Spectrum of the Sun vs black body

[fluidistic]

In wikipedia I've read that the Sun's surface temperature is about 5700K.
The emission spectrum graph can be seen there: http://en.wikipedia.org/wiki/File:EffectiveTemperature_300dpi_e.png.
I don't understand why the irradiance (or intensity I guess) of the Sun is greater than the one of a black body at 5700K. This would imply that the emissivity of the Sun is greater than 1, which is impossible. What am I missing?

[chemisttree]

Must be that the Sun's emission spectrum has more going on than just blackbody radiation.

[JeffKoch]

In wikipedia I've read that the Sun's surface temperature is about 5700K.
The emission spectrum graph can be seen there: http://en.wikipedia.org/wiki/File:EffectiveTemperature_300dpi_e.png.
I don't understand why the irradiance (or intensity I guess) of the Sun is greater than the one of a black body at 5700K. This would imply that the emissivity of the Sun is greater than 1, which is impossible. What am I missing?

It's just that the sun is not a blackbody. In other portions of the same curve, the ratio is less than 1. The 5700 number comes from a single-parameter fit to the actual spectrum, and does not mean that the sun really is a blackbody with a temperature of 5700K. At short wavelengths especially, it looks nothing like a blackbody.

[fluidistic]

It's just that the sun is not a blackbody. In other portions of the same curve, the ratio is less than 1. The 5700 number comes from a single-parameter fit to the actual spectrum, and does not mean that the sun really is a blackbody with a temperature of 5700K. At short wavelengths especially, it looks nothing like a blackbody.

I understand. But the Sun's surface is at 5700K, right? If I consider the Sun as a gray body then I should expect its irradiance to be lesser than the one of a blackbody at the same temperature, for all wavelengths. Am I right on this? If so, I still have trouble understanding the graph.