Table 1 Parameters of distance from ARs, sampled depth layer, and their interaction term for number of OTUs or each response variable of eDNA quantities estimated from a BIC minimum model based on generalized liner modelling.

From: Quantitative assessment of multiple fish species around artificial reefs combining environmental DNA metabarcoding and acoustic survey

 

Estimate

SE

Five dominant species

Estimateb

SEb

Estimatec

SEc

Number of OTUs

Beryx splendens eDNAa

Intercept

2.86

0.06

Intercept

3.02

0.36

-3.30

1.02

Distance

log(Distance + 0.1)

Depth layer (middle)

−0.24

0.10

Depth layer (middle)

−1.64

0.50

Distance : Depth layer

log(Distance + 0.1) : Depth layer (middle)

 

Total fish eDNA

Parapristipoma trilineatum eDNAa

Intercept

5.42

0.41

Intercept

3.01

0.57

−3.30

1.02

log(Distance + 0.1)

−0.30

0.06

log(Distance + 0.1)

−0.23

0.10

Depth layer (middle)

−1.25

0.36

Depth layer (middle)

0.55

0.84

log(Distance + 0.1) : Depth layer (middle)

log(Distance + 0.1) : Depth layer (middle)

−0.41

0.15

Demersal fish eDNA

Scomber spp. eDNA

 

Intercept

5.08

0.41

Intercept

−1.86

0.41

log(Distance + 0.1)

−0.34

0.06

log(Distance + 0.1)

−0.15

0.07

Depth layer (middle)

−1.17

0.35

Depth layer (middle)

−1.86

0.41

log(Distance + 0.1) : Depth layer (middle)

log(Distance + 0.1) : Depth layer (middle)

Pelagic fish eDNA

Pargus major eDNAa

Intercept

4.10

0.45

Intercept

3.21

0.47

−6.16

3.60

log(Distance + 0.1)

−0.23

0.06

log(Distance + 0.1)

−0.39

0.08

0.62

0.56

Depth layer (middle)

−1.34

0.39

Depth layer (middle)

−5.19

0.69

3.18

1.24

log(Distance + 0.1) : Depth layer (middle)

  

log(Distance + 0.1) : Depth layer (middle)

0.46

0.13

Trachurus japonicus eDNAa

Intercept

1.83

0.52

−3.30

1.02

log(Distance + 0.1)

−0.24

0.09

Depth layer (middle)

0.14

0.72

log(Distance + 0.1) : Depth layer (middle)

−0.31

0.13

  1. aGamma hurdle model was used. bAn estimate and error of Gamma component when using Gamma or Gamma hurdle model. cAn estimate and error of binomial component when using Gamma hurdle model.