As I wrote in my previous post, TEPCO said at least 300 tonnes of highly contaminated (mostly beta nuclides) waste water after the RO (reverse osmosis) treatment may have leaked, because the water level of the tank that leaked was found to have dropped by 3.4 meters.
Well, not so fast.
Thanks to mostly TEPCO's extremely sloppy management of these steel tanks that hold highly contaminated RO waste water, it turns out we really don't know how many tonnes of this waste water has leaked from the tank No.5 in the area H4. It could be more, it could be less, but no one really knows.
How so?
Here are some of the topics discussed in the yesterday (August 21, 2013)'s long meeting of Nuclear Regulatory Authority's Working Group to deal with contaminated water at Fukushima I Nuclear Power Plant and which didn't get reported in the media. (Information from NRA's youtube video, presentations at the meeting, blogpost by @tsokdba, togetter by @mtx8mg)
1. Is it really "300 tonnes" leak?
"300 tonnes" of RO waste water leak assumes the tank No.5 had been actually close to full (about 96% of the capacity) before the leak started. But this turned out to be just an "assumption" (sound familiar?) on the part of TEPCO.
Why?
Because there is only one water gauge installed for the group of tanks. As TEPCO's diagram shows, the tank No.5 belongs to the tank group I which consists of the tanks No.5, 7, 8, 9, 10. The water gauge for the group is installed on the tank No.7. Note the tiny yellow star on the side of the tank No.7:

When the waste water is transported from the Reverse Osmosis Apparatus, it is fed to the tanks No. 5, 7, 8, 9, 10 evenly, because the tanks are connected to each other via the feed line. So when the tank No.7 becomes full, all the other tanks are assumed to be also full.
TEPCO had never measured the water level of the tank No.5 until this leak was found, and the company had to send workers on top the tank to stick in the tape measure.
Mr. Kinjo from the Nuclear Regulatory Agency (secretariat of NRA) asked if TEPCO had confirmed that the tank No.5 had been full. The answer was no.
2. Where has this "300 tonnes" (if it was actually 300 tonnes) of waste water been leaking from?
"300 tonnes" of RO waste water was supposed have leaked over a month, according to the media report. However, that's 10 tonnes, or 10,000 liters a day. Even for TEPCO, the amount is too large not to have spotted it much sooner. Workers who are sent out daily to patrol the site have been spotting leaks with few hundred liters. TEPCO says the workers didn't find anything until August 19.
TEPCO hasn't found any leak from the visible part of the tank itself, but the level of waste water in the tank continues to decrease, albeit at rates that are not consistent with the constant leak.
TEPCO measured the water level of the tank No.5 on August 20, and
From 9:30AM to 3:30PM, the water level dropped by 5 centimeters.
In the next 6 hours, it dropped by 1 centimeter.
So in 12 hours it dropped by 6 centimeters, which translates into 10 tonnes in 24 hours.
(Hmmm. Why are there two different, drastically different rates? How could they extrapolate this into 24-hour rate?)
Could the leak be from the bottom of the tank, which is made of five 16-millimeter thick SS400 steel sheets fastened together? (They did use epoxy sealer inside to prevent corrosion.)
(From TEPCO's presentation to NRA Working Group, 8/21/2013)

Could the concrete platform have cracks through which the waste water has been leaking?
TEPCO couldn't answer, as there is no way to check the leak at the bottom of the tank unless the water is first emptied.
3. Has the water leaked into the open water?
TEPCO says they have found locations along the drain that have high radiation levels, and says the possibility cannot be denied.
However, the analysis of seawater on August 20 near the location where the C drain drains to the ocean shows cesium-137 at 0.0018 Bq/cm3, cesium-134 and all-beta were ND. If the waste water has been leaking into the ocean, the measurement at this location should show the high level of all-beta. It hasn't, so far, this time.
So, in conclusion, neither TEPCO nor NRA know exactly what's happening, and but at this point it is more likely than not that this waste water with high beta hasn't leaked much into the ocean, at least not through the drains located nearby.
There is a possibility that the tank No.5 wasn't full, and that it wasn't 300 tonnes that leaked. And there is a possibility that the leak is from the bottom of the tank, and it has been leaking through a crack in the concrete (if there's a crack). In case of the latter, radioactive materials, especially all-beta, will show up in the groundwater sooner or later if TEPCO monitors the wells for the groundwater bypass plan.
TEPCO has finished emptying the tank No.5, but the radiation level inside the tank is just too high right now for closer inspection, according to NHK.
But the facts (or I should say the lack thereof) don't matter much any more to most people, and the Japanese media (like NHK) is almost gleeful that the attention of foreign media like BBC, Reuters, Wall Street Journal, etc. is again on the plant. The most recent entity to weigh in is IAEA, saying it "views this matter seriously and remains ready to provide assistance on request."