**Worst day**

I tried to work out the surface of PV needed for the worst day this year so far - with a daily maximum of 2,025 GWh consumed in February (according to RTE).

With a population of 66 million, this translates into a consumption of 30.67 kwh per capita in a day. That's over 4x the daily average, according to the World Bank (7.300 kwh in France)

According to ADENFI, France gets 53,3 kwh/m2 of horizontal solar irradiance in a day, in February (average over 2012-14).

I don't know how to take the panel inclination into account, so I'll translate this directly into 0,41876 kwh/m2 for an entire day, with a 22%-efficient panel (cf. Solarcity latest module).

From there, one could say France needs at 73.24 m2 of PV to provide enough energy on the peak day.

**Is it big?**

I couldn't find the total area of roof in France.

The average housing area was 40 m2 in 2006, according to INSEE, but I don't know how to convert that into roofing (depends on the number of floors). I also don't know the area of offices, parking, industrial and commercial spaces. That might double the total area of roof.

Some very-ballpark estimations: 1 m2 of space for every m2 of housing and all buildings have 2 floors on average. This makes 40 m2 of roof per inhabitant. If 100% of the roof area could get a 22% efficient solar panel, we would need 33 m2 of ground-level PV per capita... or a full 13.82 kwh backup battery (but how to charge them the day before with so little solar energy?).

**How reliable and insightful is this estimation?**

Any advice to correct/improve this calculation, especially to estime

- the impact of panel angle on the PV efficiency %
- the total area of roof that could get solar panels
- the energy available per capita from a backup powerwall battery.

What does it say about a 100% "solar+panels" future? And how could house-mounted wind turbines compensate for the lack of solar power in a peak day? How to tackle the electrification of cars? Fortunately, we already have the TGV!