Salish Sea Marine Survival Project

40 97% 79% 55% 71% 84% 66% 50% 29% 25% 0% 20% 40% 60% 80% 100% 0 10 20 30 40 50 ESTIMATED SURVIVAL MIGRATION DISTANCE (KM) HATCHERY 2017 - High Flow HATCHERY 2016 - Low Flow DOWNSTREAM SURVIVAL Freshwater Survival Several downstream survival estimates for Chinook were created from RST and PIT tag data collected in 2014- 2017. For 2015, the RST survival estimate for hatchery fish from the lake release to the bottom of the river was 14%, and for wild fish was only 27%. These results indicate significant in-river mortality. Historically, in the absence of in-river mortality estimates, this mortality was included in marine survival estimates. This would result in an under-estimate of the 'true' marine survival estimates of this important stock. In 2016, groups of Chinook smolts were tagged and released at multiple locations along the river to deter- mine if survival improves with shorter migrations (Figure 3). Survival rates were again very low from the upper river (approx. 25%) and showed a linear decline along the river (Figure 4) for both wild and hatchery fish. In 2017, in-river survival was much higher for Chinook (71%). River discharge was six times higher during 2017 compared to 2016 (Figure 5). Thus, there appears to be a relationship with river discharge. The study was repeated one more time in 2018 under moderate flow conditions and losses were again found to be significant from upper river sites (Figure 6a). Two independent detection locations 5 km apart showed similar results. Survival data from each release was also converted to an estimate of abundance (fish remaining) at each site from an initial release of 1,000 (Figure 6b). Losses in the upper portion of the river (Road Pool to Stoltz, ~30 km) were found to be 10X higher per km than through the lower 17 km. In-River Predation Studies 2017 In April 2017, 5,494 hatchery Chinook were PIT tagged and released at various locations along the Cowichan River. PIT Tag antennas (with motion-activated trail cameras) were set up at four investigation sites to detect tagged fish consumed by predators (e.g., North American River Otter, North American Raccoon, Pacific Pine Marten and Common Merganser). PIT tag antennas were monitored for six weeks, post-release. Based on photo and video images, a minimum of 197 predators crossed through an operational PIT tag detection field but not a single tag was detected in any predator. However, PIT tag scans were completed underneath the Pacific Great Blue Heron rookery at Cowichan Bay (100 nests in 2017) and 428 PIT-tags were detected in the guano. Pit tags from all four years of tagging in the Cowichan River were present. Over 50% of the recovered tags were from 2016, a very low-flow year (Figure 7). In total, 71% of the PIT tags at the Cowichan Bay Heron rookery were from hatchery-reared Chinook smolts— and 90% of the recovered tags were from smolts released in the river. Figure 3. Locations of smolt release along Cowichan River for assessment of river segment mortality. Figure provided by Kevin Pellett, DFO. Figure 4. Survival rates for wild (2 size bins) and hatchery Chinook smolts from different release sites along Cowichan River, 2016. PIT tagged fish are recorded when they pass over the Biomark array at the bottom of the river (River Km 0). Figure provided by Kevin Pellett, DFO. Figure 5. Relationship between estimated freshwater survival rates for hatchery fish in Cowichan River during a low flow (2016) and higher flow (2017) year. Figure provided by Kevin Pellett, DFO. STRUCTURED TAG RELEASES 0 km 5 km 10 km Road Pool (1501H, 641W) 70.2 Trestle (604H, 862W) Skutz Falls (726H, 604W) Stoltz Pool (597H, 872W) Vimy Road (599H, 817W) Parkers Biomark Array (Allenby) MJ Channel South Arm North Arm 47 km 40 km 34 km 25 km 13 km 7 km 0 km 101% 73% 67% 47% 29% 84% 66% 50% 29% 25% 89% 64% 41% 34% 14% 0% 20% 40% 60% 80% 100% 0 5 10 15 20 25 30 35 40 SURVIVAL MIGRATION DISTANCE (KM) Wild 60-90 mm Hatchery 60-90 mm Wild 50-59 mm Predation rate estimates in 2016 were 5.4% and 1.9% on hatchery and wild fish, respectively. Heron predation rates are likely underestimated as herons typically defecate as they take flight. The relationship with low flows is likely due to the fact that herons cannot forage successfully at a maximum depth over ~0.5m.

• Cover