Figure 1.
The study site, sampling stations and habitats at Sandy Bay, Ningaloo Reef, Western Australia.
Panels show (a) sampling stations 1 to 7 across the reef (see Wyatt et al. [23] for more details on the location and hydrodynamics); (b) reef zonation and bathymetry between locations derived from hyperspectral imagery; and (c) representative images of the benthic habitat demonstrate the shift from the tabulate hard coral dominated reef flat (station 1), to patch reef (3) and sandy lagoon (5).
Figure 2.
Fish tissue isotopes compared to gut content isotopes.
Data shows isotope ratios at the time of sampling for (a) nitrogen (δ15N, n = 126) and (b) carbon (δ13C, n = 111). Solid lines represents significant linear regression relationships for nitrogen (Tissue δ15N = 0.325×Gut δ15N+7.535; r2 = 0.138; F[1,124] = 19.93, p<0.001) and carbon (Tissue δ13C = 0.547×Gut δ13C−6.472; r2 = 0.583; F[1,109] = 152.2, p<0.001).
Table 1.
Average tissue δ15N, ΔN between tissue and gut δ15N, tissue δ13C, and ΔC between tissue and gut δ13C (all in ‰) for individuals from 26 species, showing stations and number collected (n).
Figure 3.
Discrimination factors versus tissue isotopic composition.
Data is shown for (a) nitrogen (ΔN vs. δ15N, n = 126) and (b) carbon (ΔC vs. δ13C, n = 111 ). Solid line represents a significant linear regression relationship for nitrogen (ΔN = 0.574×δ15N–3.32; r2 = 0.226; F[1,124] = 36.23, p<0.001). There was no relationship for carbon (r2 = 0.005; F[1,109] = 0.0569, p = 0.452).
Figure 4.
Discrimination factors by trophic group.
Data is shown for herbivorous, planktivorous, carnivorous and detritivorous trophic groups in terms of (a) nitrogen (ΔN) and (b) carbon (ΔC). Data are mean ± s.e. (n as per Table 1); * and ** denote significantly different trophic groups (no significant seasonal differences or season×trophic group interactions).
Figure 5.
Seasonal variation in isotope discrimination factors.
Plots show average nitrogen discrimination factors (ΔN, bars, left axis), tissue δ15N (solid circle, right axis) and gut δ15N (hollow circle, right axis) during May and Nov 2008 for (a) Pomacentrus moluccensis and (b) Acanthurus triostegus from 1 and 6a; (c) Parupeneus signatus sampled from 2, 3, 4, 6 and 7; and (d) Chromis viridis sampled from station 1. All ΔN and gut δ15N differences significant, no significant tissue δ15N differences or sample period×location interactions. See Figure 1 for locations. Fish images obtained from Froese & Pauly [39]; copyright J.E. Randall, 1997, used with permission. Data are mean ± s.e. (n as per Table 1).