Supplementation 1: Fish growth during the experiments in 2015
To evaluate the effect of fish growth during our experiments, we measured fork length (FL) and body weight (BW) of each 1+ fish before and after the experiments in 2015 (June 11th and July 11th, respectively). They showed 11.5% increase in FL and 40.6% increase in BW for the Age 1+ group (BEFORE: FL =95.7±15.5 mm, BW =9.8±1.7 g,AFTER: FL =106.7±14.2 mm, BW =13.8±2.9 g, n = 32).T–test suggested significant growth in their body size (p 0.001 for FL and BW).
Since the fish growth might have had a significant influence in our estimation, we also compared environmental DNA (eDNA) concentrations (per fish in 2µL template DNA samples) in water samples from June 21st and that from July 11th, 2015 (the first and the last aquarium experiment, respectively, in 2015). The eDNA concentration showed a slight difference between them, but the differencewas statistically non-significant (average eDNA concentrations were4.36±1.34 copies fish-1 and5.20±1.37 copies fish-1 in June 21st and July 11th, respectively. n = 4, p = 0.22).
In comparison, there were also significant difference in BWof age > 20+ adults in 2015 and FL and BW of age 2+ juveniles in 2016 during the rearing experiments (p < 0.01), but thesewere rather limited(> 20+: 0.3% increase in FL, 8.3% decrease in BW. 2+: 3.9% increase in FL, 12.4% increase in BW).
Supplemental Table 1. Body size information of age 1+ juveniles in 2015. The same 32 individuals were measured in June 11th and July 11th that were the before starting and after finishing experiments, respectively.Data is shownin the ascending orderfor FL in each measuring day and individual data on BW corresponds to that on FL.
Supplementation 2: Comparison of the eDNA concentration between tank water samples from three days and five days after fish introduction
For the validation of the equilibrium state in our experiments, we compared the eDNA concentration between samples from three days and five days after age 1+ fish introduction in June 19th and 21st, 2015, respectively. Two, four, eight and 16 individuals were introduced into four separate 40L-tanks and rearing water was kept pouring and overflowing in each tank. Procedures for collecting water samples, DNA extractions and qPCRs were exactly same as our experiments mentioned in the main text. Results of qPCRs were converted to the eDNA concentration per fish in 2µL template DNA samples. T–test was employed for checking the equilibrium state with a null hypothesis that there was no difference in the eDNA concentration between tank water samples from three days and five days after age 1+ fish introduction.
The average eDNA concentrationsper fish were very similar between samples from three days and five days after fish introduction(20.64±4.09copies and 29.51±4.83 copies, respectively.Both n=4). T–test failed to reject the null hypothesis (p = 0.15), indicating that the eDNA concentration of the tank water reached plateau after three dayssince fish was introduced in the tank.
Supplementation 3: List of checked species and these accession numbers
Supplemental Table 2Sequence information from 52 species of salmonid fish species on the NCBI nucleotide sequence database were used for our designing ofSakhalin taimen species–specific primers and probe.
Supplementation 4: Sample sizes
First fish were introduced into each tank in June 11th, 2015 and July 11th, 2016. After five days and three days from fish introduction, two 1L water samples were collected from each tank in2015 and 2016, respectively. We repeated these samplings according to the time tables in 2015 and 2016 with changing the fish density and tanks (Supplemental Tables 3, 4).
All filter samples were applied to qPCR with triplicate, and average eDNA concentration data in each tank was used for our analyses. Sample sizes for each age and tank were; 0+ juvenile in 40L-tank = 1, 0+ juvenile in 2000L-tank = 1, 1+ juveniles in 40L-tank = 21, 2+ juveniles in 40L-tank = 20, 2+ juveniles in 2000L-tank = 14, >20+ adults in 2000L-tank = 10, total = 67. However, there were no detections in the samples that were collected from 2000L-tank with a 0+ juvenile, so we finally used 66 tank samples for our analyses (0+ in 40L-tank = 1, 0+ in 2000L-tank = 0, 1+ in 40L-tank = 21, 2+ in 40L-tank = 20, 2+ in 2000L-tank = 14, >20+ in 2000L-tank = 10).
SupplementalTable 3Sampling scheme in 2015. First fish were introduced on June 11th. After fish introduction, we collected water samples in five-day intervals. All 40L-tanks and 2000L-tanks were for age 1+ juveniles and >20+ adults, respectively.
SupplementalTable 4Sampling scheme in 2016. First fish were introduced on July 11th. After fish introduction, we collected water samples in three-day intervals for age 2+ juveniles. From July 11th–August 4th, all 40L-tanks and 2000L-tanks were only for 2+ juveniles. After sampling on August 4th, we removed2+ juveniles from all tanks and introduced one age 0+ juvenile to one 40L-tank and 2000L-tank in 6 days.
1