Report on the Status of Salmon and Steelhead in the Columbia River Basin - 1995
This report was funded by the Bonneville Power Administration (BPA), U.S. Department of Energy, as part of BPA's program to protect, mitigate, and enhance fish and wildlife affected by the development and operation of hydroelectric facilities on the Columbia River and its tributaries. The views in this report are the author's and do not necessarily represent the views of BPA.
Duane A. Anderson, Pacific States Marine Fisheries Commission
Ray Beamesderfer, Oregon Department of Fish and Wildlife
Bob Woodard, Washington Department of Fish and Wildlife
Mike Rowe, Shoshone-Bannock Tribes
Gary Christofferson, Pacific States Marine Fisheries Commission
Jerome Hansen, Idaho Department of Fish and Game
Northwest Power Planning Council
851 S.W. Sixth Avenue, Suite 1100
Portland, OR 97204-1348
BPA Project Number 88-108-04
Contract Number 95BI65130For hard copies of this report, write to: Bonneville Power Administration Public Information Center - CKPS-1 P.O. Box 3621 Portland, OR 97208 Please include title, author, and DOE/BP number 65130-1 in the request.
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
ABBREVIATIONS AND ACRONYMS USED IN THIS REPORT
Information on fish populations, fisheries, and fish habitat is crucial to the success of ongoing programs to protect, recover, enhance, and manage fish resources in the Columbia River Basin. However, pertinent data is often difficult to locate because it is scattered among many agencies and is often unpublished. The goal of this annual report is to bring many diverse data types and sources into a single comprehensive report on the status of anadromous fish runs in the Columbia River Basin and the environmental conditions that may affect that status. Brief summaries are provided to identify the type and scope of available information. This synopsis is intended to complement other more detailed reports to which readers are referred for comprehensive treatment of specific subjects.
This first report focuses mainly on anadromous salmon and steelhead (primarily through 1994) but we intend to expand the scope of future issues to include resident species. This is the first of what we intend to be an annual report. We welcome constructive suggestions for improvement.
In this report, we identify available information but make no attempt to evaluate its implications. Inclusion does not represent endorsement of methods or results. When applying the information, it is incumbent on the reader to understand the limitations of the data imposed by the method of collection and related assumptions. We do attempt to flag controversial issues. Most of the summary data in this report was generated using the StreamNet (formerly Coordinated Information System and Northwest Environmental Data Base) database system. The StreamNet Distributed System (DS) is a PC based database application containing fully referenced data and a user friendly interface to query, report, or export the data. Contents of the DS are shown in Appendix A (to receive a copy, contact Duane Anderson at 503-595-3100). As with any summary information derived from a database, this report represents conditions to the best of our knowledge. No warranty for the correctness, accuracy, or usefulness of this data is expressed or implied. If errors or inaccuracies are discovered please contact any of the author's of the report. Data in the report that came from sources other than the StreamNet database are cited in the bibliography.
This report is a product of the StreamNet project which is funded by the Bonneville Power Administration (BPA), U.S. Department of Energy, as part of BPA's program to protect, mitigate, and enhance fish and wildlife affected by the development and operation of hydroelectric facilities on the Columbia River and its tributaries. The project is called for in the Fish and Wildlife Program of the Northwest Power Planning Council. The project's objective is to promote exchange and dissemination of information in a standardized electronic format throughout the basin. This project is administered by the Pacific States Marine Fisheries Commission with active participation by tribal, state, and federal fish and wildlife agencies.
To facilitate the presentation of large amounts of data in this report, the Columbia River Basin has been divided into four regions. Included in each region are both the mainstem Columbia and/or Snake rivers and their adjoining tributary systems. The regions are defined as follows: Below Bonneville - the Columbia River and its tributaries below Bonneville Dam; Bonneville to Priest Rapids - the Columbia River and its tributaries between Bonneville Dam and Priest Rapids Dam; Snake - the Snake River and its tributaries up to Hells Canyon Dam; and Priest Rapids to Chief Joseph - the Columbia River and its tributaries between Priest Rapids Dam and Chief Joseph Dam.
Since 1938, the minimum number of salmon and steelhead, including jacks, entering the Columbia has ranged from a high of 3.2 million fish in 1986 to a new low of 856,500 fish in 1994 (Figure 1)1994 Columbia River commercial landings were the second lowest in history (Figure 1).
Following a significant increase in the early 1980's, total escapement to the various Columbia River regions has been in decline since 1986 (Figure 2). Total escapement to Columbia River Basin was just over 700,000 adults and jacks in 1994. Adult and jack escapement to regions above Bonneville Dam comprised only about 441,000 fish of that total. Total escapement of spring chinook was the lowest in recorded history (Figure 3).
Adult upstream survival rates have been estimated for various stretches of the system by members of the U.S. vs. Oregon Technical Advisory Committee (Table 1), personal communication, Ray Beamesderfer, report in progress). Using dam counts, harvest estimates, and estimates of numbers of fish returning to tributaries between dams, it is possible to estimate adult survival between dams. These survival rates are also known as conversion rates and they vary substantially from year to year and have a significant impact on adult escapement. For Snake River spring chinook stocks, the adult conversion rate for the river section from Bonneville Dam to Lower Granite Dam has averaged only about 60% since 1979.
Comprehensive estimates of the total number of natural spawners are not available in the Columbia River Basin at this time. The reasons for this are both institutional (different management agencies have differing monitoring and estimation techniques), and geographic (spawning in the Columbia River Basin occurring in thousands of stream miles for long time periods making it unfeasible to count all of the spawners in all of the streams). Data for many individual populations are available, however, and are shown in the example population section of this report.
While estimates of total spawners are not consistent between management agencies or stocks of fish, these estimates hold the best promise for monitoring overall natural spawning trends. Idaho Department of Fish and Game and Oregon Department of Fish and Wildlife, for example, have been counting spring chinook redds in Snake River drainage subbasins since the late 1950's (Figure 4). In 1994, index areas in 44 streams were surveyed averaging only 3 redds per stream, the lowest level since surveys began.
As the StreamNet data compilation process becomes more sophisticated, it may prove advisable to establish key indicator streams throughout the Basin that would be surveyed in a pre-defined, consistent manner in order to provide more accurate estimates of spawning populations, life stage survival, and production basin-wide.
The majority of hatchery rack returns occur in the region below Bonneville Dam (Figure 5). This is predictable given that the majority of the Basin's hatcheries are in this region, as are the majority of hatchery releases.
Hatchery returns by species and run are shown in Figure 6. Coho comprise the majority of returns in most years but have declined substantially in the last five years. Fall chinook make up the next most abundant returns followed by spring chinook and summer steelhead.
Detailed information on migrant abundance, condition, and behavior is provided by the Smolt Monitoring Program (SMP) whose primary objective is to provide up-to-date information for management of a water budget and spill agreement. The SMP is administered by the Fish Passage Center and conducted by federal agencies, state agencies, and Indian tribes (FPC 1995). Chinook salmon and hatchery-reared juveniles comprised the majority of the almost 1.5 million migrants sampled at various sites in 1994 (Tables 2 and 3).
A passage index of juvenile salmonid abundance is estimated based on number collected in juvenile diversion systems at dams with a correction for the proportion of flow which is spilled rather than passed through the powerhouse. (This index thus represents an underestimate of total juvenile abundance because it does not account for fish that pass through the turbines.) Annual passage indices varied substantially among years, species, and dams (Figure 7) depending on hatchery releases, natural production, number transported, and survival rates of migrants.
Peak migration periods are May through June for steelhead, sockeye, coho, and age 1 chinook juveniles and June through July for age 0 chinook juveniles. Migration timing at most sites was roughly comparable between 1994 with the average for the previous three years (Figure 8).
Migration speed is often considered an index of survival rate with shorter travel times corresponding to higher survival rates although the strength of this relationship is a question of considerable debate. Smolt travel time is closely correlated with water particle travel time. Water particle travel time in the Columbia and Snake river mainstems increased with dam construction which has increased the cross-sectional area of the river and decreased flow. Average water particle travel time through the Snake River has increased ten-fold since 1962 while average discharge has been reduced by less than half (Figure 9). Travel times for summer migrants such as subyearling chinook are typically longer than those of spring migrants including yearling chinook and steelhead (Table 4). Travel times also vary between hatchery and wild fish and seasonally in relation to changes in flow and degree of smoltification (FPC 1995).4. Fish Passage Efficiency
Fish passage efficiency or FPE refers to the proportion of juvenile migrants which pass a dam by means other than turbines. Passage mortality is generally thought to be reduced by increasing passage efficiency to avoid turbines which impose an approximate 10-15% mortality rate per dam. Passage efficiency is improved by increasing the proportion of river flow which is passed over spillways and by increasing the proportion of fish which are diverted from turbines by bypass systems such as submersible traveling screens. Fish guidance efficiency (FGE) refers to the proportion of migrants which pass via the powerhouse but are diverted from turbines by the bypass systems. Passage efficiency and guidance efficiency are affected by stage of smoltification which varies seasonally. Benefits of improving FPE with high spill rates are controversial because of the resulting gas supersaturation which can also be lethal to fish. Passage efficiencies in 1994 generally fell below the 70 or 80% levels typically recommended by state, tribal, and federal fishery management agencies (Figure 10).
In an attempt to avoid passage mortality through the mainstem Columbia and Snake rivers, over 15 million juvenile salmon and steelhead were collected in 1993 (Figure 11) at Lower Granite, Little Goose, Lower Monumental, and McNary dams and transported by barge or truck to release sites downstream from Bonneville Dam (Hurson et al. 1995). Fish are collected out of turbine intake bypass systems where they are diverted with screens. Transportation began on an experimental basis in 1968 and has been conducted by the U. S. Army Corps of Engineers since 1981 (BPA et al. 1994, Harmon et al. 1995). Comparisons of the relative number of transported and non-transported marked fish observed in fisheries, hatcheries, and other sample sites have been used as an index of transportation benefits (Harmon et al. 1995) but interpretation of this information is extremely controversial (Mundy et al. 1994).
During 1993 and 1994, the National Marine Fisheries Service (NMFS) and the University of Washington (UW) tested methods for estimating survival probabilities of individual yearling chinook salmon and steelhead in the Snake River using passive integrated transponder (PIT) tags (Table 5), Iwamoto et al. 1994; Muir et al. 1995). These probabilities (Table 6) are related but not equivalent to survival rates (K. Steinhorst, University of Idaho, unpublished). Survival estimates remain controversial with unresolved questions related to the validity of statistical assumptions; inferences to other river reaches, river conditions, and portions of the outmigration including saltwater entry; and impacts of associated fish capture and handling.
Minimum survival estimates were also produced by radiotelemetry studies on hatchery chinook salmon juveniles used in evaluation of the transportation program (Schreck et al. 1994). Up to 70% of radio-tagged smolts transported from Lower Granite Dam and released downstream from Bonneville Dam were detected 160 km downstream from the release site. These estimates are conservative because not all fish retain tags and not all tags are detected.
Predation by northern squawfish is a significant problem for migrating salmon and steelhead juveniles. Efforts to control northern squawfish by fishing have been underway in the Columbia and Snake mainstems since 1990 (Willis and Young 1995). Squawfish are harvested by agency employees who electrofish, angle, and gillnet at dams and hatchery release sites, and by recreational anglers who are paid rewards for each squawfish turned in to check stations. In 1994, 14 check stations were operated 7 days a week from May 1 through September 25 (Figure 12). Registered anglers logged 40,800 days of effort and averaged 3.2 fish per day (Smith et al. 1995). Over 700,000 fish were removed from 1990 through 1994. In 1994, exploitation rate (% of the population harvested) increased to 13% from a program low of 9% in 1993 (Table 7). The 1994 exploitation rate was within the 10-20% annual goal for the program.
Using adult abundance information we looked at general trends in escapement (adult returns) throughout the region. Only those trends with at least 15 years of data having at least one data point in the 1990's were used. We divided the average of the last five years of each trend by the average of the first five years of the trend to get the trend value. Values near one indicated little change, less than one indicated declines in abundance, and greater than one indicated increases in abundance. We performed two analyses, one for indicators of natural spawning escapement and one for hatchery rack escapement. The results of the natural trends are shown in Figure 13. The results of this exercise are fairly consistent for all chinook stocks, with those in the lower and mid Columbia regions doing better than those in the Snake. Summer steelhead, however, are doing slightly better in the Snake than in the Lower Columbia. Coho populations below Bonneville (not shown) were nearly all in decline (95% in the <0.75 category).
Hatchery stocks exhibit similar behavior to natural stocks (Figure 14). Chinook stocks are generally performing better in the Columbia River as compared to the Snake River, while steelhead stocks exhibit the converse pattern.
Salmon and steelhead travel great distances during the ocean phases of their life history. Generalized ocean migration patterns are shown in Figure 15, Figure 16, and Figure 17 (CDFO).
Salmon and steelhead stocks utilize thousands of miles of streams throughout the Columbia River Basin (Table 8). Chinook and summer steelhead stocks generally inhabit all major portions of the Basin that are currently accessible (Figure 18, Figure 19, and Figure 20). Winter steelhead and coho are confined primarily to areas below Bonneville Dam (Figure21 and Figure 22), although some remnant coho populations do exist in the Columbia Basin above Priest Rapids dam.
The Stock Summary Reports (Hymer et al. 1992, Kiefer et al. 1992, Olsen et al. 1992) identify 287 populations of anadromous fish within the Columbia River Basin (Table 9). These populations are discreet species/run groups of fish that inhabit a particular drainage basin and do not necessarily represent genetically unique stocks. Natural (including wild) runs make up 45% of these stocks while the remaining 55% are hatchery or mixed runs. Detailed run timing information is shown in Table 10.
Hydroelectric and other purpose dam development in the Columbia River Basin has been widespread and ongoing for over 100 years. There are at least 145 hydropower dams in the basin and over 900 other purpose dams greater than 10 ft. in height, (Table 11).
Dam development in the Columbia River Basin has affected anadromous fish production in a variety of ways including: 1) complete loss of upstream habitat due to blockage, 2) direct mortality caused by the dams to both downstream and upstream migrants, and 3) indirect mortality caused by alteration of the environment (change in flow patterns and travel time, etc.).
Grand Coulee Dam on the Columbia (completed in 1941) and Hells Canyon Dam on the Snake (completed in 1967) completely blocked fish passage. There are currently 13 passage mainstem dams operated in Basin at this time (Table 12). The only truly free flowing section of the Columbia River that remains above Bonneville Dam is the Hanford Reach.
Hydro-development, and the increased storage capacity that resulted from it, as well as irrigation has had a significant impact on seasonal flows in the Columbia River Basin. While the annual average flow of the Columbia has not changed significantly, spring and summer flows have been reduced (Figure 23) and winter flows increased. This reduction in spring and summer flows has been aggravated by lower than normal run-off in 8 of the last 10 years.
Reservoir storage capacity in the Columbia River Basin has reached over 100 million acre feet (Figure 24). Total storage available increased over 50% in one year alone (1973) with the completion of Libby Dam in Montana, Dworshak dam in Idaho, and the Mica Dam in British Columbia. Mica dam has an incredible storage capacity of over 23 million acre feet.
Table 13 shows daily average total project flows and percent spill for the spring period (April 16 to May 31 for all projects except the four lower Columbia dams (Bonneville, The Dalles, John Day, and McNary) and May 1 - June 15 for those projects). Summer flows and percent spill are shown in Table 14. The summer period was defined as June 1 through July 31st for upriver projects and June 16th through August 31st for the four lower Columbia dams.
Average project flow and spill rates vary dramatically from year to year based on run-off conditions, operational mandates, and storage capacity in the Columbia River Basin. We averaged the flow and spill at the dams existing in a given year between the Columbia River mouth and Snake River spawning grounds for spring and summer time periods (Figure 25). Both spring and summer flow and spill levels have decreased significantly since the early 1960's. Flow reduction has been significantly influenced by the increased storage capacity developed in the Columbia River Basin during that period (Figure 24), more than doubling from about 50 million acre feet in 1961 to over 105 million acre feet in 1995. Spill reductions have been influenced by reduced flows and by increased powerhouse capacity from additional turbine installations made at the mainstem dams.
The average spring and summer spill levels of the mainstem dams between the Columbia River mouth and Snake River spawning grounds (expressed as a percentage of average total flow) declined through the 1960's and 1970's. Spill levels have steadily increased since 1988 (Figure 26), though they are still far below those of the early 1960's.
Freshwater habitat for anadromous fish in the Columbia River Basin has been severely depleted by hydroelectric development. For the U. S. portion of the Columbia River Basin, we calculate that over 18,700 miles of historically accessible streams have been blocked by hydroelectric dams (based on 1:250,000 digital line file). This represents nearly 38% of the estimated historical range of 49,300 miles. The allocation of this habitat loss varies widely throughout the Basin with the Snake River area sustaining the largest loss (Figure 27). Currently accessible habitat is approximately 30,600 miles (Figure 29) although a little over half of that habitat is actually in use by anadromous fish at this time (Figure 28).
As the figures indicate, while the Snake River Region has sustained the largest historical loss of habitat due to hydro development, it still represents the majority of available habitat in the U.S. portion of the Columbia River Basin. Note that Grand Coulee Dam blocked significant historic production areas in Canada. These losses are not reflected in the figures below.
In the late 1980's, agencies and tribes participating in the Northwest Power Planning Council's subbasin planning process subjectively rated habitat quality for all currently utilized salmon and steelhead production areas in the Columbia River Basin. Habitat conditions were rated based on relative comparisons of the present fish producing potential of habitat within a given subbasin (not based on comparisons of habitat to that in other subbasins.) Excellent habitat was defined as that which would support the highest productivity for a species within a subbasin. Poor was the classification for habitat which would support the lowest level of productivity. Good and fair were used to describe habitats that were intermediate relative to the other two categories (NWPPC 1989). We summarized these habitat ratings for salmon and steelhead by Columbia River Region. Figure 30 shows the results for salmon. The area below Bonneville was the only region where more than half of the salmon habitat was rated as good or excellent.
Results for steelhead are shown in Figure 31. While the contrast between the Columbia River and Snake River regions for salmon was not that great, the picture is different for steelhead. While below Bonneville and the Snake regions were found to possess 60% or greater excellent or good habitat, only 25% of the Bonneville to Priest Rapids regions received excellent or good ratings.
While hydro development and harvest have played major roles in the decline of Columbia River salmon and steelhead, habitat degradation has also been significant. Sedimentation problems linked to poor land use practices, are prevalent throughout the Basin (Table 15). High instream temperatures, loss of large woody debris, degraded instream and streambank conditions, loss of habitat due to channelization, and low flow levels have also reduced the productivity of many of the Basin's streams.4. Habitat Changes
McIntosh et al. (1994) identified changes in fish habitat over a 50-year period in selected Columbia Basin tributaries by comparing the frequency of large pools and coarse woody debris from two time periods based on surveys in 1934-42 and 1990-92 (Table 16). The frequency of large pools increased in managed and unmanaged watersheds of the mid-Columbia River, with the increase twice as great in unmanaged watersheds. Large pool frequency declined in managed watersheds of the Snake River, except for the Tucannon River where there was a significant increase. Coarse woody debris was generally more common in unmanaged than in managed areas. Large pools were identified as key rearing habitat for juveniles and resting habitat for adults. Coarse woody debris creates and maintains high-quality fish habitat by providing cover, enhancing pool development, and reducing erosion.
These data suggest a reduction in damaging land use practices in the mid-Columbia watersheds during the last 50 years and continuing effects of more recent activities in the Snake watersheds.
A large-scale program to install new fish screens on unscreened irrigation diversions and to upgrade or replace existing fish screens has been under way since 1991 in an attempt to improve survival of juvenile salmon and steelhead in Columbia and Snake river tributaries upstream from Bonneville Dam (FSOC 1996). With funding from the Federal Mitchell Act and from the Bonneville Power Administration, 163 screens have been constructed in 1991-94 to National Marine Fisheries Service's current design criteria. By 2002, fish screens will be installed at over 300 unscreened diversions and 602 old fish screens will be replaced or upgraded. To reduce fish passage delay and fish screen operation and maintenance costs, irrigation ditches are also being consolidated or replaced with pumps or groundwater wells. Finally, this program also is screening irrigation pump intakes and is upgrading fish ladders of tributary obstructions.
Table 17. Gravity diversion screens constructed or replaced in Columbia River Basin tributaries, 1985-1994 (Hawkes, Columbia Basin Fish and Wildlife Authority, personal communication). Totals include sites eliminated by consolidation or conversion to ground water. Totals do not include intake pump screens or fishways constructed or replaced by this project.
Ocean conditions are highly variable and can have a significant impact on production and survival of anadromous fish. Coastal upwelling conditions are generally thought to influence early ocean smolt survival with higher levels of upwelling associated with more favorable fish conditions. Figure 32 shows the mean March - September upwelling anomaly for sites off the Oregon and Washington coasts. Values greater than 0 represent greater than average levels of upwelling while negative values represent less than average. In 7 of the last 10 years upwelling has been below normal.
Another commonly accepted measure of ocean conditions is the Southern Oscillation Index-SOI which is related to El Niño events. The El Niño/Southern Oscillation (ENSO) phenomenon is an atmosphere-ocean coupling across the central tropical Pacific Ocean which influences climate in many regions of the Earth. Values of the SOI that are less than minus one are generally thought to be related to El Niño events; the lower the SOI, the stronger the event. Much of the North American continent is influenced to some extent by the ENSO phenomenon and fish production in the Pacific is also affected. Again, the SOI indicates that ocean conditions have been less than optimal in the majority of years since 1977 (Figure 33).
Another general indicator of ocean conditions is sea surface temperature (Figure 34). Again, the data is highly variable but shows a general tendency for increasing temperatures in southerly coastal areas, a condition generally recognized as being unfavorable to the production and growth of anadromous species which inhabit these areas.
Hatchery releases have been widespread and numerous in the Columbia River Basin. Only a few watersheds in the Columbia which have not received hatchery plants since 1975 (Figures 37-42). Notable areas that have not received plants during that time include the John Day Basin and portions of the Salmon River Basin. Hatcheries releasing fish in the Columbia Basin since 1980 are listed in Table 18 by management agency.
Average annual hatchery releases into the Columbia River Basin have exceeded 185 million fish since 1980. During this time, the region below Bonneville Dam has received, on average, about 60% of all hatchery plants. The Bonneville to Priest Rapids Region of the basin has received about 26%, the Snake River Region about 9%, and the area above Priest Rapids has received about 4% of all hatchery releases (Figure 35). Releases in 1994 were about 25% lower than releases in the years 1990 through 1992.
Since 1980, fall chinook account for the majority of hatchery releases (Figure 36), with 53% of the total releases, followed by coho with 22% of the releases, spring chinook with 18%, and summer steelhead with 5%. Releases of coho in 1994 were down more than 40% from 1990 levels while spring chinook and summer steelhead releases in 1994 were both down more than 33% from totals in 1990.
Figure 37. Hatchery spring/summer chinook releases by USGS Cataloging Units for the Columbia Basin (PSMFC 1995, based on data from the RMPC).
Figure 38. Hatchery fall chinook releases by USGS Cataloging Units for the Columbia Basin (PSMFC 1995, based on data from the RMPC).
Figure 39. Hatchery summer steelhead releases by USGS Cataloging Units for the Columbia Basin (PSMFC 1995, based on data from the RMPC).
Figure 40. Hatchery winter steelhead releases by USGS Cataloging Units for the Columbia Basin (PSMFC 1995, based on data from the RMPC).
Figure 41. Hatchery coho releases by USGS Cataloging Units for the Columbia Basin (PSMFC 1995, based on data from the RMPC).
Figure 42. Hatchery sockeye releases by USGS Cataloging Units for the Columbia Basin (PSMFC 1995, based on data from the RMPC).C. Hatchery Authorization and Funding
The majority of hatchery releases in the Columbia River Basin have been authorized and funded either through federally mandated programs like the Mitchell Act and the Lower Snake River Compensation Program, or as part of specific dam mitigation programs. These two broad categories of authorization have been responsible for about 93% of the total hatchery releases since 1980 (Figure 43).
Records of coded wire tag (CWT) recoveries are a useful index of the incidence of straying. Straying of hatchery fish varies by stock (Table 19, Table 20, Table 21). Individual fish from some stocks such as Hood River chinook in the lower Columbia have been recovered as far away as the Snake River in Washington.
Columbia River harvest is divided into 3 broad categories; sport, commercial, and tribal. Sport harvest is typically constrained to the lower river and estuary. Non-sport harvest is regulated by 6 defined areas or zones on the river; Zones 1 through 5 define the area from the mouth to Bonneville Dam and are typically reserved for non-Indian commercial fisheries. Zone 6 is defined as the area from Bonneville Dam to McNary Dam and is typically reserved for Indian harvest.
Since 1960, harvest peaked in 1986 when 1.6 million fish were taken (Figure 44). Since 1986, harvest has dropped dramatically.
In most years, coho and fall chinook comprise the majority of harvest (Figure 45). Since 1960, coho harvest has averaged 36% of the total while fall chinook has averaged 35%. Steelhead harvest represents about 16% of the total, but the number of steelhead harvested has been increasing for the past 10 years. Spring/summer chinook harvest has averaged 9% of the total since 1960, but only 2% of the total harvest since 1974.
Average total harvest in the mainstem Columbia was around 550,000 in the 1960's and 1970's, rose to around 720,000 in the 1980's, and has declined to about 440,000 so far in the 1990's (Figure 46).
The allocation of that harvest has changed dramatically (Figure 46). The proportion of Columbia River harvest attributed to commercial fishing (zones 1-5) has declined from over 80% in the 1960's to 40% in the 1990's. Sport harvest has increased five-fold from 7% in the 1960's to 37% in the 1990's. Tribal harvest increased markedly from the 1960's to the 1970's (10%-21%) but has remained about constant since that time.
Most harvest in tributaries is attributed to sport fishing. Sport harvest in the tributary systems is concentrated in subbasins below Bonneville Dam (Figure 47). Since 1975, sport harvest below Bonneville Dam has comprised, on average, nearly 71% of the total.
Species composition of the sport harvest is shown in Figure 48. Summer and winter steelhead comprise, on average, nearly 65% of the sport harvest in the Columbia River Basin. Spring chinook average about 24% of the catch while coho and fall chinook comprise about 7 and 3% respectively.
Ocean salmon harvest is regulated by the respective states within 3 miles of shore and by the Pacific Fishery Management Council (PFMC) from 3 to 200 miles from shore (PFMC 1995). PFMC divides the Washington, Oregon, and California coast into four management areas of which only the northern-most, Cape Falcon to the Canada border, affects Columbia River stocks (Figure 49). Landings and fishing effort between Cape Falcon and Canada were much reduced in 1994 over previous years (Figure 50). Some Columbia River stocks are also intercepted in Canadian and Alaskan fisheries (Table 22). Ocean exploitation rates in recent U. S. and Canadian ocean fisheries average 46-58% for tule chinook stocks, 24-39% for upriver bright chinook stocks, 24% for Willamette spring chinook, and less than 5% for upriver spring and summer chinook (Table 23)
Economic values of salmon fisheries can be described by prices paid to commercial fishers for their landings (exvessel value) and total personal income associated with fisheries. Exvessel values in 1994 were only 24% of the 1981-93 average for combined commercial fisheries for salmon and steelhead in the U.S. controlled portion of the ocean and in the Columbia River (Table 24). Personal income values in 1994 were only 30% of the 1986-93 average for combined U.S. ocean salmon fisheries (Table 25).
BPA is the primary parties involved with mitigation activities in the Columbia River Basin. BPA's fish protection, restoration, and enhancement projects in the Basin have totaled nearly $370 million from 1981-1993 with funding distributed throughout the Basin (Figure 51).
The types of projects funded and the amount spent have changed dramatically since 1990 (Figure 52). Total spending for 1995 is over $80 million dollars on over 200 projects.
The U.S. Army Corps of Engineers is another major player in anadromous fish mitigation activities in the Columbia Basin. Primary activities funded by the Corps include project modifications aimed at improving juvenile and adult fish passage, hatcheries (Lower Snake River Compensation Program), research, spillway modifications, and juvenile fish transportation. Through fiscal year 1987, the Corps has spent nearly $545 million in the Columbia Basin on fish mitigation measures (Figure 53, Mighetto 1994).
Research activities have been funded by the Corps since 1953 with total research expenditures exceeding $63 million through 1993 (Figure 54, Mighetto 1994).
Corps expenditures for the operation, maintenance, and research operations for Lower Snake River Compensation Program (LSRCP) hatcheries currently exceed $12 million per year (Figure 55).
The Mitchell Act, passed by Congress in 1938, funds state and federal hatcheries on the lower Columbia River. Its objective was to offset the impacts to fish resulting from the construction of Bonneville and Grand Coulee Dams, as well as the effects of logging and pollution (Mighetto 1994). Funds are also used to pay for large irrigation diversion screening programs. The Columbia River Fisheries Development Program (CRFDP) was authorized under the Mitchell Act in 1949 and is currently administered by the Environmental and Technical Services Division (ETSD) of the National Marine Fisheries Service (NMFS) in Portland. The CRFDP is a cooperative effort between NMFS and the Oregon Department of Fish and Wildlife (ODFW), the Washington Department of Fish and Wildlife (WDFW), Idaho Department of Fish and Game (IDFG), and the U.S. Fish and Wildlife Service (Delarm 1990). Between 1949 and 1989, the program has expended over $183 million dollars, primarily on the construction and operation of hatcheries (Figure 56). The program is currently authorized to expend approximately $10 million dollars per year.
Example populations for which detailed information is presented were selected from areas where information was readily available (Figure 57). Future editions of this annual report will expand this section to include key populations from throughout the Basin including all index populations for listed endangered species.
Table 1. Fisheries Data Dictionary for the September, 1995 version of the Columbia River CIS Distributed System
Table 2. Non-Fisheries Data Dictionary for the September, 1995 version of the Columbia River CIS Distributed System