June 24, 2019
The article “The World Will Get Half Its Power From Wind, Solar by 2050” by Millicent Dent and Chris Martin optimistically suggests that the world power demand will be more than 50% met by renewables by 2050. While the premise of the article is reasonable, and the predictions acceptable, the authors miss several important points.
First, the cost of wind and solar power is indeed less than coal power in many jurisdictions, but by no means is wind/solar power easily implemented in all jurisdictions. Where the wind is good and the degree of insolation (sunny hours) is excellent, this may be the case. In regions such as the Canadian North, or in regions where the wind is mild, provision of power from renewables is far more challenging than suggested in the article. Undoubtedly, the conclusions in this article might apply well to areas such as the Gulf of St Lawrence littoral, where the winds are highly consistent throughout the year, or to the high plains of Mexico, where the sunshine hours are exceptional. Not so much in Northern Ontario.
Second, the issue with renewable wind and solar power may not be the cost, but its availability and dispatchability. For example, if a local grid has a high proportion of solar power, what can be done when a statistically improbable event such as a week of cloud cover takes place? Where have you stored the energy for this event? Are you tied in to distant grids that provide an “insurance policy”? How expensive is your insurance policy? These are not trivial questions because if you have to have full back-up of the variable and unpredictable renewable power with a system of natural gas turbines (for example), it becomes a very expensive insurance policy against low probability events.
Dispatchability means being available when the grid needs it, as fast as it needs it. Clearly, all variable and intermittent power sources such as wind and solar have poor dispatchability. They may be cheaper than other power sources (hydro, coal, nuclear, natural gas), but they are usually not available when you need the power.
Then, there is the throw-away comment in the article about batteries, implying that batteries will solve any energy storage issues that arise. Right now, the costs of power storage using batteries is four to five times the cost of power storage with pumped hydro or CAES (compressed air energy storage). Furthermore, whatever battery system is to emerge in the future, or if we stay with Li-Ion batteries, the environmental issues of mining the valuable minerals (lithium and the rare earths for example), of the short life of batteries (6-10 years, not an issue for cars, but a major issue for grid-scale batteries), of the recycling costs, and other environmental risks, make batteries a dubious choice for grid-scale storage of renewable energy. Especially if you are trying to store enough energy to get you over those periods of calm wind or no sun. Of course, at some time in the future, all those electric cars we are going to buy will help as storage sites. Batteries are nonetheless critically important for grid operation, but not as the single energy storage system for the grid.
We can integrate far more renewable power into our grids of various sizes, providing we can achieve economic storage capacity. The storage function not only allows us to transfer the power to the period when it is valuable, storage “cleans up” the very irregular nature of wind and solar power, making the variable part into smooth dispatchable energy. Without large-scale energy storage, the idea of 50% renewables in the grids in North America will remain a pipe dream. …and batteries are not the answer for that scale.