(UPDATE 2/21/2014) Someone may have opened the wrong valve, not the valve malfunctioning as initially claimed by TEPCO, causing the leak. See my latest post.
According to TEPCO's alert for the press (2/20/2014), there has been a fresh leak of highly contaminated RO (reverse osmosis) waste water at Fukushima I Nuclear Power Plant, the same kind of water that was found leaking in August 2013.
The densities of radionuclides in the highly contaminated water that was found leaking on February 19, 2014 from the top of a tank that stores waste water after reverse osmosis treatment are:
Cesium-134: 3,800 Bq/L
Cesium-137: 9,300 Bq/L
Cobalt-60: 1,800 Bq/L
Manganese-54: 1,300 Bq/L
Antimony-125: 41,000 Bq/L
All-beta: 230,000,000 Bq/L
There are 1,000 liters per tonne, and TEPCO says about 100 tonnes of this water leaked.
So, in the 100 tonnes of water, there are:
23,000,000,000,000 Bq, or 23 terabecquerels of all-beta (including strontium-90)
How did this happen? It was not through some cracks or loose rivets with deteriorating packing this time.
The short answer is: TEPCO let it overflow by negligence.
The longer answer, as I extract from TEPCO's photos and videos library (2/20/2014) document:
1. The tank is a tall, riveted tank to store RO waste water in H6 area. The tank was almost full.
2. For some unknown reason, the valves of the pipe that feeds RO waste water to the tank were open, which should have been closed. The waste water kept coming and filling the tank.
3. At 2:01PM on February 19, 2014, an alarm went off, indicating the water level is dangerously high.
4. But there was no transfer of the water scheduled at that time. So TEPCO and the workers thought either the alarm was malfunctioning or the water gauge was malfunctioning. They did nothing.
5. The waste water started to overflow from the flange at the top of the tank.
6. The waste water flowed into the rain gutter installed on the perimeter of the tank top to collect rainwater.
7. The gutter is connected to a drainpipe to guide rainwater OUTSIDE the dam that surrounds the concrete pad on which the tanks stand.
8. The extremely contaminated waste water flowed through the drainpipe onto the ground.
9. Finally a worker from a TEPCO affiliate company found the leak during a patrol at 11:25PM on February 19, 2014.
10. The water stopped at 5:40AM on February 20, 2014 after the valves were closed.
The ground outside the dam looks well-saturated: