SEARCH

SEARCH BY CITATION

FilenameFormatSizeDescription
geb12075-sup-0001-si.docx7601K

Figure S1 Growth (left) and mortality (right) survey plots from the NFIs used to calibrate growth and mortality models, respectively. Growth rates correspond to trees with d.b.h. < 200 mm at NFI2 and are shown as d.b.h. differences between NFI3 and NFI2 averaged per plot. Density growth functions are also shown for each species, x-axis represent the growth of the species in mm yr−1 and y-axis the frequency per pixel. Mortality measurements recorded in NFI3 (red) are shown together with the current species distributional (grey) taken from the Euforgen database (http://www.euforgen.org/distribution_maps.html).

Figure S2 Predicted growth (mm yr−1) versus predicted mortality (normalized between 0 and 1) in mainland Spain (green). The range of the species computed by the maximization of the TSS of the growth and mortality models is delimited by the blue area.

Figure S3 Relative importance of each factor on the growth and mortality models (FULL) according to the random forest algorithm. The y-axis has been normalized between 0 and 100 for comparison between variables and species. The variables included in the models were: slope, aspect, insolation (ins), annual mean temperature (tmed), minimum temperature of the coldest month (tmmin), maximum temperature of the warmest month (tmmax), mean spring temperature (tmspr), mean summer temperature (tmsum), mean autumn temperature (tmaut), mean winter temperature (tmwin), total annual precipitation (ptot), spring precipitation (pspr), summer precipitation (psum), autumn precipitation (paut), winter precipitation (pwin) and annual potential evapotranspiration (etrm), basal area in larger trees (BAL), canopy cover fraction (FccArb), plot tree density (td2) and land use.

Table S1 Relative explanatory relevance of independent variables on tree growth and mortality models (from most to least relevant). The score of a given variable in the model is estimated as the increase in the mean square error in the random forest algorithm corresponding the CLIM model (i.e. only environmental variables included). Variables considered are: slope, aspect, insolation (ins), annual mean temperature (tmed), minimum temperature of the coldest month (tmmin), maximum temperature of the warmest month (tmmax), mean spring temperature (tmspr), mean summer temperature (tmsum), mean autumn temperature (tmaut), mean winter temperature (tmwin), total annual precipitation (ptot), spring precipitation (pspr), summer precipitation (psum), autumn precipitation (paut), winter precipitation (pwin) and annual potential evapotranspiration (etrm).

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.