2.3 Prey determination in harbor seal scat
The diet of harbor seals was determined by combining DNA and hard part data. The DNA prey identification and quantification were completed following the procedure outlined in Thomas et al. (2016). Briefly, for all locations the scat matrix DNA (obtained as described above) for each sample underwent a multiplex PCR using primers for a 16s mtDNA barcoding fragment (~260 bp) described by Deagle, Chiaradia, McInnes, and Jarman, (2010). Amplicons were labelled using a combination of unique F and R primer tags, in addition to indexed, post-PCR ligated Illumina TruSeqâ„¢ adapter sequences (for details see Thomas et al. 2016). An Illumina MiSeq was then used to sequence the amplified DNA fragments. Lastly, a custom BLAST database comprised of publicly available reference sequences specific for known prey species was used to produce identifications to the lowest taxonomic level possible for each amplified sequence.
Extraction and preparation of prey hard parts were completed by Thomas et al. (2017) for Belle Chain, Comox, Cowichan Bay, and Fraser River 2012 and 2013 samples, by one of the authors (BAN) for Cowichan Bay 2014, and by the first author (MRV) for Baby Island samples. Each scat was placed in a set of nested sieves, and then rinsed and stirred until all that was left in the sieves were prey hard parts. All hard parts, except cephalopod beaks, were transferred to 20 ml scintillation vials with 70% ethanol. They were allowed to sit for a minimum of two weeks before the liquid was poured off and the hard parts were allowed to dry. The cephalopod beaks were transferred to separate 20 ml scintillation vials with ethanol. All diagnostic prey hard parts were identified to the lowest taxonomic level possible using reference sets of prey bones from Washington and British Columbia by Thomas et al. (2017) for Belle Chain, Comox, Cowichan Bay, and Fraser River, and by collaborators at Long Live The Kings for Baby Island samples. Published keys for both fish bones and cephalopod beaks were used as described in Thomas et al. (2017). Notably, this analysis allowed differentiation between the proportion of adult and juvenile Oncorhynchus spp. consumed. The percentage of juvenile versus adult salmon was determined using the method described in Thomas et al. (2017).