What this suggests unfortunately is that radionuclides collected in
the sediments continue to be remobilized back into the food chain.
This can continue for years and it suggests that it will become
necessary to dredge the affected sea bottom. Since we are generally
dealing with heavy metals, this could even be a practical solution.
This also implies that it remains largely localized and can be dealt
with.
The real problem of course is that our present solutions for handling
radioactive waste remain unsatisfactory. Reasonable solutions are
possible but naturally unproven.
I would dissolve the material at high temperature in a molten
refractory to manufacture bricks. Those same bricks I would then
encase in a larger block to reduce the radiation effect to a safe
level. Best choices need to be determined but that is cheap and
simple. The blocks them selves could then be packed into a coal mine
which can be allowed to recompress. This surrounds the material with
a natural sponge for errant atoms that might leak out.
That is about as practical a solution as I am able to devise. Once
it is set up and operating, special solutions can obviously be worked
up also for irradiated hardware. If it gets down to it, we can
operate a coal mining front with simple concrete blocks containing
irradiated hardware replacing the coal in the back space and simply
leaving no pillars.
Fishing for answers
off Fukushima
by Staff Writers
Cape Cod MA (SPX) Oct 31, 2012
WHOI senior scientist and marine chemist Ken Buesseler (foreground) checks a CTD sampler prior to deploying the instrument to collect data and water samples from the ocean off the coast of Fukushima, Japan. Credit: Ken Kostel, Woods Hole Oceanographic Institution.
Japan's
"triple disaster," as it has become known, began on March
11, 2011, and remains unprecedented in its scope and complexity. To
understand the lingering effects and potential public health
implications of that chain of events, scientists are turning to a
diverse and widespread sentinel in the world's ocean: fish.
Events
on March 11 began with a magnitude 9.0 earthquake, the fourth
largest ever recorded. The earthquake in turn spawned a
massive 40-foot tsunami that inundated the northeast Japanese coast
and resulted in an estimated 20,000 missing or dead. Finally, the
wave caused catastrophic damage to the Fukushima Dai-ichi nuclear
power plant, resulting in the largest accidental release of
radiation to the ocean in history, 80 percent of which ended up
in the Northwest Pacific Ocean.
In
a Perspectives article appearing in October 26, 2012, issue of the
journal Science, WHOI marine chemist Ken Buesseler analyzed data made
publicly available by the Japanese Ministry of Agriculture, Forestry
and Fisheries (MAFF) on radiation levels in fish, shellfish and
seaweed collected at ports and inland sites in and around Fukushima
Prefecture. The picture he draws from the nearly 9,000 samples
describes the complex interplay between radionuclides released from
Fukushima and the marine environment.
In it, Buesseler shows
that the vast majority of fish caught off the northeast coast of
Japan remain below limits for seafood consumption, even though
the Japanese government tightened those limits in April 2012.
Nevertheless, he also
finds that the most highly contaminated fish continue to be caught
off the coast of Fukushima Prefecture, as could be expected, and
that demersal, or bottom-dwelling fish, consistently show the highest
level of contamination by a radioactive isotope of cesium from the
damaged nuclear power plant.
He
also points out that levels of contamination in almost all
classifications of fish are not declining, although not
all types of fish are showing the same levels,
and some are not showing any appreciable contamination.
As a result, Buesseler
concludes that there may be a continuing source of radionuclides into
the ocean, either in the form of low-level leaks from the reactor
site itself or contaminated sediment on the seafloor. In addition,
the varying levels of contamination across fish types points to
complex methods of uptake and release by different species, making
the task of regulation and of communicating the reasons behind
decision-making to the fish-hungry Japanese public all the more
difficult.
"To
predict the how patterns of contamination will change over time will
take more than just studies of fish," said Buesseler, who led an
internationalresearch cruise in 2011 to study the spread of
radionuclides from Fukushima. "What we really need is a better
understanding of the sources and sinks of cesium and other
radionuclides that continue to drive what we're seeing in the ocean
off Fukushima."
To
help achieve this, Buesseler and his colleague Mitsuo Uematsu at the
University of Tokyo are organizing a scientific symposium in Tokyo on
November 12 and 13 to present the most current findings available
about how radionuclides from Fukushima Dai-ichi have affected the
ocean, marinelife, seafood, policy decisions, and media coverage
to date.
The event will also
include a free public colloquium in Tokyo on November 14 to help
spread information about the lessons learned to the broadest possible
audience.
No comments:
Post a Comment