Environmental conditions lead to shifts in individual communication, which can cause cascading effects on soundscape composition

Abstract

Climate change is increasing aridity in grassland and desert habitats across the southwestern United States, reducing available resources and drastically changing the breeding habitat of many bird species. Increases in aridity reduce sound propagation distances, potentially impacting habitat soundscapes, and could lead to a breakdown of the avian soundscapes in the form of loss of vocal culture, reduced mating opportunities, and local population extinctions. We developed an agent-based model to examine how changes in aridity will affect both sound propagation and the ability of territorial birds to audibly contact their neighbors. We simulated vocal signal attenuation under a variety of environmental scenarios for the south, central semi-arid prairies of the United States, ranging from contemporary weather conditions to predicted droughts under climate change. We also simulated how changes in physiological conditions, mainly evaporative water loss (EWL), would affect singing behavior. Under contemporary and climate change-induced drought conditions, we found that significantly fewer individuals successfully contacted all adjacent neighbors than did individuals in either the contemporary or predicted climate change conditions. We also found that at higher sound frequencies and higher EWL, fewer individuals were able to successfully contact all their neighbors, particularly in drought and climate change drought conditions. These results indicate that climate change-mediated aridification may alter the avian soundscape, such that vocal communication no longer effectively functions for mate attraction or territorial defense. As climate change progresses, increased aridity in current grasslands may favor shifts toward low-frequency songs, colonial resource use, and altered songbird community compositions.

Keywords: agent‐based model; aridity; climate change; evaporative water loss; house finch; vocal communication.

FIGURE 1

Pictogram of questions, hypotheses, and predictions for the ABM. The questions, variables, and values outlined in the orange box represent the FAB1 hypothesis, while the blue box represents the FAB2 hypothesis.

FIGURE 2

Virtual environment for a population of virtual birds. Each hexagon represents a bird's territory in which the bird moves around. Each bird has six adjacent neighbors, except for the hexagons on the grid edge, which were not included in the final calculations since they could not contact all six neighbors. A bird will either sing, move, or rest until all neighbors were contacted. Birds that have contacted all six neighbors will turn gray and that bird will stop exhibiting behavior for the rest of the day.

FIGURE 3

Heatmaps of frequencies affected by extreme temperatures across different territory sizes. We applied the ABM to multiple territory sizes ranging from 25 m radius territories to 1500 m radius territories across the audible bird song frequencies for the differences in drought versus contemporary weather data (a), and the differences in climate change drought versus climate change weather data (b). Cooler colors represent frequencies and territory sizes that would lead to fewer birds successfully contacting all neighbors under extreme conditions in both the extreme and climate change drought data, suggesting that selection may drive bird populations toward smaller territory sizes and higher frequency songs.

FIGURE 4

Population completion percentages across four different environment conditions: (a) mean contemporary, an average of air temperature (TAIR), relative humidity (RELH), and air pressure (PRES) from the ERIC Mesonet station from 2010 to 2019. (b) Contemporary drought, the weather data from the 2011 drought from the ERIC Mesonet station. (c) Mean climate change, 7.5°C TAIR increase, and 6% RELH decrease to the mean contemporary data. (d) Climate change drought, 7.5°C TAIR increase, and 6% RELH decrease to the contemporary drought weather data. Three frequencies that span the songbird frequency bandwidth were tested (4 kHz: Orange, 8 kHz: Blue, 12 kHz: Gray). Models without (solid) and with (dotted) the evaporative water loss (EWL) equation are included.

FIGURE 5

Total contacts decrease as territory size increases under extreme climate change conditions (i.e., the “worst case scenarios”) for 8 kHz. A total number of contacts is represented on the y‐axis and time (min) is represented on the x‐axis. We tested the model under these conditions without (solid) and with (dotted) the evaporative water loss (EWL) equation.