21 December 2012
Christopher Clement, CHP
International Commission on Radiological Protection (ICRP)
280 Slater Street
Ottawa K1P 5S9
Subject: SENES’ Comments on the ICRP Draft Report: Annals of the ICRP, Occupational Intakes of Radionuclides (OIR) Part 3 (Chapter 12: Radon)
We are pleased to submit SENES’ comments on the subject ICRP draft document.
Overall, we recognize the huge amount of detailed and challenging work that has gone in to the preparation of the OIR draft report. However, notwithstanding this large effort, there are some concerns about Chapter 12 of the OIR draft report which addresses radon and radon progeny. In particular, at present, it seems premature to apply the dosimetric model and its output (dose coefficients) to specific work environments. This is particularly valid for uranium mines and especially for regulatory purposes. A few of the reasons that support this position include the following:
Dosimetric model: The main issue is more related to the dosimetric model itself and its application as opposed to a given numerical dose coefficient value (dosimetric model output) per se.
A key issue of concern is that existing knowledge on the input required for the dosimetric model is outdated for mines and is not representative of modern mines. [This is acknowledged in the OIR draft report: e.g. page 200: lines 9163-64; and page 207, lines 9530-33.] Moreover, it can be expected to take several years to gather updated information on the input parameters that are representative of modern mines. This time period could serve to further improve the underlying dosimetric model.
Particle size is a key issue and there is a lack of readily available instruments for making measurements in the nanometer size range, the size of most interest to dosimetry of radon progeny. Designing a measurement program and testing such a program is quite challenging and certainly would require a few years to collect meaningful data.
It is also important to understand that the mine environment is quite variable from location to location and changes with mining practices and ventilation. It is important to understand that mine environments can change rapidly over short distances and hence, there are real concerns with differences in aerosols that are measured versus those that miners actually breathe. The issue of interpretation will require some careful consideration.
Another key issue of concern is that the dosimetric model output does not seem consistent with the results from epidemiological studies, or perhaps consistent by chance in some cases. In particular, current dosimetric models are not capable of fully correcting for the effect of smoking which is by far the major cause of lung cancer. This is especially important as the current epidemiological radon risk models are relative risk models and thus very sensitive to smoking patterns in the exposed populations to radon. Moreover, smoking prevalence has been declining worldwide and the implications of this should be clearly acknowledged.
Obviously, the issue of insufficient confidence in the dosimetric model input parameters and consequent dose coefficient output for uranium mines is an important consideration as both industry and regulators are being challenged these days and it is important to have the key elements of radiological protection, including the dose coefficient and/or dose conversion convention (DCC), as clear and defensible as possible. (In other words, a numerical dose coefficient value that is not well substantiated would be challenged.)
We understand that the uranium mining industry supports the concept of using a default value of dose coefficient to generally represent a generic mining environment, in as much as the results of the dosimetric model agree with the results from epidemiological studies. This should imply that the dose coefficients adequately reflect (a nominal) smoking prevalence in populations.
Numerical values of dose coefficient or of dose conversion convention: Regarding a dose coefficient value of 11 mSv/WLM, we note that it is about two times higher than the dose conversion convention (DCC) current value of 5 mSv/WLM. At first glance, this appears consistent with recent epidemiological results which show a doubling of the risk (in terms of health detriment) per unit of exposure to radon progeny. However, it is important not to confuse this risk with the DCC as the later accounts for the “nominal” characteristics of the reference populations that underpin this risk, and notably from smoking prevalence which is the dominant factor that influences the risk of lung cancer.
A careful review of the most recent data, including trends in smoking supports a DCC value of 6 to 7 mSv/WLM. Overall, a DCC of 10 mSv/WLM is likely to be increasingly conservative in the future and out of line with the anticipated future overall population risk. Ultimately, the DCC value (or does coefficient value) to be recommended by ICRP needs to be well supported and to adequately account for a reasonable “nominal” level of current smoking prevalence and trend.
The above position is elaborated in the following recent paper (October 2012) which has been submitted to the ICRP: International Debate on the Changes to Radon Risk: Key Issues for Implementers, (Saint-Pierre, Chambers, Takala, Harris, Huffman and Mason). We invite ICRP to consult this more detailed contribution.
In complement, some more detailed comments are provided in the Annex.
While we acknowledge that much work has gone in to the development of the OIR Chapter 12 on radon and radon progeny, we respectfully suggest that the move to a fully dosimetric approach for radon and radon progeny is premature, especially for modern uranium mine environments. We would be pleased to elaborate on our comments should the Commission be interested in it.
We appreciate this opportunity to provide our views to ICRP on this important matter.
Yours very truly,
SENES Consultants Limited
Sylvain Saint-Pierre, P. Eng.
Vice-President, Marketing, Energy and Nuclear
(WNA, Consultant for the WNA Working Group on Uranium Mining Standardization)
c.c. ICRP: J. Cooper, A. Gonzalez, J-F. Lecompte, M. Trimarche, J. Harrison, J. Marsh
WNA UMSWG members (radon project)
c.c. D. Chambers (SENES), J. Takala (Cameco), F. Harris (Rio Tinto), D. Huffman (Areva) and C. Mason (Bhp billiton).
International Debate on the Changes to Radon Risk: Key Issues for Implementers, (Saint-Pierre, Chambers, Takala, Harris, Huffman and Mason).
SENES Comment on the ICRP Draft Report:
Annals of the ICRP, Occupational Intakes of Radionuclides (OIR
Part 3 (Chapter 12: Radon)
Some More Detailed Comments
To illustrate this key issue, it is noted that the dose coefficient numerical value of 11 mSv/WLM is derived from the following input parameter values: 0.01 (or 1%) for the unattached fraction of radon progeny aerosols, and an equilibrium ratio value of 0.2 between the concentrations of radon progeny and radon. The value of 0.01 seems derived from tests carried out in the late 1990’s in wet underground uranium mines in northern Saskatchewan (Canada) [page 206: lines 9464-92.] whereas the value of 0.2 seems derived from Olympic Dam underground uranium mine in South Australia (Australia). We also note that the dosimetric model seems quite sensitive to the value of unattached fraction of radon progeny aerosols. We therefore wonder if the above values are representative of any specific modern mine and if it is relevant to use such values as default values. As the OIR draft report indicates values of unattached fractions of 3 to 4% for the Olympic Dam underground mine, if this is correct, it would therefore be difficult to justify a default value of 1%. Such issues points at a significant level of uncertainty in the dose coefficients that would be inappropriate for a comprehensive regulatory system of radiological protection for underground uranium miners.
The value of 0.2 (or 20%) for the Olympic Dam underground mine appears a little higher than usual for active workplaces in modern underground uranium mines. We wonder if it corresponds to an average for the mine active workplace areas or for the mine as a whole which combines both active and inactive workplace areas. Typically, an equilibrium ratio of a few percentages would be expected for the mine active workplace areas. The implications of this should be closely examined in terms of relevant default values for the dosimetric model input parameters and output.
As mentioned earlier, the dosimetric model seems quite sensitive to the value of unattached fraction of radon progeny aerosols which consists of fine particles (e.g. nanometre diameter). We therefore wonder if this is compatible with the observed significant reduction in lung cancer risk that resulted from ventilation and radon control improvements implemented in underground mines since at least the 1970’s. It would be expected that the levels of nanometre aerosol particles vary widely in any active underground mine environment. For example, it would be the case for a basic mining sequence that includes: drilling, loading explosive, blasting, scaling and bolting, scooping and mucking, and hauling ore and waste rock. All factors considered, would the significant reduction in lung cancer for underground uranium miners be fortuitous or not?
Page 220: lines 9163-64:…”Also, the above equation may underestimate fp in situation where the radon progeny is far from equilibrium as is the case in some modern mines.” Is the equation relevant or not for modern underground uranium mines?
Page 206: Lines 9464-92: “The unattached fraction (fp) was about 1%”. But other results in this paragraph seem to go up to 6%. What should be relevant values for modern underground uranium mines is therefore unsettled.
Viewed together the following two statements from the OIR draft report suggest that the proposed default values for modern underground uranium mines are premature at this stage, and that they can be a source of significant instability for the regulatory system of radiological protection for uranium miners: