The objective of this project was to develop relationships and models for the propagation of impulsive noise generated by military operations into communities such that the resultant vibration of and potential damage to structures can be assessed. Key elements of the objectives included establishing the importance of ground-borne propagation versus air-borne propagation, to identify or establish models to predict the waveforms that could impact a community, to predict the response of structures to impulsive noise, to collect response properties of a variety of buildings exposed to impulsive noise sources, and to establish the probability of damage of different structural types and materials.

The technical approach taken was a combination of analytical and field studies. The basic mechanisms of air-borne and ground-borne propagation were reviewed. Existing air-borne propagation models – Blast Noise Prediction (BNOISE) and Sound Intensity Prediction System (SIPS) – were identified. An approach to assessing ground versus air propagation paths, based on the relative timing of primary and secondary ground motion relative to air wave timing, was identified. A single degree of freedom model for structural response was prepared, and the key elements of a probability of damage model outlined. Propagation and response measurements were conducted on eleven buildings at three military facilities. Sources included artillery and tank gun firing and explosive ordnance disposal. Propagation data collected consisted of pressure wave measurements at heights of 4, 20 and 40 feet above the ground, and motion of the ground. These measurements were made near each structure and at several intermediate points between the source and the structure. Vibration was recorded on key elements - walls, windows and window frames - of all eleven structures. Modal analysis measurements were made on a selected subset of the buildings.

Results of the field measurements included demonstration that propagation was dominated by air-borne mechanisms (even for the shallow buried ordnance disposal sources) and a database of structural response data. Structural response measurements were processed into frequency response functions and resonant frequencies identified. These are key elements in the single degree of freedom structural response model. The response model extended to multimodal for damage assessment, where the additional detail was necessary for the estimation of peak stress. A Probability of Damage (POD) model was prepared, including data for a wide range of structural element types. The POD model has two parts: prediction of the incident wave and its statistics, and prediction of the resultant probability of damage. If the incident wave and its statistics are known (e.g., from SIPS or BNOISE), then that becomes the input to the structural part. If the incident wave is not known, then a source model based on distance and equivalent trinitrotoluene (TNT) weight is used.

The benefit of this research is quantification of the response of structures to military impulsive noise. An understanding of propagation mechanisms and structural response is presented. The models developed permit planners and range managers to perform risk analysis of proposed activities and provide guidance for impulsive noise criteria in communities.