OOS 7-3 - A spatial-temporal framework for analyzing soundscapes across climate change and land transformation gradients

Tuesday, August 13, 2019: 8:40 AM
M104, Kentucky International Convention Center
Bryan Pijanowski1, Kristen Bellisario1, Benjamin L. Gottesman1, Dante Francomano1, David T. Savage1 and Amandine Gasc1,2, (1)Department of Forestry and Natural Resources, Purdue University, West Lafayette, IN, (2)Europôle Méditerranéen de l'Arbois, Institut Méditerranéen de Biodiversité et d'Écologie, Cedex, France
Background/Question/Methods

Soundscapes are the acoustic signatures of a given place at a defined moment of time. Sounds from the biological community, the geophysical environment and those produced by humans compose a soundscape. Reduced biological sounds have been shown to reflect the reduction of species richness and abundances strongly correlating with ecological factors such as natural variation of vegetation structural complexity, human driven land transformations, overfishing (in marine systems), and the occurrence of invasive species, among others. Climate change is known to shift the timing of plant phenological events such as flowering and plant species composition, and some early research in soundscape ecology suggests that the biological composition of a soundscape, as reflected in all animal communication, may shift as well. In this presentation, we demonstrate a spatial-temporal soundscape framework that will allow soundscape ecologists to measure how climate change and land transformations may alter the composition of sounds from animal communities. We use a selection of studies from a Purdue soundscape database of over 4,000,000 recordings from a diverse set of biomes to demonstrate the efficacy of this framework.

Results/Conclusions

The spatial-temporal soundscape analysis framework presented here contains several new measures of change that have not been suggested in the literature currently. These included those that can be examined long-term (over an entire year) and over a short period of time (hourly, summarized over time) that represent climate, land transformation and climate-land interaction effects on a soundscape. Several big data visualizations have allowed us to match the theoretical framework with that of long-term, hourly audio recordings. These include alluvial diagrams, stream graphs, parallel coordinate graphs and false color “string” spectrograms. We suggest several field experimental designs, software tools, and data handling techniques that would allow for others to explore the role that climate change and land transformations may have on soundscapes and ultimately biodiversity.