More info in English about food test results can be found on the Ministry of Health, Labor, and Welfare's (MHLW) site:
http://www.mhlw.go.jp/english/topics/2011eq/index.html#corrected15Aug2012
Data for the entire period March 2011 to now is available; in most cases it's in both .html and .pdf form, with several reports in searchable/sortable .xls files
Corrected Test results of radionuclide in foods sampled since 19 March 2011 to 31 March 2012:
http://www.mhlw.go.jp/english/topics/2011eq/index.html#corrected15Aug2012
Levels of radioactive contaminants in foods tested in respective prefectures.
Latest is 18 September, 2012:
http://www.mhlw.go.jp/english/topics/2011eq/dl/18Sep2012.pdf
Some people in Italy have made a useful front end to the MHLW database:
http://www.contaminazione-alimentare.tk/dom-en.html
It's not necessarily up to date, but very good nonetheless.
==========
The Japan Fisheries Agency (JFA) keeps its own testing stats for fish and seafood:
http://www.jfa.maff.go.jp/j/sigen/housyaseibussitutyousakekka/
None of this appears to be available in English.
JFA data:
3/30/2012 results, (covers 12/27/2011- 3/30/2012)
3386 items total
41 between 500 and 1000Bq/kg (1.2%) , 28 items over 1000Bq/kg,(0.82%)
8/10/2012 results (7/1-8/10/2012)
2214 items total,
17 between 500 and 1000Bq/kg (0.76%), 2 over 1000Bq/kg(0.09%)
9/18/2012 results (7/14-9/18/2012)
(I made a little more detailed breakdown)
3971 items total
1756 44.2% ND
2215 55.8% detected
241 6.0% over 100Bq/kg
1974 49.7% less than 100Bq/kg limit
211 5.3% 100-500 Bq/kg
27 0.6% 500-1000 Bq/kg
3 0.07% over 1000 Bq/kg
(highest 3300 Bq/kg "kurodai" = black sea bream, caught off Miyagi pref)
=====
Adding info from MHLW on fish:
Jan 1-Feb 1 2012 (several sample dates),
307 samples total:
ND 108 35%
1-100 Bq/kg 118 38%
under 100Bq/kg total 226 74%
100-500 Bq/kg 55 18%
500-1000Bq/kg 15 5%
over 100 total 81 27%
over 500 total 26 8.4%
over 1000 11 3.7%
The highest was rockfish, at 3100Bq/kg; there were 2 other rockfish samples over 1000 Bq/kg.
5 samples of greenlings were over 1000Bq/kg. Sea bass and poacher also had samples over 1000Bq/kg.
========
March 1-April 1 2012 (tested March 7)
451 samples total:
ND 126 28%
1-100 Bq/kg 203 45%
under 100Bq/kg total 329 72.9%
100-500 101 22.4%
500-1000 19 4.2%
over 100 total 125 27.7%
over 500 total 28 6.2%
over 1000 total 9 2%
The highest was land-locked salmon from the Niida river, Iitate at 18,700Bq/kg.
Other items over 1000Bq/kg include poacher and greenling; there were several instances of skate (spelled "skete") over 500Bq/kg. Other high items included flounder, rockfish, greenling, land-locked salmon, char, poacher.
========
June 1-July 1 2012 (tested June 5)
421 samples total:
ND 170 40%
1-100 170 40%
under 100Bq/kg total340 80%
over 100 total 81 20%
100-500 75 17.8%
500-1000 5 1%
over 500 total 6 1.4%
over 1000 total 1 (rockfish, 1600Bq/kg)
over 2000 none
High items include rockfish, greenling, flounder.
========
July 31, 2012 data
ND 38.7%
1-100Bq/kg 44.8%
under 100 total 83.3%
over 100 total 15%
over 500 0.75%
over 1000 (0)
highest level found: 640Bq/kg
=======
Basically, from looking at these and other reports, it appears that the percentage of samples over 100Bq/kg have been declining, as have percentages over 500 and 1000Bq/kg; the percentage ND is fairly steady and probably not really increasing (around 40%), and the percentage detected but less than 100Bq/kg also continues to fluctuate around 40%.
The number/percentage of really "high" items, i.e. over 1000Bq/kg seems to be declining, but the levels found are not. Some species, like rockfish and greenling, exhibit strong bioaccumulation, and we'll probably continue to see occasional high levels in other species like the kurodai as well. This is just the first year, so we'll need to see the results of a few cycles to get a better grasp of what kinds of changes are seasonal, connected to migration, etc.
Azby