Report comment

Report comment

Report comment

Obviously, the further south and the fewer clouds, the more energy from solar. Dry climates have few clouds. But wet climates have another problem, which is that the water holds heat and so decouples heat from direct irradiation. So maybe solar is perfectly aligned with peak demand in Spain and California, but not in Germany and New Jersey. But it is certainly better than a random source like wind.

Ryan, this is very misleading:

This of course means twice the CO2 emissions. These emissions are also not accounted for as emissions from wind/solar generation.

If solar requires more peaking plants, then of course solar should be blamed for the cost of those plants. But that is a capital cost. The pollution from an additional peaking plant is negligible because the peaking plant almost never runs, I point you yourself mention in the follow-up. So the inefficiency of the peaking plants is irrelevant. (But building the plant may itself cause pollution.)

Report comment

Report comment

Second, at least some ISOs (like CAISO) pay direct compensation for load-following capacity, much as they do for black start capacity, rather than relying entirely on the LMP market incentives.

Report comment

Report comment

Meh, call it “not as impossible as it sounds” maybe?

I’d love to see papers on utility plans for plant construction as well. It would be the very best evidence.

Yes the real gigantic advantage of solar is that its peak in output logically (and as far as I know empirically) correlates with peak demand. Most of the time. It’s the rest of the time that worries me (say if we start scuttling peaker plants because the sun has been shining very consistently lately, then clouds decide to be unpredictable again and we end up with brownouts).

Wind on the other hand empirically tends to anti-correlate with demand, as crazy as that sounds. At least in Texas anyway:

http://www.ercot.com/gridinfo/generation/windintegration/

They do a report of wind generation over time in various pdf’s. So on the issue of peaker plant construction plans I would expect wind power to have a larger effect (if any effect is observed).

One other issue about the correlation of solar generation and peak demand is the second order effect it will have on peaker plant economics. The spot price of electricity spikes with afternoon demand, and the peaker plants sort of rely on it being much higher than normal prices. It’s already hard for it to be profitable to build a power plant that in extreme cases may only operate 3 hours a day for 14 days every year. I’ve seen instances in Germany where solar ramps up so much that the spot price of electricity actually approaches negative territory. It’s hard to tell if they can deal with that competition.

What we really need is for Commander LeForge to finish whipping up the low cost high capacity storage device he’s working on. Then it’s all gravy.

Report comment

Getting more down-to-earth, I’ve experienced a lot of power outages in the last couple of years: one this morning in a community center where I volunteer, one a couple of weeks ago in my apartment building, and several over the last year or two, including one in my house that lasted three weeks and led to my neighbors marching in the streets. (My entire neighborhood smelled like rotting corpses for a week, because everyone had to throw away the meat that had been in their refrigerators.) Every single one of these blackouts was a result of inadequate distribution capacity. Generation and transmission are able to meet demand, but the distribution infrastructure is underbuilt and poorly maintained. The vast majority of blackouts I’ve experienced in my life, in middle- and upper-income countries, have been results not of grid instability but of failures of local distribution. I also lived through the 2001 rolling blackouts in California, which were failures not of system capacity but rather of politics and dysregulated energy markets. The majority of large-scale blackouts that I have heard about during my lifetime, none of which have I experienced personally, have been results of failures of transmission, not generation or distribution.

To state what I hope is obvious, photovoltaic generation reduces the load on transmission infrastructure and decentralizes power; distributed photovoltaic additionally reduces the load on distribution infrastructure and the impact of a distribution blackout; PV should therefore reduce the risk of transmission, distribution, and political blackouts, which are orders of magnitude more common at present than the grid instability problems you’re worried about due to PV’s poorer load-following capacity. There are people whose job it is to worry about grid instability (a rather surprising number of them, if you aren’t familiar with the industry), they are doing a great job at preventing it, and even if PV turns out to make their jobs harder (as you suggest) rather than easier (as I have argued), it will take a really major change before the new unreliability outweighs the existing unreliability from the other sources I’ve called out above. I say, let them freak out for now, and worry about the problem in five years, at which point the US will have 40 or 50 GW of photovoltaic capacity installed, about 3–5% of the total. That’s the point at which we can start budgeting for new peaker plants if it turns out they’re actually needed at all.

Report comment

The WHO number given on that page for 2008 outdoor air pollution deaths in the USA is 56618, which is more than the 24000 number you think is implausibly high, and also more than deaths due to car accidents. That’s because, in the US, most of the outdoor air pollution is no longer due to coal-fired power plants.

Report comment