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 2 / 13 PLOS ONE Photosynthetic oxygen bubble stream sounds from aquatic macrophytes Sound production by marine algae on coral reefs [13,14] and by seagrass beds [17,18] are perhaps the best-known examples. Algae in coral reefs release oxygen bubbles, creating ping- ing [14] and ticking sounds [10,13]. Similar to algae, freshwater macrophytes release oxygen as gas bubbles, either through the leaf surface (a process that aquarists refer to as “pearling”) [14] or as bubble streams from damaged tissues. Some aquatic beetles, known as plant breath- ers, pierce aquatic plants to replenish the oxygen in their plastron (physical gill) from the aerenchyma of the plant [19–21]. From damaged tissue, oxygen is able to escape, potentially contributing to what Rountree et al. refer to as a “cacophony of sounds” [12]. Bubble streams can be used to assess photosynthetic activity [9,10,13–15,22,23]. However, there has been a greater focus on controlled laboratory settings and the physics of regular bubble streams [15] than on the situation in the field. Lab settings allow for precise measure- ments of gas composition, bubble size, and the timing of bubble release. However, in natural environments, bubble production is subject to changing light regimes, wave action, damage to plant tissues [15], plant tissue elasticity, and other factors [24–28]. Determining the origins of aquatic sounds is a complicated process. Unknown sounds in

freshwater are often attributed to fishes or insects [9–12], perhaps due to acoustic similari- ties to the sounds of known fishes, or those of terrestrial insects such as crickets and cicadas. Generally, however, soundscape studies are conducted in deeper water, where sport fish occur, often in sites too deep for rooted macrophytes. In shallower settings, the stridulations produced by aquatic insects have attracted some research attention, especially those of water boatmen (Hemiptera: Corixidae) [29–31]. Most aquatic insect sounds attenuate quickly over distance in water [32–34], making the notion of a soundscape highly localised for most organ- isms, especially shallow water organisms. In the course of prospecting for undescribed insect sounds in shallow freshwater lakes in Alberta, Canada, we encountered additional sounds produced by aquatic macrophytes, and set about characterising these sounds and confirming their origins. Our results contrast somewhat with those of other published studies, with consequences for soundscape ecology in similar environments. Materials and methods Recordings were made using an Aquarium Audio H2a-XLR hydrophone, a Sound Devices Mix Pre microphone preamp, and a Tascam Linear PCM Recorder/Mixer DR-60D with gain settings maximised on both the preamp and the sound recorder. The sampling rate was 44100 Hz. Spectrograms and waveforms were created from unprocessed selected exemplar record- ings using Raven Lite2 software [35]. Spectrograms were produced with a short-time Fourier transform and a percent overlap (hop size) of 50%.

Recordings were made while wading in the shallows of Gull Lake. The hydrophone was handheld, and thus slight changes in its position affected the amplitude of recorded sounds, since these sounds attenuated markedly at small distances from the hydrophone. The hydrophone was typically held within the bottom 10 cm of the water column and sus- pended in a way that it was not touching plants or the substrate, to avoid friction sounds. Visual assessments were conducted during recordings via direct observation from above the water, through polarised sunglasses (the water is typically clear enough for easy viewing), as well as underwater with waterproof cameras, including a GoPro Hero, Safari 3 HD Action Camera, and a Panasonic Lumix TS20. The date, time, location, and in situ assessment of the apparent sound sources were noted as a verbal “slate” on the recording. Once we realised the possibility that most of the sounds were produced by bubble streams emanating from sago pondweed, we obtained simultaneous hydrophone and underwater video recordings (using a Lumix TS20 camera) of five sound-source sago pondweeds, and five silent sago pondweeds for Data availability statement: All relevant data (i.e., sound files) are fully available without restriction, as uploads to Xeno-canto, a public repository (https://xeno-canto.org/). The sound files referenced in the manuscript have the following Cat.nr.: XC920087, XC893189- XC893193. They are all found under the following user profile: https://xeno-canto.org/ contributor/ZXWJFVRVKA?dir=0&order=xc Funding: The author(s) received no specific

funding for this work. Competing interests: The authors have declared that no competing interests exist.