Although four species of odontocete and four species of baleen whale have been recorded in Prydz Bay,their vocalizations have been rarely investigated.Underwater vocalizations were recorded during March 2017 in Prydz ...Although four species of odontocete and four species of baleen whale have been recorded in Prydz Bay,their vocalizations have been rarely investigated.Underwater vocalizations were recorded during March 2017 in Prydz Bay,Antarctica.Bio-duck sounds,downsweeps,inverted“u”shape signals,whistles,pulsed sounds,and broadband clicks were recorded.Bio-duck sounds and downsweeps were associated with Antarctic minke whales(Balaenoptera bonaerensis)based on visual observations.Similarities between inverted“u”shape signals,biphonic calls,and clicks with vocalizations previously described for killer whales(Orcinus orca)lead us believe the presence of Antarctic killer whales.According to sound structures,signal characteristics,and recording location,Antarctic type C killer whales were the most probable candidates to produce these detected calls.These represent the fi rst detection of inverted“u”shape signals in Antarctic waters,and the fi rst report of Antarctic killer whale in Prydz Bay based on passive acoustic monitoring.The co-existence of Antarctic minke and killer whales may imply that minke whales can detect diff erences between the sounds of mammal-eating and fi sh-eating killer whales.Our descriptions of these underwater vocalizations contribute to the limited body of information regarding the distribution and acoustic behavior of cetaceans in Prydz Bay.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
Accurate estimations of animal population size are pivotal for implementing management strategies properly.Recapture technique based on sounds as a specimen identifcation mark has barely been used for marine mammals.H...Accurate estimations of animal population size are pivotal for implementing management strategies properly.Recapture technique based on sounds as a specimen identifcation mark has barely been used for marine mammals.However,inferring abundance estimates from acoustic methods could enhance the accuracy and precision of population size assessments.Here,we tested the possibility of using signature whistles as individual marks for estimating the size of common bottlenose dolphin(Tursiops truncatus)populations.Data were continuously collected for 326 days in 2015–2016,by using a fxed acoustic device located in the Sicily Strait(Italy).The SIGID method was applied to identify Signature Whistles Types(SWTs)over 7,000 h of recordings.Eighty SWTs were detected as long as their stereotyped fundamental frequency contours were repeated in bouts of at least 6 renditions.The mean SWTs monthly recording rate resulted in 0.19(Standard deviation=0.16),with 20 SWTs recorded over 5 or more different encounters(until a maximum of 30 encounters).The Jolly–Seber model(with POPAN formulation)was run in Mark software to estimate the population size.The estimated population size resulted in 171 bottlenose dolphins(95%confdence interval=137–215).Even if the detection and identifcation of signature whistles required crucial precautions,and animals could be detected differently from visual techniques,the population size estimate obtained was comparable with previous results based on physical marks data.These outcomes demonstrated that signature whistles can be considered a strongly effective tool for integrating traditional mark-recapture techniques with fnely estimated dolphins’population abundances.展开更多
Passive source imaging can reconstruct body wave reflections similar to those of active sources through seismic interferometry(SI).It has become a low-cost,environmentally friendly alternative to active source seismic...Passive source imaging can reconstruct body wave reflections similar to those of active sources through seismic interferometry(SI).It has become a low-cost,environmentally friendly alternative to active source seismic,showing great potential.However,this method faces many challenges in practical applications,including uneven distribution of underground sources and complex survey environments.These situations seriously affect the reconstruction quality of virtual shot records,resulting in unguaranteed imaging results and greatly limiting passive source seismic exploration applications.In addition,the quality of the reconstructed records is directly related to the time length of the noise records,but in practice it is often difficult to obtain long-term,high-quality noise segments containing body wave events.To solve the above problems,we propose a deep learning method for reconstructing passive source virtual shot records and apply it to passive source time-lapse monitoring.This method combines the UNet network and the BiLSTM(Bidirectional Long Short-Term Memory)network for extracting spatial features and temporal features respectively.It introduces the spatial attention mechanism to establish a hybrid SUNet-BiLSTM-Attention(SBA)network for supervised training.Through pre-training and fine-tuning training,the network can accurately reconstruct passive source virtual shot records directly from short-time noisy segments containing body wave events.The experimental results of theoretical data show that the virtual shot records reconstructed by the network have high resolution and signal to noise ratio(SNR),providing high-quality data for subsequent monitoring and imaging.Finally,to further validate the effectiveness of proposed method,we applied it to field data collected from gas storage in northwest China.The reconstruction results of field data effectively improve the quality of virtual records and obtain more reliable time-lapse imaging monitoring results,which have significant practical value.展开更多
During the Yangtze Freshwater Dolphin Expedition 2012,Yangtze finless porpoises(Neophocaena asiaeorientalis)were acoustically monitored in 9 port areas at night.During 6566 min of nocturnal monitoring,porpoise sonar w...During the Yangtze Freshwater Dolphin Expedition 2012,Yangtze finless porpoises(Neophocaena asiaeorientalis)were acoustically monitored in 9 port areas at night.During 6566 min of nocturnal monitoring,porpoise sonar was detected for 488 min(7.43%of the total time).Of all 81 encounters,the longest echolocation span obtained was 102.9 min,suggesting frequent and prolonged porpoise occupation of the port areas.A combined total of 2091 click trains were recorded,with 129(6.2%)containing minimum inter-click intervals(ICIs)below 10 ms(termed a buzz).Buzzes with a decrease in ICIs and search and approach phases that resembled feeding echolocation signals accounted for 44.2%(N=52)of all buzzes.Buzzes with an increase in ICIs,suggesting a mirrored prey capture phase,accounted for 20.2%(N=26)and could reflect attempts to locate escaped prey because they were followed by approach-phase feeding buzzes.Anecdotal evidence of porpoises fleeing the proximity of vessels was observed.The recordings indicating clusters of porpoises feeding near the port areas suggest a forced choice for feeding due to the relatively higher prey availability in the port areas compared to other areas in the Yangtze River that are probably overfished.展开更多
The associations between feeding activities and environmental variables inform animal feeding tactics that max-imize energetic gains by minimizing energy costs while maximizing feeding success.Relevant studies in aqua...The associations between feeding activities and environmental variables inform animal feeding tactics that max-imize energetic gains by minimizing energy costs while maximizing feeding success.Relevant studies in aquatic animals,particularly marine mammals,are scarce due to difficulties in the observation of feeding behaviors in aquatic environments.This data scarcity concurrently hinders ecosystem-basedfishery management in the context of small toothed-cetacean conservation.In the present study,a passive acoustic monitoring station was deployed in an East Asianfinless porpoise habitat in Laizhou Bay to investigate potential relationships between East Asianfinless porpoises and their prey.The data revealed that porpoises were acoustically present nearly every day during the survey period.Porpoise detection rates differed between spring and autumn in concert with activities offish choruses.During spring,fish choruses were present throughout the afternoon,and this was the time when porpoise vocalizations were the most frequently detected.During autumn,whenfish choruses were absent,porpoise detec-tion rates decreased,and diurnal patterns were not detected.The close association betweenfish choruses andfin-less porpoise activities implies an“eavesdropping”feeding strategy to maximize energetic gains,similar to other toothed cetaceans that are known to engage similar feeding strategies.Underwater noise pollution,particularly those maskingfish choruses,could interruptfinless porpoises’feeding success.Fisheries competing soniferousfishes withfinless porpoise could impactfinless porpoise viability through ecosystem disruption,in addition tofishing gear entanglement.展开更多
Passive acoustic monitoring has the potential to be a useful tool for population estimation of sound-producing fish and mammals(mostly whales).Previous work on population estimates of callers employed a simple cross-c...Passive acoustic monitoring has the potential to be a useful tool for population estimation of sound-producing fish and mammals(mostly whales).Previous work on population estimates of callers employed a simple cross-correlation technique with recordings from two acoustic sensors,and the current work extends the technique to two configurations of a 3-acoustic sensors array using two different sounds,i.e.,chirps which is commonly generated by damselfish(Dascyllus aruanus),humpback whales(Megaptera novaeangliae),dugongs(Dugong dugon)etc.,species,and grunts which is commonly generated by Japanese gurnard(Chelidonichthys kumu),Grey gurnard(Eutrigla gurnardus),gulf toadfish(O.beta),etc.,species.We compared simulated results from this technique with values determined by theoretical approach.We have found that an increasing number of cross-correlation function(CCF)provide better results using this technique.However,the technique has some limitations including negligence of multipath interference,assuming the delays to be integer.展开更多
基金Supported by the National Natural Science Foundation of China(No.41906170)the Indian Ocean Ninety-east Ridge Ecosystem and Marine Environment Monitoring and Protection(No.DY135-E2-4)+1 种基金the Cooperation of Top Predators Observation in the Southern Ocean(No.QT4519003)the China-ASEAN Maritime Cooperation Fund。
文摘Although four species of odontocete and four species of baleen whale have been recorded in Prydz Bay,their vocalizations have been rarely investigated.Underwater vocalizations were recorded during March 2017 in Prydz Bay,Antarctica.Bio-duck sounds,downsweeps,inverted“u”shape signals,whistles,pulsed sounds,and broadband clicks were recorded.Bio-duck sounds and downsweeps were associated with Antarctic minke whales(Balaenoptera bonaerensis)based on visual observations.Similarities between inverted“u”shape signals,biphonic calls,and clicks with vocalizations previously described for killer whales(Orcinus orca)lead us believe the presence of Antarctic killer whales.According to sound structures,signal characteristics,and recording location,Antarctic type C killer whales were the most probable candidates to produce these detected calls.These represent the fi rst detection of inverted“u”shape signals in Antarctic waters,and the fi rst report of Antarctic killer whale in Prydz Bay based on passive acoustic monitoring.The co-existence of Antarctic minke and killer whales may imply that minke whales can detect diff erences between the sounds of mammal-eating and fi sh-eating killer whales.Our descriptions of these underwater vocalizations contribute to the limited body of information regarding the distribution and acoustic behavior of cetaceans in Prydz Bay.
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.
基金the project“BIOforIU project PONa3_00025-Multidisciplinary Infrastructure for the Study and Development of Marine and Terrestrial Biodiversity in the Perspective of Innovation Union.”Part of the analysis was funded under the National Recovery and Resilience Plan(NRRP),Mission 4 Component 2 Investment 1.4-Call for tender No.3138 of 16 December 2021,rectifed by Decree n.3175 of 18 December 2021 of Italian Ministry of University and Research funded by the European Union-NextGenerationEU,Award Number:Project code CN_00000033,Concession Decree No.1034 of 17 June 2022 adopted by the Italian Ministry of University and Research,Project title"National Biodiversity Future Center-NBFC".
文摘Accurate estimations of animal population size are pivotal for implementing management strategies properly.Recapture technique based on sounds as a specimen identifcation mark has barely been used for marine mammals.However,inferring abundance estimates from acoustic methods could enhance the accuracy and precision of population size assessments.Here,we tested the possibility of using signature whistles as individual marks for estimating the size of common bottlenose dolphin(Tursiops truncatus)populations.Data were continuously collected for 326 days in 2015–2016,by using a fxed acoustic device located in the Sicily Strait(Italy).The SIGID method was applied to identify Signature Whistles Types(SWTs)over 7,000 h of recordings.Eighty SWTs were detected as long as their stereotyped fundamental frequency contours were repeated in bouts of at least 6 renditions.The mean SWTs monthly recording rate resulted in 0.19(Standard deviation=0.16),with 20 SWTs recorded over 5 or more different encounters(until a maximum of 30 encounters).The Jolly–Seber model(with POPAN formulation)was run in Mark software to estimate the population size.The estimated population size resulted in 171 bottlenose dolphins(95%confdence interval=137–215).Even if the detection and identifcation of signature whistles required crucial precautions,and animals could be detected differently from visual techniques,the population size estimate obtained was comparable with previous results based on physical marks data.These outcomes demonstrated that signature whistles can be considered a strongly effective tool for integrating traditional mark-recapture techniques with fnely estimated dolphins’population abundances.
基金supported by the CNPC-SWPU Innovation Alliance Technology Cooperation Project(2020CX020000)the Natural Science Foundation of Sichuan Province(24NSFSC0808)the China Scholarship Council(202306440144).
文摘Passive source imaging can reconstruct body wave reflections similar to those of active sources through seismic interferometry(SI).It has become a low-cost,environmentally friendly alternative to active source seismic,showing great potential.However,this method faces many challenges in practical applications,including uneven distribution of underground sources and complex survey environments.These situations seriously affect the reconstruction quality of virtual shot records,resulting in unguaranteed imaging results and greatly limiting passive source seismic exploration applications.In addition,the quality of the reconstructed records is directly related to the time length of the noise records,but in practice it is often difficult to obtain long-term,high-quality noise segments containing body wave events.To solve the above problems,we propose a deep learning method for reconstructing passive source virtual shot records and apply it to passive source time-lapse monitoring.This method combines the UNet network and the BiLSTM(Bidirectional Long Short-Term Memory)network for extracting spatial features and temporal features respectively.It introduces the spatial attention mechanism to establish a hybrid SUNet-BiLSTM-Attention(SBA)network for supervised training.Through pre-training and fine-tuning training,the network can accurately reconstruct passive source virtual shot records directly from short-time noisy segments containing body wave events.The experimental results of theoretical data show that the virtual shot records reconstructed by the network have high resolution and signal to noise ratio(SNR),providing high-quality data for subsequent monitoring and imaging.Finally,to further validate the effectiveness of proposed method,we applied it to field data collected from gas storage in northwest China.The reconstruction results of field data effectively improve the quality of virtual records and obtain more reliable time-lapse imaging monitoring results,which have significant practical value.
基金suported by grants from the Knowledge Innovation Program of Chinese Academy of Sciences(No.KSCX2-EW-Z-4)the National Natural Science Foundation of China(No.31170501 and 31070347)+1 种基金the Special Fund for Agro-scientific Research in the Public Interest of the Ministry of Agriculture of China(No.201203086)the Ocean Park Conservation Foundation,Hong Kong.Some logistic support was provided by Wuhan Baiji Conservation Foundation and Societe Generale de Surveillance S.A.Special thanks are also extended to the academic editor and anonymous reviewers for their helpful critique of an earlier version of this manuscript.
文摘During the Yangtze Freshwater Dolphin Expedition 2012,Yangtze finless porpoises(Neophocaena asiaeorientalis)were acoustically monitored in 9 port areas at night.During 6566 min of nocturnal monitoring,porpoise sonar was detected for 488 min(7.43%of the total time).Of all 81 encounters,the longest echolocation span obtained was 102.9 min,suggesting frequent and prolonged porpoise occupation of the port areas.A combined total of 2091 click trains were recorded,with 129(6.2%)containing minimum inter-click intervals(ICIs)below 10 ms(termed a buzz).Buzzes with a decrease in ICIs and search and approach phases that resembled feeding echolocation signals accounted for 44.2%(N=52)of all buzzes.Buzzes with an increase in ICIs,suggesting a mirrored prey capture phase,accounted for 20.2%(N=26)and could reflect attempts to locate escaped prey because they were followed by approach-phase feeding buzzes.Anecdotal evidence of porpoises fleeing the proximity of vessels was observed.The recordings indicating clusters of porpoises feeding near the port areas suggest a forced choice for feeding due to the relatively higher prey availability in the port areas compared to other areas in the Yangtze River that are probably overfished.
基金supported by grants from the China National Offshore Oil Corporation foundation(grant number CF-MEEC/TR/2021-12)the Central Public-interest Scientific Institution Basal Research Fund,CAFS(grant number 2019ZD0201)the Bureau of Fisheries,the Ministry of Agriculture and Rural Affairs of the People’s Republic of China(grant number 125C0505),The research project was permitted by the Ministry of Agriculture and Rural Affairs of the People’s Republic of China.All procedures strictly adhered to Chinese law and ethical guidelines.
文摘The associations between feeding activities and environmental variables inform animal feeding tactics that max-imize energetic gains by minimizing energy costs while maximizing feeding success.Relevant studies in aquatic animals,particularly marine mammals,are scarce due to difficulties in the observation of feeding behaviors in aquatic environments.This data scarcity concurrently hinders ecosystem-basedfishery management in the context of small toothed-cetacean conservation.In the present study,a passive acoustic monitoring station was deployed in an East Asianfinless porpoise habitat in Laizhou Bay to investigate potential relationships between East Asianfinless porpoises and their prey.The data revealed that porpoises were acoustically present nearly every day during the survey period.Porpoise detection rates differed between spring and autumn in concert with activities offish choruses.During spring,fish choruses were present throughout the afternoon,and this was the time when porpoise vocalizations were the most frequently detected.During autumn,whenfish choruses were absent,porpoise detec-tion rates decreased,and diurnal patterns were not detected.The close association betweenfish choruses andfin-less porpoise activities implies an“eavesdropping”feeding strategy to maximize energetic gains,similar to other toothed cetaceans that are known to engage similar feeding strategies.Underwater noise pollution,particularly those maskingfish choruses,could interruptfinless porpoises’feeding success.Fisheries competing soniferousfishes withfinless porpoise could impactfinless porpoise viability through ecosystem disruption,in addition tofishing gear entanglement.
文摘Passive acoustic monitoring has the potential to be a useful tool for population estimation of sound-producing fish and mammals(mostly whales).Previous work on population estimates of callers employed a simple cross-correlation technique with recordings from two acoustic sensors,and the current work extends the technique to two configurations of a 3-acoustic sensors array using two different sounds,i.e.,chirps which is commonly generated by damselfish(Dascyllus aruanus),humpback whales(Megaptera novaeangliae),dugongs(Dugong dugon)etc.,species,and grunts which is commonly generated by Japanese gurnard(Chelidonichthys kumu),Grey gurnard(Eutrigla gurnardus),gulf toadfish(O.beta),etc.,species.We compared simulated results from this technique with values determined by theoretical approach.We have found that an increasing number of cross-correlation function(CCF)provide better results using this technique.However,the technique has some limitations including negligence of multipath interference,assuming the delays to be integer.
