Hierarchical method for single-observer, continuous sound source observer logging
Richard D. Horonjeff, Grant S. Anderson · The Journal of the Acoustical Society of America · 2007
Sound environments, whether urban or remote, often contain a variety of audible sound sources, each with different onset and offset times. Obtaining a continuous log of these sources can be a daunting task, especially for a single observer. In situations where it is impractical to deploy several observers to a single location, a hierarchical, forced-choice method has been developed. The method stratifies the sources into layers, each layer containing sources with common attributes, such as human-related or natural-indigenous. The method collects sufficiently detailed information for study objectives, without taxing the observers’ cognitive and recording abilities, and thereby introducing large observation or recording errors. So long as the onsets and offsets of audible sources in one layer are independent of those in another, statistical descriptors, such as the percent of time audible, can be determined for each layer. The highest layer in the hierarchy becomes an exhaustive sample, whereas lower layers become statistical samples with random blocks of time removed. The method has been applied in a number of national parks over the past 15 years, by several organizations and sponsors. Advantages and disadvantages of the method, along with inter-observer logging consistency results, are presented.