Using Airborne Geophysical Data to Predict Radon Risk Areas in Ireland

Aerial measurements of outdoor radon levels in Northern Ireland were conducted to produce risk-exposure mapping at fine resolution for household and building safety.

J. A. Hodgson
S. Carey
Geological Survey of Ireland

Figure 1: Radon risk map of Ireland based on indoor measurements, produced by the Radiological Protection Institute of Ireland. Data are presented as an estimated percentage of the likelihood of a dwelling within a defined area exceeding a reference level of 200 Bq/m3 radon gas indoors.

Figure 1: Radon risk map of Ireland based on indoor measurements, produced by the Radiological Protection Institute of Ireland. Data are presented as an estimated percentage of the likelihood of a dwelling within a defined area exceeding a reference level of 200 Bq/m3  radon gas indoors.

Introduction

Radon is a radioactive element which occurs naturally from rocks and soils as a daughter product of Uranium 238 and Radium 226. Exposure to high concentrations is known to increase the risk of lung cancer. As a reference point, The European Union and Irish government agencies recommend action be taken with concentration greater than 200Bq/m3, while WHO recommends a national reference level of 100 Bq/m3) (UNSCEAR 2009; WHO, 2009). Accurate mapping of its occurrence can help identify households that are at potential risk of radon exposure and highlight areas where buildings require radon mitigation measures. Within Ireland, radon risk maps are based solely on indoor radon measurements carried out by the Radiological Protection Institute of Ireland (RPII). Although numerous measurements (more than 40,000) have been carried out nationwide, map resolution is low at 10 km2 (Figure 1). Indoor measurements are typically concentrated in areas of perceived radon risk or areas were radon awareness campaigns have been undertaken. Therefore, many areas have little or no data from which to map the natural radon distribution.

It has been shown that airborne geophysical data can be successfully used to help map radon (Appleton et al., 2008, 2011). A gamma-ray spectrometry instrument mounted on an aircraft measures the real-time natural radiation of potassium, thorium and uranium emitted from surface rocks and soil at one second intervals. This equates to a measurement approximately every 60 meters along each flight line. The combined mapping of uranium along with other geological data can be modeled to help provide an indicator of the likely emission of radon gas.

Tellus Border Survey

Figure 2: Twin Otter aircraft used for geophysical survey with electromagnetic coils located in pods at end of the wings and magnetometer housed at end of the nose cone. Gamma-ray spectrometry crystals are located with the body of the plane. Image Credit: Geological Survey of Ireland/Geological Survey of Northern Ireland.

Figure 2: Twin Otter aircraft used for geophysical survey with electromagnetic coils located in pods at end of the wings and magnetometer housed at end of the nose cone. Gamma-ray spectrometry crystals are located with the body of the plane. Image Credit: Geological Survey of Ireland/Geological Survey of Northern Ireland.

The Tellus Border project carried out a state-of-the-art airborne geophysical survey of the six northern counties of Ireland (Donegal, Leitrim, Sligo, Cavan, Monaghan and Louth) (Hodgson and Ture 2013) and provided an opportunity to help model radon risk within this region. The project was funded by the INTERREG IVA program of the European Regional Development Fund and was an extension of the successful Tellus project of Northern Ireland (Beamish and Young 2009). Airborne data were collected using a fixed-wing De Havilland DHC-6 Twin Otter aircraft (Figure 2). Survey lines were flown at an altitude of 60 meters with a line spacing of 200 meters. More than 900,000 radiometric measurements at approximately 60-meter intervals along flight lines were recorded. This equates to about 200 times more measurements than indoor radon readings for this region. Figure 3 shows the distribution of equivalent uranium across the border counties of Ireland. Uranium highs are typically associated with granitic and shale-rich rocks. Overlying water bodies and thick peat deposits result in low measured values. All survey data is freely available from the project website www.telluborder.eu.

The location and transport of radon gas

The presence and concentration of radon gas in homes is a function of its source, its transport to the surface, and the nature and construction of the building.  The source of radon is associated with the local geology and the concentration of uranium within rocks and soils. For example, highly radioactive granitic rocks present within the region produce strong uranium anomalies and are associated with “radon highs.” However, although radon is a uranium derivative, it is important to note that simply mapping its distribution does not necessarily give an accurate indication of the levels of radon gas at the ground surface – which is critical for decision-making.

Thus, also capturing the pathways that radon takes to the surface (which are identified by mapping the underlying geological structures) is necessary.   Knowing the substrate layers and groundwater flows allows for a more accurate understanding of radon distribution.  For example, understanding these features is important when considering that: 1)thin permeable soils aid the migration of the gas, whereas thick non-permeable or waterlogged soil impedes its movement; and 2) radon is soluble in water and therefore can travel great distances from its source within groundwater bodies. Within Ireland, significant groundwater is transported through karstified aquifers (rock formation characterized by cavities which have been eroded by dissolution). These formations are associated with elevated radon gas levels (Gammage et al., 1992; Cliff and Miles, 1997) and therefore their mapping can assist in predicting radon highs.

Figure 3: Equivalent uranium concentrations in parts per million (ppm) derived from Tellus Border airborne geophysical data, 2011-2013. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

Figure 3: Equivalent uranium concentrations in parts per million (ppm) derived from Tellus Border airborne geophysical data, 2011-2013. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

Although information is not available about house types and their construction (i.e., presence of radon barriers), by mapping both the source and pathway of radon from geological and geophysical data we can better understand the natural distribution of radon gas. These improved high-resolution maps can then be used to highlight the potential risk homeowners face and assist in the planning and construction of new houses and the remediation of existing dwellings if required.

Methodology

To model not only the source but also the transport of radon to the surface, a methodology for modelling radon risk was developed which took into account airborne uranium data, the degree of karstification of the bedrock and thickness and permeability of the subsoil. Groundwater recharge data already available from the Geological Survey of Ireland was used as a proxy dataset for subsoil permeability and thickness seeing as it models these properties in relation to aquifer recharge. Numerous other geological and geochemical datasets were tested but were determined to be less significant based on calculated p-values (the probability of obtaining a test statistic at least as extreme as the one that was actually observed).

Figure 4: Model of radon potential risk derived from airborne uranium, groundwater recharge coefficient and karst values. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

Figure 4: Model of radon potential risk derived from airborne uranium, groundwater recharge coefficient and karst values. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

This methodology involved assigning all data directly to 1-kilometer grid squares in ArcGIS 10.1. These data could then be modelled against indoor radon readings falling within the same grid square using multivariate linear regression. The derived model equation was then applied to data in all grid squares.

Indoor radon data falling within each grid square was also summed and averaged. However, as the radon data is log normally distributed, it was necessary to transform the data so it could be correlated with the other data sets. Most radon maps are displayed as an estimated percentage of the likelihood of a dwelling within a defined area exceeding a reference level of 200 Bq/m3.  Therefore, a radon percentage reference level was calculated from the derived geometric mean and standard deviation based on the methodology employed by Fennell et al., (2002) for the Irish National Radon Survey.

Model results

The goodness of fit for each model was determined by a derived R2 value (the coefficient of determination, R2 indicates how well data points fit a line or curve) using Minitab software. R2 values improved when data was filtered to use only grid squares with a minimum number of measurements. A maximum R2 value of 0.66 was achieved for the models.

Figure 5: Number of indoor radon measurements by 1-kilometer grid used in models. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

Figure 5: Number of indoor radon measurements by 1-kilometer grid used in models. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

Figure 4 shows the results of the model applied to all grid squares in the region. Significantly more detail is resolved when compared to the existing radon map (Figure 1). Potential radon highs are seen in a number of zones coincident with the existing map, but new radon potential highs where 20 percent of values exceed the reference level of 200 Bq/m3 have also been modeled.

Model evaluation and future research

One of the limitations of the model is the limited number of grid squares with a high number of radon readings which were used to construct the model.  Also many of these grid squares with the most numerous readings are clustered together in urban areas. This may result in bias of the models toward the geological terranes occurring in these areas, because of the geological factoring. Other uncertainties relate to the accuracy of the indoor radon measurement as well as possibility that the karst parameter may result in over-predicted values in areas of thicker subsoil deposits. 

Figure 6: Radon data for new points shown in Bq/m3 used for model evaluation. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

Figure 6: Radon data for new points shown in Bq/m3 used for model evaluation. Image Credit: Geological Survey of Ireland/Ordnance Survey Ireland Licence No. EN 0047214.

To help evaluate the model, a subset of indoor radon measurements, not used in the initial modeling, (Figure 6) was gridded. This new data was limited in its distribution and quantity but did show similar trends as those predicted from the model (Figure 4), which gives some confidence to the model.

The model is being adapted to incorporate new data including additional information on the thickness of soil deposits. Further work is ongoing to fully evaluate the model. However, the initial model, using airborne uranium data along with geological information, has helped improve the resolution of radon risk potential across the area. This work is being further developed with the RPII and will contribute to national radon mapping strategies that will help inform people of their risk of exposure to radon gas.

 

 References

Appleton, J.D., Doyle, E., Fenton, D. and Organo, C. 2011. Radon potential mapping of the Tralee-Castleisland and Cavan areas (Ireland) based on airborne gamma-ray spectrometry and geology. Journal of Radiological Protection, 31, 211-235.

Appleton, J.D., Miles, J.C.H., Green, B.M.R and Larmour, R. 2008. Pilot study of the application of Tellus airborne radiometric and soil geochemistry data for radon mapping. Journal of Environmental Radioactivity, 99, 1687-1697.

Appleton, J.D. and Ball, T.K. 2001. Geological radon potential mapping. In: Bobrowsky, P.T (Ed), Geoenvironmental mapping: Methods, Theory and Practice. Balkema, Rotterdam, 577-613.

Beamish, D. and Young, M.E. 2009. The geophysics of Northern Ireland: the Tellus effect. First Break, 27, 43-49.

Cliff, K. D. and Miles, J.C.H. (Eds.) (1997). Radon Research in the European Un-ion, EUR 17628, National Radiological Protection Board, Chilton, UK.

Fennell, S.G., Mackin, G.M., Madden, J.S., McGarry, A.T., Duffy, J.T., O’Colmain, M., Colgan, P.A. and Pollard, D. 2002. Radon in dwellings. The Irish national radon survey. Radiological Protection Institute of Ireland.

Gammage, R.B., Dudney, C.S., Wilson, D.L., Saultz, R.J. and Bauer, B.C. Subterranean transport of radon and elevated indoor radon in hilly karst terrains. Atmospheric Environment Vol. 26A, No.12:2237-2246; 1992.

Hodgson, J.A. and Ture, M.D., 2013. Tellus Border Airborne Geophysics Survey Logistics Report Version 3. Geological Survey of Ireland / Geological Survey of Northern Ireland joint report.

UNSCEAR, 2009. United Nations Scientific Committee on the Effects of Atomic Radiation
(UNSCEAR). UNSCEAR 2006 Report. Annex E. Sources-to-Effects Assessment for Radon in Homes and Workplaces. United Nations, New York.

WHO, 2009. WHO Handbook on Indoor Radon: A Public Health Perspective. World Health Organisation (WHO), Geneva.

 

Topic: