But even that would still be within the annual regulatory target (22 trillion becquerels of tritium) for Fukushima I Nuclear Power plant.
The information about the amount of tritium released since the accident came from TEPCO during the August 2 meeting of the NRA working group to deal with contaminated water at Fukushima I Nuclear Power Plant.
So far, reports of radioactive materials in water have been focused mostly on I-131 (no longer detected) and radioactive cesium. TEPCO does measure other nuclides including tritium and strontium in seawater and fish, but the total estimates haven't been presented, until yesterday for tritium (August 2, 2013).
First, the news from Kyodo (8/2/2013):
Dozens of trillions of becquerels of tritium may have leaked from Fukushima I Nuclear Power Plant
Regarding the problem of contaminated water leaking into the ocean from Fukushima I Nuclear Power Plant, TEPCO disclosed the estimate on August 2 showing the amount of tritium in the contaminated water that leaked since May 2011 to be between 20 to 40 trillion becquerels. The estimate was reported during the meeting of the working group set up within Nuclear Regulatory Authority to deal with the contaminated water [at Fukushima I Nuclear Power Plant].
TEPCO will also estimate the total amount of radioactive strontium that has leaked from the plant, as strontium is easily accumulated in body and may affect health. The company will start the survey of fish and shellfish in the ocean near the plant.
The very last sentence in Kyodo News is not true. The survey has been ongoing. TEPCO has admitted so far only to the leak into the open culvert in front of the water intake canals, and not into the water in the plant harbor.
Before we collectively freak out on the huge number like 40 trillion becquerels, let's take a look at what it was before the accident for Fukushima I Nuke Plant and what it probably continues to be at all the other nuclear power plants in Japan when it comes to releasing tritium.
Here's a page (page 38) from the 2011 report by Nuclear and Industrial Safety Agency (NISA) under the Ministry of Economy, Trade and Industry regarding the nuclear waste management at Japan's nuclear power plants, showing the amount of tritium released yearly by nuclear power plants (English labels added by me; click to enlarge):
Before the accident, Fukushima I Nuclear Power Plant was releasing tritium in the order of trillion becquerels per year. In 2009, the amount was 2.0E+12 becquerels, or 2 trillion becquerels of tritium.
Nuclear power plants with pressured-water reactors release more tritium (the numbers in 2011 circled in blue in the table). Hokkaido's Tomari Nuclear Power Plant, for example, released 3.8E+13 becquerels, or 38 trillion becquerels of tritium in 2011.
The total amount of tritium released from Japan's commercial nuclear power plants in 2011 was, according to the table, 3.1E+14 becquerels, or 310 trillion becquerels. The order of magnitude is the same in all years in the table, at 100s of trillions becquerels per year.
The pre-accident target set for Fukushima I Nuke Plant for tritium release per year was 2.2E+13, or 22 trillion becquerels, according to TEPCO's document on tritium released on February 28, 2013. (Here's the results of 2006, showing ND for all nuclides tested except for tritium.)
So, according to TEPCO's estimate, release of tritium after the accident in 40 trillion becquerels, or 20 trillion becquerels per year, would be still within the company's pre-accident annual target, and about the same as amounts released by nuclear power plants with pressured-water reactors.
The density of tritium in waste water that can be released is 60,000 Bq/Liter, according to the law and regulations governing the operating nuclear power plants.
What's more worrisome is strontium, which many suspect has exceeded the target.
(H/T Kontan-Bigcat for NISA/METI document, TEPCO's document)