2003 Juvenile Salmonid Production Evaluation and Adult Escapement: Intensively Monitored Watersheds (IMW) Annual Report

Since the Endangered Species Act listing of numerous salmon and steelhead populations in the Pacific Northwest in the 1990’s, millions of dollars have been dedicated to the restoration of freshwater habitat. Little is known about the effectiveness of these efforts in restoring salmon populations. Scientists have concluded that the most effective means of determining the contribution of restoration projects to salmon recovery is to implement experimental, watershedscale evaluations that include the measurement of freshwater (smolt) production. Several organizations in the Pacific Northwest have begun to establish such projects. The Intensively Monitored Watersheds (IMW) Project evolved in 2003 from the joint Washington Department of Fish and Wildlife and Washington Department of Ecology Index Watershed Monitoring Project. A complete description of the watersheds and progress made on the IMW project during its first year is described in IMWSOC (2004). IMW monitoring activities include the measurement of freshwater production and escapements into IMW streams. This report presents the 2003 smolt production estimates for the Hood Canal and Lower Columbia IMWs and the 2003 escapement estimates for the Hood Canal IMWs. It also details the field work and analytical steps taken to produce these estimates.

Hood Canal IMWs
The Hood Canal IMWs are comprised of Big Beef, Little Anderson, Seabeck, and Stavis Creeks located in western Kitsap County. Downstream migrant (smolt) trapping is conducted using temporary fence weirs located near the mouths of Little Anderson, Seabeck, and Stavis Creeks. A permanent weir is located at the mouth of Big Beef Creek and three fan traps attached to the weir during the spring capture all downstream migrants. Adult coho and chum salmon returning to Big Beef Creek are trapped and counted each fall.

2003 Downstream Migrant Production
Traps were installed in late March or early April in all streams and were operated until the end of May or early June when smolt catches declined. A total of 34,463 coho smolts were captured in Big Beef Creek. Based on typical migration timing at this site, we estimated that an additional 800 coho migrated before/after the period of trap operation. Adding these to the catch along with an estimated 797 naturally-reared smolts from the University of Washington Fishery Research Institute (FRI) channels and ponds results in a total production estimate of 36,060 smolts. Of these, 31,355 were coded wire tagged (CWT) prior to release. Coho catches in Little Anderson, Seabeck, and Stavis Creeks were 226, 1,518, and 7,454 smolts, respectively. Assuming the same run timing as Big Beef coho, we estimate total coho production at 240 smolts for Little Anderson Creek, 1,565 smolts for Seabeck Creek, and 7,757 smolts for Stavis Creek. Steelhead and cutthroat smolts were also caught at all of the sites. As with coho, the steelhead production from Big Beef Creek was much larger than in the other streams (1,232 smolts vs. 83 combined smolts from the other three streams). The cutthroat migration was similar in all streams with 1,206, 891, 405, and 1,557 smolts caught in Big Beef, Little Anderson, Seabeck, and Stavis Creeks, respectively. In addition to these catches, 24,000 chum and 350 chinook fry were captured in Big Beef Creek. Chum are present in the other three streams, but the fence weirs are designed to preclude capturing them. Chinook found in Big Beef Creek are of FRI origin and are not found in the other streams.

Coho fork length was measured from a sample of the captured fish. Big Beef coho averaged 104 mm, whereas those from Seabeck and Stavis Creeks averaged 5mm smaller and those from Little Anderson Creek averaged 9mm smaller.

2003 Escapements
The adult trap was operated from mid-August 2003 through January 2, 2004. All fish entering Big Beef Creek were enumerated by species, age, sex, mark status and condition before being released upstream. All adipose marked (ad-marked) hatchery coho were killed to preclude their spawning in Big Beef Creek. Unmarked coho were checked for a CWT. Of the unmarked coho that tested positive for a CWT, approximately 5% of the males and 25% of the jacks were sacrificed to estimate the number of tagged Big Beef fish returning and the incidence of unmarked, tagged hatchery fish. Tags were also recovered from carcasses on the spawning grounds. All unmarked coho not sacrificed for CWT recovery were passed upstream. Fork length and a scale sample were taken from approximately 20% of the unmarked returning coho. Scale sampling was used to estimate the number of unmarked hatchery fish that were passed upstream as well as for age determination.

A total of 5,105 adult coho and 361 jacks were captured between September 20 and December 13, 2003. These included 357 adults and 11 jacks that were ad- marked. Roughly two thirds of the unmarked adult coho (3,185) contained CWTs, as did 257 unmarked jacks. Of these, we sacrificed 96 adults and 72 jacks for tag recovery. A total of 4,647 unmarked adults and 279 unmarked jacks were released upstream. Five adult coho died either in the trap or below the weir.

Scale sampling indicated that 98.6% of the unmarked adult and 96.9% of the unmarked jack returns were naturally-reared fish. CWTs were recovered from 96 unmarked/tagged adults from the trap and 237 from adult carcasses on the spawning grounds. Of these tags, 334 were from natural origin Big Beef coho and 3 were from hatchery fish. Similar results were found from the unmarked/tagged jacks with 80 of the 81 recovered tags being from natural origin Big Beef coho. Good corroboration was found between scale-based and CWT-based estimates of the unmarked naturally- reared and hatchery components of the run. Both approaches estimated the survival from smolts to adult returns for the 2000 brood at 18.3%. Some 2000 brood Big Beef coho were harvested in fisheries. Marine survival (survival from smolts to age-3 recruits) was estimated at 19.3% based on preliminary tag data.

In addition to coho, 3,744 adult chum salmon, 896 summer chum and 2,848 fall chum, returned to Big Beef Creek. The chum migration began on September 6 and extended to December 24, 2003. Two adult steelhead were captured in the Big Beef trap in late December and 69 cutthroat were captured between October 10 and December 24. These represent an unknown portion of the total steelhead and cutthroat returns since the trapping ceased on January 2, before their migrations were completed. Also, a number of adult chinook returned to Big Beef Creek. These were the progeny of FRI releases and the adults were returned to their facility.

Adult coho escapements into Little Anderson, Seabeck, and Stavis Creeks were estimated by multiplying their respective 2000 brood smolt productions by the ratio of the total Big Beef adult return to the 2000 brood Big Beef smolt production. This approach estimates the 2003 escapements at 52, 316, and 1,480 adults into Little Anderson, Seabeck, and Stavis Creeks, respectively. These estimates assume that hatchery stray rates into these streams were the same as estimated for Big Beef Creek. It is further assumed that smolt-to-adult survival is the same for all of the streams.

Lower Columbia IMWs
The Lower Columbia IMWs consist of Mill, Aberna thy, and Germany Creeks located in Cowlitz and Wahkiakum Counties west of Longview, Washington. The downstream migrant production of coho and steelhead are estimated from each stream using a 1.5- m diameter rotary screw trap located near its mouth. Unlike the traps in the Hood Canal IMWs, these traps capture a portion of the outmigrating smolts. Production is estimated by conducting a series of mark-recapture experiments to determine the proportion of the downstream migrating juveniles from each species captured in each trap.

The traps were operated from April 4 to June 19, 2003 on all three streams. The traps operated continuously except on Abernathy Creek, when debris prevented the trap from operating on four occasions.

Catches in the Abernathy trap totaled 2,324 coho, 3,779 steelhead (603 natural origin and 3,176 hatchery), and 139 cutthroat smolts. Of the hatchery steelhead smolts, 124 were tagged with Passive Integrated Transponder (PIT) tags, 2,815 were tagged with CWTs, and 237 were unmarked. Interpolation was used to estimate the catch that would have occurred during the four periods when debris halted trap operation. This analysis estimated 10 coho and 2 steelhead of natural origin would have been caught during those periods. To estimate trap efficiency, 2,145 coho and 530 natural origin steelhead captured in the trap were marked using partial fin clips and released upstream. To estimate cutthroat efficiency, we were assisted by USFWS who PIT tagged 110 cutthroat captured in the trap and released them above their instream antenna array at river kilometer 2.9.

The trap position was moved on April 22 and weir panels were added on May 19 to divert more flow and fish into the trap. These changes in trap operations resulted in three trap efficiency strata: original position, 2nd position, and 2nd position with weir panels. Average trap efficiency for coho and steelhead was estimated for each efficiency stratum based on results from the markrecapture experiments. Total migration during each efficiency stratum was estimated and summed across strata to estimate total production. Using this approach we estimated 9,626 coho, 4,141 natural origin steelhead, 21,713 hatchery steelhead, and 531 cutthroat emigrated from the stream.

Catches in Germany Creek totaled 2,832 coho, 1,859 steelhead, and 178 cutthroat smolts over the trapping period. A total of 2,560 coho and 1,757 steelhead were marked with a partial fin clip and released above the trap to assess trap efficiency. Too few cutthroat were captur ed to assess trap efficiency for this species, so the steelhead efficiency was used since cutthroat and steelhead smolts were similarly sized. The season average trap efficiency was used to estimate a total production of 5,775 coho, 5,936 steelhead, and 563 cutthroat smolts.

On Mill Creek, 4,168 coho, 253 steelhead, and 115 cutthroat smolts were captured. Of those captured, 3,363 coho and 229 steelhead were marked with a partial fin clip and released upstream to assess trap efficiency. As with Germany Creek, too few cutthroat were captured to assess cutthroat efficiency directly, so steelhead efficiency was used to estimate the cutthroat production. Weir panels were installed upstream of the trap on May 6 to divert more flow and fish through the trap. This change in operation resulted in two efficiency strata: before weir panels and after weir panels. Mean trap efficiency was calculated for each stratum and used to estimate migration. Total production was estimated at 10,514 coho, 1,383 steelhead, and 574 cutthroat smolts.