TABLE 3.
Top linear regression models (ΔAICc < 2) for explaining tuber density and tuber biomass at Shahu Lake, a sublake of Poyang Lake, China
| ID | Models | df | AICc | ΔAICc | ωi |
|---|---|---|---|---|---|
| Tuber density | |||||
| 1 | Last winter temperature + spring water level | 4 | 126.12 | 0.00 | 0.13 |
| 2 | Last winter temperature + autumn temperature + spring water level | 5 | 127.19 | 1.07 | 0.08 |
| 3 | Last winter temperature + summer water level | 4 | 127.67 | 1.55 | 0.06 |
| 4 | Spring water level | 3 | 127.88 | 1.76 | 0.06 |
| Tuber biomass | |||||
| 1 | Last winter temperature + autumn temperature | 4 | 107.16 | 0.00 | 0.14 |
| 2 | Last winter temperature + autumn temperature + summer water level | 5 | 108.90 | 1.74 | 0.06 |
| 3 | Last winter temperature + autumn temperature + spring temperature | 5 | 108.94 | 1.78 | 0.06 |
Abbreviations: AICc, Akaike information criterion for small sample size; df, degrees of freedom; ΔAICc, the difference between each model's AICc and the lowest AICc; ωi, Akaike weights.