Photosynthetic oxygen bubble stream sounds from aquatic macrophytes, and their consequences for acoustic biodiversity inventories and acoustic communication in shallow freshwater settings

PLOS ONE | https://doi.org/10.1371/journal.pone.0317424 April 23, 2025 3 / 13 PLOS ONE Photosynthetic oxygen bubble stream sounds from aquatic macrophytes comparison. Recordings (both audio and video) were assessed in person; sounds were classi- fied into broad categories. Five POBS were selected to illustrate the diversity we encountered. For these exemplars, the peak frequency, duration of repeated sections, and other relevant qualities of the sounds were characterised. These were manually determined in Raven Lite2 based on the peak frequency range of sounds on the spectrogram. Terms suggested by Baker and Chesmore [36] were used to characterise the sound. For example, a syllable is a single burst of sound that is separate from other syllables. The primary recordings for this study were made in the sandy shallows of Gull Lake, Alberta, Canada, adjacent to the Summer Village of Gull Lake, in water 0.3 m deep or less. Recording took place during the daytime, under sunny conditions when macrophytes were actively photosynthesising, beginning between the hours of 10:00 and 14:00 in June, July, and August of 2021 and 2023. Additional recordings were obtained at other shallow, freshwater lakes and slow-moving waters in Alberta (e.g., Open Creek Reservoir, Blindman River near Bentley, Guinevere Park (Edmonton), and Wabamun Lake). Recordings were also made in aquaria indoors. A “50-gallon” (c. 32 cm X 121 cm basal dimensions X 50 cm high) glass aquarium with c. 20 cm water and 5+ cm of sand sub-

strate, sloped up towards the sides, was used to record corixid stridulations. A smaller “five-gallon” glass aquarium (c. 21 cm X 40.5 cm basal dimensions X 27 cm high) with c. 24 cm water depth and a c. 3 cm sand substrate was used for sago pondweed aquarium recordings. The latter was illuminated by an overhead halogen desk lamp and light from a north-facing window c. 15 cm away. Two gas samples, c. 1.0 ml each, were collected from sago pondweed bubble streams in Gull Lake, and in the five-gallon aquarium. Gas chromatography with thermal conductivity detection (GC-TCD), was used to quantify oxygen (O2) concentrations. This method provides a reasonably accurate estimate of O2 content, particularly when O2 is present at or near 100%, due to TCD’s sensitivity to thermal conductivity differences between gases. However, other potential gases in the sample, such as carbon dioxide (CO2) and nitrogen (N2) may lie below the detection limits of the method (100–1000 ppm). Results In total, we obtained and examined 309 hydrophone recordings, including 98 from the sandy shallows of Gull Lake (Photo 1). The Gull Lake recordings cover more than 160 minutes and in addition to the focal sounds they also record our own vocal slating, wave sounds and sounds of shifting sands (on windy days), and the sounds of large gas bubbles emanating from the sediments in pocket (pock) marks kept free of vegetation by occasional gas upwellings. We interpreted the latter as methane craters, although we did not analyse the gasses produced, and more were less than one metre in diameter.

The clear shallow waters on the south end of Gull Lake made our work relatively straightforward, since direct observation of potential sound sources was possible. In terms of biological sounds, we determined that the majority of sounds on our recordings were produced by bubble streams emanating from the tissues of sago pondweed plants (Stucke- nia pectinata). In ten simultaneous hydrophone and underwater video recordings, bubble streams were visible from all five of the sound producing plants, and on none of the five soundless plants. The only species of macrophyte apparent to us in the Gull Lake sites was sago pondweed, a common plant in lakes and ponds in the region. Once we knew what to look for, we were able to identify a bubble stream coming from a sago pondweed plant each time we heard a POBS on the hydrophone. We also recorded the sound of one bubble-stream-producing sago