Parentage-based tagging improves escapement estimates for ESA-listed adult Chinook salmon and steelhead in the Snake River basin

Parentage-based tagging (PBT) is a nonlethal, genetic tagging method that has been successfully applied in hatchery-supplemented populations to manage hatchery brood stock and monitor hatchery harvest and straying rates. We show that PBT can also improve the accuracy of escapement estimates by significantly reducing the number of hatchery-origin fish falsely classified as natural-origin. Unlike conventional abundance estimates, which use physical marks and tags to distinguish hatchery individuals from their wild counterparts, PBT identifies origin independent of physical form. We applied PBT to populations of Chinook salmon (Oncorhynchus tshawytscha) and steelhead (Oncorhynchus mykiss), which are classified as Threatened under the Endangered Species Act and subject to extensive hatchery supplementation efforts. For spawn years 2014–2018, 16 511 adipose-intact Chinook salmon and 21 953 adipose-intact steelhead were sampled, and PBT identified 19.6% of returning Chinook salmon and 8.3% of steelhead were of hatchery-origin, despite having no physical or mechanical marks. The 90% confidence intervals for escapement estimates of natural-origin Chinook salmon and steelhead made with and without corrections using PBT were nonoverlapping for nine of ten comparisons, indicating that failing to account for unmarked, untagged hatchery-origin fish would result in a significant overestimation of natural abundance.

returning to the Snake River basin to identify source of origin (hatchery-or natural-origin).

172
Using biological data and genetic samples collected from adults trapped at Lower Granite Dam, 173 Washington, we quantified the number of putatively natural-origin fish (adipose-intact) without     (Table 1). Ad-intact fish were scanned for a CWT and inspected for a   retained parentage assignments we checked for discrepancies between the phenotypic/genotypic 309 sex of parents as well as assignment to unexpected brood years to ensure the accuracy of inferred 310 parent-offspring relationships. Unexpected brood years would be those that fall out of the known 311 age distribution for hatchery-reared Chinook Salmon and Steelhead (e.g., <2 or >5 years of age).

312
Returning adults were ultimately decomposed by rearing type as either hatchery (H), 313 natural-origin (N), or adipose-intact hatchery (HNC). We used results from parentage analysis to 314 estimate the rearing type of natural-origin fish. In other words, the expectation was that natural-315 origin fish would fail to assign to hatchery broodstock. Returning adults that were confidently 316 assigned to hatchery broodstock were identified as hatchery-origin and flagged as HNC to denote 317 the sample represented an unmarked, untagged hatchery fish.

318
To test whether PBT would falsely identify natural-origin fish as hatchery-origin (false     Figure 3A). Detection of coded wire tags were the primary method of identifying Chinook 391 Salmon as hatchery-origin without the use of PBT, with only a small fraction classified using 392 physical marks (Figure 4A). 393 Estimates of natural-origin, adult Chinook Salmon escapement calculated using SCOBI 394 without PBT were higher than those with the use of PBT across all spawn years (mean = 21,357,  Figure 3B). Hatchery-origin Steelhead identified without the use of PBT were 415 identified using marks, fin erosion (e.g., dorsal fin erosion) and coded-wire tags (Figure 4B). 416 The observation of marks other than coded wire tags was much more common for Steelhead than 417 for Chinook Salmon.

418
Estimates of escapement for adult natural-origin Steelhead calculated with and without 419 the use of PBT were less variable than for Chinook Salmon ( Table 3). Natural-origin 420 escapement estimates generated without PBT were higher than those with the use of PBT across Salmon. As a result of this species-specific mark, PBT was responsible for identifying a far 520 greater number of hatchery-origin Chinook Salmon without a mark or tag than Steelhead (Table   521 2). That we observed annual variation in the number of unmarked, untagged adults identified as      mykiss) determined to be hatchery-origin based on coded wire tags or marks. Marks included fin clips (e.g., left ventral) in either Chinook Salmon or Steelhead and dorsal fin erosion in Steelhead only. The yellow bars are relatively high for Steelhead since dorsal fin erosion is a common mark to identify hatchery-origin fish for this species, but it is not an effective mark for Chinook Salmon.
D r a f t D r a f t Table 3. Model estimates of total escapement of natural-origin spring/summer Chinook Salmon (Oncorhynchus tshawytscha) and Steelhead (O. mykiss) entering the Snake River basin made with and without the application of PBT to determine source of origin among adipose-intact returning adults (outputs from SCOBI). Window counts represent the raw number of all adults passing the observation window at Lower Granite Dam. LCI and UCI correspond to lower and upper 90% confidence intervals on estimates of escapement. Escapement difference and the percent change columns display the inflated proportion of returning natural-origin adults without the use of PBT. Values in the escapement difference column marked with an asterisk represent years in which the 90% confidence intervals for escapement estimates with and without PBT were non-overlapping.