In my last post, I argued that we’re working in the dark when it comes to figuring out how to replace the electricity generated by Washington’s only coal-fired power plant. One key unanswered question is how much of that energy is really necessary to keep the lights on here in the NoLight bulbrthwest.

But let’s ignore that missing piece of information for now and presume—for the sake of this exercise—that you want to replace every dirty electron of coal-fired power that the Centralia plant produces with something cleaner. Your next questions would likely be threefold: 1) What are the alternatives? 2) Can we tap into existing resources? And 3) Do we need to build new generating capacity?

Encouraging people to use energy more wisely is probably the cheapest option, although regional power planners are already counting on conservation to help meet new demand for electricity over the coming decades. Wind or solar installations may become an important piece of the puzzle, but unlike coal plants they tend to generate power intermittently. And many of those renewable energy projects are being developed east of the Cascades and won’t necessarily help keep the electric grid stable on the demand-heavy west side of the mountains, as the Centralia plant does now.

That leaves natural gas plants, which still run on a polluting fossil fuel. But they’re much cleaner than Centralia’s coal burners, which release more greenhouse gases than any other single source in Washington. And state negotiators aiming to cut those emissions are considering ways to help TransAlta, the plant’s owners, line up buyers and financing for a new natural gas plant, and possibly a large one. But we already have gas plants on the west side of the Cascades that are sitting idle during parts of the year.

So couldn’t we just use those existing gas plants to help replace Centralia’s coal power?

Unfortunately, we don’t really know the answer to that key question, since so far, no one at the state level has done a detailed analysis. After spending some time sifting through databases, I came up with a rough cut that begins to shed some light, and leaves me wishing we had much better information.

My analysis (which comes with the many caveats, below) suggests there are months of the year when we might be able to replace Centralia’s coal power without doing much of anything but turning up nearby natural gas plants. And even during peak months—when electricity demand is high and river flows are low and many power sources in the region are running at full tilt—you still might not need as much new generation as you might first assume.

I’ll dive into the numbers shortly, but here’s my real point: the people negotiating on behalf of the public with a multi-billion energy company should be armed with better answers. (Incidentally, the Olympian reported this week that a new Legislative task force will be asked to offer input on reducing Centralia’s use of coal.) Dialing down coal-fired power would be one of the most effective ways to shrink Washington’s carbon footprint, and state officials are absolutely right to zero in on this particular problem. But to craft a plan with the greatest climate benefits and the least burden on Northwest residents, we ought to know how much energy we need, and how much we can wring out of the existing system.

But right now we don’t know those things.

  • To begin getting a handle on those questions, I looked at 2008 electricity generation data compiled by regional power planners, which showed Centralia’s two coal-fired burners produced 1054 average megawatts of electricity during the year.

    Interestingly, it also showed between 1,200 and 1,300 average megawatts of unused capacity at existing combined-cycle natural gas plants in southwest Washington. In other words, those plants could have generated 1,200 average megawatts or more of additional power if they were running all the time rather than sitting idle for parts of the year. That might lead one to think: Problem solved! We’ll just turn off the coal burners tomorrow and fire up these gas plants no problem!

    But it doesn’t work like that, as this month-by-month comparison illustrates:

    gas v coal

    [Larger version of the chart here.] The chart depicts Centralia’s output compared to the excess capacity of natural gas-powered generators in southwest Washington. (I’ve omitted natural gas plants in other parts of the Northwest because there’s some evidence that, owing to transmission bottlenecks, power from other regions might not be able to effectively supplant Centralia’s power generation.)

    As the chart shows, Centralia doesn’t operate at full capacity every month of the year. The coal-fired burners are capable of producing roughly 1,400 average megawatts of power (and do at peak times), but sometimes they’re shut down or undergoing maintenance, or simply not generating for economic reasons. Just so, the output of the gas plants in southwest Washington also varies widely depending on the time of year; they respond to market conditions and get turned when utilities need more power, or when prices rise high enough.

    In April through July—when rivers are running high and flooding the region’s electric grid with hydropower—there’s ample unused capacity in existing gas plants that, at least in theory, could be available to replace Centralia’s coal power at that time. And again in November and December, the unused gas capacity also exceeded Centralia’s coal output.

    But in late winter and summer—when Northwesterners are cranking up the heat or Californians are running their air conditioners—Centralia is running full bore and the gas plants are working hard too. In those months, there isn’t a lot of excess power in the system. And whatever extra electricity the gas plants could have generated wouldn’t have been enough to cover Centralia’s coal output. Depending on the month, there would have been a shortfall of 200 to 600 average megawatts, which suggests we would need some new sources of power if we wanted to replace every electron of Centralia’s coal-fired electricity during those times.

    But this raises yet another important question: Do we really need a plan to replace all 1,400 megawatts of Centralia’s coal power, as the state appears to be seeking to do, or could we get by with a leaner, faster, and more scaled-down approach?

    To truly answer that question, you’d want to do an hour-by-hour comparison on the most power constrained days of the year. And you’d need to factor in transmission bottlenecks or other limiting factors in the electric grid. And you’d need to worry about whether it would make economic sense for these particular gas plants to operate, and whether it’s possible to
    develop incentives for a half dozen of them to work together.

    Lastly, it would also be useful to know how much of Centralia’s electricity is essential to keeping our power grid stable (which you’d definitely want to replace) and how much primarily reaps profit for a private power producer (which the public might not miss so much). Once you factored in all those variables, the graph above might look entirely different. So I’m not actually arguing that these are the correct numbers. What I am arguing is that it’s shortsighted not to have them at all.


    Notes on Sources: This analysis compares the 2008 output and capability of Centralia’s two coal-fired burners with the following natural gas plants located roughly from Olympia to Portland: Centralia’s gas units, Chehalis Generation Facility, Frederickson Power Facility, Mint Farm Generating Station, Port Westward, River Road and the Grays Harbor Energy Center (Satsop), during the 6 months that it operated that year.

    Capability data comes from the Northwest Power and Conservation Council’s Existing Power Plants database. Monthly generation data comes from the EPA’s Clean Air Markets Data and Maps.

    Thanks to Nancy Hirsh at the Northwest Energy Coalition, consultant and economist Jim Lazar and Jeff King of the Northwest Power and Conservation Council for additional information.


    Light bulb photo courtesy of flickr user Caveman 92223 under a Creative Commons License