Detection algorithms identify significant sound segments in recordings from passive acoustic recorders, suggesting the presence of various aquatic species.
Long-term passive acoustic recorders (LTPARs) are useful for studying changes in underwater soundscapes by capturing variation in sounds produced by fish mating choruses, marine mammals, and invertebrates (biophony); human activity (anthrophony); and the physical environment (geophony). With advances in storage, LTPARs can record for weeks, months, or longer, capturing information about changes in soundscapes on multiple scales. However, researchers cannot listen to and analyze these recordings in reasonable timeframes. Therefore, we use detection algorithms to obtain reduced sets of recordings for detailed analysis. Here we describe the Power Spectral Band Sum (PSBS) used to identify segments of interest in a recording where target sounds, such as fish choruses, are present or where target sounds change in loudness. The PSBS adds the power spectrum components for all frequencies in a band characteristic to a sound of interest. These quantities are computed without difficulty for long recordings. We give examples of PSBSs of long-term recordings and present the “unknown buzz” recorded off NC, leading to understanding the presence of Atlantic midshipmen (Porichthys plectrodon) populations off NC and FL, and perhaps a related species of midshipmen in the Indian Ocean. The PSBS could help identify and locate these and other never-recorded fish choruses.
No takes yet. Share an insight, caveat, or question.
Sprague et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: