But the level of contamination is less than water in Shaft C that connects to Reactor 2 turbine building.
TEPCO announced the results of nuclide analysis of the water sample taken from the Shaft B (see the diagram below, from TEPCO's 7/26/2013 document, with added English labels) at different water levels in the shaft. Shaft B and the seawater intake pipe trench that connects to the shaft go to Reactor 3's turbine building.
In an effort perhaps to make the data more accessible to mere mortals, in addition to announcing the results in liter TEPCO also spelled out the numbers instead of using "x.xE+x" format. (I wish they just used Bq/cm3. Too many zeroes, and I couldn't grasp the numbers right away.)
From TEPCO's 8/1/2013 handout for the press in Japanese (English labels are by me):
Comparing Shaft B for Reactor 3 and Shaft C for Reactor 2, you will note:
Chloride content is one order of magnitude bigger in Shaft B than in Shaft C.
Cesium content is one order of magnitude smaller in Shaft B than in Shaft C.
All-beta content is one order of magnitude smaller in Shaft B than in Shaft
Therefore, unlike the water in Shaft C and seawater pipe trench for Reactor 2, the water in Shaft B for Reactor 3 is not likely to be the highly contaminated water that leaked to the ocean in April 2011. Rather, the density is in the same order of magnitude as in the water that is currently in the Reactor 3 turbine building basement.
According to the most recent analyses of the contaminated water inside the turbine buildings,
Water in the Reactor 3 turbine building basement (as of 6/13/2013) has:
Cs-134: 1.5E+04 Bq/cm3, or 15,000 Bq/cm3
Cs-137: 3.1E+04 Bq/cm3, or 31,000 Bq/cm3
For reference, water in the Reactor 2 turbine building basement (as of 7/22/2013) has:
Cs-134: 1.1E+04 Bq/cm3, or 11,000 Bq/cm3
Cs-137: 2.5E+04 Bq/cm3, or 25,000 Bq/cm3
The numbers are comparable to water in the Reactor 3 turbine building basement but one order of magnitude smaller than the water sitting in the Shaft C and trenches connected to it.
Put the information yet another way for easier comparison, the numbers in Bq/cm3 instead of Bq/liter, except for chloride (click to enlarge):