A prediction method for radon in groundwater using GIS and multivariate statistics

Radon (222Rn) in groundwater constitutes a source of natural radioactivity to indoor air. It is difficult to make predictions of radon levels in groundwater due to the heterogeneous distribution of uranium and radium, flow patterns and varying geochemical conditions. High radon concentrations in groundwater are not always associated with high uranium content in the bedrock, since groundwater with a high radon content has been found in regions with low to moderate uranium concentrations in the bedrock. This paper describes a methodology for predicting areas with high concentrations of 222Rn in groundwater on a general scale, within an area of approximately 185×145km2. The methodology is based on multivariate statistical analyses, including principal component analysis and regression analysis, and investigates the factors of geology, land use, topography and uranium (U) content in the bedrock. A statistical variable based method (the RV method) was used to estimate risk values related to different radon concentrations. The method was calibrated and tested on more than 4400 drilled wells in Stockholm County. The results showed that radon concentration was clearly correlated to bedrock type, well altitude and distance from fracture zones. The weighted index (risk value) estimated by the RV method provided a fair prediction of radon potential in groundwater on a general scale. Risk values obtained using the RV method were compared to radon measurements in 12 test areas (on a local scale, each of area 25×25km2) in Stockholm County and a high correlation (r=−0.87) was observed. The study showed that the occurrence and spread of radon in groundwater are guided by multiple factors, which can be used in a radon prediction method on a general scale. However, it does not provide any direct information on the geochemical and flow processes involved.