It isn't what you asked, but on islands where they often import diesel for power generation, solar is probably cheaper across the board. Even if it only offsets fuel and not the capital cost of a diesel plant, imported fuel is stupid expensive. The arguments that solar is cheaper than natural gas (or anything else) seem to center around retail costs of energy. So an argument might be that a homeowner can install rooftop solar for a cost of $4/watt. With a broad assumption of a thirty year lifetime and average production of 15%, and ignoring the investment cost, they can produce energy for about $0.10/kWh, a favorable price in most of the US when we're at a customer meter. Where solar falls short is in broader system planning. There are minimum capacity requirements. Utilities, whether investor owned or non-profit municipals, must be able to meet load demands. Solar has limited ability to meet peak load demands, so a natural gas plant is needed anyway. Since the power into the grid has to equal the power out, something controllable is necessary. Again that falls to natural gas. When we get up onto the wholesale side of things, typical energy prices are in the mid-$20/MWh. Xcel Energy recently finished a 100 MW solar plant in Minnesota at a cost of $180M. Making the same assumptions as before, we have a cost of $45/MWh ($0.045/kWh). On the wholesale side it's less competitive, though its immunity to fuel cost fluctuations are a pro. But again it lacks capacity and controllability. So my take is one can make arguments based on fact that will show solar is cheaper than natural gas, but those arguments have a limited scope and are deliberately structured to give the desired result. That doesn't mean solar is bad; its cost reductions have been impressive. But there are challenges to continued growth. On the topic of cost reductions, much of the savings has been in panel costs. It seems to be to the point where labor is the biggest cost. That explains part of the gap in cost between a utility scale plant like Xcel's (labor efficiencies) and rooftop solar.
Isn't this really the core problem that needs to be solved in any discussion of power? We have plenty of sources for generating power, but we have very few effective ways to store power for use later. The entire system is designed to support peak loads and generate power on demand, rather than storing potential energy in some way so we can pull it out when need it. Isn't that really the core problem that needs solving? I keep visualizing two huge spheres connected with a tube, and a turbine in the tube. When the sun is out, power is used to suck the air our of one sphere, creating a vacuum. When we need to pull that stored potential energy back out of the sphere, you release the air back into the other sphere (which is now the vacuum), spinning the turbine, and generating power. (Using solar power to create the vacuum means it doesn't matter what it costs, or how efficient it is, as long as the power to run the compressor motor does not exceed the power generated by the solar panels.) So: Isn't STORAGE the problem? Not generating capacity? Since the power into the grid has to equal the power out, something controllable is necessary.
Storage could be a solution for sure. Compressed air has been discussed, and I think there are a couple real examples. There's a Wikipedia article on it. A couple projects are mentioned in the History section. The issue I understand with compressed air is heat. It gets hot when compressed, and that heat energy is lost. Then when decompressed, heat needs to be added. In a way, storage can be thought of as generation capacity. The system today works on intermittent energy supply (e.g. coal train deliveries a few times a week) that can be stored (pile of coal) and converted to electric energy on demand (by burning it in a boiler). Doing the same with wind or solar (intermittent energy supply), pumped into storage (whether air, batteries or hydro dams) and converted on demand (storage output) essentially provides the same functionality.
I guess my implied intent is to reduce the impact of our power generation on the environment. If a majority of homes had solar panels trickling energy into the grid, and the utility then stored that extra power for later use, the need for coal drops. That eliminates transport costs, environmental damage from mining and burning, etc, etc, etc.
That can work on paper, and the technology exists to do it (compressed air, pumped hydro, batteries). The challenge is the cost. It's surprisingly cheap to dig up coal and burn it for energy. There are a million terrible side effects of fossil fuels, but faced with them consumers almost exclusively will pick the cheaper option.
As long as the general public continues to pay for all the side effects of the "cheaper" option. It's only "cheaper" because the majority of the costs are subsidized by the public. Make coal producers pay for the FULL cost of the product - probably with a carbon tax - and I think the math pencils out very differently.
My general caution with energy is people often talk past each other. One side will say "(blank) is currently impractical and wildly expensive" while the other side will say "no you're wrong, long term forecasts expect the cost to come down with technology improvements likely to expand functionality." Both sides miss the time scale the other is using. I'm probably just as guilty of this. Utilities tend to be in the first camp. Nobody wants serial number 1. It's too much risk for an industry that demands extremely high uptime.
I don't know much about McKibben, but Googling him I see he's a part of 350.org. Metric, a band I like, had 350.org at shows in the lobby a year or so ago. Just stealing the line from his Wikipedia article: I agree with that entirely. If the goal is carbon reductions, nuclear can play an important role at the expense of nuclear waste and risks. But many of the people actively concerned about carbon emissions are similarly concerned about nuclear waste and risks. And my thought on that is holding firm on no nuclear directly results in more coal and natural gas use. When we were being taught multiplication and division, I remember being shown a film (may have been actual film; this was about 1988) with a skit where someone was hired to cut a 10" piece of candy into three, 3" pieces. And if they wasted any, they'd be fired. In the skit, the factory said they just couldn't find someone to do this simple task. So the employee cut three pieces and ate the leftover 1". I think of that when I see calls for low or no carbon wind and solar energy supplies. While coming from a good place, it's impractical (at today's costs) and ends up with a very flawed system meeting nobody's goals.McKibben has been quoted as saying that he personally believes increased use of nuclear power is necessary to reduce carbon emissions, yet he is reluctant to publicly promote nuclear energy because such a position “would split this movement in half”.
Founder of 350.org. He's got about a chapter and a half in this book (starts on pp. 57) where his argument is that nuclear is nice and all, but much of the economies it gains are due to subsidies and externalities. He quotes a study that puts the real-world capital expenditure cost of new nuclear plants at 17 to 22 cents per kWH. Not only that, but from a real world standpoint you have to consider the community response and regulatory antagonism nuclear power faces which push nuclear reactors years late and over budget."Bottom line: building enough conventional nuclear reactors to eliminate a tenth of the threat of global warming would cost about $8 trillion, not to mention running electricy prices through the roof. You'd need to open a new reactor every two weeks for the next forty years and, as the analyst Joe Romm points out, you'd have to open ten new Yucca Mountains to store the waste. Meanwhile, uranium prices have gone up by a factor of six this decade because we're - you guessed it - running out of the easy to find stuff and miners are having to dig deeper."