Radon Surveys in Israel – Problems and Solutions

In many countries, including in Israel, mass statistical surveys are conducted to assess the impact of indoor radon on the population. Since the radon concentration in buildings has a lognormal distribution, the main parameters of the radon potential hazard are: (a) arithmetic mean (AM) of radon concentration in buildings - to estimate the average radon dose, and (b) geometric standard deviation (GSD), that characterizes dispersion of radon spatial distribution and allows estimating the part of buildings with high radon (above the reference level). An important condition for a reliable assessment of the AM and GSD is a representative sampling of buildings. The main requirements for representativity are: (i) the random sampling principle, and (ii) the distribution of measurement points in proportion to the population density, otherwise the estimates of AM and GSD could be biased (unreliable).

In Israel, the largest numbers of measurements were conducted in radon surveys in 1998, 2007, 2008 and 2011 (more than 1000 per year). However, the AM and GSD values obtained in different years sometimes differ significantly within the same territory. This fact is primarily due to the absence of a representative sample. Thus, the issues arise that are relevant not only for Israel:

• How to ensure the representative sampling of buildings?
• Can short-term measurements be used to evaluate the AM and GSD, and if yes - how much will their accuracy differ from the accuracy of the long-term measurements?
• How many measurement points are needed to reliably evaluate the AM and GSD?

Involving schoolchildren in the radon surveys seems to be easy and effective instrument, since the dwellings to be covered correspond to both conditions (i) and (ii), regardless of the territory size. In addition, older schoolchildren are quite capable to perform a simple procedure for recording conditions and passive exposure, using simple sensors, such as charcoal canister or track detector. However, the organizational and logistical work, as well as laboratory measurements, have to be carried out by professionals. Moreover, the participation of schoolchildren in the habitat security control can be regarded as a national educational project providing important information to the public about the “radon problem”.

The Monte-Carlo calculations lead to the new interesting findings, which show that in order to obtain reliable estimates of the AM and GSD by mass survey of a representative sample of buildings, also short-term and even spot measurements could be successfully used instead of long-term measurements, which are usually considered as the most reliable. The AM uncertainty does not exceed 10% or even 5%, if the number of measurements is at least 1000 or 5000, respectively, regardless of the measurement duration and GSD value (taking into account possible seasonal variations of AM and GSD). A disadvantage of spot and short-term measurements may be the overestimation of GSD, but this problem can be easily solved by modeling to restore the true value of GSD, taking into account the contribution of radon temporal variations to the dispersion of radon spatial distribution. The dependence of radon temporal dispersion GSD(t) on the mode and duration of measurements based on the results of annual radon monitoring in experimental rooms is determined for the first time. The values of GSD(t) are usually lower than the dispersion of radon spatial distribution, especially with increase of the measurement duration.

Thanks to the EU funding, a mass radon survey with the involvement of older schoolchildren in Haifa is planned in 2018-2020 using the charcoal method. In addition, in 10-12 buildings with high radon, a whole-year monitoring will be carried out to verify the tabulated values of radon temporal variations, as well as the values of GSD(t).