The Impact of Dams on Natural Flows

Dams have unquestionably altered the natural flow of the Columbia and Snake Rivers. Flow proponents point to changes in the natural “hydrograph”—a graph of average flow over the course of a year. Hydropower production is optimized when the peak on this hydrograph (in the spring) is smoothed, and the water saved in storage reservoirs for release during lower flow times (the fall and winter). Flow proponents argue that the dams, by storing spring runoff for use in generating power in the fall and winter, have an adverse effect on salmon.

Flow theorists have succeeded in destroying a principal benefit of reservoirs: storing spring runoff for fall and winter electric power production. Instead, they require the dam operators to drain the reservoirs in the spring and summer to release water for salmon. In an attempt to “mimic the natural hydrograph”, they require smaller reservoir releases in the fall and winter, to save the water until the spring.

This wastes money that the Bonneville Power Administration could otherwise make selling power generated in the fall and winter. Then, in the spring, there is too much water coming down the river for the turbines to handle. As a result, much of the water goes over the spillways instead of generating power, another source of waste. And in some high-flow years, like 1996-97, too much power is generated all at once in the spring for BPA to sell, yet another source of waste.

One obvious problem with the "changes in the hydrograph killed salmon" theory is that the salmon runs in the worst shape are in the Snake River Basin, where there has not been that much alteration of the shape of the natural hydrograph. The Columbia River runs are in far better shape, yet the alteration of the natural river hydrograph is far greater (mostly because of storage reservoirs in Canada). Here are two graphs that the National Research Council used to show natural vs. regulated flows at the Snake River at its mouth (left) and at the Columbia River at The Dalles (right):

Figure 1: Effects of Reservoir Storage on Hydrographs21
As one can see, the natural flows in the Columbia River (denoted with crosses) had a substantially sharper peak than the present regulated flows (denoted with circles)—but natural flows in the Snake didn’t.

The National Research Council recently noted that it was a “common misperception” in the Pacific Northwest “that there has been a major shift in the mean discharge hydrograph of the Snake River”. “Because there has not been a major shift in the Snake River hydrograph”, the Council continued, “it is doubtful a priori [i.e., even before looking at evidence] that the declines in Snake River stocks are due to or reversible by changes in the seasonality of the flow regime of the Snake River alone”.22

If flow changes were of much importance, we would expect the mid-Columbia stocks to be in much worse shape than the Snake stocks. But upriver salmon runs are in worse shape in the middle Snake River, and healthiest in the mid-Columbia River.

This “misperception” about the significance of flow later came to dominate the thinking of United States District Judge Malcolm Marsh in a series of cases concerning salmon and dams, discussed in Chapter 9. In the first in a series of opinions unwittingly promoting the Great Salmon Hoax, Judge Marsh opined that a “primary” cause of the decline in endangered Snake River salmon is a “failure to meet guidelines, such as water budgets set aside to improve juvenile salmon migration”.23 In a footnote, Judge Marsh lamented that in a recent year, dam operators had only provided 440,000 to 480,000 acre-feet of water despite a fishery agency request for 1,190,000 acre-feet.24 There was no scientific evidence before him, however, that adding such additional water to the Snake River would accomplish anything for salmon.

21 NRC, Upstream at 194 (Prepub. ed.) (Figure 9-1).

22 Id. at 193.

23 PNGC v. Brown, 822 F. Supp. at 1487.

24 Id. n.16.

Previous PageTable Of ContentsNext Page

This Web page was created using a Trial Version of HTML Transit 3.0.