change is a major challenge.Agroforestry an ecologically sustainable land-use system that integrates trees,crops,and sometimes livestock offers significant promise by enhancing biodiversity,ecosystem services,and agricultural productivity.A central concern in such systems is pest management,which traditionally relies on chemical pesticides.However,their excessive use has led to environmental degradation,pest resistance,and health hazards.This review explores the potential of insectivorous birds as natural pest control agents in agroforestry systems.It focuses on how habitat features,vegetation complexity,and species-specific behaviors influence bird-mediated biological control.Insectivorous birds manage pest populations through direct predation,targeting a range of insect pests including caterpillars,beetles,and grasshoppers.Their foraging activity helps maintain pest populations below the economic threshold.Vegetation strata comprising ground cover,shrubs,understory,and canopy offer diverse foraging niches and nesting habitats that enhance bird diversity and functional roles.Pest control efficiency is closely linked to seasonality,resource availability,and habitat structure.Differentiating between beneficial(predatory)and pestiferous birds is essential to maximize ecosystem services and minimize crop losses or damage to beneficial insects.Conservation of beneficial bird species,informed vegetation planning,and regular monitoring are vital to strengthening multitrophic interactions and achieving sustainable pest control.Future research should focus on bird behavior,predator-prey interactions,and habitat management to optimize bird-friendly pest regulation strategies in agroforestry landscapes.
基金supported by the Beijing Natural Science Foundation (5252014)the National Natural Science Foundation of China (62303063)。
文摘Bird vocalizations are pivotal for ecological monitoring,providing insights into biodiversity and ecosystem health.Traditional recognition methods often neglect phase information,resulting in incomplete feature representation.In this paper,we introduce a novel approach to bird vocalization recognition(BVR)that integrates both amplitude and phase information,leading to enhanced species identification.We propose MHARes Net,a deep learning(DL)model that employs residual blocks and a multi-head attention mechanism to capture salient features from logarithmic power(POW),Instantaneous Frequency(IF),and Group Delay(GD)extracted from bird vocalizations.Experiments on three bird vocalization datasets demonstrate our method's superior performance,achieving accuracy rates of 94%,98.9%,and 87.1%respectively.These results indicate that our approach provides a more effective representation of bird vocalizations,outperforming existing methods.This integration of phase information in BVR is innovative and significantly advances the field of automatic bird monitoring technology,offering valuable tools for ecological research and conservation efforts.
