The application of an actual stratified geoacoustic model for sound propagation is quantitatively assessed using joint geoacoustic and hydroacoustic survey experimental data from the South China Sea.Three sound propag...The application of an actual stratified geoacoustic model for sound propagation is quantitatively assessed using joint geoacoustic and hydroacoustic survey experimental data from the South China Sea.Three sound propagation models,namely,the ray model BELLHOP,the parabolic equation model RAM,and the normal mode model KRAKENC,are applied individually to characterize an actual complex,stratified geoacoustic model and subsequently calculate the sound transmission loss along the survey line.The simulated transmission losses from the three sound propagation models are compared with the measured results to assess their capabilities,particularly computational accuracy and efficiency,in handling complex geoacoustic environments.The influence of the seabed’s stratified structure and its diverse geoacoustic parameters on sound propagation is also analyzed by calculating the root mean square error between the theoretical and measured sound propagation losses.The compatibility of the actual stratified geoacoustic model with sound field predictions is enhanced through a sensitivity analysis of the above parameters,ultimately leading to the development of a finely calibrated geoacoustic model.展开更多
The structural characteristics of underwater acoustic fields are governed by propagating modes.Underwater acoustic signal processing requires accurate identification of both the number and types of these modes,which a...The structural characteristics of underwater acoustic fields are governed by propagating modes.Underwater acoustic signal processing requires accurate identification of both the number and types of these modes,which also has significant potential applications in underwater target detection,underwater communication,and ocean environment inversion.For a vertical line array(VLA),the conventional singular value(SV)method can estimate the mode number in high signal-to-noise ratio(SNR)conditions,but it fails to discriminate modal types and suffers rapid performance degradation in practical low-SNR environments.To overcome these limitations,a novel optimal projection residual(OPR)method that simultaneously identifies the number and types of acoustic propagating modes in the acoustic field of a VLA was developed.Comprehensive numerical simulations and experimental data demonstrate two distinct advantages of the proposed OPR method compared to the conventional SV method:1)the OPR method enables more accurate estimation of the number of propagating modes,and 2)it can efficiently identify both the number and types of acoustic propagating modes.展开更多
Sound is considered an important aspect of an ecosystem and acoustic methods have emerged as effective tools for ecosystems research.Xincun Lagoon,Hainan Island,is an important ecosystem characterized by dense seagras...Sound is considered an important aspect of an ecosystem and acoustic methods have emerged as effective tools for ecosystems research.Xincun Lagoon,Hainan Island,is an important ecosystem characterized by dense seagrass,which has been declining due to increased human activities,raising great concerns.Previous studies have identified various threats to seagrass,including heavy metal pollution,poor quality water,and so on.In this study,we investigate sources and levels of noise in seagrass beds and attempt to point out potential threats from noise pollution.A line array of six hydrophones was deployed over a period of seven days,from January 15 to January 21,2024.The recordings captured various sounds from marine life,human activities,and natural processes.Biological sounds,such as fish sounds and whale calls,were the most prevalent.Low-frequency noise from wind and tide were often recorded.Xincun Bay hosts more than 1500 fishing vessels;however,due to bad weather conditions that kept most vessels docked during the recording period,only one segment of boat noise was recorded;it lasted for 7 minutes,exhibiting strong energy over a broad frequency band.This event underscores the necessity of long-term monitoring of noise to identify and evaluate not only boat noise but other noise sources,especially ones that are intermittent but strong,that were not encountered during the limited period of observation on which this report is based.展开更多
Forward scattering detection in shallow-water environments presents many challenges,particularly the issues of environmental uncertainties and direct blast,which is an intense sound wave that propagates directly from ...Forward scattering detection in shallow-water environments presents many challenges,particularly the issues of environmental uncertainties and direct blast,which is an intense sound wave that propagates directly from the source to the receiver without interaction with the target.In this paper,we account for environmental uncertainties and extend the generalized likelihood ratio detector(GLRD)for forward scattering detection in a known environment to uncertain environments.In a suitable bistatic sonar configuration where the source is positioned on the broadside of a large aperture horizontal linear array(HLA),the GLRD exhibits good resistance to direct blast.Moreover,the GLRD demonstrates a certain degree of robustness against environmental uncertainties,particularly when the sampling uncertainty sets of the direct blast/signal wavefront are large enough—including both the real direct blast wavefront and the real signal wavefront.Despite facing the challenge of direct blast in forward scattering detection,the GLRD still performs well in this scenario and demonstrates its effectiveness as a method for forward scattering detection in uncertain shallow-water environments.展开更多
The acoustic properties of seafloor sediments are crucial for accurate acoustic field prediction,seafloor resource exploration,and marine disaster prevention.However,traditional prediction equations,often based on lab...The acoustic properties of seafloor sediments are crucial for accurate acoustic field prediction,seafloor resource exploration,and marine disaster prevention.However,traditional prediction equations,often based on laboratory-measured sound speeds,suffer from low precision and discrepancies with in situ measurements.To address these issues,we employed eXtreme Gradient Boosting(XGBoost)machine learning algorithms to develop high-precision in situ sound speed prediction models for seafloor sediments.The models were constructed using in situ sound speed and sediment physical property data(density,water content,porosity,median grain size,and grain group content)from 48 sites in the East China Sea shelf.Through feature parameter reduction and hyperparameter optimization,the optimal XGBoost model achieves training and validation R2 values of 0.989 and 0.977,respectively,having hyperparameters set at n_estimators=49 and max_depth=6.Compared to other machine learning models and empirical equations,the XGBoost model based on density,water content,sand content,and median grain size exhibited the lowest mean absolute error(MAE)and mean absolute percentage error(MAPE)at 5.603 m/s and 0.366%,respectively.This represents significant improvements over existing models,with MAE reductions ranging 2.165–118.903 m/s and MAPE reductions 0.137%–7.657%.This study thus provides an innovative and highly accurate method for predicting the in situ sound speed of seafloor sediments.展开更多
基金supported by the National Key Research and Development Program of China(No.2023YFC 3107703)the National Natural Science Foundation of China(No.42206195)the Development of an in situ System for Measuring Marine Sediment Acoustics based on High-Frequency Microvibration Injection Technology(No.U200620147).
摘要The application of an actual stratified geoacoustic model for sound propagation is quantitatively assessed using joint geoacoustic and hydroacoustic survey experimental data from the South China Sea.Three sound propagation models,namely,the ray model BELLHOP,the parabolic equation model RAM,and the normal mode model KRAKENC,are applied individually to characterize an actual complex,stratified geoacoustic model and subsequently calculate the sound transmission loss along the survey line.The simulated transmission losses from the three sound propagation models are compared with the measured results to assess their capabilities,particularly computational accuracy and efficiency,in handling complex geoacoustic environments.The influence of the seabed’s stratified structure and its diverse geoacoustic parameters on sound propagation is also analyzed by calculating the root mean square error between the theoretical and measured sound propagation losses.The compatibility of the actual stratified geoacoustic model with sound field predictions is enhanced through a sensitivity analysis of the above parameters,ultimately leading to the development of a finely calibrated geoacoustic model.
基金supported by the National Natural Science Foundation of China(Nos.12304504,12304506,and U22A2012)the Youth Innovation Promotion Association,Chinese Academy of Sciences(No.2021023)+1 种基金the Strategy Priority Research Program(Category B)of Chinese Academy of Sciences(Nos.XDB0700100 and XDB0700000)the Natural Science Foundation of Tianjin(No.22JCYBJC00070).
摘要The structural characteristics of underwater acoustic fields are governed by propagating modes.Underwater acoustic signal processing requires accurate identification of both the number and types of these modes,which also has significant potential applications in underwater target detection,underwater communication,and ocean environment inversion.For a vertical line array(VLA),the conventional singular value(SV)method can estimate the mode number in high signal-to-noise ratio(SNR)conditions,but it fails to discriminate modal types and suffers rapid performance degradation in practical low-SNR environments.To overcome these limitations,a novel optimal projection residual(OPR)method that simultaneously identifies the number and types of acoustic propagating modes in the acoustic field of a VLA was developed.Comprehensive numerical simulations and experimental data demonstrate two distinct advantages of the proposed OPR method compared to the conventional SV method:1)the OPR method enables more accurate estimation of the number of propagating modes,and 2)it can efficiently identify both the number and types of acoustic propagating modes.
基金supported financially by the Director General’s Scientific Research Fund of Guangzhou Marine Geological Survey(Grant Number:2023GMGSJZJJ00029).
摘要Sound is considered an important aspect of an ecosystem and acoustic methods have emerged as effective tools for ecosystems research.Xincun Lagoon,Hainan Island,is an important ecosystem characterized by dense seagrass,which has been declining due to increased human activities,raising great concerns.Previous studies have identified various threats to seagrass,including heavy metal pollution,poor quality water,and so on.In this study,we investigate sources and levels of noise in seagrass beds and attempt to point out potential threats from noise pollution.A line array of six hydrophones was deployed over a period of seven days,from January 15 to January 21,2024.The recordings captured various sounds from marine life,human activities,and natural processes.Biological sounds,such as fish sounds and whale calls,were the most prevalent.Low-frequency noise from wind and tide were often recorded.Xincun Bay hosts more than 1500 fishing vessels;however,due to bad weather conditions that kept most vessels docked during the recording period,only one segment of boat noise was recorded;it lasted for 7 minutes,exhibiting strong energy over a broad frequency band.This event underscores the necessity of long-term monitoring of noise to identify and evaluate not only boat noise but other noise sources,especially ones that are intermittent but strong,that were not encountered during the limited period of observation on which this report is based.
基金Project supported by the National Natural Science Foundation of China(Grant No.12004335)。
摘要Forward scattering detection in shallow-water environments presents many challenges,particularly the issues of environmental uncertainties and direct blast,which is an intense sound wave that propagates directly from the source to the receiver without interaction with the target.In this paper,we account for environmental uncertainties and extend the generalized likelihood ratio detector(GLRD)for forward scattering detection in a known environment to uncertain environments.In a suitable bistatic sonar configuration where the source is positioned on the broadside of a large aperture horizontal linear array(HLA),the GLRD exhibits good resistance to direct blast.Moreover,the GLRD demonstrates a certain degree of robustness against environmental uncertainties,particularly when the sampling uncertainty sets of the direct blast/signal wavefront are large enough—including both the real direct blast wavefront and the real signal wavefront.Despite facing the challenge of direct blast in forward scattering detection,the GLRD still performs well in this scenario and demonstrates its effectiveness as a method for forward scattering detection in uncertain shallow-water environments.
基金Supported by the National Natural Science Foundation of China(Nos.u2006202,42376076,42074140,42106072)the Shandong Province Higher Education Youth Innovation and Technology Support Program(No.2024KJG033)。
摘要The acoustic properties of seafloor sediments are crucial for accurate acoustic field prediction,seafloor resource exploration,and marine disaster prevention.However,traditional prediction equations,often based on laboratory-measured sound speeds,suffer from low precision and discrepancies with in situ measurements.To address these issues,we employed eXtreme Gradient Boosting(XGBoost)machine learning algorithms to develop high-precision in situ sound speed prediction models for seafloor sediments.The models were constructed using in situ sound speed and sediment physical property data(density,water content,porosity,median grain size,and grain group content)from 48 sites in the East China Sea shelf.Through feature parameter reduction and hyperparameter optimization,the optimal XGBoost model achieves training and validation R2 values of 0.989 and 0.977,respectively,having hyperparameters set at n_estimators=49 and max_depth=6.Compared to other machine learning models and empirical equations,the XGBoost model based on density,water content,sand content,and median grain size exhibited the lowest mean absolute error(MAE)and mean absolute percentage error(MAPE)at 5.603 m/s and 0.366%,respectively.This represents significant improvements over existing models,with MAE reductions ranging 2.165–118.903 m/s and MAPE reductions 0.137%–7.657%.This study thus provides an innovative and highly accurate method for predicting the in situ sound speed of seafloor sediments.