Geospatial sound modeling

Mennitt, Daniel

Georeferenced maps of expected environmental sound levels across the United States.These maps were generated from Random Forest (Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32) models that investigated relationships between long term measurements of the ambient sound pressure level and non-acoustic geospatial features such as topography, climate, hydrology, and anthropogenic activity. These models were used to project expected sound levels under existing conditions and a natural scenario on landscape scales.The existing sound pressure level is the condition as measured; it includes all acoustic energy. A natural level was estimated by systematically minimizing contributions from all anthropogenic model inputs, leaving only biotic and abiotic sources of sound. The impact level is defined as the difference between the existing and natural sound pressure levels; this is most informative on local scales, like within park boundaries. The L50 is the sound pressure level exceeded half of the time and A-weighting is an adjustment that reflects how the human ear perceives sound. All conditions are predicted for a typical summer daytime hour.This project contains links to documents and geospatial data products. The primary products included are georeferenced maps of the expected existing, natural, and impact A-weighted hourly L50 sound pressure level dB re 20 uPa on a typical summer day across the contiguous United States, Alaska, and Hawaii regions at resolutions of 270m, 250m, and 270m respectively. Raster datasets for the explanatory variables considered are linked to this reference as product references.Because the existing sound level includes contributions from all biotic, abiotic, and anthropogenic sound sources, translation to other metrics are not advised. Maps of additional metrics such as the L10, L90, and Leq are forthcoming. For more information, please visit http://www.nature.nps.gov/sound/soundmap.cfm

Geospatial sound modeling

Products

Collections