Spatial association between residential radon conc

Authors: Xun Shi David J Hoftiezer Eric J Duell Tracy L Onega

Publish Date: 2006/05/05

Volume: 51, Issue: 1, Pages: 65-71

PDF Link

Abstract

Using a large database of residential shortterm radon measurements in New Hampshire this study evaluated the ability of expertassigned bedrock radon potential for predicting residential radon concentration First each bedrock type was assigned a radon potential level by a geologist familiar with the local geology Then using residential radon measurements a continuous surface of radon concentration was generated through a kriging process The mean residential radon concentration within the spatial extent of each bedrock type was then calculated based on that surface The Spearman Rank Correlation Coefficient was calculated between the two ranks of the bedrock types one based on the expertassigned potential level and the other based on the mean residential concentration A strong correlation between the rank correlation and the area of the bedrock type was found When only the 15 largest bedrock types were used the Spearman Correlation Coefficient reached 06 Geological knowledge is concluded to be useful in predicting and mapping residential radon concentration but the prediction should be interpreted with caution especially for areas in which the underlying bedrocks are highly localizedThis work was supported by the National Institutes of Health Grant P20 RO18787 The authors are grateful to David Chase at NH Department of Environmental Services who made substantial contribution to this paper but kindly refused the authorship He provided the data of both the geologist’s classification and the indoor measurements and also helped to formalize the idea of comparing the two Geostatistical Analyst ArcGIS and Spatial Analyst are registered trademarks of ESRI Redlands CA USA

Keywords: