Estimation of blast noise sound fields over large regions using noise monitors and geostatistical models

Abstract

This study explores the feasibility of accurately estimating blast noise levels over a large region between and beyond noise monitoring stations using geostatistical models. The potential improvements over propagation modeling include faster computations, fewer assumptions, and improved accuracy. The estimation models explored include kriging, simple interpolation, and models that include meteorological and terrain parameters commonly incorporated into outdoor sound propagation models. The estimation models are evaluated using both experimentally measured and simulated noise monitor data gathered under various atmospheric conditions in several large regions (e.g., greater than 16 km2). The performance of using geostatistical-based estimation models is discussed in terms of the uncertainty of sound pressure field estimates, sensitivity to atmospheric variability, sensor density and geometry configuration, and model validation.