Last time I started to unpack the inherently dismal performance and underlying economics of wind power in Alberta. The whole reason Alberta has started this dysfunctional relationship with wind is because it has a dream that wind will somehow replace coal for electricity generation. This dream is shared by wind advocates across the globe. Can it ever achieve reality?
Recall last time I showed each grid-connected Alberta wind farm’s output over the period October 24 2018–June 25 2019. I presented the data in the form of non-exceedance curves for each wind farm. I repeat that figure below. (If the individual plots are too small you can enlarge them; they won’t pixelate.)
All the curves have the same general shape: they all slope steadily upward from left to right. They all look pretty much the same. Ho-hum, kind of boring.
Now, let’s look at the non-exceedance curves for each of the sixteen Alberta coal plants, over the same period.
Notice how the coal curves are much different in shape than the wind ones? In particular, look at the plots for the Genessee, Keephills, and Sheerness facilities—those are the ones that, going by the capacity factor (CF) figure in the plot titles, saw most of the action over the sample period. Note the location of the dashed median lines in the plots for those facilities. Compare those positions with the median line positions in every single one of the wind plots.1
That difference in the shape of the curves and the location of the median lines is, in a nutshell, why wind absolutely will not replace coal in Alberta. The location of the median lines in the coal plots is determined by the will of the system operator, instructing the operators of the individual coal plants at time 
to put 
megawatts of power into the system at or by time 
. The coal plant operator receives the order from the system operator, and throttles the boiler accordingly—if the system operator wants more power, then more pulverized coal is fed to the firebox.
The location of the median lines in the wind plots, on the other hand, is a function of the inherent intermittence of wind. There’s nothing the wind farm operator can do to change that. Wind blows when wind blows. How much does it blow in Alberta? Look at the wind plots. Wind farm electrical generation is a rough proxy for the fluid dynamics of wind at that location over a period. The non-exceedance curves in the individual plots give probabilities, based on the output values of the sample. That’s how much the wind blew over the roughly 344,000 minutes October 24 2018–June 25 2019 in each of the locations where the Alberta wind farms are.
Wind advocates’ answer to the impossibility of matching wind output to system demand on a timely basis is typically twofold:
- Scale up the number of wind farms so they collectively produce as much energy over a year as the coal plants collectively do.
- In periods where wind output exceeds demand, store the surplus in batteries. Use the stored energy to cover the shortfall during periods where wind output falls short of demand. That way, non-exceedance curves of the combination of wind plus battery output would, in the aggregate, resemble those of those coal plants that saw the real action over the period.
The first is a political non-starter. Wind turbines occupy around 28 hectares per installed megawatt (Meredith Angwin in her excellent Campaigning for Clean Air arrives at 34.5 hectares, which is likely more reliable; I’ll stick with 28 because all my calculations are based on it and I’m too lazy to change). To replace coal with an 80:20 mix of gas and wind would require over 1,000 square kilometers of territory for just the wind turbines—that’s greater than the amount of land distrurbed by oilsands operations. To completely replace coal’s annual output with wind would require plastering Alberta with 5,000 km
of wind turbines (more if Meredith’s number is the right one). That won’t happen.
And the second part of the twofold wind advocate answer to matching wind to demand—using batteries to store surplus wind output—is even more risible. I’ll explain why next post. For now, recognize that grid-scale battery storage capacity costs around CAN$278 per kilowatt-hour.
NOTE: The recent Alberta provincial election that saw Jason Kenney replace Rachel Notley as premier puts Alberta in line to become one of the federal “backstopped” provinces—i.e. one of the provinces that does not have a carbon tax that is to the liking of the current federal government. All of the aforementioned Alberta coal plants meet the criteria for a “covered facility” under the Output Based Pricing System (OBPS)—i.e., they all emit more than 50,000 tons of CO2 per year. Genessee Unit 1 threw more than 30 times that amount into the air just in the roughly 8-month period studied here. Every coal unit with a mean output greater than 10 MW broke the 50,000-ton limit in the 8 months.2 That’s every plant except Sundance 3 and 5. It will be interesting to see how or whether the Alberta coal plants get tagged with the federal backstop tax.
- Recall from the last article that the median lines in these curves are at the output (x-axis) value above or below which there’s a fifty-fifty probability the farm will produce at any given time.
- Mean MW
5,735 hours (344,112 minutes 
60 minutes per hour 
5,735 hours) megawatt-hours over the period. Multiply that figure by 800 grams per kilowatt-hour (the approximate average CIPK of Alberta coal-fired power generation) to find total grams of CO2; divide by 1,000,000 to obtain tons.)
Steve – thanks for the article series. I’m looking forward to part 3.
Sadly, none of this is news to anyone who’s ever looked at daily and annual demand curves.
What’s downright depressing is that you can explain this to Greens, and they dismiss it out of hand. Green isn’t a political or energy position, it’s a religion. We can hope that the inherent contradiction between “climate change is a global threat” and “we must rely on renewables even if they are unreliable” will kick some out of their romantic delusions, but I fear that the majority of them are beyond reach of arguments based on mere information.
As of today, seems even less likely that wind will replace coal, as they are restoring the energy-only market. “A key difference between the markets is how electricity generators are paid.
In energy-only markets, electricity generators are paid only for the electricity they produce in real time when electricity is required.
In capacity markets, electricity generators are paid for their overall ability to produce electricity, in addition to the electricity they produce and sell in real time.
Government will table the necessary legislation and amend regulations to stop the implementation of the capacity market as soon as possible.” https://www.alberta.ca/release.cfm?xID=642387D0ECA3E-ED8E-6B02-885D35312EBBB3EE
Thanks Todd. I’m an Albertan who hasn’t been keeping up with what my government is doing. It’s good to see that we haven’t fallen for any of the market distortions that have been tried in other places. Maybe we could even set up a sensible plan to reduce carbon emissions while maintaining reliable electricity.
For wind turbines “land use” is not the same as other generation methods like coal because that figure incorporates the distance needed between turbines. The land can still be dual used for farming for example. The actual footprint is just the turbine base and minimal infrastructure. That’s actually what makes it an attractive option as you can develop wind farms as a secondary source of revenue for land owners. They can either own the turbine and sell the power or lease their land. From that perspective the land usage is not an issue. Alberta alone has 23 million acres (93 million sq km) of farmland. Also, thete is still innovation happening in this area. They are building larger wind turbines that produce more energy so that footprint is actually decreasing.
Eddie, it’s good to focus on that portion – required land area. It is the most arguable point of this data. The rest is somewhat difficult to refute.
To dumb it down further. They are stating that we would need a whole lot of windmills.
And if having built all those windmills, there is no cost effective storage method for storing said windmill electricity to be used during peak periods of demand. There would also be large periods of overproduction of said windmill electricity that would have to be stored when capacity is not required.
I know that they are wind turbines, and not windmills, but its more romantic to think of Europe and the legend of Don Quixote with the backdrop of windmills then the modern giant airplane propellers of today’s world.
Who cares what the area/windmill is at the end of day as that is essentially only a function of the effectiveness of the current machinery.
The points are:
The large number of windmills required. Who really wants to see these everywhere in large quantities?
Windmills cannot supply electricity when peak demand is.
There could be days or weeks with little or no wind power.
Without backup storage of windmill electricity, which is cost prohibitive, to add additional capacity – how can this occur?
this scaling problem is not insignificant, as you point out. For me, storage cost, astronomical though it is, is secondary to land use when we start considering the sheer numbers of machines required just to meet electricity demand.
And the problem just gets worse. Add heating demand, then add transportation energy demand, to electricity demand. That’s one super-important grid, as if it isn’t super important already. My prediction is, long before the latter expansions occur people will have figured out what a joke it is to suggest that 100 percent RE is viable.
So your argument boils down to the number of devices needed. I don’t see that as a very strong argument for why it’s not viable. I would actually say the reverse is true.
In fact decentralization of power generation would increase reliability and resiliency. As would offering a mix of renewal energy power sources.
Imagine the extreme case where each individual can generate electricity with solar panels or wind turbines and store or feed excess production into a smart grid. In this scenario you would never worry about a fault with an individual device. This is far more reliable than depending on a single power supplier.
This strategy is also more flexible as increasing capacity can be done in smaller increments rather than having to invest in mega power plants that take years to plan and construct. Shorter time frames also allows you to take advantage of the latest innovation in the technology. Turbines and solar panels are continuing to improve every year.
Storage is an issue that can also be addressed in numerous ways.
Besides centralized batteries provided by the utility, leverage batteries people will have in their electric cars. Pay people for their extra unused capacity when needed.
Besides chemical batteries there are other gravity based options like pumped hydro systems and similar ideas that store energy as potential energy.
Other options include molten salts and compressed gases in mines, etc.
Storage is not limited to lithium batteries only and is an area ripe for innovation. This is where we can create new industries if we treat it as an opportunity.
Although Alberta can readily replace coal with wind, storage and gas on an 80:20 basis, there is a much better strategy.
Sitting right next to Alberta is BC, which is 95% hydro, with a lot of that behind dams with a high storage capacity, or run of river fed by dams higher up in the river system. The right approach is to build a few GW of transmission links between the two, then use BC hydro and Alberta wind to complement each other.
Alberta has similar average wind speeds to Texas, where capacity factors of new wind farms with larger turbines are typically over 50%. Though central West Texas was over 50% even back in 2012, based on figures I have seen. So wind power in Alberta is pretty cheap at around $37/MWh. I assume this is in US not Canadian dollars? So Alberta wind must provide cheaper power than new gas plants, and possibly even than existing Alberta gas plants.
Hydro is too valuable to use as base load power. Alberta should configure enough wind capacity to replace coal, but also to export wind power at windy times to BC, allowing BC to save hydro water. When the wind isn’t blowing in Alberta, BC should ramp up hydro production, using the water previously saved, and export power back to Alberta. BC has a high hydro capacity to average demand ratio.
There’s a potential issue if Alberta wants BC hydro power at the time when both Alberta and BC have evening peak demand. And here grid battery storage can help out. With a few hours of battery storage, power for the evening peak could be stored in advance in grid batteries in both BC and Alberta. Then the hydro capacity is only constrained by something like average hydro demand over the peak 24 hours, not peak hour hydro demand.
There may still need to be some additional gas generation built in Alberta, but much less. And the load factor of that gas generation should end up far lower than 20%.
In this approach, the storage of water in BC behind hydro dams with large capacity is being tapped to provide the large quantities of storage Alberta needs to firm up wind power. A scheme like this is already in operation between Norwegian hydro and Danish wind (including offshore wind).
It needs a grid simulation to sort out the details, but if the data is available that would not take too long. I’m not volunteering, although I’ve simulated an all-renewable grid for Texas. Stephen already has the Alberta wind data, so could easily do it, though it also needs BC + Alberta demand and BC hydro information. If not, I know a BC hydro operator who has modelled the US grid and just might be interested in something nearer home. That’s assuming no-one has already modelled it.
like you Table A-1 showing Ontario energy consumption and Co2 produced. Can you do one like that for Alberta?
I could, I have the data, might take a while to get it into this site. I show hourly aggregates on emissiontrak.com, including for AB, but it’s graphical not tabular. thanks for the prompt, I’ll try to get a snapshot into it.