|
Indoor radon measurements in the Walloon region in Belgium have been used to establish a radon risk map. The data were collected by the "Agence fédérale de Contrle nucléaire, AFCN-FANC" on the one hand and by the "Institut Supérieur Industriel de Bruxelles (ISIB)" on the other hand. The ISIB data, about 5000, are short-term measurements collected in houses with charcoal canisters. The FANC data, about 7500, are long-term indoor radon measurements collected using track-etch detectors. The composite database has been created only after the application of the t-test, that showed that the geometrical means of the two groups are not statistically different from each other.
As variance analysis has demonstrated, the biggest percentage of the variation in indoor radon concentration can be explained by local geology, in comparison to other factors relative to the house characteristics such as house type, double-glazing, floor level, date of construction. For this reason we decided to consider the data having the some geological context separately. The land surface is divided in cells of one square kilometre. The local geological information is associated to the centre of each cell using digital geological map data. We associate to each cell the corresponding value of the geometrical mean of radon concentration calculated by considering all data having the same geology.
Our database can be approximated with a log-normal distribution. Using the smoothed geometrical mean and the geometrical standard deviation it is possible to calculate the proportion of the distribution above the Action Level in each cell. The logarithmic standard deviation (LSD) is taken constant for each geological group. It is obtained as a weighted average of the LSD of FANC data and the LSD of ISIB data corrected for the higher variability of short-term measurements. In this way the affected areas are easily identifiable.
|
