Phantom rivers filter birds and bats by acoustic niche

D G E Gomes,C D Francis,H J Cole,C A Toth,J R Barber

doi:10.1038/s41467-021-22390-y

Abstract

Natural sensory environments, despite strong potential for structuring systems, have been neglected in ecological theory. Here, we test the hypothesis that intense natural acoustic environments shape animal distributions and behavior by broadcasting whitewater river noise in montane riparian zones for two summers. Additionally, we use spectrally-altered river noise to explicitly test the effects of masking as a mechanism driving patterns. Using data from abundance and activity surveys across 60 locations, over two full breeding seasons, we find that both birds and bats avoid areas with high sound levels, while birds avoid frequencies that overlap with birdsong, and bats avoid higher frequencies more generally. We place 720 clay caterpillars in willows, and find that intense sound levels decrease foraging behavior in birds. For bats, we deploy foraging tests across 144 nights, consisting of robotic insect-wing mimics, and speakers broadcasting bat prey sounds, and find that bats appear to switch hunting strategies from passive listening to aerial hawking as sound levels increase. Natural acoustic environments are an underappreciated niche axis, a conclusion that serves to escalate the urgency of mitigating human-created noise.

Highlights

Natural sensory environments, despite strong potential for structuring systems, have been neglected in ecological theory
Birds with a peak vocalization frequency closer to the median of the background spectrum showed lower abundances, with declines of 10.0% (5.1–15.3%) for each 2 kHz increase in spectral overlap (Fig. 2B). These overlap-mediated effects interact with sound level in a diminishing way (Table S1): higher amplitude background noise resulted in weaker relationships between spectral overlap and bird abundance
We focused our analysis on bat species that employ both strategies and predicted that high sound-level acoustic environments would hinder bats’ use of passive listening and result in heightened use of sonar[18,26]

Summary

Introduction

Despite strong potential for structuring systems, have been neglected in ecological theory. There are 150,000 km of marine shoreline (NOAA 2014) and 5.6 million km of rivers and streams in the United States alone (US EPA 2014) that expose adjacent environments to the sounds of moving water We hypothesize that such intense natural acoustic sources have the power to structure habitat use[11]. We broadcast whitewater river noise from speaker arrays powered by solar panels and banks of batteries at five additional, naturally quiet streams using acoustic recordings taken from the highest sound level control sites. To understand the mechanisms underlying responses to the acoustic environment, we create a gradient of background spectra by broadcasting shifted river noise of an identical temporal profile, but shifted upwards in frequency (4.8 ± 1.3 kHz) at five additional quiet-stream sites (Fig. 1B). Our experimental design allows us to explicitly test the effects of sound level separately from those of background spectra

Methods

Results

Conclusion