首页> 外文会议>Oceans MTS/IEEE Conference >Up close and personal: recording humpback whale song at close ranges (10-50m)
【24h】

Up close and personal: recording humpback whale song at close ranges (10-50m)

机译:关闭和个人:在关闭范围(10-50米)的录音驼背鲸歌曲

获取原文

摘要

Summary form only given. Male humpback whales (Megaptera novaeangliae) produce long, complex vocalizations consisting of a hierarchically-structured sequence of distinctive broadband 'units', each of a few seconds' duration. The function of this 'song' remains unclear. Singers are often observed in a stationary position with their body inclined downwards while vocalizing for about 10-12 minutes. The posture is suggestive of the singer attempting to optimize song propagation, but without calibrated measurements of the sound speed profile, source characteristics, range and orientation to the singer at close range, such ideas are difficult to test. Previous recordings have been made at considerable distance, where absorption, multipath and other propagation effects are imposed on the signal and the orientation to the singer is unknown. Prior recordings are also usually of low bandwidth. Recent work, however, has suggested that, close to the singer, song units may contain significant energy up to 20 kHz. The Acoustic Research Laboratory, Tropical Marine Science Institute at the National University of Singapore in collaboration with the Kewalo Basin Marine Mammal Laboratory of the University of Hawaii have deployed two types of acoustic recording systems in close proximity to singers in addition to CTD casts taken by the US Geological Survey. The first acoustic system is a digital video camera in an underwater housing with built-in hydrophone, recording at 44.1 kSa/s and operated by rebreather divers who are able to simultaneously ascertain both orientation and range to the singer using an ultrasonic depthsounder. The second is a custom-built self-contained high-frequency system with four hydrophones in an array that recorded at 250 kSa/s per channel. Both systems provide extremely high signal-to-noise ratio (typically 60 dB) calibrated recordings without propagation artifacts. Individual song units have been analyzed for power spectral density. Results indicate that most of the energy is located below 5 kHz with occasional harmonics up to 8 kHz. Individual units from a 6-minute sequence of 134 units were classified into 12 types with four classes containing 80% of the units. Comparisons of the power spectral densities within and across the four most common classes were conducted. Variations within classes were found to be within 7 dB when normalized for total power. Differences between the classes were small (up to 5dB) in the range of 500-1800 Hz and larger (up to 14dB) in the range from 1800-3500 Hz. The high-frequency recording system provides independent verification within the overlapping frequency range of both systems as well as information about the frequency content above 22 kHz up to 120 kHz.
机译:摘要表格仅给出。雄性驼背鲸(兆瓦Novaeangliae)生产长,复杂的发声,由分层结构化的独特宽带'单位'序列组成,每秒持续时间。这个'歌'的功能仍然不清楚。歌手经常在静止位置观察,他们的身体向下倾斜,同时发声约10-12分钟。姿势暗示歌手试图优化歌曲传播,但没有校准测量声速曲线,源特征,范围和方向近距离,这些想法难以测试。以前的距离已经在相当远的距离,其中对信号产生了吸收,多径和其他传播效应,并且对歌手的方向未知。之前的记录通常也是低带宽的。然而,最近的工作表明,靠近歌手,歌曲单位可能包含高达20 kHz的重要能量。新加坡国立大学热带海洋科学研究所的声学研究实验室与夏威夷大学的Kewalo盆地海洋哺乳动物实验室合作已经部署了两种类型的声学记录系统,除了由CTD铸件采取的CTD铸件美国地质调查。第一个声学系统是一个数字摄像机,在水下外壳中,内置水下的水箱,在44.1 ksa / s录制,并由rebreanther divers操作,他们能够使用超声波深度使用超声波深度来同时确定歌手的方向和范围。第二种是定制的自包含高频系统,其中阵列中有四个流水声,记录在每通道250 ksa / s处。两个系统提供极高的信噪比(通常为60 dB)校准记录,无需传播伪影。已经分析了各个歌曲单元以进行功率谱密度。结果表明,大多数能量位于5kHz以下,偶尔次谐波高达8 kHz。从6分钟的134个单位序列的单位分为12种类型,其中四种含有80%的单位。进行了四个最常见的类内部和跨越的功率谱密度的比较。当总功率标准化时,发现类内的变化在7 dB内。课程之间的差异在500-1800 Hz的范围内小(最多5dB),更大(高达14dB)的范围为1800-3500Hz。高频记录系统在两个系统的重叠频率范围内提供独立的验证,以及关于高于22kHz的频率内容的信息高达120 kHz。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号