SEARCH

SEARCH BY CITATION

Figure S1. Simulated and measured gross ecosystem production (GEP) for the 6 flux tower sites. Confidence intervals are generated from the 1000 parameter set combinations and following simulations.

Figure S2. Simulated and measured ecosystem respiration (Reco) for the 6 flux tower sites. Confidence intervals are generated from the 1000 parameter set combinations and following simulations.

Figure S3. Spatial pattern of changes in soil carbon from 1981/00 mean reference period and 2081–2098 projection for all 8 climate models (SRES A2 emissions storyline).

Figure S4. Spatial pattern of ranked partial correlation coefficients (RPCC) for all parameters used in full Basin-wide simulations. The first column show the changes in soil carbon stocks between 2001/10 reference period and 2091/98 projection period for the standard parameter set (Std), the average of all parameter sets (Avg), and the standard deviation of soil carbon changes for all parameter sets (stdv). The first row corresponds to parameters related to carbon allocation, the second row for photosynthesis, and the third row contains parameters related to water use. The RPCC correlations are shown for the Hadley CM3 and GISS ER climate projections.

Figure S5. Spatial pattern of ranked partial correlation coefficients (RPCC) for all parameters used in full Basin-wide simulations. The first column show the changes in tropical evergreen cover between 2001/10 reference period and 2091/98 projection period for the standard parameter set (Std), the average of all parameter sets (Avg), and the standard deviation of tropical evergreen cover changes for all parameter sets (stdv). The first row corresponds to parameters related to carbon allocation, the second row for photosynthesis, and the third row contains parameters related to water use. The RPCC correlations are shown for the Hadley CM3 and GISS ER climate projections.

Table S1. List of parameters, original value, and parameter ranges, used in flux site simulations and Basin-wide simulations (boldfaced).

Table S2. Ranked partial correlation coefficients for individual flux sites (fulfilling criteria of being greater than a value of 0.2). The sites are ordered from left to right in terms of dry season length. A positive correlation indicates that variability in vegetation carbon storage is positively correlated with changes in the parameter value. These parameters were selected for the full Basin-wide simulation parameter set.

Table S3. Ranked partial correlation coefficients for individual flux sites (fulfilling criteria of being greater than a value of 0.2). The sites are ordered from left to right in terms of dry season length. A positive correlation indicates that variability in soil carbon storage is positively correlated with changes in the parameter value. These parameters were selected for the full Basin-wide simulation parameter set.

Table S4. Ranked partial correlation coefficients for individual flux sites (fulfilling criteria of being greater than a value of 0.2). The sites are ordered from left to right in terms of dry season length. A positive correlation indicates that variability in annual GPP storage is positively correlated with changes in the parameter value. These parameters were selected for the full Basin-wide simulation parameter set.

Table S5. Ranked partial correlation coefficients for individual flux sites (fulfilling criteria of being greater than a value of 0.2). The sites are ordered from left to right in terms of dry season length. A positive correlation indicates that variability in tropical evergreen cover is positively correlated with changes in the parameter value. These parameters were selected for the full Basin-wide simulation parameter set.

Table S6. Ranked partial correlation coefficients for changes in soil carbon (1981/00 to 2081/98) for the Eastern and Western Amazon Basin. The RPCC values must be interpreted in the context of the directly change of soil carbon for each model, these are illustrated in Fig. S3.

Table S7. Ranked partial correlation coefficients for changes in tropical evergreen cover (1981/00 to 2081/98) for the Eastern and Western Amazon Basin. The RPCC values must be interpreted in the context of the directly change of tropical evergreen cover for each model, these are illustrated in Fig. 5a.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
GCB_2157_sm_suppinfo.doc1016KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.