Figure Captions
Figure 1 The global distribution map of Tridacna maxima (A) and the study area map (B). The red dots represent existing occurrence records of the T. maxima worldwide, while the yellow and blue dots represent occurrence records of the East Indian Ocean-South China Sea population (EIOS) and the West Pacific-Indonesian population (WPI), respectively.
Figure 2 Results of collinearity analysis of nine predictors. Scatter collinearity analysis (A), Pearson’s correlation analysis (B), and variance inflation factor analysis (C) results for nine predictor variables. Depth - ocean depth; LandD - land distance; Cv - current velocity; Do - dissolved oxygen; Sal - salinity; Tmean – temperature mean; Trange – temperature mean; Lb - Light at bottom; Pp - Phytoplankton.
Figure 3Ecological niche differentiation. (A) Percentage of explained variance of each principal component of principal component analysis for the nine selected predictors. (B) The niches of the two populations ofTridacna maxima quantified via four-dimensional hypervolumes. To visualize the shape and boundary of the hypervolumes in two dimensions, a random selection of 20,000 stochastic points for each hypervolume was used. The large blue and orange points indicate the mean niche position (niche centroid) of EIOS and WPI, respectively. (C) Contribution of environmental predictors to each principal component (PC). The number in the bar indicates the contribution rate (%) of each predictor to PC axes, and only values>15% are shown. Depth - ocean depth; LandD - land distance; Cv - current velocity; Do - dissolved oxygen; Sal - salinity; Tmean – temperature mean; Trange – temperature mean; Lb - Light at bottom; Pp - Phytoplankton.
Figure 4Model evaluation and importance of environmental factors. Predictive abilities of the ten modeling algorithms in projecting the distribution ofTridacna maxima at the population and species levels. (A) the True Skill Statistics (TSS) value; (B) the Area Under the receiver operating characteristic Curve (AUC) value. The black horizontal lines indicate the cutoff values of the AUC (0.8) and TSS (0.7) of the single model used to build the ensemble model. (C) Relative importance of the nine predictor variables in the three ensemble models built at population and species levels. Data are expressed as mean ± standard error. Depth - ocean depth; LandD - land distance; Cv - current velocity; Do - dissolved oxygen; Sal - salinity; Tmean – temperature mean; Trange – temperature mean; Lb - Light at bottom; Pp - Phytoplankton.
Figure 5The response curves of Tridacna maxima occurrence probability against the two most important driving factors based on the population-level (A1, A2) and species-level model (including the Eastern Indian Ocean – South Sea population (B1, B2) and the Western Pacific – Indonesia population (C1, C2)). LandD - land distance; Tmean – temperature mean; Lb - Light at bottom.
Figure 6Habitat suitability maps of Tridacna maxima predicted by species and population level integrated models under current climate scenarios. Panels (A1, A2) show the corresponding continuous and binary maps for the species; panels (B1, B2) show the corresponding maps for EIOS; panels (C1, C2) show the corresponding maps for WPI.
Figure7Future predictions and changes. Habitat suitability maps ofTridacna maxima predicted by an integrated model established at both species and population levels under future climate scenarios. Panels (A1, B1, C1) are binary maps for each species, and panels (A2, B2, C2) show the predicted changes in suitable habitats for the 2100s under RCP 8.5 (0085). The category “loss” represents areas projected to be suitable under current climatic conditions but unsuitable under future climatic conditions; “stable” represents areas projected to be suitable under both current and future climatic conditions; “gain” represents areas projected to be unsuitable under current climatic conditions but suitable under future climatic conditions; and “unsuitable” represents areas projected to be unsuitable under current and future climatic conditions.
Figure 8 Analysis of Marine Protected Area (MPA) Gaps. (A) Analysis of suitable habitats for Tridacna maxima in the current climate scenario and existing MPA gaps. (B) Analysis of suitable habitats forTridacna maxima in the 2100s under the RCP 8.5 climate scenario and existing MPA gaps.
Table 1The nine environmental variables selected for this study