Hi RL

Solar, in space, is about 1370 W/sq.m at Earth's average orbital distance. After plowing through all the atmospheric crap it's 1000 W/sq.m even at zenith. Lower the angle the more crap to absorb it before it hits your panel. The best I've seen is about 12,200 kW.hr/year in space and 2500 kW.hr/year on the ground. So SPS would have a huge advantage over ground based if the transmission efficiency was high enough. Many cells won't even supply voltage if the incident light drops below a certain level, so clouds can really screw up the figures for some arrays. Realistic weather modelling gives figures of 1800 - 1200 kW.hr/year for a lot of locations in Europe and the USA.

An average American or Australian - we're the "worst" energy users globally, but in both countries we have big transportation costs compared to Europe - uses about 10 kW - 240 kW.hr per day. If the whole world used at our level about 65 terawatt.years would be used. If we had solar collectors running at 50% efficiency, hooked up to a superconducting powergrid and based in the sunniest climes then they'd be generating 1250 kW.hr/year per sq.metre. Total collector area would need to be 456,000 sq. km - pretty huge. Even more if we're using 10% efficient (about twice the current lab best) hydrogen making cells for making vehicle fuel.

A long, long way to go. Current (well 2003) non-hydro renewables is about 310 billion kW.hr generated - 35 GW installed. Solar would about 5 GW worldwide. So a growth of 13,000 fold is needed. At 10% p.a. that'd be achieved in 95 years. At 20% a mere 47, so 2053... almost in time for the fusion demo-reactor...

Adam