Table 3. Regression coefficients and comparison between the five hypotheses used to explain the distribution of fish species richness in streams.
Hypotheses | r2 | p | AIC | Δ AIC | Moran’s I | r2 for partial regression results | |||
---|---|---|---|---|---|---|---|---|---|
A.B | A:B | B.A | 1-(A+B) | ||||||
Temporal Heterogeneity | 0.401 | <0.001 | 4510.500 | 0.000 | 0.093 | 0.031 | 0.054 | 0.316 | 0.599 |
Energy | 0.376 | <0.001 | 4535.200 | 24.700 | 0.115 | 0.007 | 0.090 | 0.279 | 0.624 |
Area | 0.301 | <0.001 | 4538.400 | 27.900 | 0.122 | 0.005 | 0.053 | 0.316 | 0.626 |
Water-Energy | 0.376 | <0.001 | 4538.900 | 28.400 | 0.114 | 0.007 | 0.143 | 0.226 | 0.624 |
Neutral | 0.369 | <0.001 | 4539.900 | 29.400 | 0.128 | - | - | - | 0.631 |
Terrestrial Productivity | 0.370 | <0.001 | 4541.600 | 31.100 | 0.131 | 0.001 | 0.001 | 0.370 | 0.628 |
r2—Coefficient of determination; p—Type one error probability; AIC—Information criterion of Akaike; Δ AIC—Akaike variation; Moran’s I—Autocorrelation index of Moran for variable; A.B—Environmental component; A:B—Shared Component; B.A—Spatial Component; 1-(A+B)–Residual