[home] [Japanese]

Detection of floating radioactive materials: its motion and timeline after the Fukushima Nuclear Accident

M. Yamauchi
Swedish Institute of Space Physics, Kiruna

cf. summmary figures

The accident at the Fukushima Dai-ichi nuclear power plant (FNPP-1) in March 2011 contaminated an area of more than 100 km in diameter with radioactive materials amounting to about 10-20% of that from the Chernobyl accident. A map of radioactive contamination levels has been published by the Japanese government after data sampling efforts by many volunteer regardless of the health risks. However, this gives only the surface-contamination information but not the motion and amount of the radioactive materials that was floating in the air. This includes both initial spread from the FNPP-1 and re-suspension from the ground. Without this knowledge, one cannot estimate the internal dose risk, which is more serious than the external dose for the general public outside the FNPP-1 area in the Fukushima case. (The external dose is more important for those directly exposed to nuclear explosion and highly contaminated area).

What so far many people believe is that
"The radioactive materials released by the explosion or bent are carried by wind and dropped by wet contamination during rain."

However, this explanation misses important points in terms of the health risk. Unlike the Chernobyl case, majority of the radioactive materials from the FNPP-1 to the atmosphere was transported by surface wind or low-altitude wind, and hence deposition was rather soft, causing high-risk of re-suspension. Although the external dose rate was not critical for the health risk, internal dose can be high enough to cause the future health problem. Namely, we have following issues:
* Floating radioactive materials that stayed in the air near the ground before wet deposition on 20 March.
* Re-suspension of the deposited radioactive materials and new inflow from the FNPP-1 area after 20 March.
* Overall high-risk period of the floating radioactive materials.
Without this knowledge, one cannot estimate the risk of the internal dose, a risk that is considered to be more serious than the external dose for the general public outside the FNPP-1 area in the Fukushima case. This knowledge is also extremely important in revising the current acute monitoring and warning system in case of future emergency situation. Unfortunately, these questions are difficult to answer by direct measurement of the floating dust because of the difficulty of the dust sampling (very little sampling during the critical period) and severe damage of all monitoring stations within 20 km from the FNPP-1 by the earthquake.

If direct data are not sufficient, one should estimate from indirect data. This is the task for scientists. Fortunately, Fukushima accident took place within relatively dense network of measurements for the first time in history. Although the accident itself is the nuclear physics problem, estimation of the spread of radioactive "dust" from these multipoint measurements is a geoscience problem. This subject of the study, namely the dynamics of radioactive dust in the atmosphere, is an unexplored field without any scientific society. Therefore, contributions from scientists that have relevant expertise are urgently needed.
As space scientists, we are one of the most relevant scientists to this subject because
(a) We deal with ion motion in the gas state. The ionizing radiation produces ions in the atmosphere during its transportation.
(b) We are expert in dealing with multi-point ground measurement to indirectly derive motions of remote ions. This technique can directly be applied to network data of the radiation dose rate.
(c) We are familiar with the effect of the radiation belt particles in satellite data. Although it is not gamma ray, the ionizing effect is the same.
Therefore, it is our duty to analyze the data in the current emergency situation with very few specialists in the world. In fact, we have written three scientific papers within 4 months from the accident (all are published).

(Link) Takeda, M., M. Yamauchi (correspondence), M. Makino, and T. Owada (2011), Initial effect of the Fukushima accident on atmospheric electricity, Geophys. Res. Lett., 38, L15811, doi:10.1029/2011GL048511.
(Link (open access)) Yamauchi, M., Takeda, M., Makino, M., Owada, T., and Miyagi, I. (2012): Settlement process of radioactive dust to the ground inferred from the atmospheric electric field measurement, Ann. Geophys., 30, 49-56, doi:10.5194/angeo-30-49-2012.
(Link (open access)) Yamauchi, M. (2012): Secondary wind transport of radioactive materials after the Fukushima accident, Earth Planets Space, 64(1), e1-e4, doi:10.5047/eps.2012.01.002.

In the published and follow-up studies, we used two methods to estimate the behavior or the radioactive dust:
(1) Using the ionizing effect of the radioactive materials, which can be monitored by the atmospheric electric field. This happened to be recorded 150 km southwest of the FNPP-1 (at Kakioka, the only observatory in Japan); and
(2) Deriving anomalies of decay curves of the radiation dose rates by comparing curves between nearby stations for more than 20 stations.
The atmospheric electric field data and radiation dose rates data show quite different changes, and therefore, they provide independent information to each other, giving us a better understanding of the motion of the radioactive materials. To supplement these rate data, cesium/Iodine ratio data from soil sampling are also examined.

(1) In a global scale, there is 200 kV difference in voltage between the ionosphere and the ground, while both the ionosphere and the ground are nearly equipotential, respectively. (This voltage is maintained by lightnings/sprites in the tropical thunderstorms.) In the normal condition, 200 kV is mainly distributed at lower part of the atmosphere, i.e., PG (voltage difference between unit distance in height) is large near the ground and small at high altitude. This causes very weak (~ pA/m^2) electric current under clear sky. (Atmosphere has very little, but non-zero ions due to natural radiation from radon, cosmic ray, etc.). The extra radioactive materials are expected to increase the ion density by the "ionizing radiation" within several meters distance from them, reducing the atmospheric resistivity near the ground. This changes the voltage distribution, reducing the PG near the ground. (Roughly speaking, 200 kV voltage generate about pA/m^2, and the increased ion density near the ground causes reduction of the atmospheric resistivity and hence the PG near the ground.) This is exactly what was observed at Kakioka, 150 km southwest of the FNPP-1. The decrease was in three steps, which was different from the Chernobyl case (only single-step). One can also use this principle to estimate the re-suspension of the radioactive materials, and the result showed substantial re-suspension until the end of April.

(2) If the decay of the radiation dose rate is solely determined by physical decay (half-life of 8 days for Iodine 131 and 2 and 30 years for Cesium 134 and 137, respectively), the decay curve should be smooth. However, actual data show anomalies, indicating occurrence of intermittent deposition/removal of the radioactive materials to/from the surface. (The removal includes weathering loss such as subsurface migration.) To find out the anomaly in model-independent way, simplest method is to compare the decay curve between nearby stations (e.g., by taking ratio). If one location decays slower than other nearby locations for a limited period, this most likely indicates new influx to that location. Alternatively, one can compare the decay curve from modeled change using the isotope ratio (between Iodine and Cesium) that is estimated from soil sampling. Both methods revealed large anomaly until end of April in the northwest direction from the FNPP-1, which is the downwind direction from daily sea wind in this area. Intermittent anomalies are observed even in middle of June. Furthermore, the ratio of the radiation dose rates often approaches unity, suggesting diffusion of the radioactive dust from highly contaminated area to lower contaminated area by frequently alternating wind in direction.

In summary, a combination of these data revealed that both the deposition and re-suspension took place at many times in different forms.
(1) Large increases of the radiation dose rate of about a few-hour scale is associated with the wind direction from the FNPP even 150 km distance. These increases are the result of passages of radioactive plumes near the ground level. This process was significant during 12-24 March. (Such events are difficult to catch in sampling, i.e., air sampling may underestimate the internal dose).
(2) Some passages caused substantial dry deposition to the surface. The amount of the dry deposition is quite significant even at Kakioka, 150 km away from the FNPP-1. On 14 March, the dry deposition at Kakioka (21 UT) caused an increase of local ion density by more than one order of magnitude.
(3) The radioactive dust after these dry deposition wes most likely suspended in the air near the ground, until the first substantial rain on 20 March. Therefore, the risk of the internal dose stayed high during 12-20 March, particularly for children whoes air taking is close to the ground.
(4) The deposited radioactive dust was most likely blown up from the surface by the strong wind. At Kakioka, the largest re-suspension event took place 02-07 UT on 16 March.
(5) Plumes are not correlated with explosions or vents at the FNPP-1 but with wind.
(6) Meanwhile, a single plume may have different I/Cs ratios at different parts of the plume. One such event is the plume traveled south on 20 March: the I/Cs ratio within the same plume turned out to be different by a factor of two between its core and the surrounding.
(7) Heavy rain on 20-23 March caused wet deposition in the entire area, and drastically decreased the amount of new ejection of large radioactive plumes from the FNPP-1.
(8) Re-suspension of the radioactive fallout and its large-scale transport was significant until the end of April, i.e., nearly 40 days after the accident. The transport direction is mainly from highly-contaminated to moderately-contaminated areas.
(9) Daily convection due to sunshine is one of the major *drivers* of such re-suspension. The down-wind direction of the daily sea wind (northwest of the FNPP-1) was found to be under the highest density of the floating radioactive materials during this period.
(10) Large re-suspension ceased by the end of April at more than 30 km distance from the FNPP-1. However, minor events of re-suspension from the area close to the FNPP-1 repeated even during June.

The analyses result indicates the importance of a dense network of the simplest ground-based observations. The analyses also showed usefulness of the electric field measurement, which almost no nuclear power plant currently has. Even portable instrument for the electric field measurement will help in understanding this risk, and therefore, it should be deployed as soon as any accident occurs.

updated 2012-3-18