The Correlation of Radon Concentration with Various Building Attributes at U.S. Air Force Bases

Abstract

A statistical analysis was conducted on radon data from the United States Air Force's Radon Assessment and Mitigation Program (RAMP). The data came from l-y alpha track detectors which were deployed at 15 U.S. Air Force installations worldwide. Sample sizes at the different installations ranged from 373 to 5801. Radon concentration was modeled at each installation utilizing multi-factor analysis of variance (ANOVA) with the following building attributes as independent predictor variables: type of structure, age, type of foundation, number of stories, type of air conditioning, type of fuel, type of heating, type of water, floor where sampler was placed, the presence of a sump pump on the lowest level, and the presence of a drain on the lowest level. In addition, a trend analysis was conducted among class levels of each individual attribute for each installation. The attributes age, type of structure and their interaction were the most strongly correlated to radon concentration, generally accounting for about one-fourth to one-half the variation of radon concentrations in the models. Other attributes which exhibited a weaker correlation with radon concentration include: type of foundation, type of fuel, the number of stories, and the floor where the sampler was placed. In general there was no correlation between radon concentration and the attributes type of water and presence of a drain or sump pump at the lowest level. The coefficients of determination, R2 ranqed from 0.191 to 0.627 which is rather poor for predictive uses and indicates other factors, such as the underlying geology, may be more important then the attributes examined in this study. The trend analyses indicated that the following attributes tend to yieli the highest radon concentrations: single family homes, single story structures, and structures built during the 40s, 50s and 60s.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1992
Accession Number
ADA258548

Entities

People

  • Scott M. Nichelson

Organizations

  • Purdue University

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Age Groups
  • Air Conditioning
  • Air Force
  • Air Force Facilities
  • Analysis Of Variance
  • Computer Programs
  • Construction
  • Construction Materials
  • Databases
  • Economic Forecasting
  • Environmental Protection
  • Information Science
  • Natural Gas
  • Regression Analysis
  • Statistical Analysis
  • Surveys
  • United States

Readers

  • Electrical Engineering
  • Groundwater Contamination Remediation.
  • Regression Analysis.