Figure legends
Figure 1. Above-ground carbon (A) and soil carbon accumulation (B) in chronosequences of tree plantations and second-growth forests, and reference forest values, in agricultural landscapes of Brazil’s Atlantic Forest. Solid and dashed lines represent, respectively, significant and non-significant power regressions. ANCOVA on log-log transformed data indicated a significant difference (P < 0.05) of estimated intercepts between plantations and second-growth forests for above-ground carbon biomass (A) and soil carbon stocks (B).
Figure 2. Soil carbon association with clay content in tree plantations, second-growth forests, and reference forests within agricultural landscapes of Brazil’s Atlantic Forest. Solid lines represent significant power regressions. Soil carbon stocks versus clay content relationship: ANCOVA on log-log transformed data indicated a significant difference (P<0.05) of estimated intercepts between restored and reference forests, and between plantations and naturally regenerated forests.
Figure 3. Graphical representation of the influence of different drivers of above-ground carbon and soil carbon stocks in chronosequences of tree plantations and second-growth forests established in agricultural landscapes of Brazil’s Atlantic Forest. Results reflect the average model developed by merging all models ∆AICc ≤ 2. Arrow color represent the sign of the average coefficient, and relative importance express the weighted proportion of the models ∆AICc ≤ 2 that contain the driver.
Figure 4. Temporal variation of restoration implementation costs (A), accumulated land opportunity costs (B), and total (C) costs for accumulating above-ground carbon stocks (i.e., cost-effectiveness) in chronosequences of tree plantations and second-growth forests established in agricultural landscapes of Brazil’s Atlantic Forest. Solid and dashed lines represent, respectively, significant and non-significant power regressions. ANCOVA on log-log transformed data indicated a significant difference (P < 0.05) of estimated intercepts between plantations and second-growth forests for implementation cost-effectiveness (A) and total cost-effectiveness (C).