Summary:
Dr. David Welch submits this research proposal to report an important addition to his understanding of where sockeye smolt mortality occurs. In prior testimony at the commission, he reported that most mortality occurred after passing the northern end of Vancouver Island, but he has since re-analyzed his previously collected data to directly compare survival rates of acoustically tagged sockeye smolts migrating in the Strait of Georgia and then in Discovery Passage/Queen Charlotte Strait and found a level of higher mortality that may explain the 10-fold decline in Fraser sockeye survival seen since 1990. To address these issues, Dr. Welch and his colleagues designed a new study building off the results from the POST prototype array.
Salmon smolts migrate quickly after entering the ocean, typically at about 1 body-length per second. For 10 cm wild sockeye smolts, this means that they move about 9 km per day, resulting in the smolts being far removed from a location where disease transfer might occur within only a very few days. This complicates the interpretation and analysis of data from smolts collected for disease studies in the vicinity of fish farms, because the prior exposure history of a collected smolt is unknown. In addition, smolts debilitated from disease simply may disappear before capture because predators target weakened individuals prior to pathological symptoms (such as lesions) being expressed.
To address these issues, my colleagues (Drs. Scott Hinch, Kristi Miller, Brian Riddell, Tony Farrell, Carl Schwartz) and I designed a new study building off the results from the POST prototype array. The draft proposal is submitted for the Commission’s (& public’s) information as part of this commentary. However, the primary reason for this submission is to report an important addition to our understanding of where sockeye smolt mortality occurs. In prior testimony at the Commission I reported that most mortality occurred after passing the northern end of Vancouver Island. This statement remains correct. However, we have since re-analyzed our previously collected data to directly compare survival rates of acoustically tagged sockeye smolts migrating in the Strait of Georgia and then in Discovery Passage/Queen Charlotte Strait.
The results are summarized in Figure 4 on page 8 of the attached submission. In 5 of 6 years of study, survival rates per week of migration were substantially lower in the Discovery Passage region than in the Strait of Georgia. (2011 data are preliminary, as data from some receivers will not be collected until early January 2012).
This is a very important finding that clarifies and extends my earlier testimony:
1) The results continue to contradict the theory that 2007 sockeye smolts died in the Strait of Georgia, before reaching the area with fish farms; in 2007 smolt survival was substantially lower in the northern area.
2) The lower survival rate the northern area (currently estimated at roughly 2/3rds of the Strait of Georgia survival rate when averaged across all years) would have profound effects on marine survival if prolonged; after 5 weeks smolts would be reduced to only ~1/10th the number that would survive in the Strait of Georgia.
3) This level of higher mortality would be sufficient to fully explain the 10-fold decline in Fraser sockeye survival seen since 1990.
4) We caution that this new result remains a correlation, not proof that the fish farms present in the northern area caused the reduced survival, because the two regions probably differ in other ways (more abundant predators are likely present in the northern region, for example).
5) As the technical calculations in the full proposal demonstrate, the survival difference observed between freely migrating smolts in the northern and southern areas would be easily measurable with an appropriately re-designed array.
6) As discussed during my earlier testimony at the Cohen Commission, the “gold standard” for scientific research is a blinded experimental test comparing the survival of smolts exposed to fish farms to that of smolts not exposed (the later act as the control group, and are treated identically except for exposure to fish farms).
7) We estimate that fully implementing all aspects of the attached experimental design, including additional genomic & physiological analyses to assess response to fish farm exposure, would cost $3M~$3.5M per year. We believe that the study would need to be annually repeated for 3-5 years to provide a definitive answer as to whether fish farms elevate mortality sufficiently to justify regulatory action by government.
8) The scientific community (and thus the Cohen Commission) has had substantial difficulty in making reasonable conclusions owing to the lack of relevant data. The approach outlined in the proposal should both (i) resolve the key regulatory question for government of whether fish farm effects are large enough to warrant regulation and (ii) provide critically needed baseline information on when and where marine survival is determined.
9) The final appendix provides a list of letters of support for the approach, from members of the scientific community not having a direct interest in the research program.
David Welch, Ph.D.
Kintama Research Services, Ltd.
Nanaimo, B.C.
http://www.cohencommission.ca/en/submissions/ViewASubmission.php?sub=1127
