Recently, UNSCEAR published its latest report, UNSCEAR 2016.
Several colleagues have pointed out that Annex C of the UNSCEAR report discusses the radionuclides of tritium and carbon-14. They have requested my views on the UNSCEAR report.
In particular, paragraph 304 on page 320 of the full report (ie plus annexes) dismisses the view that the childhood leukemias observed near nuclear power stations (NPPs) worldwide are due to releases of these nuclides, as discussed in my 2014 article published in the peer-reviewed Journal of Environmental Radioactivity.
Mainly between 1954 and 1962, 550 atmospheric bomb tests were carried out which released large amounts of, inter alia, tritium and carbon-14 into the atmosphere. Para 304 of the UNSCEAR report argues that because there was no consequent evidence of raised childhood leukemias therefore nuclear power plant (NPP) emissions of tritium and C-14 could not have caused the increased child leukemias that have been observed world-wide near NPPs.
However there were in fact observed increases in childhood leukemias arising at the time of the atmospheric test bombs. The best collection and analysis of the not inconsiderable evidence is contained in the UK Government’s CERRIE Report (2004) on internal emitters. Pages 70 to 72 of the Report (paras 33 to 42) examine the evidence of increased leukemias following exposures to test bomb fallout in the Nordic Countries, in the US, and in the UK. And while some puzzling aspects remain (eg increased effects in areas of dry fallout rather than wet fallout), the Committee concluded that
“Overall, the studies of childhood leukaemia and fallout from atmospheric nuclear weapons testing suggest an increased risk due to the exposure…” (para 42)
Secondly, the comparison between NPP emissions and test bomb explosions is inappropriate. As well as H-3 and C-14, the atmospheric test bombs emitted many other hazardous nuclides, including Cs-137, Cs-134, Sr-89, and Sr-90 and others. All of these could have had effects on bone marrow as well as tritium and C-14. As far as I’m aware, no scientific evidence exists that the observed health effects from test bomb fallout may be due to H-3 and C-14 alone, so it is strange for UNSCEAR to appear to be making this claim.
In addition, the two scenarios are quite different in their volumes. NPP emissions are restricted to regional, relatively low height plumes, whereas bomb test emissions were ejected high into the troposphere (ie up to 15 km) and on occasions into the stratosphere (>15 km), then circulated around the world initially in the northern hemisphere. The important matter here is the resulting tritium concentrations in air from both scenarios. However there are few data on tritium in air concentrations from these scenarios. In situations such as this, ie almost no data, scientists often try to make order-of magnitude estimates to test a hypothesis (here, that tritium air concentrations from the two scenarios are comparable).
With several assumptions and caveats, it is possible to do this. See appendix below. These estimates indicate that tritium air concentrations throughout the troposphere from the bomb tests are about the same as the air concentrations near NPPs after refuelling, although there is considerable variability in both.
As for estimated doses, the picture is more complex. Hamby (1993) estimated annual exposures to airborne tritiated water at the US Savannah River Plant to be 0.63 µSv (10-6 Sv). However this military plant was known to have emitted large amounts of tritium in the 1980s and 1990s – larger than those from most NPPs. On the other hand, it is not known when or where the tritium concentrations were measured at the plant, This is important as the author states that the doses were directly estimated from his measured tritium concentration.
The problem is that Hamby’s dose estimate is one of the few reliable estimates we have of a tritium dose from a nuclear facility. And it indicates a considerably (~70x) higher dose than the average estimated dose from tritium from the bomb tests of 9 nSv (10-9 Sv) as stated by UNSCEAR (2000). However it is lower (~10x) than the estimated average tritium dose in 1962 of about 7 µSv from the atmospheric tests cited by Bennett (2002). 1962 was the worst year for bombs.
It is concluded that UNSCEAR’s assertion of no leukemia increases after the atmospheric bomb tests is incorrect. And that its hypothesis (viz, since none were observed from tritium released during the bomb tests therefore tritium NPP releases could not have caused the observed leukemia increases near them) is without foundation.
There were, in fact, observed leukemia increases, and in any event many other hazardous radionuclides were ejected into the atmosphere during the bomb tests apart from tritium.
A rough analysis indicates that average tritium in air concentrations were of the same order of magnitude for both scenarios. At least one estimated tritium dose from a nuclear facility is about the same as or larger than those from test bomb fallout.
The outstanding reason for the widely observed leukemia increases near NPPs remains their radionuclide emissions (Fairlie, 2014).
Appendix: Tritium in air concentrations from both scenarios
It is difficult to be accurate due to the lack of data, but with some assumptions the following order-of-magnitude concentrations for tritium to air emissions over one year can be estimated for the two scenarios. Tritium in air concentrations will vary a great deal due to distance from the NPP/bomb location, the time since release, and weather conditions. The following estimates are therefore broad averages, and should be treated with caution. It is stressed that this is an order-of-magnitude comparison only.
From NPP releases
UNSCEAR 2000 indicates great variability among annual figures for tritium emissions for NPPs around the world. From UK emissions data (RIFE, 2015), annual tritium emissions to air from UK nuclear power stations in 2015 were, on average, about 1 TBq (1012 Bq). We assume that most of this is released during refuelling episodes which occur about once a year, and that the tritium release forms a plume about 1 km wide x 1 km high x 20 km long with a volume of approximately 20 x 109 m3. . This gives an average tritium concentration of roughly 1012 Bq divided by 20 x 109 m3 = approximately 50 Bq per m3 after the annual NPP refuelling discharge.
To check this, I examined the literature for other measurements of tritium in air concentrations. Few exist. Hamby (1993) estimated tritium concentrations in air moisture but not in air volumes. The CNSC (2001) published a report “Tritium in the Canadian Environment” which showed air concentrations of tritium near Canadian NPPs. These varied between 0.01 and 40 Bq per m3 depending on the distances from NPPs. It is recalled that Canadian heavy water reactors are prolific sources of tritium, producing perhaps as much as 100 times more tritium per annum than PWR or BWR reactors. On the other hand, it is not stated whether the stated measurements were averaged values or when the measurements were taken, eg during annual outages or not.
From Atmospheric Bomb Tests
It is assumed that all bomb test fallouts were within the troposphere, ie limited to 15 km and restricted to the northern hemisphere. The volume of the earth’s troposphere in the northern hemisphere can be calculated to be 4 x 1018 m3 assuming a maximum height of 15 km.
Bennett (2002), who was a main author of the 2000 UNSCEAR report cited by UNSCEAR 2016, estimated that 185 EBq (1018 Bq) of tritium was released over a ~10 year period into this volume. For one year, this gives an average annual tritium concentration of 185 x 1018 Bq divided by ~10 years and by a volume of 4 x 1018 m3 = about 50 Bq per m3.
This concentration is similar to and, of more relevance, is the same order of magnitude as that from NPP emissions.
Bennett BG (2002) Worldwide dispersion and deposition of radionuclides produced in atmospheric tests. Health Phys 82:644–655.
CERRIE (2004) Committee Examining the Radiation Risks of Internal Emitters. Report of the Committee. HMSO.
CNSC (2001) Tritium in the Canadian Environment: Levels and Health Effects. Report RSP-0153-1. Prepared for the Canadian Nuclear Safety Commission under CNSC contract no. 87055-01-0184 by Ranasara Consultants and Richard Osborne.
Hamby DM (1993) A probabilistic estimation of atmospheric tritium dose. Health Phys. 1993 Jul; 65(1):33-40.
RIFE (2015) Radionuclides in Food and the Environment. Environment Agency. HMSO.
UNSCEAR (2000) United Nations Scientific Committee on the Effects of Atomic Radiation. New York, 2000.