Let's pay attention to the fission product /ratios/ rather than the absolute
amounts.
Let's see if we can understand this. We start with the hypothesis:
the distribution of fission products is a more-or-less flat
function of atomic number Z and baryon number A. [1]
That means that we should start out with roughly equal amounts of 131-I, 134-Cs,
and 137-Cs ... where "amount" can be measured in moles. Call the amount "N".
Note that I'm assuming that iodine compounds and cesium compounds are highly
soluble in water. Seems like a safe assumption.
For each nuclide, the activity should go like
R = N/τ [2]
where τ is the half-life. The activity can be measured in becquerels. Note
that I'm blowing off factors of ln(2). The lifetimes are
134 Cs 2.0652 years 65171363.5752 s
137 Cs 30.17 years 952072457.42 s
131 I 8.0197 days 692902.08 s
Now the funny thing is that the TEPCO data shows almost the same activity for
all three nuclides. That is inconsistent by many orders of magnitude, inconsistent
with hypothesis [1] in light of equation [2]. Even if we compute the initial
activities, by correcting for the fact that the 131-I has been decaying since
March 11th, the initial activities are still little more than an order of magnitude
apart, which is still wildly inconsistent with hypothesis [1].
The hypothesis [1] is rejected ... but then we are left with an amazing puzzle:
Why are the activities almost the same?
The obvious new hypothesis is:
The produced amount (N) for each nuclide is
roughly inversely proportional to the half-life [3]
Is this a coincidence that affects only these three nuclides, or is there some deep
physics principle that leads to rule [3] more generally?
The 134-Cs and 137-Cs activities are comparable, and the 131-I activity is just a
little bit higher.
===================================
Now, here is an interesting bit of analysis. Consider the isotope /ratios/
rather than the absolute amounts, and consider only the /relationship/ of the
seawater data to the spent fuel pond data.
The relationship is a match ... which is what you would expect in accordance
with the following hypothesis:
Both samples had similar histories. [4]
However, we are told that unit 4 was shut down in late November 2010. That is
some 145 days before the measurement in the spent fuel pool. We would expect
that the 131-I would have decayed by more than 5 orders of magnitude in that
time.
So we must consider two more hypotheses:
The fuel pins in the spent fuel pool in unit 4 are damaged. The fuel
has gone critical in the last few weeks, so as to produce a huge quantity
of fresh 131-I, and enough fresh Cs to swamp any "old" Cs. This is the
dreaded "recriticality accident". [5]
Or:
The radioactivity measurements do not (by themselves) indicate any damage
to the fuel pins in unit 4, but instead are almost entirely explained by
nasty stuff from units 1, 2, and/or 3 that got washed into unit 4. [6]
On the other hand, the fact remains that unit 4 suffered an explosion and
fire. The explosion was small compared to unit 3, but not small in absolute
terms, so there must have been quite a bit of hydrogen. That all by itself
strongly suggests that there has been a lot of damage to the fuel pins in
unit 4.
Bottom line: There are at least two different ways in which I am unable
to make sense of the data.