I’m finalising my PhD thesis, so forgive me for the inactivity. But something caught my eye this week that I thought would be worth a short blog post.
An “independent report” – the BBC’s description – for the lobby group RenewableUK has got some media attention for its claims about reductions to fossil fuel imports due to wind farms. In essence, it is claiming that fossil fuel imports have been reduced significantly, with the bill being lowered by £579 million in 2013.
The report, however, takes some excessively simplistic assumptions about the electricity sources displaced by wind farms, claiming that without wind farms there would have needed to be 16.5 TWh extra electricity generated by coal power plants and 11.9 TWh from gas power plants. These numbers seem to be based purely on the proportion of Britain’s electricity that came from gas and coal. But this is not how the thing would work in reality.
When a wind farm is generating electricity it pushes the most expensive source of electricity of the grid. In wonkish language this is called the marginal source of electricity. However, RenewableUK’s analysis assumes that the marginal source of electricity in 2013 was coal more often than it was gas, which simply wasn’t the case. This can be shown by looking up prices.
Alternatively, you can look at actual generation data. If wind farms are displacing gas or coal, then all things being equal, higher wind power should equal lower gas or lower coal. And we can see from actual generation data that higher wind power equals lower gas.
The figure below shows instantaneous output from wind farms versus gas power plants. The blue line shows a simple linear regression relating gas power to wind power.
The trend is relatively clear. More wind power equals less gas power, all things being equal.
What about coal? The figure for wind versus coal is shown below.
This indicates the opposite. Higher wind power coincides with higher coal power output. So, little evidence that wind farms displace coal.
However, the astute reader will notice a problem with the above. It tends to be windier in winter, when demand is higher. So, perhaps it is better to do a model which factors in demand.
My original model – the blue lines in the above figures – was of the form,
coal power = m * wind power + c.
Let’s try the following instead,
coal power =m * wind power + n * demand + c.
This should largely factor in the fact that coal and gas power are higher in winter simply because demand is higher. Likewise, it will also factor in the fact that its windier in winter.
So, what does this equation look like? What we really care about is the value of m. For gas, m is -0.98. In other words, an increase of 1 GW of wind power means gas power is reduced by 1 GW (but see caveat at the bottom). For coal, m is -0.066. So, a 1 GW increase in wind power will reduce coal power by around 0.07 GW. Wind therefore appears to displace much more gas than coal.
This is, of course, a relatively simple statistical analysis. A more elaborate one would likely give slightly different results. However, it makes it very clear that it is wrong to assume that wind farms displace the typical fuel mix on the electricity grid. Clearly wind farms displace more gas than coal, and RenewableUK’s assumptions are therefore wrong.
This should be reflected by any claims about the fuel or emissions savings of wind farms, at least for historical accounts. Fuel prices change, and higher coal, cheaper gas or higher carbon prices may result in coal being the marginal fuel in future. But in recent years it seems that wind farms have more or less replaced nothing but gas.
Caveat and notes on data
I pulled historical data in from the website Grid Watch, which gives instantaneous electricity data every 5 minutes. However, its data does not show all wind farms in Britain. Some are not visible to the grid. The actual numbers I calculated above ought to be re-jigged slightly to account for this. However, that should not alter the conclusion about whether wind farms are primarily displacing coal or gas.