Different Species of Drawdown

There are four main species of drawdown, ranging in severity. Reservoirs have, by engineering design, a variable operational range, from full to minimum operating pool. Minimum operating pool, or MOP for short, is the lowest water level at which the dams were expected to operate. Both fish and commerce require dams, at least in their present configuration, to be run at MOP. The fish passage facilities will not generally work below MOP, and barges will run aground.

As of 1997, the reservoir behind John Day Dam, Lake Umatilla, is the focus of efforts to reduce reservoir levels to MOP. Irrigation interests in the area would prefer to operate at minimum irrigation pool (MIP), which is the lowest level at which they can pump water out of the reservoir without installing new pumping facilities. In Lake Umatilla, MIP is about five feet higher than MOP. New pumping facilities would cost $25 million.

The effects on salmon from reducing the water level by five feet are essentially unknown. We do know that if the entrances to fish ladders are not deep enough, adult passage suffers.¹⁶ There is evidence that the ladders are substantially less effective at MOP than at higher reservoir levels because of reductions in orifice passage efficiency,¹⁷ but fishery managers have never bothered to quantify the adverse effects on adults from MOP operation. Drawdown proponents point to a slight increase in water particle travel time as a supposed benefit to salmon.

Under threats of litigation from the states and tribes, the National Marine Fisheries Service included a recommendation for MOP drawdown of Lake Umatilla in its 1995 Biological Opinion on Federal Columbia River Power System operations. In 1997, NMFS’ Donna Darm admitted that the only reason the MOP drawdown was still on the table was that to withdraw the recommendation would require a lengthy and “unwieldy process”.¹⁸

Promoters of the Great Salmon Hoax, like former Congressman and Northwest Power Planning Council member Mike Kreidler, proclaim the John Day drawdown as “the mainstay of salmon recovery”, and urge its immediate adoption.¹⁹ Kreidler’s continued support for the MOP drawdown provoked his irrigation constituents to label his statements “totally false, irresponsible, and a political embarrassment to the new administration being formed by Governor Locke in Washington State”.²⁰

Most of the pressure, however, particularly at the four dams on the lower Snake River, is for deeper drawdowns. The next level down is “spillway crest” drawdown. Drawing down the dams below the top of the spillway would force all migrating juvenile salmon through turbines. Thus fish advocates have pressed for drawdown to spillway crest as an intermediate step before total dam removal. Spillway crest drawdown is typically 35-50 feet below MOP, depending on the dam. With a spillway crest drawdown, the length of the typical reservoir would be reduced by about one-third, river-borne navigation would cease, and no adults could pass upstream at all without expensive modifications to the fish ladders. Juvenile bypass systems would be completely inoperative because the entrance orifices would be out of water. Power production would not be feasible.²¹

Ken Casavant, the Chairman of the Northwest Power Planning Council’s Fish and Wildlife Committee, persists in promoting the Hoax that spillway crest drawdown “could allow for continued barge transportation through the reservoir”.²² As in many statements offered by leading promoters of the Great Salmon Hoax, there is a tiny kernel of truth to what he says. Before the dam was constructed, there was a seven foot channel used by steam boats and other shallow draft vessels. But none of the commercial barges and tugs operating today could make it upriver through a seven foot channel; they draw up to fourteen feet of water. A huge dredging program, coupled with a complete reconstruction of the locks at John Day Dam, might permit navigation to continue with a spillway crest drawdown. It could cost another half a billion dollars, or more if channels had to be blasted through the solid basalt that forms the riverbed in some places.

To mitigate adverse effects on the economy, spillway crest drawdown is usually discussed as a “seasonal” alternative. The idea is that every year, during some portion of the salmon migration season, the reservoirs would drop down, and then refill later. No one has bothered to make a comprehensive estimate of the total costs of spillway crest drawdown. The U.S. Army Corps estimates the construction costs to modify the four lower Snake River dams as $1.033 billion, with an estimated time to implementation of ten years.²³ That is probably a substantial underestimate.

By 1996, few in the Pacific Northwest other than flow zealots seriously defended the concept of seasonal drawdowns; they defended it only as a tactical step on the way to natural river drawdown. Parties ranging from the National Research Council²⁴ to the state and tribal scientists leading the “Plan for Analyzing and Testing Hypotheses” (PATH),²⁵ all recognized that drawdowns short of “natural river” level—removing the dam, in effect—were unlikely to have much effect on salmon.

Natural-river drawdown involves the most drastic alteration, reducing reservoir levels about 100 feet below MOP.²⁶ There are a variety of engineering schemes for natural-river drawdown, most of which involve digging a channel around or through the existing dams. No one proposes to remove them; they would continue to stand idle as the water rushed by.

As a matter of engineering, a permanent natural river drawdown has the lowest construction costs (other than drawdown to MOP) of the drawdown alternatives, because the dams would be simply decommissioned.

Unlike seasonal drawdown, there would be no need to construct new means of passing fish and commerce by the dams during periods of seasonal refill. Commerce would cease, and the fish would be left to make it around the dams as in natural rapids.

The U.S. Army Corps of Engineers has concluded that the construction costs for bringing the four lower Snake River dams to “natural river” level” would be about $533 million, and the project would take about five years.²⁷ This number is widely believed to be fanciful. Among other things, it was based on multiplying by four the cost of building a channel around a dam built on dirt. Two of the Lower Snake dams are set in sold rock, and it will cost far more to build a channel around them. Compared to the cost of the concrete for building the dams, however, dynamite is cheap.

16 NMFS, Biological Opinion on FCRPS Operations, Mar. 2, 1995, at 114.

17 W. Ebel, pers. comm. (May 5, 1997).

18 B. Rudolph, “NMFS to Army Corps of Engineers: Study Deep Drawdowns”, NW Fishletter, Jan. 21, 1997.

19 M. Kreidler, quoted in “John Day Plan Left Off Budget, Tri-City Herald, Feb. 7, 1997.

20 Memo, Columbia Snake Irrigators Association Board of Directors to Mike Kreidler, Feb. 10, 1997, at 1.

21 U.S. Army Corps of Engineers, “Interim Status Report”, at ES-4 to ES-5.

22 K. Casavant, “New Thinking about Columbia salmon”, Seattle Times, Jan. 30, 1997.