Table 1 Variable importance for how each parameter type influences the deforestation emission baseline, as well as statistical tests (one-way ANOVA and Tukey’s HSD post-hoc) comparing relative variabilities between the different levels of each parameter, for n = 2794 jurisdictions
From: Uncertainties in deforestation emission baseline methodologies and implications for carbon markets
Parameter | Variable Importance | Statistical tests | ||
|---|---|---|---|---|
Mean | SE | One-Way ANOVA | Tukey’s HSD post-hoc | |
Deforestation rate | ||||
Projection approach | 10.2 | 0.18 | F(6,19524) = 1678, p < 0.01 | All pairs p < 0.05, except global-regional_s (p = 0.1) and global_s-regional_s (p = 0.07) |
Forest dataset | 4.41 | 0.14 | F(4,12791) = 430, p < 0.01 | All pairs p < 0.05, except hansen15-30 (p = 0.97) |
Historical reference years | 2.22 | 0.16 | F(10,30749) = 82, p < 0.01 | All pairs p < 0.05, except 7-8, 8-9, 9-10, 9-11, 10-11, 10-12, 11-12, 11-13, 12-13, 12-14, 13-14, 13-15, 14-15. |
Carbon | ||||
Forest dataset | 0.74 | 0.04 | F(4,13565) = 2.5, p = 0.042 | All pairs p > 0.05. |
AGB | 1.31 | 0.04 | F(2,8379) = 8.4, p < 0.01 | gedi-soto & gedi-spawn (p < 0.01); soto-spawn (p = 0.99) |
BGB | 0.50 | 0.02 | F(2,8385) = 26.8, p < 0.01 | ipcc-soto & soto-spawn (p < 0.01); ipcc-spawn (p = 0.98) |
SOC | 0.23 | 0.02 | F(1,5594) = 0.3, p = 0.556 | esdac-olm (p = 0.556) |
