May 13, 2004

Rodney McInnis
Acting Regional Administrator
NOAA Fisheries, Southwest Region
501 West Ocean Boulevard, Suite 4200
Long Beach, CA 90802-4213

By letter dated April 1, 2004, NOAA Fisheries transmitted to the Klamath Basin Area Office of the Bureau of Reclamation a revised incidental take statement (ITS) for NOAA Fisheries’ May 31, 2002 biological opinion (BO) regarding impacts of the Klamath Project on coho salmon. This letter provides the comments and objections of the Klamath Water Users Association (Association) to the revised ITS.

As you know, the Association and its members are vitally interested in the application of the Endangered Species Act (ESA) to the Klamath Project. We have sought at every opportunity to be heard. We have made known our desire to be involved and provide input to the agencies and have repeatedly asked for meaningful consideration of our views.

Against this background, the appearance of the revised ITS is startling. We had no notice that it was under preparation. We were not afforded opportunity to provide any input. We were not furnished with a draft ITS for comment. Our community has in the past suffered from unjustified regulation that ignored our comments and concerns, and we had hoped this could not occur again. However, the ITS simply appeared, sending the message that Reclamation and the affected communities will live with what NOAA Fisheries dictates, without dialogue. This is unacceptable.

Notwithstanding NOAA Fisheries’ avoidance of the Association’s concerns or interests, we are here providing our comments and objections to the revised ITS. As described below, the revised ITS is fatally flawed from a legal perspective, and biased and arbitrary from a technical perspective. A number of shortcomings perpetuate those in the 2002 BO. These deficiencies should be cured in a new ITS and any future BO.

In general, the ITS identifies a "surrogate" method for defining the amount or extent of incidental take. The surrogate is the "maintenance of certain flow-related habitat conditions." If these flows are maintained, incidental take is deemed to be "within the acceptable range." If they are not, the assumed "take" is assumed to be unacceptable. In addition to identifying these flow conditions, the ITS prescribes specific " Reasonable and Prudent Measures" (RPMs) to minimize take resulting from the action, and "terms and conditions" to implement the RPMs.

The ITS assumes "take" is acceptable if certain flow conditions are maintained. It utterly fails to explain how take is determined. In turn, this failure implicates two major deficiencies.

First, the ITS avoids any genuine determination of take at all. Nowhere is there any articulation of whether or how some incremental change in flow (say, from 1200 to 900 cfs) supports a conclusion that take or increased take occurs or is likely. This point is reinforced in our discussion of technical issues, below.

Second, there is no discussion or analysis that explains what flow condition is being compared to some other condition in order to determine whether and what take may occur. Stated another way, there is no identified baseline for the determination of take.

The ITS assumes some level of take from the proposed action and RPA. Necessarily, in the absence of the "action," there would be no take resulting from the action. To determine take would require comparison of "no action" to the action. The ITS does not, however, make such a comparison or even try to do so. For example, the ITS refers to reduced rearing habitat, elevated water temperature, and increased smolt run timing. The obvious question is: "reduced," "elevated," or "increased" as compared to what? There is no answer to be found in the ITS.

The Association has repeatedly pointed to this type of deficiency in NOAA Fisheries’ documents and the Association has repeatedly been ignored. In our May 23, 2002 comments on the April 25, 2002 draft BO for operation of the Klamath Project, we emphasized the need for a coherent logic in determining the effects of an action. While focusing primarily on determinations for purposes of section 7(a)(2) of the ESA, the discussion is equally applicable to determining the take associated with an action. The determination of effects requires an evaluation of the action against an environmental baseline. 50 C.F.R. § 402.02. So, too, does the determination of take require an evaluation of the action against a baseline, or no-take, condition.

In its February 25, 2002 biological assessment (BA), Reclamation described the proposed action as including diversion of water to direct use, diversion of water to storage, and use of previously stored water. It supplemented the action with a water bank to ensure certain flows. NOAA Fisheries has never defined the action, and has avoided any approach that actually seeks to determine the effects of the action. Notably, Reclamation’s BA properly concludes that use of previously-stored water has no effect on the species because it results in no change as compared to the environmental baseline. NOAA Fisheries has ignored this issue altogether. It continues to regard the ESA as a vehicle to mandate flows, which it manifestly is not. In the end, the ITS results in an attempt to prescribe flow conditions. This is far beyond the authority to impose RPMs and terms and conditions, as described below.

The ITS contains RPMs and terms and conditions to implement the RPMs. The majority of such provisions are unlawful and outside NOAA Fisheries’ authority, as described below.

First, RPMs and terms and conditions must be limited to minimizing the incidental take of the action (and any RPA). 16 U.S.C. § 1536(b)(4)(B), (C)(ii); 50 C.F.R. § 402.14. Numerous provisions of the ITS relate to reducing purported take of activities that have nothing to do with the operation of the Klamath Project. These include, but are not necessarily limited to, provisions requiring augmentation of water supplies and provisions related to diversions outside the Klamath Project.

Second, obviously, there must actually be take to be minimized by the RPMs and terms and conditions. As described above, there is no defensible analytical framework in the ITS even for determining whether or how take occurs under the action. Similarly, there is no basis to conclude that discharge from the Straits Drain causes take or that recycling of Straits Drain water would minimize take. See also Arizona Cattle Growers Association v. Fish and Wildlife Service, 273 F.3d 1229, 1244, 1246 (9^th Cir. 2001).

Third, even if they were otherwise proper, the RPMs and terms and conditions exceed the authority provided in the ESA. RPMs, and terms and conditions, "cannot alter the basic design, location, scope, duration, or timing of the action, and may involve only minor changes." 50 C.F.R. § 402.14(i)(2). The "surrogate" for take (flow levels) is, in effect and in reality, specified as a (binding) RPM or term and condition. It fundamentally changes the action and is improper.

The terms and conditions actually labeled as terms and conditions additionally exceed the limitations on the scope of RPMs and terms and conditions. They do not involve "minor changes" in the action. Rather, they impermissibly require the undertaking of entirely new activities.

In addition to the fundamental, fatal defects described above, the ITS is laden with speculation and ambiguity, and lacks objectivity. It is particularly illustrative to review the sources cited in the ITS, and consider the sources not cited. We find extensive citation to the draft "Hardy Phase II" Report (even though NOAA Fisheries acknowledges this draft Report is undergoing revision) and a draft California Department of Fish and Game Report. Tellingly there is no reference to the "draft" undepleted flow study completed by the Bureau of Reclamation. Nor is there reference to any number of reports or studies that lead to different conclusions than the ITS. Shockingly, there is no reference whatever to the final report of the National Research Council of the National Academies (NRC Committee Report). The NRC Committee Report, of course, was commissioned by the Departments of Interior and Commerce. Its dismissal by NOAA Fisheries clearly signals that the Report does not serve the policy objectives of the authors of the ITS. The Association strenuously objects to this omission.

Below, we address specific subjects discussed in the ITS.

The ITS states that a "reduction" in spring mainstem flows during the months of March through June is expected to result in take of young-of-the-year and yearling coho that are attempting to rear in the mainstem Klamath River by decreasing the area of edgewater habitat downstream of Iron Gate Dam (IGD). Consistent with our previous comments provided to NOAA Fisheries on the topic, the NRC Committee stated in its April 30^th letter to Dr. Hogarth:

and

and

In addition, the NRC Committee, in its final October 2003 report, concludes "rearing of coho in the Klamath main stem is much less important than rearing of coho in tributaries, which are the preferred rearing habitat of coho."

Despite this information, the ITS (like the 2002 BO) continues to speculate on the potential benefits to coho fry derived from edge habitat in the Klamath River. There are no data to suggest that edge habitat in the mainstem Klamath River is limiting the coho population. In fact, if NOAA Fisheries had properly used the results of the modeling exercise from the draft Hardy Phase II report, it would have shown that existing Klamath River edge habitat is more than adequate for coho fry in the mainstem. Additionally, based on information presented to the NRC Committee during its March 2003 meeting in Medford, Oregon, the fishery agencies have been assuming that chinook salmon fry can serve as a surrogate for coho fry for modeling rearing habitat. Even if this assumption was true, fry have much more habitat available in river channels than that surmised by the BO. Mr. David Vogel (Natural Resource Scientists, Inc.) provided this information to the NRC committee (and NOAA Fisheries staff in attendance) at the Medford meeting based on his extensive research and observations on salmon fry rearing in mainstem river channels. He also showed the committee underwater video footage demonstrating this fact.

As we have described in extensive detail in our previous comments to NOAA Fisheries, the available information demonstrates that the mainstem Klamath River in the river reach most affected by the Klamath Project is not a principal factor limiting coho populations. Here again, NOAA Fisheries has ignored this highly relevant information and the NRC Committee’s response and, instead, suggests that somehow, through unknown mechanisms, coho fry rearing habitat is limiting the populations.

The ITS and BO that precedes it do not describe why maximizing coho fry rearing habitat in the mainstem immediately downstream of IGD is necessary. Using a hypothetical example, if one coho fry already has 1000 sq. ft. of habitat available and needs only 100 sq. ft. to survive well, why does the fry need 2000 sq. ft. of habitat? Additionally, the BO and ITS do not describe if (and if so, how) coho fry rearing habitat in the mainstem reach most affected by Iron Gate Dam is limiting the SONCC ESU. This is particularly relevant when one considers the vast proportions of coho spawners, fry, juveniles, and their habitats that are found elsewhere in the Klamath Basin and in other drainages (i.e., coastal streams between Cape Blanco, Oregon, and Punta Gorda, California).

The ITS states that, between March and June, "shallow edgewater and side-channel habitat provides young fish refuge from excessive flow velocities, while overhanging riparian vegetation protects juveniles from predation and contributes insect prey favored by young salmonids." Many of the desirable characteristics for juvenile salmonid rearing habitat are also desirable areas for predatory fish. The ITS falsely presumes that young coho can only escape predation through the provision of edge habitat. In fact, coho rearing areas are available in many other areas of the mainstem and in the tributaries.

While the ITS notes that "the complex interaction between juvenile coho survival and mainstem Klamath River temperatures is not fully understood at this time", it nevertheless concludes that "flow releases lower than those outlined within the RPA are not likely to offer the same level of relief and could potentially result in increased coho take." The ITS assumption that Iron Gate flows—particularly in "critical summer months"—will impart positive temperature benefits to coho salmon is completely speculative.

There is no evidence to indicate that increasing upper Klamath reservoir releases during late summer or early fall would benefit salmon. In fact, because of a variety of meteorological, physical, and biological reasons, artificially increasing flows at that time would probably be harmful. This is due to the fact that IGD discharges are unsuitably warm for salmon through early September. The gradual declining temperatures in the Klamath River downstream of IGD during the fall are primarily attributable to normal seasonal declines in ambient air temperatures, not river flow. The NRC Committee found that higher summer flows from IGD increase minimum temperatures and may increase the temperature of cold water refuge areas in the mainstem. The NRC Committee also discussed strategies to reduce the temperature of IGD releases and concluded:

Vogel and Marine (1994) and Vogel (2002) recommended that any increased flows from IGD, pulsed or otherwise, to benefit adult salmon should occur during late September or early October to coincide with normal seasonal declines in air temperatures and concomitant cooler river flows.

Increased Smolt Run Timing

The ITS assumes a spring flow/coho smolt survival relationship. NOAA Fisheries has failed to acknowledge that there are no data to demonstrate that this factor is limiting the production of coho in the Klamath River. The assumption cited in the NOAA Fisheries BO was largely derived from studies of salmon migrating through the large Columbia River reservoirs, not the free-flowing Klamath River. Additionally, the BO suggested that the draft Hardy Phase II report supports the NOAA Fisheries premise on this topic. This assumption is incorrect because the draft Hardy Phase II report used a computer modeling exercise to predict theoretical habitats such as rearing and spawning, not outmigration habitat or outmigration survival.

The NRC Committee has also spoken on this matter. The April 30, 2002 letter from the NRC Committee to NOAA Fisheries states:

In its final October 2002 report, the NRC Committee noted: "…mainstem flow may directly affect the coho population at the time of downstream migration of smolts…there is some probability of benefit for the smolts to be derived from minimum flows at the time of smolt migration, as expressed in the NOAA Fisheries biological opinion of 2002."

The NRC Committee’s statements are consistent with our previous technical comments provided to NOAA Fisheries. Here again, NOAA Fisheries has ignored this highly relevant information and the NAS Committee’s response. Instead, NOAA Fisheries suggests that somehow, through unknown and un-quantifiable mechanisms, spring flows are limiting coho populations.

The ITS discusses "transit" between IGD and Seiad Valley. It does not, however, discuss whether there are even significant numbers of listed coho smolts in this reach. Nor does it recognize the annual variability in flows that will occur with or without any RPA.

The NRC Committee urged "adaptive management principles could be applied to this issue". Using increased smolt run timing as a justification for using flow variability as surrogate indicator of potential take provides yet another opportunity for advocates of high main stem flows to allege a potential ESA violation in times of low flows at Iron Gate Dam. Increasing the rigidity for minimum flows at Iron Gate Dam is not consistent with an adaptive management philosophy. Again—we reiterate what the NRC Committee stated in its October 2003 report:

NOAA Fisheries admits that estimating potential coho salmon stranding and the resultant incidental take is "difficult". Nevertheless, the ITS concludes that juvenile coho stranding will increase if IGD flows are ramped down faster than those outlined within the RPA. According to the ITS, yearling and fry coho salmon "could" become stranded in isolated pools and side channels as flows are reduced below IGD. As is apparent throughout the ITS document—and the BO upon which it is based—NOAA Fisheries fails to separate facts from assumptions. We are concerned about the widespread use of vague or generalized statements like these in the ITS. Accurate identification, description, and analysis are required to justify these statements.

NOAA Fisheries must clearly justify that they have clear evidence to support that Klamath Project operations have directly resulted in fish stranding below Iron Gate Dam. In May 2002 media accounts, for example, environmentalists quickly blamed fish strandings below IGD on water diversions to farmers. U.S. Interior Department officials ultimately disputed the reports as erroneous, pointing to an unusual snow/rainstorm that they say contributed to the river’s sudden rise and descent.

The ITS claims that "low IGD flows during the mid-August through October period may result in take of adult coho salmon through delayed upstream migration. Any delay in the natural migration rate could leave adult coho more susceptible to estuarine predators…while increasing the risk of mortality associated with density-dependent diseases… ." (emphasis added). The final NRC Committee report, however, questioned the degree of influence of the Klamath Project on the estuary, located 200 miles downstream:

NOAA Fisheries in its ITS relies on its 2002 BO findings to conclude it "does not anticipate take of adult coho salmon at flows equal to and above 1,000 cfs". The 2002 BO incorporated the inconsistent logic applied to this issue in the draft Hardy Phase II study. In that study, the authors excluded a recommended flow regime less than 1,000 cfs "under any circumstance". On the one hand, NOAA Fisheries theorizes that somehow mimicking the unimpaired hydrograph downstream of the Iron Gate Dam will create ideal conditions (for highly ambiguous reasons). But on the other hand, it concludes that good conditions for fish can only occur at the high end of the flow scale (never at the low end of the scale) and therefore increased flows must be released from artificially created storage reservoirs. Further, the assumption that a particular flow for mid-August through October is a necessary precaution is simplistic and arbitrary. It is clear that a variety of factors interact, in changing ways, through this period. Critical among them are air temperature and the timing of runs. An inflexible minimum throughout this period may be detrimental.

The ITS should be realistic and based on the current situation on the Klamath River and its tributaries and the opportunities for sustaining biological productivity as it may be affected by mainstem river flows.

The ITS is fundamentally flawed for the reasons stated above. These deficiencies should be cured in a new ITS and any future BO.

Sincerely,

Dan Keppen
Executive Director

cc: U.S. Senator Ron Wyden

U.S. Senator Gordon Smith

U.S. Rep. Wally Herger

U.S. Rep. John Doolittle

U.S. Rep. Greg Walden

U.S. Rep. Mike Thompson

Sue Ellen Wooldridge, DOI

Dave Sabo, USBR

Kirk Rodgers, USBR

Irma Lagomarsino, NOAA Fisheries

Steve Thompson, USFWS