In 2008, the Portland District of the U.S. Army Corps of Engineers (USACE) contracted with the Pacific Northwest National Laboratory (PNNL) to conduct an acoustic telemetry study to estimate the survival rates of juvenile Chinook salmon and steelhead passing through Bonneville Dam (BON) and its spillway. Of particular interest was the relative survival rate of smolts detected passing through end spill bays 1-3 and 16-18, which had deep flow deflectors immediately downstream of spill gates, versus the survival rate of smolts passing through middle spill bays 4-15, which had shallow flow deflectors.Yearling Chinook salmon (YC), steelhead (STH), and fall Chinook salmon (FC) longer than 95 mm were collected from routine smolt monitoring samples at John Day Dam (JDA) and held overnight before surgery so that they were not overly stressed. Smolts longer than 95 mm were surgically implanted with Juvenile Acoustic Telemetry System (JSATS) and passive integrated transponder (PIT) tags and held another night to allow time for fish to recover from surgery. Fish tagged the previous day were released by a PNNL team three times per day (morning, midday, and night) in the JDA pool near Arlington, Oregon, and about 2.5 km below JDA in spring and summer. The team also released FC about 3 km below The Dalles Dam (TDA) in summer. Releases of live JSATS-tagged smolts in the Columbia River upstream of BON totaled 3431 YC and 3430 STH in spring. In summer, 5909 FC were released upstream of BON. Releases were spread out over 28 consecutive days (April 29 through May 27) during spring and over 29 consecutive days (June 15 through July 13) in summer. All of the "treatment" fish released above BON had the opportunity to be detected and regrouped by a BON forebay entrance array of autonomous nodes or by dam-face arrays at the BON spillway (this study) or BON Powerhouse 2 (B2) ) before they passed the BON project. An array is a group of underwater listening devices deployed to detect acoustic tags in fish passing through a forebay or an entire cross section of the river above or below a dam. Non-spillway passage routes were assigned based on detections of PIT tags in the B2 Corner Collector (B2CC) and B2 Juvenile Bypass System (B2 JBS) and acoustic detections on a cabled JSATS array in the B2 forebay (from Faber et al. 2009). Routes of spillway passage included end bays 1-3 and 16-18 with deep flow deflectors downstream of spill gates and middle bays 4-15 with shallow flow deflectors. Powerhouse 1 was not monitored.A National Marine Fisheries Service (NMFS) team released 826 live and 50 dead tagged YC in spring and 1020 live and 52 dead tagged FC smolts in summer into the downstream end of the BON tailrace near the USACE boat launch three times per day (about 0600, 1300, and 2100 hours PST) to serve as reference releases (controls) for virtual releases of treatment fish from upstream of the dam. Reference releases were made daily from April 30 through June 2 and from June 16 through July 22. The NMFS team also released 826 live YC and 1020...