Noise Exposure in U. S. Protected Areas

Background: Noise has increased rapidly in the past four decades, such that most places in the United States are exposed to the impacts of noise.  The spatial spread of noise has accelerated due to the pace of exurban development, energy production, and growth of transportation networks.  More than 250 studies published in the past decade have documented negative consequences of noise for wildlife and outdoor recreation. No protected area is impervious to noise exposure. An integral task for managing noise in protected areas is to understand the spatial distribution and sources of noise.  Effective tools to support this task must support spatial planning on enormous landscape scales.  We present geospatial model predictions of sound levels in the contiguous United States to examine patterns of noise exposure across different conservation land designations. Methods: Geospatial models of environmental sound pressure levels were constructed using a suite of physiographical and anthropogenic explanatory variables in a Random Forest machine learning framework.  Response data, obtained from the archive of National Park Service recordings, included A-weighted L₉₀, L₅₀, and L₁₀ exceedance metrics calculated from 1.5 million h of acoustical measurements from 492 sites in the contiguous United States. To estimate levels of anthropogenic noise exposure, we calculated the difference between predicted existing sound levels and estimated natural sound levels. We summarized resulting noise exposures in protected areas, defined as land managed for the preservation of biodiversity and other natural, recreational, and cultural uses.  Boundaries were obtained from the US Geological Survey’s Gap Analysis Program. Results: Of all protected areas, we found that noise exposures were highest in city and county protected land; these areas experienced similar noise levels as surrounding unprotected land.  Wilderness areas experienced the least noise, though exposure increased with proximity to urban areas.  Within National Park units, local noise sources related to park operations (e.g. grounds care and generator motors) were common, which are likely to be the easiest to mitigate.   Although critical habitat, designated for species listed under the U.S. Endangered Species Act, generally experience the same noise exposure level as most federally protected land, 17 threatened animal species experience more than 10 times higher noise energy than predicted natural levels. Conclusions: These analyses document the expanse of noise exposure in protected areas, quantify the levels of noise that individual units experience, and identify key sources of noise.  This information enables managers to identify targets and areas where noise management could yield the greatest benefits, and can contribute to a decision analysis framework for conservation.