Development of edge computing underwater sound recorder to monitor deep sea soundscape

Sound propagates at a greater range than light in marine environments. Anthropogenic sound and noise pollution caused by deep sea mining and offshore wind farms are increasing, which may affect the marine ecosystem. However, sound datasets in oceanic environments are quite limited. In this paper, we summarize the requirements of a deep sea soundscape monitoring device by reviewing currently available instruments, and we propose a cost-effective and noise reductive recording system to the monitor marine soundscape to identify temporal variations in the soundscape at various scales. The development of this system is ongoing, and we summarize the results of two trial deployments in a huge aquarium tank and off the pier of a marine station. The results demonstrate that the noise generated and the battery consumption of the proposed system are comparable to conventional systems. However, the proposed system is highly versatile due to the scalability of the IDE of the system’s software. Our work to develop a cost-effective, noise-reductant and long-term recording system will contribute to the acquisition of soundscape data in marine environments, which is expected to enable the identification of baseline sound properties in various marine environments and reconstruct temporal and spatial variations of marine biodiversity and ecosystem dynamics.