I’m an evidence-based kind of guy, and I’ve been looking for actual measurements of emissions savings for several years. This report initially looked very promising. It works with real measurements on a real grid (actually two different grids) and is the first study I’ve come across that does so. Unfortunately, it does not include the entire grid in its analysis, concentrating on coal and omitting natural gas. It does confirm many of my suspicions and much of my research that wind energy does not produce significant emission savings and in some cases makes them worse. Because of the use of real measurements it is perhaps the most significant study to date on the topic.
Origin of the Report
Bentek is an energy research consultant that seems to specialize in natural gas production. They were engaged by the Independent Petroleum Assoc. of Mountain States to study emissions savings of coal plants under real-world conditions, where they must be continually cycled due to the variability of wind production. The executive summary is available from Bentek with this backup. The full report is 77 pages and 8mb long, and is available from Bentek directly with this backup.
Since the study was sponsored by a fossil fuel organization right away I’m looking for what their interest is. The most obvious agenda would be to discredit wind energy. Most of the negative response so far has in fact been from the wind industry. By the way, I find the wind industry’s complaints about the study pretty weak, especially when they complain about Bentek having only limited data when it is the industry that refuses to release it. In general the study looks to be well-put-together.
Their Real Agenda
If you haven’t studied this topic very much, you might be surprised to learn that this study isn’t anti-wind at all. Let’s review the basics. The natural gas industry loves anything that increases the use of gas, and thus loves wind energy. Only gas (and hydro, which isn’t at issue here) can effectively offset wind’s variability. Why do you think Pickens was so hot to build wind farms? Please tell me you didn’t really think he was worried about saving emissions. Historically gas has been somewhat more expensive than coal, so it was typically limited to “peaker” use, while coal and nuclear provide “baseline”. As the use of wind increases, the use of gas increases. This study was really anti-coal. For my purposes, this is where it falls down. I would have liked to have seen this study performed against gas generators as well as coal. The closest I’ve seen so far to a gas study is the work by Katzenstein and Apt, but that falls short of a real-world examination.
Overview of the Study
The study looked at two areas – Colorado and Texas – where wind energy has been introduced into the grid and used CMS emissions data (as in actually measured going up the stack!) correlated with generated output from wind and coal plants to arrive at the general conclusion that SO2 and NO2 emissions increased while CO2 emissions were unchanged due to the arrival of wind energy. They did this by studying “wind events” where an increase of wind generation could reasonably be shown to cause a decrease in coal generation. They then looked at the emissions data to see if the extra emissions caused by changing the coal generation would be made up by the lower coal generation while the wind was generating. They go into more details with many more charts, but the following chart is representative and serves as a good summary of what they found.
You would expect when the generation (the area in blue) went down that the emissions (the three lines) would also go down. CO2 looks like it may decrease a little,but certainly not in proportion to the decrease in generation, plus see below. SO2 and NO2 both increase. SO2 – the stuff of acid rain – increases by quite a bit and stays at an elevated level waiting for the scrubbers to recalibrate.
They also ran two basic models in both locations with different time-frames in an attempt to establish high and low bounds for the emission savings. The chart above shows the best case for co2 savings. Regardless of the assumptions, the co2 savings/increases were bounded within a +/- 1% range.
They then figured out how many of these events happened in a year and calculated the annual savings. Regardless of the location or the underlying assumptions, CO2 savings were essentially non-existent, while SO2 and NO2 emissions increased.
Comments from Bentek
An employee of Bentek was kind enough to send me some comments, which I’m pleased to include here.
Thank you for your review of our study. I wanted to point out one thing to you: Gas-Fired combined cycle facilities are used as “base load” generation in many areas. While they are still above coal and nukes on the dispatch curve, the way they are run would define them as either “base load” or “shoulder load”. These units are much more efficient than gas-fired combustion turbine facilities, which are used to meet peak load.
Also, during 2009 the crash in gas prices due to vast supply increases caused for the economics of generation to shift in some regions (the Northeast and Southeast). This resulted in gas fired combined cycle generation becoming cheaper than coal fired generation, referred to as coal-to-gas switching in the industry. As prices are expected to be suppressed for the near term, this fuel switching should continue to take place.
As I wrote back to him – it’s nice that wind is helping to force the move from coal to gas, but why bother with wind at all?
This study pretty much settles the issue of the effects of wind on emissions, at least for coal. It is the first study I’ve come across that uses real numbers off real grids, and as such it represents an important part of the evidence that continues to build the case for wind energy not producing any non-trivial CO2 savings.
It also leads into some interesting thinking about what combination of production methods would ever lead to a savings in emissions due to wind. But that’s a topic for another day.