Multi-frequency (18-, 38-, 70-, 120-, 200-, and 333-kHz) General Purpose Transceivers (Simrad EK60 GPTs), were configured with split-beam transducers (Simrad ES18-11, ES38B, ES70-7C, ES120-7C, ES200-7C, and ES333-7C, respectively). The transducers were mounted on the bottom of a retractable keel or “centerboard”. The keel was retracted (~ 5-m depth) during calibration, and extended to the intermediate position (~7-m depth) during the survey. Exceptions were made during shallow water operations, when the keel was retracted to ~ 5-m depth; or during times of heavy weather, when the keel was extended to ~9-m depth to provide extra stability and reduce the effect of weather-generated noise. Nighttime trawl sampling was conducted where echoes from CPS schools where observed earlier that day. Trawls were towed at ~ 4 kn for 45 min. The total catch from each trawl was weighed and sorted by species or groups. From the catches with CPS, up to 75 fish from each of the target species were selected randomly. Those were weighed (g) and measured to either their standard length (Ls; mm) for sardine, northern anchovy, and herring, or fork length (Lf ; mm) for jack mackerel and Pacific mackerel. Regional species composition was estimated from the nearest trawl cluster, i.e., the combined catches of up to three trawls per night, separated by ~ 10 nmi. During the day, fish eggs were collected using CUFES (Checkley et al., 1997), which collects water and plankton at a rate of ~640 l min-1 from an intake on the hull of the ship at ~ 3-m depth. The particles in the sampled water were sieved by a 505 μm mesh. All fish eggs were identified to lowest taxa, counted, and logged. Typically, the duration of each CUFES sample was 30 min, corresponding to a distance of 5 nmi at a speed of 10 kn. Because the initial stages of the egg phase is short for most fish species, the egg distributions inferred from CUFES indicate the nearby presence of actively spawning fish CalCOFI Bongo Oblique (CalBOBL, or bongo) nets (71-cm diameter; 505-μm mesh) were used to sample ichthyoplankton and krill at each station. Where there was adequate depth, 300 m of wire was deployed and then retrieved at 20 m min-1, at a nominal wire angle of 45. Paired vertical egg tow (Pairovet; formerly CalCOFI vertical egg tow or CalVET) (Smith et al., 1985) nets (25-cm diameter; 150-μm mesh) were used to sample fish eggs from a depth of 70 m to the sea surface at a rate of 70 m min-1 in areas where their densities exceeded a threshold of > 0.3 eggs min-1. Day and night, conductivity and temperature versus depth to 350 m were measured with calibrated sensors on a CTD probe cast. These data were used to estimate the time-averaged sound speed (Demer, 2004), for estimating ranges to the sound scatterers, and frequency-specific sound absorption coefficients, for compensating the echo signal for attenuation during propagation of the sound pulse from the transducer to the scatterer range and back (Simmonds and MacLennan, 2005). The CTD also provided indication of the depth of the upper-mixed layer, where most epipelagic CPS reside during the day. The survey comprised 16 east-west transects totaling 2623 nmi, and 43 Nordic trawls, which were used for acoustic-trawl biomass estimation. The survey spanned an area from approximately Newport, OR to north of Bodega Bay, CA. The concentration of adaptive samples off Oregon and damage to the trawl gear precluded sampling to the southern extent of the survey area.