Comparing passive acoustic data of Aural and Sono.Vault recordings off Elephant Island for September 2013
In this study, we compare marine mammal acoustic presence data from two passive acoustic recording devices, Aural and Sono.Vault that recorded simultaneously off of Elephant Island northwest of the Western Antarctic Peninsula, to evaluate the effect of sampling rate and recording duty cycle on acous...
| Main Authors: | , , , |
|---|---|
| Format: | Article in Journal/Newspaper |
| Language: | unknown |
| Published: |
2020
|
| Subjects: | |
| Online Access: | http://oops.uni-oldenburg.de/4620/ http://oops.uni-oldenburg.de/4620/1/hots_ir_oops_2020.pdf |
| Summary: | In this study, we compare marine mammal acoustic presence data from two passive acoustic recording devices, Aural and Sono.Vault that recorded simultaneously off of Elephant Island northwest of the Western Antarctic Peninsula, to evaluate the effect of sampling rate and recording duty cycle on acoustic observations. Passive acoustic recordings sampled with two different sample rates (2.5 kHz and 16 kHz) and two recording duty cycles (five minutes per hour vs continuous) were compared. The aim of the study was to explore how a standard-used sampling rate and duty cycle compares to continuous recordings in terms of information about marine mammal acoustic presence. Applying a recording duty cycle can enable a longer recording period while at the same time effectively reducing data analyzing time. The analysis shows that there is no large difference in the results, when comparing duty cycled Aural data sampled with a frequency range of 16 kHz versus 2.5 kHz. Only killer whales (Orcinus orca) were detected more often in the data with higher sample rate, due to their high frequency vocalizations. The comparison of continuous Sono.Vault versus subsampled Aural data with a frequency range of 2.5 kHz displays that, as expected, with the continuous recording of Sono.Vault species are detected more often than in the sub-sampled recording of Aural. As a conclusion for this research site a sample rate of 2.5 kHz is suggested to be sufficient. To obtain reliable results on the occurrence of species, recordings may be continuous, although higher duty cycles than the one investigated here may also produce good results, but need to be investigated. |
|---|