One of wind power's presumed strongest suits is the CO2 savings. When reading industry pamphlets, they generally assume that every mw-h of wind will displace a mw-h of coal. Almost needless to say, this presents the savings in the best possible light, roughly a metric tonne of savings per mw-h. But the actual savings depends greatly on what is being displaced. A wind farm in Norway makes no CO2 sense, as they use hydro almost exclusively. What is the situation in Ontario? The IESO makes complicated decisions, balancing economic, technical and political criteria about what sort of generation to make use of. But could we make use of the historical record to see how much variance there is in the controllable supplies and infer from that what the expected savings might be?

So I picked three days at random and ran a standard deviation on coal, hydro and other (i.e. Lennox) to see how much they varied during the day, figuring whatever criteria the IESO used (and would presumably continue to use) would be embodied in those numbers. Running a std dev on the fourth source, nuclear, can be done, but nuclear is the baseline, and almost always is running very close to capacity. The results? Coal represents about 40% of the variance, while hydro and "other" each represent about 30%. Of course it varied from day to day, but not by so much that I thought my three-day sample was too far from a complete study.

If we accept my logic and my numbers, each mw-h of wind would save ((.900*.4)+(.0*.3)+(.360*.3))= .468 tonnes (opg's coal figures + hydro + open-cycle gas figures). This is roughly in line with various European values.

By the way, my first attempt at figuring out the co2 savings was to look at times when the wind output rose or fell sharply and see if I could correlate that with a change in one of the other sources. It turns out that the current wind contribution is so small that it disppears into the seemingly random fluctuations of the other three sources.