Comments collated and edited by K Jones (Kelly.jones@phe.gov.uk) and
Rick Tanner (Rick.Tanner@phe.gov.uk)
Public Health England welcomes the opportunity to comment on this ICRP document. As you are aware Public Health England contributes significantly to the work of ICRP and as the UK authority with responsibility for providing advice on radiological protection, is always keen to review ICRP publications which may affect the advice it provides.
The comments enclosed below were collated from contributions from several members of staff. If you need to discuss any of these comments in detail please contact Ms K Jones or Mr R Tanner (email given above).
Many thanks to the authors for all their efforts in compiling this report.
There appears to be too much information of a general nature that makes some assumptions on the types of accidents that may occur in the future or of limited relevance to the main purpose of the report. Examples include reference to external doses from cosmic radiation which are not covered in this report and mention of the importance of external doses after accidents if food restrictions cannot be applied. It is suggested the report maintains focus on the key purposes of the report rather than trying to cover areas outside of the intended scope.
The pressure to use age dependent phantoms is inevitable when considering exposures to the public. In this case there are competing effects for the different phantoms. There is organ size, as noted in the text, but also the organs are on average shallower in the smaller phantoms, which causes higher effective dose for smaller phantoms. Whilst it is understood that the data for age dependent wT may not be robust or available, the approach here opens up the question as to why age is considered for geometry but not for the potentially more significant change in radiosensitivity. This of course has two distinct parts; the change in radiosensitivity with age and the amount of life left. The latter would be easy to account for – in this case if the adult has on average 40 years of life left and the newborn 80 years, there is a factor of 2 which is bigger than the geometry effect.
The use of monoenergetic sources limits the ultimate usage of these data. Users will have to interpolate the data to get the values they want, which is easy for some sources and a lot of work for more complicated ones. It would have been better to pick the most significant nuclides and use those as sources throughout. The energy range would not have needed to be so wide. The dose per Bq for 10 keV is insignificant – over 3 orders of magnitude down on Cs-137 and the highest energies are not very relevant. 8 MeV is a problem because of neutron capture on nitrogen, but neutron sources are not included. It might be needed near nuclear plants, but the water immersion data go far higher in energy than is needed. The high energy data are not needed for any scenarios other than air immersion, but in this case data for neutrons would also be needed as well. It would be more useful if the report included data for the 50 most significant radionuclides for example.
In general, the use of the kerma approximation is necessary. However, an 8 MeV electron travels 4 cm in tissue, so the use of 8 MeV photons incident on a newborn is contentious – in the report the doses do not drop below those of an adult but they should because many electrons will escape from the phantom.
The sentence ‘The coefficients are given as dose rates normalised to radionuclide concentrations in environmental media, such as radioactivity concentration, in units of nSv h-1 Bq-1 m-2 or nSv h-1 Bq-1 m-3 and can be re-normalised …’ needs correction of the units to ‘nSv h-1 Bq-1 m2 or nSv h-1 Bq-1 m3’
Lines 97-98 - presumably if Auger electrons are included, then Coster-Kronig are also, though their contribution to dose is negligible.
Line 132 ‘Concentration of radioactivity per unit volume per of air or water’ should be changed to ‘Concentration of radioactivity per unit volume per of air or water’
Lines 134-135 Activity areal concentration is referred to in the glossary. This appears to be a spelling mistake and there is no reference to activity area concentration in the text so should this be removed?
Line 400 – suggest change to ‘‘The Publication 103 definition…’
Lines 404-407 – It is true that the organ equivalent doses depend on the size of the person, which will correlate with the size of the person; this is obvious because the ratio of the energy deposited to the mass of the organ will be dependent on the mass of the organ, except for the most penetrating fields. However, the size of the person will also affect how shielded the organs are, especially for less penetrating radiations. But, a more significant factor may be the dose deposited in the stem cells of that organ, which may not correlate so strongly with organ size so this should also be mentioned.
Line 429 – suggest change to - should be ‘members’
Lines 506, 920-924 and Figure 5.2 – It is not possible to simulate a “fully infinite planar source” in Monte Carlo. It should be stated that this has been approximated by using 5 mfp which is reasonable given that time spent sampling at an large distance from the scoring region would be prohibitive.
Line 508 – similarly it is not possible to simulate a “fully infinite radionuclide source in water” and what would be the point, as the water would completely attenuate the radiation.
Lines 516-518 – the data stored for each particle on the “coupling cylinder” should be noted – presumably x, y, z and two angles?
Lines 522-523 – air kerma and ambient dose equivalent are strictly point quantities, so point detectors could have been used, but it is understood that these have potential problems. Since air spheres have been used instead, the radius of the spheres should be given. If the radius is too large and the vertical profile incorrect, for weakly penetrating radiation, the results could be wrong. Also, what did you tally for the electrons, since kerma is only defined for uncharged particles. Did you still use an air sphere for the water immersion?
Lines 528-530 – clearly the coupling cylinder is not really usable for the water immersion. It is understood that the coupling cylinder means that all the phantoms can be calculated from one simulation of the geometry, but the least interesting of the geometries (water immersion) is the only one that is done using the best method.
Line 532 – change second “to” to “on”; “possible impacts on the public”
Line 562 – “and extension” should probably read “and by extension”
Lines 640-643 – this is a comment in general on the ICRP definition of organ dose. The mathematical definition of organ dose does not reflect the way that it is calculated. Either the total energy deposited is divided by the mass of the tissue or organ, or the same is done for each voxel and there is an summation over all organs or tissues that contribute to the organ.
Line 683 – ‘Sievert’ should be ‘sievert’
Lines 713-727 – it would be advisable to make this definition consistent with that used in the current revision of the operational quantities which uses collision kerma.
Fig 5.2 – This figure clarifies that the infinite plane is 5 mfp. This is probably sufficient, but it is not infinite as noted on the comments for Line 506 etc.
Line 1084 – “never interact with the air” should be replaced “did not interact with the air or soil”.
Fig 5.3 – the energy axis should extend to higher than 0.5 MeV as the residual peak is almost unnoticed.
Line 1108 – Suggest rename to ‘Submersion in contaminated air’
Lines 1112-1115 – Gaussian plume modelling is a simplistic method of modelling the dispersion in air and may not be appropriate especially close to the release point. It is suggested that there does not need to be reference to particular models but instead refer to situations close to the release point not being well-mixed so the semi-infinite approach is not appropriate. It is only when the radioactive material becomes well mixed at ground level that a semi-infinite approach can be used.
Lines 1124-1126 – the lack of energy dependence on the effect of increasing air density should be anticipated if the mean Z of the air does not change. If Bellamy et al 2018 did not account for the increased H in humid air their results may be slightly wrong.
Lines 1228-1236 – Do you mean the 30 cm diameter sphere represents the size of a human torso? This is the size of the ICRU sphere, which was calculated as giving a reasonable representation of the protection quantities, but was never intended to represent the size of a person. H*(10) is actually defined at a point so the 30 cm sphere used here is rather large.
Lines 1258-1261 – the IEC do specify performance requirements for instruments and dosemeters used to estimate H*(10).
Fig 6.13 – it is hard to understand why effective dose in the newborn is higher than skin dose x 0.01from 30 keV – 90 keV. Very little bremsstrahlung should be generated in that energy range. So what radiosenstitive organs are the electrons reaching other than skin? Is the sensitive depth of skin the same in the newborn and the adult?
Figure 6.14-6.18 – these data would be better presented as a ratio. It is almost impossible to see the differences or appraise whether they are significant or not. There are some quite large differences, especially for low energy photons?
Lines 1733-1743 Given the widespread use of FGR15 (and its predecessor FGR12) it is important to give these comparisons. It would be helpful to include text on these comparisons in the Conclusions sections.
Figs 6.19-6.21 – these data are very hard to compare in figure form. What is needed is the three quantities on one graph for one mfp.
Lines 2142-2158 - Suggest removal of the section on Application of dose rate coefficients for remediation planning – is this not outside the scope of the report?
Line 2205 to 2210 – Is this a conclusion of the report? Suggest removing this section and replacing with more information about comparison of dose coefficients with other work. The reference looks incorrect. Should it refer to IAEA 2000b rather than 2000a? The location factor looks very high if accounting for shielding from external exposure. Typical building shielding factors in IAEA 2000b range from 0.005 to 0.4. The reference is an IAEA TecDoc, is there a peer-reviewed IAEA reference that can be used instead, eg GSR or GSG document?
Line 2282-2285 - full reference to IAEA documents should included.
Line 2398 - NRA reference should be completed
Lines 2888 to 2889 - The units are incorrect as should be ‘nSv h-1 Bq-1 m2 or nSv h-1 Bq-1 m3’