Radon programme
The approach for new buildings
The approach for exist. buildings
Czech experience

Czech experience with the evaluation of radon potential of foundation soils
- Building site characterization (concise information)

Decision making in the pre-construction phase
The decision whether or not a dwelling needs radon protection can be based not only on radon risk mapping or modelling, but also directly on the assessment of radon potential at a given place - on a building site characterization.
The soil characteristics are measured in-situ and protective measures are designed with respect to the measured properties of the soil and to the dwelling design (Czech Republic, since 1991).
Main advantage of the method:
It is a site-specific, individual approach that enables to propose an optimal preventive strategy corresponding to local conditions.
Description of the method:
Radon index
of the
building site
Radon concentration in the soil gas
 gas permeability of soils

Soil-gas radon concentration:
  • samples collected at a depth of 0.8 m
  • at least 15 measurements (if the building area is larger than 800 m2, measurements are made at a grid 10 x 10 m)
  • the third quartile of the data set is the decisive value, values smaller than 1kBq.m-3 are excluded from the evaluation
  • Note: In case of larger sites (over 800 m2) the distribution of soil-gas radon concentration must be taken into consideration to decide if the radon index can be based on the third quartile of the whole data set or if the area should be divided into several smaller subareas characterized by a different radon index.
Soil permeability:
  • several methods available
  • current practice: expert geological description of the vertical soil profile is combined with grain size analysis of soil samples or with direct in-situ measurements of soil-permeability
  • quality of permeability classification strongly depends on personal experience

Results of the building site characterization are used for the design of protective measures
This approach is obligatory in accordance with the Czech Atomic Law
If the radon index is other than low (i.e. medium, or high), the building must be protected against radon.
Protective measures are designed and installed in accordance with the Czech National Standard (CSN 730601 Protection of houses against radon from the soil).
Modelling can be very useful for a verification and corrections of the whole system, but it is not used for designing protective measures
Note: The experience in the field of building site characterization is very large in the Czech Republic. There are more than one hundred private firms dealing with this job. RADON v.o.s. is one of them. Since 1990 we have measured building areas for more than 17000 buildings and we have realized more than 310 000 soil-gas radon concentration measurements.
Comparison of large-scale radon risk maps and results of detailed in-situ measurements
The method described above is not assigned for radon risk mapping, but for a building site characterization. In the Czech Republic, radon risk maps are used mainly to direct the search of existing (old) houses with elevated radon levels, not to design preventive protective measures in new buildings. On the other hand, results of building site characterization are commonly used for the construction of radon risk maps.
Two comparisons were made in 1992 and 1995. They concerned 630 and 968 building sites, respectively.
Results of detailed in-situ measurements corresponded to the expected radon potential at a large number of areas.
the bedrock formed by Cretaceous sediments or by granites or granodiorites, and/or the cover formed by loess and loess loams a satisfactory coincidence between the results of building site characterization and the radon risk map prediction (about 70% of cases)
- large-scale radon risk maps are generally reliable
Significant differences were found in areas with larger variability of soil environment (river terraces, or layers influenced by anthropogenic activity).
Quarternary cover formed by river terraces; or some special rock types - chlorite-sericite phylite detailed radon surveys did not confirmed the expected risk at more than 60% of building sites
- observed differences confirmed the usefulness of a direct building site characterization
Soil-gas radon concentration - Variation with depth
The research dealing with this topic was performed in 1993, 1994
Eight reference areas with various geological settings were chosen with respect to the characteristics of the underlying bedrock, origin, permeability and stratification of soil deposits, and depth and variability of groundwater level.
Soil-gas radon concentration as well as soil permeability were determined at depths of 0.3, 0.6, 0.9, 1.2 and 1.5 m.
Fifteen measurements were made at each depth and at each reference area - average values were evaluated.
Variability of radon concentration with depth was different at areas with different geological characteristics.
Almost constant soil-gas radon concentration was observed at areas with relatively homogeneous and highly permeable Quarternary fluvial sands covered by clayey loams, or in case of highly permeable eluvial deposits of Cretaceous sediments - sands covered by sandy loams.
Insignificant increase of radon concentration with depth was measured at areas with relatively homogeneous and medium permeable loess loam.
Significant increase of radon concentration with depth was observed at areas with granodiorites characterized by extensive weathering and by high radon potential.
The similar results were measured during the research at four areas in 1999
- variability of radon concentration with depth depends on geology
Soil-gas radon concentration - Temporal variations
As the protection of new buildings against radon penetration from the ground in the Czech Republic is based on in-situ soil-gas radon concentration measurements, possible influence of temporal variations of the measured parameter on the evaluation results is very important.
Several long-term research projects dealing with this problem have been carried out.
1990 - 1992: a 14-month observation of soil-gas radon concentration using two permanently installed probes at two test sites (distanced several meters) with similar geological structure - sandstone weathering residuum covered by fine-grained sand with loamy admixtures Results and conclusions:
Site 1:
from 34.1 kBq.m-3 to 91.8 kBq.m-3 (number of measurements: 270, arith. mean = 60.2 kBq.m-3, SD = 11.3 kBq.m-3, SD/mean = 0.19)
Site 2:
from 3.0 kBq.m-3 to 6.3 kBq.m-3 (number of measurements: 270, arith. mean = 4.3 kBq.m-3, SD = 0.6 kBq.m-3, SD/mean = 0.15)
The general temporal pattern was represented by higher winter/spring and lower summer/autumn values, but seasonal variations were not so large. Short-term changes of radon concentrations at site 1 and at site 2 were often different, or even opposite.
1992 - 1993: repeated soil-gas sampling at five reference areas with different geological structure and soil permeability two different soil-gas sampling methods were tested:
(a) "probes" = standard sampling method using small-diameter probes (the probes were pounded into the ground anew every measuring day, i.e. the location of measuring points was not exactly constant.)
(b) "chambers" = sampling from cylindrical chambers permanently installed in the soil.
Measurements were repeated nine times at 25 measuring point at each reference area, using both methods. Average values of soil-gas radon were compared.
Results and conclusions:
- weak relationship between average soil-gas radon concentration and soil temperature, and between radon concentration and soil moisture
- it seemed to be different for different sampling methods
- different correlations coefficients were obtained at areas with different geological structure
Observed temporal variations of soil gas radon concentrations often correspond not only to real variations of the measured physical parameter, but also to fluctuations and errors connected with sampling and measuring techniques.
The influence of changing meteorological conditions may not be as important as the influence of changing sampling depth, the influence of geological heterogenities, and the influence of sampling and measuring errors.
The highest values of average soil-gas radon concentration obtained at any area during a 1-y period were about two times the lowest ones (more exactly from 1.5 to 2.5). The ratio of the limits that separate different categories ranges from 3 to 3.5.
The probability that any area will be classified as a low risk area and after several months as a high risk are is therefore very low.
The reproducibility of the radon risk classification seems to be satisfactory.
1994 - 1995: measurements at the test site on the surface of the uranium mill tailings
confirmation of previous results
temporal variability (SD/mean) decreased with the depth below the surface:
SD/mean < 0.10 at the depths of 1.5 and 0.8 m
SD/mean ranging from 0.13 to 0.25 at the depths from 0.5 to 0.3 m
2000 - 2001: repeated measurements of soil-gas radon concentrations at several test sites
A satisfactory reproducibility of the evaluation based on in-situ soil-gas radon concentration measurements was confirmed again.
The sampling depth of 0.8 m represents a relatively good compromise between the required reduction of errors caused by the radon concentration variability and the required reduction of weather effects on the one side and the practicability of the method under field conditions on the other side.
It is evident that all conclusions are valid for the geological and meteorological conditions that are typical for the Czech Republic, or maybe for Central Europe. The situation in other countries with a substantially different climate or geology may be different.
Soil-gas radon concentration - Spatial variations & Statistical evaluation of data
In our experience, spatial variations of soil-gas radon concentrations are more important than temporal ones. The soil-gas radon concentrations may vary, often greatly, over a small distance. The occurrence of heterogenities may indicate the presence of faults or tectonic zones.
As the areas with a homogeneous geological structure and with a homogeneous soil-gas radon concentration are rare, any evaluation based on a single measurement is almost worthless.
On the other hand, the occurrence of spatial variations represents a serious reason, why protection of a building should be based on the results of a building site characterization and not on large scale radon risk maps.
The frequency distribution of measured soil-gas radon concentrations need not be normal.
1993: almost 90 randomly chosen data sets were analysed.
The distribution of a significant part of data sets was heterogeneous, i.e. neither a normal nor a log-normal models were found to give an adequate fit to the radon concentration data.
Robust nonparametric estimates such as the median or the third quartile were therefore suggested for the description of the soil-gas radon concentration data.
Soil-gas radon concentration - Variations introduced using different methods
A perfect sealing of all parts of the equipment is required especially when soil-gas samples are taken from a low permeable soil. A "contamination" of the soil-gas sample by the atmospheric air is a frequent cause of variations.
Soil structures should not be disrupted during sampling
2001: short-term intercomparison measurement using different measuring techniques
Observed temporal changes were different in different measuring points, or, more probably, these changes were caused mainly by fluctuations connected with measuring methods themselves and they did not reflect real variations of the measured parameter.
More problems are connected with the determination of the second main parameter: SOIL PERMEABILITY
Current practice: expert geological description of the vertical soil profile is combined with grain size analysis of soil samples or with direct in-situ measurements of soil-permeability
Quality of permeability classification strongly depends on personal experience
If the evaluation of soil permeability is based on direct in-situ measurements using the simple probe technique, larger number of measurement should be required to get knowledge on spatial variations at a given building area (probably the same number as the number of measuring points for soil-gas radon concentration measurement).
In-situ permeability measurement: no standardization available!
Radon exhalation from the ground surface
2000 - 2001: a long-term follow-up at four reference areas
spatial variations similar as spatial variations of soil-gas radon concentration
measured values of radon exhalation rate are strongly affected by conditions on the soil surface (significantly lower values are observed when the soil surface is frozen or covered by water; a decrease of radon exhalation can be observed after a compression of the upper soil layer caused by heavy machines during the harvest, etc.)
The temporal variability of radon exhalation rate from the ground surface is significantly higher than the temporal variability of soil-gas radon concentration if soil-gas samples are collected from a depth of 0.8 m below the surface.
Research project dealing with radon risk classification of foundation soils
Results of detailed radon surveys as well as results obtained from research studies, carried out in the Czech Republic in the period 1994 - 2000, indicated that the uniform method should be improved.
In 2000, RADON v.o.s. corp. in cooperation with other authors of the original uniform method, Milan Matolín (Charles University Prague, Faculty of Science) and Ivan Barnet (Czech Geological Survey), prepared a project dealing with this topic. Concise information about this project you can find here.
The research was finished in the end of 2002. The definitive version of the uniform method for radon risk classification of foundation soils was accepted in 2004. (PDF)

RADON v.o.s., Novákových 6, 180 00 Praha 8    e-mail :