The fact that a zone exists does not mean the zone is contaminated, and it does not mean a release has occurred. You can confusing what can happen, the reason, with the evacuation, the precaution and assuming that an evacuation means a health significant release has occurred. I don't know how else to explain it. The fact that some radiation release has occured does not mean the region is contaminated or that this is the reason the evacuation was ordered. Evacuations are done before releases occur, if possible, and in this case thats what Japan did. They declared an emergency when it was clear it would not be possible to get a cold shutdown, thats a precaution in case things get out of control to prevent exposure to the public.
So, the data, the NNSA data as of April 29 is airborne readings:
http://blog.energy.gov/content/situation-japan/
This shows some caesium detected using airborne testing (not uranium or plutonium as someone else claimed). In those areas, the vast majority of the area the hr dose is below 0.1 mrem/hr. The annual normal dose a person gets in the US gets every year is about 360 mrem. If you fly or live in higher altitudes your annual dose is considerably higher. If you work at a nuclear powerplant your annual dose is around 1000 mrem a year. If you live in Denver, CO it would be 700 mrem/yr. If you spend a year on the beaches of Brazil your dose would be 5000 mrem/yr. Incidentally, 5000 mrem/yr is the legal occupation dose limit. Its a bit conservative.
The maximum rate measured is in an area north west of the plant on a diagonal approximately 30km long. The rate measured was between 1.9-19 millrems/hr. If you were is one of the areas, and getting a maximum dose at 19 mrem/hr it would take approximately 19 hours to get Us average, or , 263 hours to get a beach dose from Brazil. This is assuming you were near an emitting source. These dose rates will not make you sick (or kill you). If you stayed there for a few weeks you may increase your chances of cancer. Even then its not a given you will get cancer, everyone is different and it depends on how healthy you are, your genetics, distance from source, shielding, etc.
http://www.nrc.gov/about-nrc/radiation/rad-health-effects.html
Based on information from the NRC, if you smoke a pack of cigarettes you will lose 6 years of life expectancy. If you are 15% overweight you will lose 2 years of life expectancy. If you get a 360 millirem dose, you will lose 18 days, so figure at 19 mrem/hr its about a day.
As most of the area was measured as being below 0.1 millrem an hour, and the other ranges around the northwest region around 1.9 mrem/hr down to 0.19 mrem/hr, you wouldn't be talking about an appreciable effect. At 0.19 mrem/hr thats nor much at all when the US average is 1 mrem a day.
My definition of significant may be different from yours. Significant to me means it will make you sick now. Is there a likelihood of increased cancers, in that strip of 1.9-19 millirem/hr potentially if you stayed there for a few weeks. Given that the evacuation was done a long time ago, no one should be in that area that does not have adequate protection. So the health effects should be minimal at this point, if people evacuated (which apparently most did).
Additionally, keep in mind that distance from the source will reduce exposure (the inverse square law). Shielding will also reduce exposure, so just because an area may have sources does not mean you will be equally exposed to them or that its an equal amount (hence the variance in that red region). Increased shielding will decrease exposure. Internal exposure can be prevented with masks and clothing.
Anyway, theres some data. I'm tired.