Pair-Trawl Detection of PIT-Tagged Juvenile Salmonids Migrating in the Columbia River Estuary, 2008 Report of Research.

In 2008, we sampled migrating juvenile Pacific salmonids Oncorhynchus spp. tagged with passive integrated transponder (PIT) tags using a surface pair trawl in the upper Columbia River estuary (rkm 61-83). The cod-end of the trawl was replaced with a cylindrical PIT-tag detection antenna with an 86-cm-diameter fish-passage opening and two detection coils connected in series. The pair trawl was 105 m long with a 91.5-m opening between the wings and a sample depth of 4.9 m. Also during 2008, we finalized the development of a prototype 'matrix' antenna, which was larger than previous antennas by a considerable magnitude. The matrix antenna consisted of 6 coils: a 3-coil front component and a 3-coil rear component, which were separated by 1.5-m of net mesh. The fish-passage opening was 2.5 m wide by 3.0 m tall and was attached to a standard-size pair trawl net. Intermittent sampling with a single crew began on 7 March and targeted yearling Chinook salmon O. tshawytscha and steelhead O. mykiss. Daily sampling using two crews began on 30 April and continued through 14 June; during this period we detected 2.7% of all juvenile salmonids previously detected at Bonneville Dam--a measure of sample efficiency. Sampling with a single crew continued through 20 August and targeted subyearling Chinook salmon. We detected 7,397 yearling Chinook salmon, 2,735 subyearling Chinook salmon, 291 coho salmon O. kisutch, 5,950 steelhead, and 122 sockeye salmon O. nerka in the upper estuary. We deployed the matrix antenna system and the older, cylindrical antenna system (86-cm-diameter fish-passage opening) simultaneously in mid-May 2008 to test matrix detection efficiency. The cylindrical antenna system had been used successfully in 2007 and early 2008. Because distribution of migrating salmonids in the estuary changes rapidly, we felt that a tandem sampling effort between the two systems was the only way to truly evaluate comparative detection efficiency. We deployed both systems within 1 km of each other during a period of high fish densities on 13, 14, and 15 May. Detections of the matrix system surpassed those of the cylindrical system by 53% in 14 h of simultaneous sampling (total detections 716 and 339, respectively). We believe that the higher detection rate observed with the matrix system was due to fewer smolts escaping the trawl entrance and to more smolts readily passing through the larger fish-passage opening. After tandem sampling, we continued exclusive use of the matrix system for the remainder of the 2008 juvenile migration season. Mean survival rates from Lower Granite to Bonneville Dam for yearling Chinook salmon and steelhead were 42% (SE = 3.7%) and 46% (SE = 1.5%), respectively. Over 358,000 PIT-tagged salmonids were transported, and we detected 4,619 of these fish