The current methods used by the U.S. Army and Department of Defense for predicting community response to blast noise—the noise produced by large weapons, heavy artillery, and explosions—have low predictive validity and do not account for the intermittent and dynamic nature of the blast noise environments that occur on and around military installations. The models in use were based on five studies conducted between 1965 and 1995, each of which shows a large variability in annoyance to blast noise. These studies were conducted with noise measurement technologies that have since been replaced by the more accurate and advanced technologies available today. In addition, each study used different procedures to quantify the blast noise environment, which may explain the drastic variability in results.

This research focused on three objectives: (1) investigate the metrics currently used by the U.S. military to assess high-energy impulsive noise (i.e., blast noise) and assess whether these metrics adequately account for the intermittent, impulsive, nature of blasts; (2) examine the extent to which individual complaints are indicative of general community annoyance; and (3) recommend scientifically defensible criteria that can be used to actively manage noise at an installation so that the sustainability of testing and training is ensured.

Technical Approach

Three main studies were conducted over a five-year period: the Complaint Survey, the In Situ Study, and the General Community Survey (GCS). These studies were conducted at three military testing and training facilities spread across the Continental United States. The research also included a reanalysis of some early blast noise complaint work that looked at the probability of receiving a blast noise compliant, and a smaller scale study, which compared the language/terminology used by people living near military installations when describing their community, environment, and blast noise to the language used in surveys that were to be administered during the three main studies.

The Complaint Survey directly addressed Objective 2. The In Situ Study and the GCS Study were more labor-intensive measurement-based efforts that addressed Objective 1. The In Situ study was designed to enhance the understanding of how individuals respond to blast events over a short time period (i.e., on the order of hours or days) while going about their daily lives, and the GCS study was designed to enhance the understanding of how communities respond to blast events over a longer time period (i.e., on the order of weeks or months).

There are some important differences between the studies in this project and previous community response to military noise studies. Primarily, unlike previous military noise studies that relied on noise propagation models to estimate the noise environment, these studies aimed to measure the blast noise level of each and every blast event that occurred over the entire study time period. This proved to be very difficult, but was instrumental to achieving Objective 3.


Across all of the studies, blast noise was clearly the most annoying noise source among the eight sources asked about on the survey. It was found that individuals, communities, and installations (i.e., the communities around an installation) all have a unique tolerance to blast noise and that the blast noise environment in each community is unique as well, which suggests that it may not be appropriate to assess and/or manage testing and training noise with universal criteria. Rather, the findings in this study suggest that it may be appropriate to assess and manage military noise on a community-by-community basis.

The complaint studies found that those who complain about blast noise report much higher annoyance than their non-complaining neighbors, which may suggest that installations should not immediately implement testing and training restrictions to appease a few individuals, and installations might be overly restrictive with their local noise management policies. However, this statement should not be interpreted that noise complaints be dismissed outright. This study also found that the probability of receiving blast noise complaints is rather low, and that high levels (e.g., unweighted peak levels > 120 dB) occur often without complaints. The complaint work did, however, find that the probability of receiving a complaint was high in a few communities, which further supports the hypothesis that each community has a unique tolerance to noise. Lastly, this study found that the C-weighted Day-Night-Level (CDNL or LCdn) calculated over a 24-hour time period better predicts the probability of receiving a complaint than the current standard metric, maximum unweighted peak level (LZpk).

The metrics currently used to assess and manage blast noise do not adequately capture the dynamic nature of blast noise environments produced by testing and training activities; however, the results from the in situ studies and GCS studies found no evidence to move away from using the LCdn as the preferred metric to predict community annoyance. Rather, it was found that there was a higher correlation with community annoyance between blast noise levels (LCdn) calculated over a short time period (i.e., 4 weeks) in comparison to blast noise levels calculated over a longer time period (i.e., 9-12 months).

The GCS studies found the following non-acoustical factors contributed to community annoyance. There was increased annoyance if the respondent (1) indicated that there was a time of day when blast noise was most disturbing, (2) spent more time at home between the hours of 6:00 and 10:00 PM than the average, (3) indicated that noise had disturbed them (irritated, startled, frightened) in the last 12 months, (4) heard blasts more frequently, (5) thought their neighborhood was an excellent place to live, and (6) ever experienced more than one object rattling in response to blast noise. There was decreased annoyance if (1) more than the average number of adults lived in the household, (2) the importance of the installation was larger than average, (3) the respondent lived in a single-family attached home, and (4) the respondent felt they could habituate to noise. The GCS studies also found that the annoyance to blast noise varied by community, thus supporting the community tolerance hypothesis; however, the variance between communities decreased once the non-acoustical factors were accounted for.


This project enhanced the current understanding of community response to blast noise. It was the first of its kind to attempt to measure each and every blast event that occurred over the entire 5-year project, which resulted in (perhaps) the most comprehensive examination of the dynamic nature of the blast noise environment to date. This was ultimately possible with the help of the knowledge and technology of ESTCP project WP-201117 and a U.S. Army Engineer Research and Development Center (ERDC) basic research project. The project also contributed advancements to the state-of-the-art in the statistical models used to assess community response to blast noise.

The project achieved all three of its objectives, which focused on making recommendations to improve the current blast noise assessment and management procedures. In summary:

  1. It is recommended that policy and decision makers move away from using a universal criteria and move towards assessing and managing blast noise on a community-by-community basis, which can be accomplished using the noise monitor technology recently demonstrated in ESTCP project WP-201117.
  2. In terms of community annoyance assessment and management, it is recommended that the LCdn be calculated over a much shorter time window (e.g., 4 weeks, LCdn, 4wk).
  3. In terms of complaint assessment and management, it is recommended that the LCdn calculated on a daily basis be used to manage complaints and that complaint risk should be managed on a community-by-community basis.
  4. In terms of the daily LCdn (LCdn, 24hr) levels, evidence was found that there is an increase in the percentage of the community that will be highly annoyed when the levels exceed 60 dB and an increase in the percentage of the community that will be not at all annoyed if levels are below 45 dB.
  5. In terms of single event levels, the project team recommends using the C-weighted sound exposure (CSEL or LCE) level. The project team suggests implementing a universal policy that no community should be exposed to an LCE of 118 dB, and that caution should be exercised if the single event levels approach or exceed 112 dB.