Appendix D. One figure and four ANOVA tables for damage to shrub leaves and stems during fires.
|
FIG. D1. Effects of fuel treatments on probability of damage to shrub leaves during fires in 1999 and 2001. Overall bar height indicates mean probability of visible damage (>10% of leaves consumed); shaded or patterned area indicates probability that observed damage was complete (>90% consumed). Values are least-square means + SE of probabilities, back-transformed from logits. |
TABLE D1. Results of generalized linear mixed-model analysis of probability of visible damage to shrub leaves.
| Source of variation | |||
| Fixed Effects: | |||
| Elevation | |||
| Fuel | |||
| Elevation × Fuel | |||
| Year | |||
| Elevation × Year | |||
| Fuel × Year | |||
| Elevation × Fuel × Year | |||
| Linear Contrasts (Fuel): | |||
| (Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition) | |||
| Controls vs. Fuel Removal | |||
| Pine Needle Addition vs. Wood Addition | |||
| Linear Contrasts (Fuel by Year): | |||
| [(Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition)] × (Post-fire 1 vs. Post-fire 2) |
|||
| [(Controls vs. Fuel Removal) × (Post-fire 1 vs. Post-fire 2) | |||
| (Pine Needle Addition vs. Wood Addition) × (Post-fire 1 vs. Post-fire2) |
Notes: Error term for Elevation = Elevation × Transect (Ravine). Error term for Fuel and interactions = Quadrat (Fuel × Elevation × Transect (Ravine)). Error term for Year and interactions = residual.
Denominator degrees of freedom based on Satterthwaite approximation. Significance level is P < 0.025.
TABLE D2. Results of generalized linear mixed-model analysis of probability of complete damage to shrub leaves.
| Source of variation | |||
| Fixed Effects: | |||
| Elevation | |||
| Fuel | |||
| Elevation × Fuel | |||
| Year | |||
| Elevation × Year | |||
| Fuel × Year | |||
| Elevation × Fuel × Year | |||
| Linear Contrasts (Fuel): | |||
| (Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition) | |||
| Controls vs. Fuel Removal | |||
| Pine Needle Addition vs. Wood Addition | |||
| Linear Contrasts (Fuel by Year): | |||
| [(Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition)] × (Post-fire 1 vs. Post-fire 2) |
|||
| [(Controls vs. Fuel Removal) × (Post-fire 1 vs. Post-fire 2) | |||
| (Pine Needle Addition vs. Wood Addition) × (Post-fire 1 vs. Post-fire2) |
Notes: Error term for Elevation = Elevation × Transect (Ravine). Error term for Fuel and interactions = Quadrat (Fuel × Elevation × Transect (Ravine)). Error term for Year and interactions = residual.
Denominator degrees of freedom based on Satterthwaite approximation. Significance level is P < 0.025.
TABLE D3. Results of generalized linear mixed-model analysis of probability of visible damage to shrub stems.
| Source of variation | |||
| Fixed Effects: | |||
| Elevation | |||
| Fuel | |||
| Elevation × Fuel | |||
| Year | |||
| Elevation × Year | |||
| Fuel × Year | |||
| Elevation × Fuel × Year | |||
| Linear Contrasts (Fuel): | |||
| (Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition) | |||
| Controls vs. Fuel Removal | |||
| Pine Needle Addition vs. Wood Addition | |||
| Linear Contrasts (Fuel by Year): | |||
| [(Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition)] × (Post-fire 1 vs. Post-fire 2) |
|||
| [(Controls vs. Fuel Removal) × (Post-fire 1 vs. Post-fire 2) | |||
| (Pine Needle Addition vs. Wood Addition) × (Post-fire 1 vs. Post-fire2) |
Notes: Error term for Elevation = Elevation × Transect (Ravine). Error term for Fuel and interactions = Quadrat (Fuel × Elevation × Transect (Ravine)). Error term for Year and interactions = residual.
Denominator degrees of freedom based on Satterthwaite approximation. Significance level is P < 0.025.
TABLE D4. Results of generalized linear mixed-model analysis of probability of complete damage to shrub stems.
| Source of variation | |||
| Fixed Effects: | |||
| Elevation | |||
| Fuel | |||
| Elevation × Fuel | |||
| Year | |||
| Elevation × Year | |||
| Fuel × Year | |||
| Elevation × Fuel × Year | |||
| Linear Contrasts (Fuel): | |||
| (Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition) | |||
| Controls vs. Fuel Removal | |||
| Pine Needle Addition vs. Wood Addition | |||
| Linear Contrasts (Fuel × Year): | |||
| [(Controls + Fuel Removal) vs. (Pine Needle Addition + Wood Addition)] × (Post-fire 1 vs. Post-fire 2) |
|||
| [(Controls vs. Fuel Removal) × (Post-fire 1 vs. Post-fire 2) | |||
| (Pine Needle Addition vs. Wood Addition) × (Post-fire 1 vs. Post-fire2) |
Notes: Error term for Elevation = Elevation × Transect (Ravine). Error term for Fuel and interactions = Quadrat (Fuel × Elevation × Transect (Ravine)). Error term for Year and interactions = residual.
Denominator degrees of freedom based on Satterthwaite approximation. Significance level is P < 0.025.