Assessing long-term Health Effects from Fukushima’s Radioactive Fallout


The quadruple-explosion, triple-meltdown accident at Fukushima Dai-ichi nuclear plant in Japan after March 11 2011 is the world’s second-worst nuclear disaster, after the 1986 Chernobyl catastrophe.

Fukushima’s second anniversary in March 2013 is an opportune moment to assess its likely long-term consequences, although the accident is by no means over given the precarious state of the four wrecked reactor buildings – especially the spent fuel pond at Unit 4.

The figure below (reproduced from the French Government’s Institut de Radioprotection et de Sûreté Nucléaire (IRSN 2011)) indicates the extent of the radioactive fallout from Fukushima’s explosions and gaseous emissions in areas near Fukushima. In addition, lower concentrations fell over other areas of Japan, over neighbouring countries and eventually circulating the Northern Hemisphere.

Fukushima caesium

Exposure to radioactive fallout will result in radiation doses to the people living in these areas and this post will attempt to estimate future cancer deaths arising from these doses. Population doses are usually termed collective doses.

Estimating collective doses crucially depends on adherence to the Linear No Threshold (LNT) theory which assumes that radiation risks decline linearly with dose until zero dose is reached, ie there are small risks at very low doses well below background levels. Much evidence supports this view and the LNT is used by virtually all of the world’s radiation authorities including the ICRP, UK-HPA, US EPA, US NRC, BEIR VII, etc.

However UNSCEAR has recently attempted to limit the assessment of collective doses by recommending that doses below global average background levels (ie about 3 mSv per year) should be ignored. This follows the UNSCEAR chairman’s early suggestion that radiation exposure from the Fukushima nuclear accident would have no health effects (Dahl, 2011). I have examined and discounted these statements in a separate post. As we shall see, several studies have also ignored UNSCEAR’s recommendation and made collective dose estimates arising from Fukushima.

Initially this assessment is restricted to the population in Fukushima Prefecture as not enough information is yet available to estimate collective doses to other areas in Japan, to neighbouring countries and to the Northern hemisphere. However these collective doses will be considerable and should be added to the total for Fukushima Prefecture.

Initial Considerations

First, it is important to realise that air emissions from Fukushima are much more important, in terms of health effects, than Fukushima’s sea discharges. Therefore what follows is an examination of health effects from aerial emissions, ie from the radioactive fallout arising from Fukushima’s several plumes.

To gauge Fukushima’s impacts, it’s a good idea to compare them with Chernobyl’s impacts. The radioactive air emissions from Fukushima were about 3 to 5 times lower than those from Chernobyl apart from the inert gas, Xe-133. (A future post will examine source terms in more detail.)

In addition, it is estimated (Masson et al, 2011) that ~80% of Fukushima’s radioactive fallout fell over the Pacific Ocean whereas most of the radioactivity from Chernobyl fell on land. Also, collective radiation to workers and local populations appear to be lower from Fukushima compared with Chernobyl due to the better safety precautions taken after the Fukushima accident. (This is not to say that official Japanese actions were perfect, but they were better than their virtual absence at Chernobyl.) On the other hand, population densities in Japan are much greater than near Chernobyl. Overall, I shall conclude that Fukushima’s doses are about 10% those from Chernobyl, but this is a very approximate estimate.

In areas near Fukushima, most of the long-term dose to the public originates from Cs-134 and Cs-137 isotopes which fell from Fukushima’s plumes to the ground. These have longish half-lives (2 years for Cs-134 and 30 years for Cs-137) and they irradiate people with gamma and beta rays. This is called groundshine. Smaller doses originate from being immersed in the plume, from eating contaminated food and drinking contaminated water. In areas further away, the situation is reversed with most of the dose coming from ingestion and lesser amounts from groundshine.

Collective Doses and Future Cancer Deaths

Many studies have examined the emission, environmental transport and deposition of the radionuclides in Fukushima’s fallout using either observed datasets or chemical transport models or both. Takemura et al (2011), Masson et al (2011), Kinoshita et al (2011), Yasunari et al (2011), Stohl et al (2011), Morino et al (2011), and Sugiyama et al (2012).

However relatively few studies have estimated the likely levels of collective doses and future cancer deaths from the Fukushima releases. The early estimate by von Hippel (2011) was essentially a scoping study which scaled up doses from land areas contaminated with Cs-137 levels greater than 37 GBq per square kilometre. Nevertheless his estimate of about 1,000 deaths near Fukushima is consistent with later estimates.

Ten Hoeve and Jacobson (2012) [TH&J] used a more detailed global model to simulate the emissions, advection, decay, dissolution, aerosol–aerosol coagulation, aerosol–cloud coagulation, aerosol-nucleation scavenging, rainout, washout, and dry deposition of radionuclides from Fukushima’s fallout.

As the largest single source of radiation exposures to the public near Fukushima is from ground contamination, ie groundshine, they used a Cs-134 and Cs-137 ground contamination model to calculate doses. Their model assumed that these isotopes disappeared from the ground with a half life of 14 days, based on historical US measurements. Using this ground contamination-to-dose model plus the US EPA’s  dose-risk coefficient of 5.8% per person Sv, their central estimate was for 130 future cancer deaths with uncertainty bounds stretching from 15 to 1100 deaths, mostly in Japan.

However the later Beyea et al study (2012) spotted that the parameter value used TH&J’s study for radiocaesium weathering in soils was out of date. Instead of relying on older US data that tracked radioactivity only while resident on vegetation, Beyea et al pointed to a more recent study (Drozdovitch et al, 2007) on radiation exposures to the European population following Chernobyl. This study tracked radioactivity beyond residence time on vegetation to include its migration into soil, using a weathering attenuation factor with two half-life time constants of 2.4 years and 38 years – both considerably longer than 14 days used by TH&J.

This vital European weathering factor was obtained from an analysis of over 300 soil samples taken during 1987–1999 in areas of Russia (Drozdovitch et al, 2007), complemented with soil measurements of radiocaesium from weapons fallout in north-eastern US and Bavaria, Germany. Measurements in other land areas had similar findings.

This use of more up-to-date data should not be taken as a criticism of the TH&J study, because this is how science works. One uses the data known to the researchers and then corrects it with later data: this is normal. Indeed Beyea et al compliment the TH&J study for the depth of its detail, and for having the temerity to make collective dose estimates which other researchers have notably shied away from: I agree with their views.

Beyea et al’s Assessment

Beyea et al then made their own assessment of collective doses from Fukushima using the longer weathering factor to adjust the modelling results of TH&J.

It’s important to realise that the use of 2.4 and 38 year soil weathering half lives means that groundshine doses from Fukushima (and Chernobyl) last more than 70 years – ie a lifetime. Many scientists remain unaware of this implication but it is based on good evidence – see above. The dose consequences are shown in graphical form in figure 1 below (reproduced from Beyea et al, 2012) which shows that, after 75 years, the total collective dose from groundshine is about 6.5 times greater than the groundshine dose received in the first year. Unfortunately the recent second WHO (2013) report on the health risks from Fukushima used a factor of 2 and not 6.5: one of a number of deficiencies in its report.

 relative dose

Figure 1. Reproduced from Beyea et al, 2012. Cumulative external ground dose from Cs-134 and Cs-137 relative to first-year external ground dose, derived using equations in Beyea et al, 2012.

Beyea et al first assessed likely external doses from groundshine near Fukushima (ie ground contaminated with Cs134 and Cs-137) using the following assumptions –

(a) soil weathering attenuation with two exponentials (half-lives of 2.4 years and 38 years) to model the effect of caesium’s soil penetration on dose rate

(b) Cs-134 to Cs-137 Bq ratio = 0.9

(c) Cs-134 dose per disintegration 2.7 times greater than that for Cs-137

(d) dividing their collective doses from groundshine by 2 (range 1 to 4) to account for the effects of restrictions on food and water, and relocation.

(e) a shielding reduction factor of 0.28 (range 0.14 to 0.56).

For internal doses, the authors divided their estimated collective external ground dose by 3.3. They also divided their collective doses from ingestion by 2 (range 1 to 4) to account for the effects of restrictions on food and water, and relocation. After adding external and internal collective doses, the authors used the US EPA’s absolute risk of fatal cancer of 5.8% per Sv (ie incorporating a DDREF of 1.5) to convert from collective dose to future cancer deaths.

The net result was a mid-range estimate of 1,000 added cancer deaths instead of the 135 estimated by TH&J, with large uncertainties also being acknowledged.

My Own Estimates

In table A below, I estimate the cumulative population dose over 70 years from groundshine in Fukushima Prefecture based on tables 1 and 2 of the IRSN report (2011) (assuming a shielding factor of 0.65) as summarized in Beyea et al (2012).

Table A. Cumulative collective doses over 70 years to Fukushima Prefecture

Dose Range Sv Average doseSv Population in dose range Collective Dosesperson.Sv Fatal Cancers applying 10% per Sv risk
no exposures assumed in evacuated zones 0 0 0
Non-Evacuated Areas
0.082-0.136 0.11 (mid-range) 43,000 4,730
0.041-0.082 0.06 (mid-range) 292,000 18,000
0.0 – 0.041 0.0065* 1,670,000** 11,000
 totals ~34,000 ~3,000

*derived (by me) from an assumed first year average dose of 1 mSv to the rest of the population of Fukushima Prefecture x 6.5 for 70 years

** derived (by me) from the 2010 population in Fukushima = 2,030,000 less 360,000 in the previous dose categories = 1,670,000

Notes to Table A

1. Table A relies on 70 year life-time exposures from radiocaesium groundshine (as used by IRSN, 2011) as these dominate all other routes and sources of radiation exposure.

2. It only applies to Fukushima Prefecture: more collective doses will need to be added for the rest of Japan, neighbouring countries and the rest of the Northern Hemisphere when this data is made available. These added collective doses will be significant.

3. Groundshine doses do not take into account exposures from other exposure pathways, such as immersion within the Fukushima fallout plume, and internal contamination from inhalation of particles in the plume, as well as internal doses already received from contaminated food ingestion. These exposures are important in assessing first year doses, but less so for 70 year doses. Similarly, doses from the deposition of I-131 and other isotopes including tellurium-132/iodine-132, rhenium-103/rhodium-103, barium-140/lanthanum-140, and niobium-95 are not included: again these are important for early exposures, but less so for 70 year exposures.

4. My collective dose estimate partly depends on my assumption of a first-year average dose of 1 mSv to the rest of the population of Fukushima Prefecture from the Fukushima fallout. The 1 mSv figure comes from data from official reports of glass badge dosimeter surveys in Fukushima Prefecture townships, compiled by the independent radiation monitoring group Safecast in Japan. This introduces added uncertainty (range 0.5 to 2 mSv) but it is in only one of six dose categories.

5. Like Beyea et al, I assume that no evacuated people moved back into the evacuation zones. This is probably incorrect (ie collective doses could be higher) but we have no information as to how many people may have returned and/or when they returned.

6. To convert from collective doses to fatal cancers, I use the ICRP’s absolute fatal cancer risk of 10% per Sv: in other words I have not applied the dose and dose rate reduction factor (DDREF) used in the other dose assessments cited here. The use of a DDREF reduces the number of estimated fatal cancers in Europe by 2, and in the US by 1.5. However, as pointed out by Beyea (2012) many epidemiology studies offer little support for the use of such a factor, certainly for solid cancers (Little et al, 2008). Also, the recent WHO (2013) report on risks from Fukushima recommends that a DDREF should not be used for longer term exposures.

7. Considerable uncertainties surround my estimates. They should only be used as rough guides. Given the uncertainties involved for fatal cancers, only a single significant figure should be used, ie 3,000. This figure lies within the uncertainty range for Beyea et al’s main calculation.

Table B sets out various collective dose and fatal cancer estimates from the studies mentioned above.

 Table B. Published estimates of collective doses and fatal cancers

Study Comments Area Collective Dose in person.Sv Estimated future fatal cancers
von Hippel (2011) based on scaling up from ground Cs concentrations Most highly contaminated not given ~1,000
IRSN (2011)draft report first four years’ dose to population of 70,000 near Fukushima NPP Fukushima Prefecture 4,400  not given
TH&J (2012) global model,first year dose World not given 130 worldwide(range 15 – 1100)
Beyea et al (2012) (A) based on TH&J global model plus 70 year groundshine doses, using Drozdovitch (2007) data World not given ~1,000(range 60 – 6,000)pers comm*
Beyea et al (2012)** (B) based on IRSN (2011) model Fukushima Prefecture 6,000*** > 350
this study based on 70 year groundshine doses Fukushima Prefecture 34,000 ~3,000

* pers comm from Jan Beyea

** NB the average shielding factor (0.28) used by Beyea et al was twice the value used by me (0.65), and Beyea et al allowed another factor of 2 reduction for evacuation and decontamination

*** 23,000 reduced to 6,000 assuming above two factors

Reasons for differences between Beyea et al’s and my estimate

The 9.7-fold difference in estimated cancer deaths between the Beyea et al ‘s estimate based on IRSN and myself is due to the following four factors:

(a) I include a collective dose estimate for people receiving doses below 41 mSv,ie increasing the collective doses by a factor of 1.43. Beyea et al did recognize and discuss the limitation in their calculations on this score.

(b) my shielding factor (0.65) is ie twice the central value used by Beyea et al (0.28),

(c) Beyea et al allow another factor of 2 reduction for decontamination in their mid-range calculation, although they state decontamination would only apply above a threshold dose and they questioned the efficiency of decontamination.

(d) Beyea et al use a risk factor of 5.8% per Sv and I use 10%, ie increasing the risk by a factor of 1.72.

Multiplying these four factors together (1.43 x 2 x 2 x 1.72= 9.8) accounts for the increase.

8. Table C for comparison sets out collective dose and fatal cancer estimates from Chernobyl’s fallout.

Table C: Chernobyl effects

Study Comments Collective DosePerson Sv Fatal Cancers
Fairlie and Sumner (2006) World-wide 600,000 30,000 to 60,000
Anspaugh et al (1988) Northern Hemisphere 960,000 17,000
Cardis et al (2006) Europe-wide Not given 16,000 thyroid + 25,000 other cancers,(NB cases not deaths)

The overall interim conclusion, based on incomplete data, is that estimated collective doses and fatal cancers from Fukushima are about an order of magnitude lower than those from Chernobyl.

I wish to thank Jan Beyea, Ed Lyman and Frank Hippel for their informative and helpful article (2012) which has been used extensively in preparing this post.


Anspaugh LR, Catlin RJ, and Goldman M (1988) The Global Impact of the Chernobyl Reactor Accident. Science 242. pp 1513-1519.

Beyea J, E Lyman and F von Hippel (2013) Accounting for long-term doses inWorldwide health effects of the Fukushima Daiichi nuclear accident” Energy Environ. Sci., 2013, Accepted Manuscript. DOI: 10.1039/C2EE24183H

Beyea J (2012) The Scientific Jigsaw Puzzle: Fitting the Pieces of the low-level radiation debate. Bull. Atomic Sci., 2012, 68, 13-28.

Cardis et al (2006) Estimates of the cancer burden in Europe from radioactive fallout from the Chernobyl accident. Int. J. Cancer, 119, 1224–1235.

Dahl F, U.N. expert sees no serious Fukushima health impact. Reuters, 18 April 2011.

Drozdovitch et al (2007) Radiation exposure to the population of Europe following the Chernobyl accident. Radiat Prot Dosimetry (March 2007) 123(4): 515-528.

Fairlie I and Sumner D (2006) The Other Report on Chernobyl (TORCH): An independent scientific evaluation of the health-related effects of the Chernobyl nuclear disaster with critical analyses of recent IAEA/WHO reports. Commissioned by Rebecca Harms, MEP, Published by Greens/EFA in the European Parliament. April 2006.

IRSN (2011) Assessment on the 66th day of projected external doses for populations living in the north-west fallout zone of the Fukushima nuclear accident – outcome of population evacuation measures, draft Report DRPH/2011-10. L’Institut de Radioprotection et de Sûreté Nucléaire.

Kinoshita et al (2011) Assessment of individual radionuclide distributions from the Fukushima nuclear accident covering central-east Japan. Proc. Natl. Acad. Sci. USA,108, 19526–19529.

Little MP, Hoel DG, Molitor J, Boice JD, Wakeford R, Muirhead CR (2008) New models for evaluation of radiation-induced lifetime cancer risk and its uncertainty employed in the UNSCEAR 2006 report. Radiat Res 2008 Jun;169(6):660-76. doi: 10.1667/RR1091.1.

Masson et al (2011) Tracking of Airborne Radionuclides from the Damaged Fukushima Dai-Ichi Nuclear Reactors by European Networks. Environ. Sci. Technol. 2011, 45, 7670–7677.

Morino et al (2011) Atmospheric behaviour, deposition, and budget of radioactive materials from the Fukushima Daiichi nuclear power plant in March 2011.Geophys. Res. Lett.,38, L00G11.

Stohl et al (2011) Xenon-133 and caesium-137 releases into the atmosphere from the Fukushima Dai-ichi nuclear power plant: determination of the source term, atmospheric dispersion, and deposition Atmos. Chem. Phys. Discuss., 11, 28319–28394.

Sugiyama et al (2012) Atmospheric Dispersion Modeling: Challenges of the Fukushima Daiichi Response. Health Phys, 102, 493–508.

Takemura T, Nakamura H, and Nakajima T (2011) Tracing airborne particles after Japan’s nuclear plant explosion. EOS Transactions American Geophysical Union, 92, 397-398, doi:10.1029/2011EO450002.

Ten Hoeve JE and Jacobson MZ (2012) Worldwide health effects of the Fukushima Daiichi nuclear accident. Energy Environ. Sci., 2012, 5, 8743-8757.

UNSCEAR (2012) Supplement No. 46 (UN document A/67/46, 14 August 2012)

von Hippel F (2011) The radiological and psychological consequences of the Fukushima Daiichi accident. Bulletin of the Atomic Scientists. 67, 27–36.

WHO (2012) Preliminary dose estimation from the nuclear accident after the 2011 Great East Japan earthquake and tsunami. (

WHO (2013) Health risk assessment from the nuclear accident after the 2011 Great East Japan Earthquake and Tsunami.

Yasunari et al (2011) Cesium-137 deposition and contamination of Japanese soils due to the Fukushima nuclear accident Proc. Natl. Acad. Sci. USA 108, 19530–19534.

This entry was posted in News and comments. Bookmark the permalink.