4.7 Summary

Nuclear waste constitutes a health hazard for several reasons. First are the reported health impacts from routine gaseous and liquid waste emissions from nuclear facilities. Second are the very large global collective doses from nuclear reprocessing. And third is the unsatisfactory and unstable condition of much of the nuclear waste already created. High-level waste (HLW) in the form of spent nuclear fuel and vitrified waste from reprocessing contains more than 90 percent of the radioactivity in nuclear waste. However, there is no fully operational HLW final disposal site in the world. The continued practise of storing spent nuclear fuel for long periods in pools at most nuclear power plants worldwide constitutes a major risk to the public and to the environment. Spent nuclear fuel contains most of the radioactivity in the world’s nuclear waste, and consists of fission and activation products.

Estimates of the impacts of an operational HLW disposal remain speculative, but HLW still poses key questions of intergenerational liability and justice. The very long time-frames involved—the half life of Pu-239 is over 24,000 years—remains the single most important factor distinguishing nuclear waste from other kinds of waste.

Reprocessing of nuclear fuel creates more accessible forms of highly dangerous radioactive wastes, proliferation problems, high exposures to workers and the public, and radioactive contamination of the air and seas.

Only few countries do publish information, for example, on nuclide inventories in wastes. Such data collection and dissemination are primarily the responsibility of national governments. The data is needed to properly assess risks from nuclear waste and develop hazard rankings which tie observed health effects to exposures. So far, no comprehensive hazard scheme exists for the radionuclides in nuclear waste.

Risks may be derived from epidemiological studies, but the few that exist are of limited quality. Some studies suggest increased cancer rates, for example, but are individually too small to give statistically significant results. Meta-analyses could combine smaller studies to generate larger datasets which do produce statistically significant findings. However, meta-analyses on nuclear waste are notable for their virtual absence. The result is that many small studies continue to be criticized for their lack of statistical significance.

Finally, in order to assess risks, it is also necessary to have accurate doses, but these are often not measured in epidemiology studies. Even if they do exist they can often be unreliable due to the large uncertainties which surround them.