There are many areas in all three states where California sea lions and Pacific harbor seals haul-out within foraging range of rivers with depressed salmonid runs. In most of these areas little or no information is available on whether seal or sea lion predation is a problem. However, because many salmonid runs are depressed, even limited predation may slow the recovery of these runs. The Working Group identified areas where there is potential for pinniped predation to affect depressed salmonid populations, and areas where research is needed to assess the magnitude of the impacts and determine whether mitigation efforts are warranted. The areas where depressed stocks are vulnerable to predation are listed in each region from north to south.

Strait of Juan de Fuca/San Juan Islands

Harbor seals are present year-round. California sea lions are present in the fall, winter, and spring. Salmonid stocks in this area are vulnerable to predation as adults from summer to winter, and as juveniles from April to June. All fall chinook salmon in this area are classified by the state as depressed or critical stocks, and adults are vulnerable to predation. Numerous coho salmon and several winter steelhead runs are depressed in this area, and both adults and juveniles are vulnerable to predation. Juvenile spring chinook salmon and steelhead, especially at Discovery and Sequim Bays, also are vulnerable to impacts from pinniped predation. Summer chum and coho salmon returning to Discovery Bay are critical stocks and are vulnerable to being impacted by pinniped predation. In addition, Dungeness and Elwha River pink salmon are critical stocks and may be impacted by pinniped predation.

Eastern Bays (Bellingham Bay, Skagit Bay)

Harbor seals are present year-round. Because of their low abundance, California sea lions are not a threat to salmonids in this area. Juvenile salmonids of all critical and depressed stocks are at least partially vulnerable to harbor seal predation during outmigration. The April-to-June timeframe is of most concern because both juveniles and adult salmonids from critical stocks are present. Numerous coho salmon stocks in this area are depressed, and adults and juveniles are vulnerable to predation. Nooksack River spring chinook salmon, Baker River sockeye salmon, and Deer Creek summer steelhead are classified by the State as critical stocks and may be impacted by pinniped predation.

Puget Sound

Harbor seals are present year-round. California sea lions are present in the fall, winter, and spring. Pinnipeds have been observed upriver in several rivers draining into Puget Sound. There is overall concern in the Puget Sound area for predation on spring-, fall-, and winter-run adult salmonids and on juvenile spring chinook salmon and steelhead. More than 1,000 California sea lions, which occur seasonally near the mouth of the Snohomish River, have been observed 8-10 miles upriver and prey on free-swimming salmonids in the estuary. As many as 300 harbor seals haul-out on log booms near the mouth of the Snohomish River in fall and winter and have been reported 15-20 miles upriver. Pinnipeds are known to depredate catch in the fall coho and chum salmon gillnet fisheries and the steelhead set-net fishery in the Snohomish River area. In the Nisqually River, both seals and sea lions are common at the mouth; sea lions have been observed preying on free-swimming salmonids and have been observed as far as 40 miles upriver near the dam at McKenna. California sea lions frequently interact with tribal coho salmon and steelhead gillnet fisheries at the mouth of the Green River (Duwamish Waterway), have been observed preying on free-swimming salmonids inriver, and have been observed as far as 20 miles upriver in the Green River. Both seals and sea lions prey on adult salmonids and smolts below and above the Ballard Locks facility, and sea lions have been observed preying on steelhead in Lake Washington up to the mouth of the Cedar River. California sea lions have been documented to have consumed 65% of the winter steelhead migrating through the Ballard Locks, and have also been observed foraging on the fall coho salmon run, the early portion of the sockeye run, and downstream migrating smolts in the spring. White River spring chinook salmon are designated as critical and may be impacted by predation. Numerous coho stocks in Puget Sound are depressed, and both adults and juveniles are vulnerable to predation.

Hood Canal

Harbor seals are present year-round. California sea lion abundance is low, but a few occur in the area in the fall, winter, and spring. Most salmonid stocks of concern migrate as subyearlings and therefore are not vulnerable to predation as juveniles. However, steelhead and coho salmon migrate as yearlings and are sufficiently large to be pinniped prey. Pinniped predation on adult salmonids from summer to winter is of concern. Quilcene, Dosewallips, Duckabush, Hamma Hamma, Dewatto, Tahuya, and Union River summer chum salmon (classified by State as critical stocks) may be impacted by pinniped predation.

Washington Outer Coast

Harbor seals are present year-round. California sea lions are present in the fall, winter, and spring. A California sea lion was observed preying on free-swimming coho near Tatoosh Island in September 1995. Sea lions also depredate catch in coastal salmon troll fisheries. Harbor seals have been observed upriver in the Sooes River. Of most concern is potential harbor seal predation on spawning adults and outmigrating juveniles from May to July. Queets and Quinault River spring and summer chinook salmon and Lake Ozette sockeye salmon are classified by the State as depressed and may be impacted by pinniped predation.

Grays Harbor

Harbor seals are present year-round. California sea lions occur infrequently, but have been observed far upriver in the Chehalis River. There is a potential for harbor seal predation on adult and juvenile salmonids from May to July and December to February. Harbor seals commonly depredate catch in the salmonid gillnet fisheries in Grays Harbor and in the Chehalis and Humptullips Rivers. Satsop River summer chinook salmon and winter steelhead populations are classified by the State as depressed and may be impacted by pinniped predation.

Willapa Bay

Harbor seals are present year-round and commonly depredate catch in the summer and fall salmon gillnet fisheries in Willapa Bay. There is concern for predation on adult salmonids in August and September. North River early fall chinook salmon are classified by the State as depressed and may be impacted by pinniped predation.

Columbia River

Harbor seals are present year-round, with peak numbers exceeding 3,000 from mid-December through mid-March. California sea lions (300-500) are present in fall, winter, and spring. The large and increasing number of pinnipeds raises concern over impacts of pinnipeds on Snake River spring/summer chinook and fall chinook salmon, which are declining and listed as threatened under the ESA. Harbor seals regularly occur more than 50 miles upriver. California sea lions occur as far as the Bonneville Dam (about 145 miles upriver) and into the Willamette River up to Willamette Falls (128 miles from the Pacific Ocean). Pinniped scarring on numbers of ESA-listed spring chinook salmon at Lower Granite Dam raises a greater concern about the level of pinniped impact on ESA-listed species. The large numbers of both harbor seals and California sea lions in the mouth of the Columbia River from late fall to early spring raises concerns for impacts on adult winter steelhead and spring chinook migrating upriver, as well as on juvenile salmonids from all stocks migrating downstream from March to June. California sea lions have been observed consuming adult salmonids far upriver near the fish ladder system at Willamette Falls since 1990. Steelhead and spring chinook passing through the Willamette fishway are depressed stocks and especially vulnerable to predation at this site. Harbor seal numbers in the Lower Columbia River begin increasing during the fall chinook migration upstream, raising concerns for impacts of pinniped predation on these populations also.

North Oregon Coast

Harbor seals are present year-round. California sea lions are present from fall through spring. All Oregon coastal coho salmon and steelhead have been proposed for listing under the ESA and are vulnerable to impact by pinniped predation at all sites where they co-occur with pinnipeds during migration. Pinniped scars have been documented on both coho and steelhead in most rivers. Groups of California sea lions have been regularly observed foraging for winter steelhead in the mouth of the Nehalem River estuary for the past 4-6 years. Between 1985 and 1992, the occurrence of pinniped scars on returning adult winter steelhead in the Nehalem River has averaged from 30% to 50%. Harbor seal abundance in the Siletz River has increased over the past 10 years, while counts of spawning adult coho salmon have declined. In the lower Alsea River, sea-run cutthroat are currently at very low numbers and no longer support a viable sport fishery. This raises concerns about the impacts of pinniped predation in the lower river because the Alsea Bay estuary has a large year-round population of harbor seals in Oregon (300-600 seals). Counts of spawning adult coho salmon have declined significantly in recent years even though high-quality spawning habitat is still available. Between 1982 and 1992, pinniped scarring rates on coho salmon in the Alsea River were reported at 11%, while 19-27% of returning winter steelhead had pinniped-caused scars.

Tillamook Bay

One of the largest aggregations of harbor seals in Oregon (500-800 seals) resides year-round in Tillamook Bay. These seals regularly interact with salmonid sport fisheries in this region. California sea lions are frequently observed foraging for salmonids in the mouth of the bay. Pinniped scars have been observed on 35% of the winter steelhead returning to the hatchery on the Trask River. Counts of spawning coho salmon have declined in recent years. Coho salmon and steelhead in this region have been proposed for listing under the ESA and are vulnerable to impacts by pinniped predation.

Umpqua River

Year-round abundance of harbor seals (600-1,000 seals) in the Umpqua estuary is second in Oregon only to the Columbia River. Umpqua River sea-run cutthroat trout are declining and have been listed as endangered under the ESA. Although information is lacking, there is great concern about this endangered population being impacted by pinniped predation, especially since pinniped scarring has been observed on Umpqua River cutthroat. Coho salmon and steelhead in the Umpqua River have been proposed for listing under the ESA and are vulnerable to impacts by pinniped predation during migration through the estuary and lower river areas.

Rogue River

The diversity and abundance of pinnipeds (harbor seals, California sea lions, and Steller sea lions) that forage in the mouth of the Rogue River may be greater than at any other coastal river. Coho salmon and steelhead in this region have been proposed for listing under the ESA and are likely vulnerable to being impacted by pinniped predation. Lower Rogue River and Illinois River fall chinook stocks are depressed and also may be vulnerable to being impacted by pinniped predation. Consumption of returning adult salmonids by pinnipeds at the mouth of the Rogue River has been reported at rates of several fish per hour during peak fish runs. Predation during fall months is of greatest concern because of the poor conditions of the salmonid runs at that time.

South Oregon Coast

Harbor seals are present year-round. California sea lions are present from fall through spring. This region has a number of small coastal rivers and streams that have low or precluded flow during some years when coho salmon and steelhead are attempting to migrate. All Oregon coastal coho salmon and steelhead have been proposed for listing under the ESA and can be impacted by pinniped predation.

Smith River

Harbor seals are present year-round, with peak haul-out abundance in the summer. California sea lions are present in the fall through spring. Coho salmon and steelhead have been proposed for listing as threatened under the ESA and are vulnerable to impacts from pinniped predation. Spring chinook (classified by State as at high risk of extinction) and fall chinook and cutthroat trout (classified by State as at moderate risk of extinction) also may be vulnerable to impacts from pinniped predation.

Klamath River

Harbor seals are present year-round, with a peak of about 400 seals in the summer. California sea lions are present in the fall and spring. Both harbor seals and sea lions have been documented feeding on salmonids in this area since the 1960s. Coho salmon and steelhead have been proposed for listing as threatened under the ESA and are vulnerable to being impacted by pinniped predation. Klamath River spring chinook are classified by the State as having high risk of extinction, and impacts of pinniped predation are of concern. Impacts of pinniped predation on fall chinook in the Klamath, Shasta, Salmon, and Trinity Rivers are also of concern as these populations have been classified by the State as having a moderate risk of extinction.

Cape Mendocino

Harbor seals are present year-round. California sea lions are present in the fall and spring. Harbor seals haul-out in large numbers (600-1,050 seals) at the mouth of the Eel River. More than 1,200 sea lions have been counted in the vicinity of Trinidad Head. Coho salmon and steelhead have been proposed for listing as threatened under the ESA and may be vulnerable to impact by pinniped predation. Mad River fall chinook (classified by State as at moderate risk of extinction) and Eel River fall chinook (classified by State as at moderate to high risk of extinction) also are vulnerable to impacts from pinniped predation.

North Coast

Harbor seals are present year-round. California sea lions are present in the fall and spring. Harbor seals and sea lions have been documented feeding on salmonids in the Russian River. Coho salmon and steelhead have been proposed for listing as threatened under the ESA and may be vulnerable to impact by pinniped predation. Mattole and Russian River fall chinook (classified by State as at high risk of extinction) also may be vulnerable to pinniped predation impacts.

San Francisco Bay/Central Valley

Harbor seals are present year-round. California sea lions are present year-round just offshore at the Farallon Islands. The San Francisco Bay/Central Valley area has many river systems and creeks, and both sea lions and harbor seals have been found well within the inland deltas and river systems. Central California coho salmon, which have been listed as threatened under the ESA, and all steelhead populations, which have been proposed for listing as endangered under the ESA, are vulnerable to being impacted by pinniped predation. Sacramento River winter chinook have been listed as endangered under the ESA. Papermill Creek, San Leandro River, Alameda Creek, and Sacramento River winter steelhead have been classified by the State as having a high risk of extinction and are of special concern for pinniped predation impacts. Other stocks classified by the State as having a moderate risk of extinction are Sacramento River spring/summer chinook, Yuba River spring chinook, and Deer Creek and Mill Creek spring chinook.

Monterey Bay Area

Harbor seals are present year-round. California sea lions are present year-round at Año Nuevo Island, peaking in the fall, with numbers well over 6,000 individuals. Central California coho salmon in the San Lorenzo River and other rivers to the north are listed as threatened under the ESA and are vulnerable to impacts from pinniped predation. Steelhead have been proposed for listing as endangered under the ESA, and impacts of pinniped predation also are of concern.

Central Coast

Harbor seals are present year-round. California sea lions are present in the summer and fall. Steelhead in this region have been proposed for listing as endangered under the ESA and there is concern about potential pinniped predation.

Southern California Bight

Harbor seals are present year-round. California sea lions are present year-round in large numbers at the Channel Islands, with peak population (81,300 sea lions) during the summer. Most historic salmonid runs are extinct in this area. Southern California steelhead have been proposed for listing as endangered under the ESA. There are no known harbor seal or sea lion haul-out sites in the southern California rivers, and pinniped predation therefore is probably not a concern. However, the presence of only a few migrant pinnipeds in one of the rivers with steelhead could impact the steelhead populations.

Salmonid life-history patterns determine the availability of salmonids to pinniped predation. Adult salmonids are most vulnerable to pinniped predation during the spawning migration through estuaries and river mouths, especially where salmonids concentrate or passage may be constricted. Predation on juvenile salmonids is affected by their size during the outmigration. Chum and pink salmon migrate to sea as fry soon after hatching, when they are too small to be pinniped prey. Pink salmon, however, may summer in nearshore ocean areas where they are vulnerable to pinniped predation. Spring chinook salmon, coho salmon, sockeye salmon, and steelhead migrate to the ocean as yearlings or older, at a size where they are vulnerable to pinniped predation. Most fall and summer chinook salmon migrate downstream as sub-yearlings when they may also be too small to be prey of pinnipeds. However, some of the early-timed summer chinook salmon migrate downstream as yearlings when they are large enough to be vulnerable to predation. Sea-run cutthroat migrate to the coastal estuaries at a variety of ages, from subyearlings to 2-year-olds, and spend most of their adult life close to shore where they could be continuously vulnerable to pinniped predation.

Chinook, chum, coho, pink, and sockeye salmon and steelhead have all been documented as prey of pinnipeds in Washington, Oregon, or California. Sea-run cutthroat have not been specifically identified in any food habits studies, but pinniped-scarred cutthroat have been observed in the Umpqua River (Beach et al. 1985) and they are likely prey of pinnipeds in estuaries. Some studies reported "trout," which may be steelhead or cutthroat. The studies show varying rates of occurrence of salmonids as prey of pinnipeds depending upon location, season, prey availability, and methods used for collecting food habits data in each particular study (Appendices F and G).

The most widely known and intensely studied pinniped/salmonid conflict is California sea lion predation on winter steelhead at the Ballard Locks in Seattle, Washington. Although California sea lions first began appearing in the Ballard Locks area on a somewhat regular basis in 1980, their predation on steelhead was not viewed as a resource conflict until 1985, when a significant decline in the wild winter steelhead spawning escapement was noted (Gearin et al. 1996). Subsequent scientific studies documented that sea lions were removing significant numbers of adult steelhead that were returning to the Lake Washington system to spawn (Scordino and Pfeifer 1993). As shown in Table 2, the number of wild steelhead consumed by sea lions between 1986 and 1992 was 42-65% of the total run (NMFS 1995b). In spite of intense sea lion deterrence and mitigation efforts from 1985 to 1995, a small number of sea lions returned to the Ballard Locks area each season and preyed on steelhead (Scordino and Pfeifer 1993). NMFS and WDFW have attempted all feasible, nonlethal approaches to reduce sea lion predation without success (NMFS 1995b). The winter steelhead population declined significantly and recent spawning escapements have been less than 150 fish, which is within the range considered to be near the threshold level below which the ability of the population to recover may be impaired (NMFS 1996a). Because of the precarious status of the steelhead population and the impact that sea lions were having on the status and recovery of this salmonid population, NMFS authorized WDFW in 1995 to lethally remove the "problem" sea lions at the Ballard Locks that could not be deterred by nonlethal means (NMFS 1995b).

The observations of steelhead predation by California sea lions at the Ballard Locks show a significant proportion (65%) of an entire salmonid run can be consumed by sea lions (Scordino and Pfeifer 1993) and clearly demonstrates that the combination of high local-predator abundance during salmonid migrations, restricted passage, and depressed fish stocks can result in significant impacts on local salmonid populations (NMFS 1995b). There are only a few areas on the West Coast, other than the Ballard Locks, where studies have documented the influence of pinniped predation on local salmonid populations. In the Puntledge River estuary, British Columbia, Bigg et al. (1990) observed Pacific harbor seals surface feeding on salmonids and documented predation rates of up to 46% of the returning adult fall chinook. In Netarts Bay, Brown and Mate (1983) found that the number of seals feeding in the area was similar in each year of their study; however, the impact of the predation was greatest when the chum salmon return was low. In 1979, the seals took more than 7% of about 550 returning chum salmon, while in 1980, the estimated consumption of nearly twice as many fish represented less than 2% of the return of more than 5,000 salmon. In the Rogue River, Roffe and Mate (1984) estimated that in the late 1960s, sea lions and seals removed less than 1% of the spring chinook and about 6% of the summer steelhead returns to the Rogue River, which was equal to about half of the annual sport catch during that time.

California sea lions and harbor seals are present in most areas where salmonid runs occur in Washington, Oregon, and California. In most places, little information is available on year-to-year changes in the seasonal abundance and daily distribution of pinnipeds near or in salmonid rivers. More information exists on the seasonal abundance of salmonids in rivers and estuaries, but, because cohort sizes of salmonids can vary dramatically, estimates from one year or season applied to another year or season may over- or underrepresent the importance of salmonids in the pinniped diet. Consequently, there is little appropriate long-term information on the effect of pinniped predation on salmonids. However, data from the Ballard Locks (Gearin et al. 1988a) and the Puntledge River estuary in British Columbia (Bigg et al. 1990) indicate that where salmonid populations are depressed and particularly where fish passage is restricted by man-made structures (e.g., dams), narrow channels, or shallow water, pinniped predation can have a detrimental effect on salmonid populations. At the Ballard Locks, California sea lions consumed as much as 65% of the wild steelhead run in Lake Washington (see Table 2). At the Puntledge River estuary, harbor seals consumed up to 46% of returning adult fall chinook. As expected, the greatest effect of pinniped predation occurs when salmonid populations have already been reduced to low numbers.

Much of the predation at the Ballard Locks was by a few California sea lions that repeatedly foraged on salmonids in spite of deterrence efforts by NMFS and WDFW. The observations at the Ballard Locks indicate the ability of individual animals to consume large numbers of salmonids. One sea lion in 1986 was observed over the course of a 7-day period to kill at least 84 steelhead in 56 hours of observations, for a combined rate of 12 steelhead killed per 8 hours per day (Gearin et al. 1986). The highest predation rates observed were 4 steelhead kills in 23 minutes during 1 day for this animal. On the same day, this sea lion killed 12 steelhead in 4.75 hours. These observations indicate the potential predation levels of California sea lions when prey is abundant and where foraging ability is enhanced by narrow feeding channels. Individual sea lion behavior was also observed during the coho salmon runs through the Ballard Locks. In 1996, a single California sea lion was observed to kill 136 coho salmon in 62 hours (2.1 coho per hour) (NMFS 1996a). The highest predation rates observed for this animal were 18 coho salmon over 4.4 hours (4.1 fish per hour). The maximum number of coho observed killed by this sea lion during any one day was 19 coho salmon in 6.9 hours (2.7 fish per hour). Similarly, one sea lion was observed killing 5 spring chinook in 3 hours in the area of the Willamette Falls fishway (ODFW unpubl. data). Another observation at the Falls was one sea lion taking 7 spring chinook in 7 hours (1 per hour). Although these observations cannot be applied to other areas nor extrapolated over time, they clearly show the potential for individual California sea lions to consume large numbers of salmonids.

In most cases where pinnipeds and salmonid smolt co-occur, it is assumed that the pinnipeds are feeding on smolt. However, because the smolt are consumed under water, it is unknown to what extent the seals and sea lions exploit that resource. At the Ballard Locks, California sea lions were observed actively foraging during the peak of smolt outmigration, and although the observers were confident that the sea lions were eating smolt, they could not quantify numbers of smolt consumed (NMFS 1996a). One recent study in Canada quantifies harbor seal predation on smolt. In the lower Puntledge River in British Columbia, harbor seals forage on chum salmon fry and coho salmon smolts at night by using the lights from bridges to silhouette the fish and aid in their capture. During the peak of predation, consumption was estimated at 140,000 chum salmon fry and 13,000 coho salmon smolt per night (P. Olesiuk, Canada Department of Fisheries and Oceans, Science Branch, Pacific Biologic Station, Nanaimo, BC V9R 5K6. Pers. commun., 1996). As is true in most areas where individual pinnipeds can be identified, most predation (53-57%) was attributable to a small number (10) of recognizable seals. Total consumption was estimated at 3.1 million chum salmon fry (7-31% of the 1995 production) and 138,000 coho salmon smolt (15% of the 1995 production) between April and June (P. Olesiuk, Canada Department of Fisheries and Oceans, Science Branch, Pacific Biologic Station, Nanaimo, BC V9R 5K6. Pers. commun., 1996).

The Working Group considered California sea lion and Pacific harbor seal foraging on salmonids in the open ocean. In a review of marine mammal-salmonid interactions in the Pacific Northwest, Fiscus (1979) suggested that mammal predation on free-swimming salmonids in the open ocean probably has a minimal impact, and consumption rates on healthy and abundant fish stocks in these situations is relatively low. Studies from the 1970s on northern fur seals offshore of Washington, however, found that salmonids (mostly immature pink, coho, and chinook salmon) were present in 20% of the samples examined (Fiscus 1980). Antonelis and Perez (1984) estimated that 11.6% of the prey consumed by northern fur seals off Washington and Oregon annually was salmonids, and they estimated an annual consumption of 3,897metric tons (t) of salmonids. Although the Working Group found no comparable information on California sea lions or Pacific harbor seals, it did note that a California sea lion had been observed taking a coho salmon in coastal waters off Washington. Nonetheless, pinniped predation on small populations of depressed or listed salmonids, whether inriver or in the open ocean, is important in assessing the impacts of predation on recovery of salmonid populations.

In many of the small coastal rivers and streams in southern Oregon and northern California, the Working Group found there is a unique situation that makes returning adult coho salmon and winter steelhead more vulnerable to pinniped predation than larger systems. In low rainfall years or when rain arrives late in the winter season, small coastal rivers do not flow with sufficient volume to open the beach crest and flow into the sea. Low-tide periods also create or confound this condition in small, low-flowing rivers and streams. During such periods, adult fish arrive and accumulate in nearshore waters just offshore of the closed-off river mouth. The adult salmonids are then exposed to days or weeks of pinniped predation at these sites until sufficient rainfall occurs or higher tides allow access to the river or stream. During successive years of drought, the situation is exacerbated because the river mouths are open only intermittently during the salmonid spawning season. Downstream migrating smolt also become more vulnerable to pinniped and bird predation in these conditions, as the fish congregate in the lagoons formed near the river mouth until it opens up to the sea.

In understanding the effects of pinniped predation on salmonids, the Working Group noted that it is important to keep in mind that not all pinnipeds at a haul-out near a salmonid run are actively feeding on salmonids. Herder (1983) found that although there were up to 200 harbor seals in the Klamath River area, only 9 seals were responsible for depredation on gillnets each day. At the Ballard Locks, only 3% of the 248 sea lions marked in the nearby Shilshole Bay entered the Ballard Locks area in 1995 to feed on steelhead (NMFS 1996a). This indicates that removing pinnipeds from nearby areas may not be an effective solution to the problem of pinniped predation in local areas.

In all but a few sites, information on direct mortality--how many pinnipeds (and whether they are seals or sea lions) are feeding on how many salmonids (which species of salmonid and whether they are adults or juveniles)--is unknown. In reviewing past data on pinniped food habits, seasonal pinniped and salmonid abundance and distribution, and salmonid mortality due to pinniped predation, the Working Group identified deficiencies that limited the use of such data in quantifying pinniped consumption of salmonids. In addition, information is lacking on changes in abundance or distribution of other salmonid predators (e.g., mackerel) which may affect salmonid populations and thus confound the effects caused by pinnipeds.

Most food habits studies were not designed to estimate overall consumption or species-specific consumption rates, and results from such studies cannot be extrapolated to estimate salmonid consumption. Some studies were conducted at a time of year when salmonids were not present; consequently, salmonid importance in the annual diet is underestimated. Smolt predation was not represented or was underrepresented in most food habits studies because the otoliths of juvenile salmonids are fragile and quickly digested; therefore, they may not be identified in stomachs or scats. The occurrence of salmonids was also underrepresented in earlier food habits studies because only otoliths were used to identify prey species. Several studies using bones, teeth, gill rakers, and otoliths to identify prey species have noted that using only otoliths will underrepresent salmonids in the diet (Gearin et al. 1988b, Riemer and Brown 1996). At this point, even though the use of other hard parts besides otoliths provides better detection of salmonids in scats, it is not possible to identify what species of salmonid was consumed.

Many of the pinniped food habits studies were conducted 10-20 years ago when salmonids, other fish, and pinniped population levels were quite different. Results from older studies may not be applicable to current conditions of increased abundance of pinnipeds and decreased abundance of many salmonid stocks. The year-to-year variation in salmonid abundance is an important factor in assessing impacts of pinnipeds. A constant number of pinnipeds consuming a constant number of salmonids will have a much greater effect on small or declining salmonid populations.

Data were also lacking for an estimate of indirect mortality due to wounds inflicted by pinnipeds in unsuccessful predation attempts. The Working Group found that scarring data cannot be used to estimate salmonid mortality or actual rates of predation. However, pinniped scarring data does serve as an indicator of trends of exposure of salmonids to pinniped predation. Where time series of annual pinniped scarring rates can be compiled, they may be a valid indicator of changes in exposure of adult salmonids to pinniped predation in specific rivers and estuaries.

The Working Group did not attempt to estimate salmonid consumption by pinnipeds because they found the data available at present were inadequate. However, they did review recent estimates of salmonid consumption by California sea lions and harbor seals in Oregon made by Kaczynski and Palmisano (1992), as well as estimates of Snake River salmon consumption by harbor seals made by Chapman et al. (1991) and Park (1993). The Working Group found similar technical and analytical weaknesses in all three reports.

Annual consumption of salmonids by California sea lions in Oregon was estimated by Kaczynski and Palmisano (1992) at 142.9 t (35,800 fish) based on the following assumptions: 1) a daily maintenance diet of about 6.8 kg for an adult male sea lion, 2) salmon comprises about 10% of the total biomass that sea lions consume, and 3) up to 2,000 sea lions are present for 3 months during migration and 200 sea lions overwinter in Oregon.

Harbor seal annual consumption of salmonids in Oregon was estimated by Kaczynski and Palmisano (1992) at 816 t (204,500 fish) based on the following assumptions: 1) a daily maintenance diet of 2.7 kg per day for a 54-kg harbor seal, 2) salmon comprises 10.8% of the biomass that harbor seals consume, and 3) harbor seal abundance in Oregon was 10,000 in 1992. Chapman et al. (1991) and Park (1993) estimated harbor seal consumption of Snake River spring chinook using the same assumptions of average size, daily food intake, and percentage of salmon in the diet as Kaczynski and Palmisano (1992). Chapman et al. (1991) calculated that 2,100 seals in the Columbia River (over the 100 days when adult Snake River spring chinook are migrating) consumed 15,700 salmon (of which they assumed 3,000 were Snake River chinook). Park (1993) estimated harbor seals in the Columbia River consumed 22,558 salmon (of which 4,500 were assumed to be Snake River chinook) based on 3,000 seals present during the spring chinook run.

ODFW (1992) reviewed the Kaczynski and Palmisano (1992) report and concluded that they failed to stratify analyses by location, species, and life histories of both salmonids and pinnipeds; inaccurately interpreted or reported scientific studies; and failed to analyze data using valid and sufficiently rigorous methods. The Working Group found that Kaczynski and Palmisano (1992) extrapolated their estimates from one or two site- and season-specific food habits studies to the entire state and failed to differentiate between species of salmonids.

The Working Group found that Chapman et al. (1991) and Park (1993) made many of the same assumptions as Kaczynski and Palmisano (1992) in using data on another species from another area to estimate predation on adult spring chinook in the Columbia River. The Working Group noted that spring chinook had not been identified in the scats or stomach contents of harbor seals in the Columbia River (Beach et al. 1985, Brown et al. 1989) and, consequently, a predation rate cannot be estimated. Although there is pinniped scarring on adult spring chinook as a result of predation attempts by seals and sea lions, the Working Group found that it is unknown to what extent harbor seals are successful in predating adult spring chinook in the Columbia River and what proportion of the salmonids that may be taken by seals or sea lions are from the Snake River.