Click Here To Get Your Site Sponsored!

News from the Front #30:

Lunatics Demand that BPA Spend $70 Million to Save a Handful of Salmon, BPA Obliges

The tail end of the juvenile fall chinook salmon run is straggling down the Snake and Columbia Rivers.  Over the course of their multiyear journey, roughly half of the adult fish will be caught and killed for human consumption, the highest harvest rate for any "endangered species".  The fish will sell for upwards of $1.50 a pound, or roughly $20 for a whole fish.  

As the fish are migrating downstream, the Bonneville Power Administration is struggling to keep the lights on in the Pacific Northwest.  The problem has been known for a long time:  in 1995, an  eminent electrical consulting firm, Merrill Schultz & Associates, advised BPA that biological opinions issued by the National Marine Fisheries Service constituted "a radical, fundamental transformation of the region's hydroelectric system".  They warned that "if the system really is forced to operate as our model shows in August and September, there is no way that load can be met through purchases in the low-quartile [that is, the 25% driest] years".  But the Clinton/Gore Administration forced BPA to agree to biological opinion operations anyway.

Last week, as conditions got drier and drier, BPA realized it was short by nearly 1,500 megawatts of power, and began a massive program to purchase additional power at prices of up to $250 per megawatt-hour (the average price paid was $200).  All told, BPA will probably spend at least $70 million to buy power in August, and if hot, dry conditions continue, BPA could spend even more in September.  (See Clearing Up, August 28, 2000, at 11)

What does BPA buy for all this money?  It buys the ability to spill water over the top of dams without generating power.  The Imperial Salmon Troopers declare that spill is good, because only 2% of fish that are spilled tend to die, while 5% of the fish that go through turbines die (they say 10%).  By spilling water, they say, we can save fish. And they are right, in theory.

In practice, however, no one seems to be able to measure any benefits from spill, as this data from NMFS, based on monitoring groups of fish downriver during low and high spill conditions,  demonstrates:

Any eighth-grade science student can appreciate what this data means:  there are no measurable survival benefits from spill in the recent, high spill ranges where this data was taken.  Unfortunately, our leaders are apparently too lazy to figure this out or too afraid to challenge the Imperial Forces who find immense significance in such clouds of dots.  We must defer to the "scientists", say the politicians and the judges.  But I digress.

It is possible to measure whether or not fewer fish go into the turbines when you spill more water over the top of the dam, even if you can't measure survival benefits. At The Dalles Dam, for example, they have two sets of measurements.  When BPA spills 64% of the water, 9% of juvenile fall chinook go through the turbines.  When BPA spills 30% of the water, 8% go through.  (Source:  Draft All-H Plan Hydropower Appendix, December 1999, at 133)  Oops, more spill means more fish go through the turbines.  

After a couple years of argument, the Imperial Forces grudgingly allowed spill this year at The Dalles to be reduced from 64% to 40%.  From time to time this year, when BPA can't buy enough power to keep the lights on, BPA has trimmed spill slightly at The Dalles and other dams, triggering storms of protest.  Washington Department of Fish & Wildlife representative Jim Neilsen, for example, declared that his Department could not support reducing spill "for any reason".  (B. Rudolph, "BPA Pulls Rank to Stay Flexible During CA Power Crunch", Clearing Up, August 14, 2000, at 3.)  The Troopers want to spill more, even when the best available data indicates (at least at The Dalles), that it puts more fish through the turbines.  It's not about the fish.  

But The Dalles Dam is atypical.  Usually, spilling more water over the top of dams puts less fish through the turbines.  At least we think so, because we haven't bothered to take the detailed measurements we've taken at The Dalles.  Under the conventional wisdom, at John Day Dam, for example, roughly a third of the juvenile fall chinook are diverted away from the turbines by screens into the bypass system, and the rest must choose between turbines and spill.

How many salmon do we save by spilling more at John Day?  For every 1000 fish approaching the project, roughly two-thirds, or 667, face the spill/turbine choice, so that spilling 10% more water will probably help 67 fish avoid the turbines.  (There are reasons to believe that the first bit of spill gets more than a proportional share of fish, so that higher and higher spill produces less benefits, but we can ignore that complexity.)  If turbine mortality is 5% (no one has measured it at John Day Dam), that saves 3.35 juvenile salmon.  

Roughly 3.5 million juvenile fall chinook make their way down the Columbia River every year, but nearly all of them are long gone in August.  Most days in August, passage at John Day Dam is typically less than 5,000 fish a day.  So we might save 17 juvenile fall chinook at day by spilling 10% more water; and since return rates are roughly 1%, this means we might save 0.17 adult fish a day--and most days it's much less.  That's why the graph above is a cloud of dots:  these benefits are too small to measure.

How much does all this cost?  John Day Dam can generate 2,160 megawatts of power when the river is running at 322,000 cubic feet/second (cfs).  Flows in late August have been running closer to 100,000 cfs.  10% spill is 10,000 cfs, or 10/322*2,160 megawatts, or 67 megawatts of power.  Bonneville has had to pay an average of $200 per megawatt hour to facilitate spill in August.  67 megawatts times $200 per megawatt-hour equals $13,400 per hour, or $321,600 per day.  And the total cost is much more, because BPA is actually spilling from 25,000 to 75,000 cfs at John Day Dam in August, not just 10,000 cfs, and it's spilling at all the dams.  That's why BPA will probably spend $70 million in August on spill--and that's only what BPA will admit to.

Anyway, our cost per fish comes out to be $321,600/0.17, or $1,891,764.  (On the frequent slow days in August, when much less than 5,000 fish are passing John Day, this number rises proportionately.)  Given that many of the fish will be sold for $20 a fish, this is a cost/benefit ratio of 0.00001.  This could be an all-time record for the Federal government--except that in September, we will begin spending millions purchasing power to store water for the Empire's spring flow augmentation program, which has an even worse cost/benefit ratio.  (And remember, at The Dalles, we are probably spending millions a fish to kill the fish.) 

And there are other costs of spill, like air pollution when we buy power made from burning fossil fuels, instead of generating it through clean, renewable hydropower.  Governor Locke recently demonstrated the link between fish spill and air pollution by ordering Washington State air pollution authorities to look the other way so that Avista Power Corporation could burn more gas and keep the power on (for frozen fish storage!).

These are rough, back-of-the envelope calculations.  I could be off by a factor of two, or even ten.  It wouldn't make much difference.  The conclusion would still be the same:  the Imperial Salmon Troopers are nuts.  It makes no sense whatsoever to spend $2,000,000 a fish for the same fish we sell off the back of pickup trucks for $20.  Remember who to blame as your power bills go up, and up, and up.  

© James Buchal, August 31, 2000

You have permission to reprint this article, and are encouraged to do so. The sooner people figure out what's going on, the quicker we'll have more fish in the rivers.