A Cambridge physicist derives systematically if and how Britain could live on locally generated renewable energy. And in which cases nuclear power or external renewables (think Desertec) need to be part of the equation.

He describes every source of energy that is worth mentioning (wind, solar, hydro, offshore wind, wave, tide, geothermal). Next to that, he looks at consumption (cars, planes, heating/cooling, light, gadgets, food, stuff, public serving). He conveniently uses one unit of measurement everywhere: kWh/day. Currently, the outlook is bleak: we consume way more than we would be able to generate. Then, he describes what he thinks are ways to narrow that gap. Finally he sketches 5 energy plans that would even out.

It is partly estimation and partly prognosis, both seem mostly reasonable and I learned a lot. I bet that he is wrong here or there, but at least it is a documented process (it also has an extended technical appendix).

And entertaining: My recommendation is the new indicator he introduces to compare security of generation technologies (deaths per Megawatt) and when he notes how house-cats kill many many more birds than windmills.