PS 72-122
Ecological implications for long term acoustic monitoring at NEON observatory sites

Thursday, August 13, 2015
Exhibit Hall, Baltimore Convention Center
Samuel Denes, Department of Biology, Syracuse University, Syracuse, NY
Susan E. Parks, Department of Biology, Syracuse University, Syracuse, NY
Pramod Varshney, Electrical Engineering & Computer Science, Syracuse University, Syracuse, NY
Kurt Fristrup, Natural Sounds and Night Skies Division, National Park Service, Fort Collins, CO

The National Ecological Observatory Network (NEON) has established a framework through which a variety of metrics are measured at stations located across the United States.  In addition to the measurements already included within the framework, continuous stereo acoustic recordings are currently being made at the Harvard Forest site (Petersham, MA), adjacent to a NEON sampling locations.  The goal of this project is to demonstrate the power of acoustical monitoring to supplement existing NEON data collection.  By utilizing long-term recordings, a wealth of data relating to species presence, animal behavior, and anthropogenic disturbance can be collected without the presence of human researchers.  The phenology of biological events such as avian, amphibian or insect chorusing activity can be determined through automated band limited energy detectors. These recordings also allow the measurement of the relative contributions of anthropogenic activity to the cumulative sound energy at these locations. The most common influence of human activity in the acoustic record is from aircraft overflights.  Aircraft can be quantified using automated detection tools in a time-efficent manner. Counts, duration of aircraft noise, and/or the cumulative energy from these flights can be used to quantify their contribution to the acoustic environment. 


Bird vocalizations are detectable throughout the year, however during spring, a change in vocalization behavior was clearly detectable in the acoustic record.  Assuming a qualitative definition of chorusing as the persistent presence of bird vocalizations, the onset of bird chorus can be identified from long term spectral plots of the acoustic data.  For the recorder located in Harvard Forest in 2014, the onset of the spring bird chorus was on May 6th.  A random sample of 10 24-hour periods was selected for analysis from the Harvard Forest location for detection of anthropogenic activity.  The dates included spanned the seasons (11/13/2013 – 6/7/2014) and included various weather events. The number of aircraft detected per hour varied within and between days.  The number of aircraft detected was highest overnight during periods without weather events such as wind and rain, often with over 10 flights per hour. These aircraft included high-altitude jets and low altitude propeller aircraft, with most low altitude propeller aircraft detected during the day. Adding acoustics to the measurements already collected under NEON protocol can provide high-resolution information on the presence and behavior of acoustically active species in birds, frogs, and insects, as well as the acoustic impact of human activity at these locations.