More than four years have passed since an earthquake and tsunami triggered damage to the Fukushima Daiichi nuclear power station. The units were scrammed prior to loss of power and there were delays before meltdown, allowing for significant decay of the shortest-lived radionuclides, which contribute most to early radiation fields. More importantly, the Fukushima Daiichi primary containment was effective in greatly limiting releases of even volatile radionuclides and ensured negligible loss of the most toxic, alpha-emitting actinides. The radiological impact was further reduced because most of the releases (80%) were dispersed over sea rather than land.

Radioactivity releases from Fukushima Daiichi were due to specific events (venting, explosions) and the resulting transport and fallout of radionuclides was complex, depending on the wind strength and direction and whether it was raining or snowing at the time. The quantity and characteristics of fallout were quite different depending on whether "dry" or "wet" deposition occurred and its subsequent behaviour depends on the local topography and land use. Although initial concerns focused on radioiodine, with an eight-day half-life, contamination outside the Fukushima site is now dominated by caesium-137, with a 30 year half-life, and rapidly decreasing activities of caesium-134, with a 2.1 year half-life.

Ideally, in a major accident releasing volatiles, stack measurements of the source term are combined with meteorological data to predict potential contamination, which is then refined by regional radiation monitoring. But at Fukushima Daiichi releases were poorly defined and much of the monitoring network was knocked out by the earthquake and tsunami. Aerial surveys of gamma dose thus played a critical role in mapping contamination, calibrated by point analysis and vehicle-born surveys where access was possible. Data were initially used to run the fallout model "SPEEDI" in inverse mode in order to quantify releases.

Resulting fallout maps show the highest activities in a zone to the northwest of Fukushima Daiichi. This information was initially used to guide evacuation, which was difficult due to the damaged transport infrastructure. Logistics were further complicated by the scale of the evacuation due to tsunami damage, which affected a larger area than radioactive contamination.

Later, the contamination maps were used to plan remediation. Local dose rates in air were manually measured at several thousand locations and soil profiles were sampled to build up a detailed 3D understanding of initial fallout and its subsequent redistribution. Most soil profiles of caesium concentration show a marked decrease with depth: almost the entire inventory is contained within the upper 5cm.

For Fukushima prefecture, vegetation is an important factor as about 70% of the land area is forested. The extent of fallout capture depends on the size and type of tree (deciduous or evergreen) and the subsequent distribution of fallout between foliage, leaf litter and soil varies significantly with time for different kinds of forest in different topographic settings.

Decontamination

In absolute terms, the levels of radiocaesium contamination in much of the evacuated zone are low and comparable to those resulting from the 1957 Windscale fire or the distant fallout from Chernobyl in Fenno-Scandinavia, the UK and the southern European Alps. In these other examples, there were some restrictions on use of local foodstuffs but no attempt at decontamination. As in Fukushima, contamination dropped at a rate greater than that expected by radioactive decay alone, due to wash-off of soil-bound radionuclides and redistribution deeper into the soil column.

In Japan regional remediation was used, predominantly for socio-political reasons, but also recognising that fear of radiation can have serious health effects and that Japan can afford the measure. The aim is to enable the rapid return of evacuees and to provide assurance that they can resume their previous lifestyles without concerns about their health and that of future generations.

The Japan Atomic Energy Agency is working with a number of Japanese and international organisations and research institutes in the research around the remediation. This proceeded in three stages. In the first, JAEA tested the technology at two locations outside the evacuated zones where contamination was low. This allowed alternative approaches for measuring contamination, removing it and handling resulting wastes to be developed and tested in a representative range of agricultural and rural settings. This work formed the basis of a larger test of technology when JAEA contracted consortia of engineering companies to carry out remediation in a wider range of settings, including urban and industrial areas and locations with higher contamination levels (Decontamination Pilot Project -DPP ). Novel decontamination approaches were also investigated.

Contractors working for the Ministry of the Environment are carrying out regional remediation. JAEA supports such work predominantly throughdevelopment of improved understanding of the long-term behaviour of caesium in this setting (project termed F-TRACE), evaluating the management of forest areas that cannot be simply decontaminated, developing quantitative models of the natural self-cleaning processes that eventually transport radioactivity through river systems into the coastal marine environments, and assessing how control measures minimise environmental impacts.

The cleanup is generating huge volumes of contaminated soil and vegetation waste, which must be managed safely and cost-effectively. Future reuse of soil for construction is an important option, if constraints in terms of allowable organic and clay content can be managed. High organic waste and other combustible material can be incinerated, but a robust concept for conditioning, packaging and storing the resulting ash is needed.

Temporary storage of simply packaged waste at the site is a challenge in itself, as stores are vulnerable to earthquakes, typhoons, heat and cold, while the waste may settle and degrade. Over a longer term, waste will be transported to interim stores in the more highly contaminated area near the site and stored for 30 years. No decision has been made on final disposal.

Extensive efforts have been made to communicate such work to the public, although it is clear that even more could be done. There are concerns over environments where active remediation is limited (forests, lakes, coastal marine areas). Here it may be important to put concerns in perspective by analogy with similar contamination (e.g. the UK Lake District, a popular tourist destination where past contamination is almost completely forgotten) or areas of naturally high radiation background that are generally considered to be healthy holiday destinations (Finland, high altitude ski resorts).

Although evacuees are returning to some remediated areas, cleanup of the highest contamination near Fukushima Daiichi will take a long time and may require new technology. Effective technical communication must be maintained over a long period of time and the tacit knowledge of the multidisciplinary teams involved captured, so modern web-based knowledge management tools are employed. A version of the communication platform (Cleanup Navi) is available in English and information has been summarised in JAEA technical reports and two international workshops.

Lessons learned

An important lesson is that basing all planning on past experience is dangerous where large populations and sensitive infrastructure is located in zones at risk from natural catastrophes. Low frequency events should be better covered in disaster response plans.

For regional radioactive contamination, planning can greatly minimise consequences and aid recovery. It should include training response teams, establishing communication channels and material, and setting up regulatory infrastructure.

A system of civil defence or disaster response already exists in most developed countries and they may only need to develop a wider range of scenarios. Learning from Fukushima, it is important that too much focus is not placed on the cause of the accident, rather effort should be placed on response to the unimaginable occurrence. A good example of this approach is the response to the accident by the Swiss nuclear regulator ENSI.
Communication was one of key issues to be improved in the response to the Fukushima Daiichi meltdowns. Problems were experienced at every level – local, regional, national and international. To be fair, no past accident of this type has been well communicated or has received such extensive media coverage, which continues to the present day. It has to be recognised that, in many cases, fear of radiation causes larger health impacts than radiation itself and so reduction of such fear is an important component of any disaster response plan. Active response is also needed to combat deliberate "fear mongering" by groups with other political agendas.

Although the Japanese government reacted quickly to pass acts for decontamination, implementation was challenging for cleanup activities and, in particular, the movement, storage and disposal of resultant wastes. For the last, an act passed on November 2014 specified that the disposal site shall be located outside of the Fukushima prefecture. Such an arbitrary decision is clearly driven by social pressure: if the regulatory framework is put in place in advance of an accident, it would not only have facilitated decontamination, but would also have been made on a more logical technical basis.

The Fukushima work demonstrates that large-scale decontamination can be carried out cost-effectively if the value of the land is sufficiently high, although Japanese boundary conditions in terms of the nature of contamination and the technology applied may not be transferable to other sites.
Fukushima work has built an extensive database on the environmental behaviour of caesium, as this is a common contaminant. Although caesium has been extensively studied in the past, the integrated regional assessment, which follows transport from forested mountains to a coastal marine environment is unique and may be useful for other countries with similar geographic and climatic conditions. ■