Check out this solar energy map for the USA:

The problem I see with this is that solar energy is given as Watt-hours per square meter per day. That is a measurement for electricity, not radiation. In other words they were measuring electricity produced. My mistake may be that I was not taking into account the efficiency of photovoltaic panels (perhaps 18% efficient). That might explain the confusion here.

Here is another graph for solar irradiance that I found:

I am thinking that this one may be the correct one, although I haven't integrated it to find the total power at sea level, so I'm not sure what the number will be. Notice that the scale on the left-hand side is not logarithmic - that will make things much easier. Indeed just looking at it about 75% of total sunlight appears to be transmitted through the atmosphere.

This will change my website in a couple ways:
1. The graph for ground-level sunlight will change, although the graph for low-earth orbit will stay the same.
2. The photosynthetic absorption of sunlight by green leaf extract will change, since the spectrum for ground-level sunlight is slightly different.
3. The actual efficiency of photosynthesis based on crop yields will be drastically reduced.
4. The size of a space-based sunlight collector for farming will increase quite a bit too.

Thanks for pointing it out, this is important stuff. I will let you know when that changes are done. Any other suggestions?