SEARCH

SEARCH BY CITATION

FilenameFormatSizeDescription
gcb2679-sup-0001-MaterialS1.docxWord document44KMaterial S1. Supplementary Methods
gcb2679-sup-0002-FigureS1.pdfapplication/PDF19KFigure S1. Seed dispersal function used in the study. The curve depicts a negative exponential seed dispersal kernel. The probability of seed dispersal decreases along the distance. Ninety-five per cent of the acorns produced by a parent oak were dispersed in a 4-km radius around it (grey surface; i.e. EffDist in the seed dispersal function). It is a simplification of the shape of the seed dispersal curve normally experienced by Quercus species due to the larger spatial resolution we used (1-km²) compared with that required to highlight the two peaks of seed dispersal (<1-km²).
gcb2679-sup-0003-FigureS2.jpgimage/jpg966KFigure S2. Trends in the difference of mean potential surface area occupied by Quercus pubescens between the climate-only and climate-plus-edaphic simulations over the French forest territory and the 21st century. (a) Comparative trends for different seed dispersal distances. (b) Comparative trends for different ages of first reproduction. A decrease and an increase in seed dispersal distance and age of first reproduction respectively lead to reduced differences in the predicted surface areas between climate-only and climate-plus-edaphic simulations. The highest difference compared with the simulations outputs based on the parameters used in the study is reported for the lowest age of first reproduction tested (10 years). However, such an age of first reproduction is likely out of the range of age of reproduction observed in situ for a long-lived species, and therefore its interpretation is limited. Hence, changes in parameters do not challenge the surface comparison of the Quercus pubescens potential distributions predicted from the climate-only and climate-plus-edaphic SDMs and performed in the study. Seed dispersal distance corresponds to the distance reaches by 95% of the acorns produced (i.e. EffDist in the seed dispersal function). We computed differences as area based on climate-plus-edaphic simulations minus area based on climate-only simulations. Results are based on the HadCM3 A2 climate change scenario and 10 simulations. Dotted lines specified no difference in area between the two simulations.
gcb2679-sup-0004-FigureS3.jpgimage/jpg2601KFigure S3. Trends in the differences of latitude shifts of the potential distribution areas of Quercus pubescens between climate-only and climate-plus-edaphic simulations in function of the seed dispersal distance (panels a, b and c) and the age of first reproduction (panels d, e and f) predicted over the 21st century. (a) and (d) Trends in the 95th quantile of latitude of the distribution area (i.e. the Northern boundary). (b) and (e) Trends in the median latitude of the distribution area. (c) and (f) Trends in the 5th quantile of the latitude of the distribution area (i.e. the Southern boundary). Changes in parameters do not challenge predictions of Q. pubescens range shifts based on the seed dispersal distance (4000 m) and the age of first reproduction (20 years) used in the study. Seed dispersal distance corresponds to the distance reaches by 95% of the acorns produced (i.e. EffDist in the seed dispersal function). We computed differences as latitudinal shift based on climate-plus-edaphic simulations minus latitudinal shift based on climate-only simulations. Results are based on the HadCM3 A2 climate change scenario and 10 simulations. Dotted lines specified no difference in latitudinal shift between the two simulations.
gcb2679-sup-0005-FigureS4.pdfapplication/PDF21KFigure S4. Moran's I spatial correlograms computed from residuals of climate-only (trends with blue triangles) and climate-plus-edaphic (trends with red circles) SDMs. Moran's I values reach 0.246 and 0.088 for adjacent pixels (distance less than 1.5 km) in residuals of climate-only and climate-plus-edaphic SDMs respectively. Considering all distance classes, Moran's I values are 0.187 and 0.068 in residuals of climate-only and climate-plus-edaphic SDMs respectively.
gcb2679-sup-0006-FigureS5.pdfapplication/PDF26KFigure S5. Trends in the altitude shift of the potential distribution areas of Quercus pubescens simulated from climate-only (dotted lines) and climate-plus-edaphic (uninterrupted lines) SDMs between 2010 and 2100 over the French forest territory. (a) and (d) Trends in the 95th quantile of altitude in the distribution area (i.e. the top boundary). (b) and (e) Trends in the median altitude in the distribution area. (c) and (f) Trends in the 5th quantile of altitude in the distribution area (i.e. the bottom boundary). Lines with and without hollow circles displayed trends computed under the HadCM3 A2 and PCM B2 climate scenarios respectively.
gcb2679-sup-0007-FigureS6.pdfapplication/PDF27KFigure S6. Trends in the spatial indexes of the potential distribution area of Quercus pubescens simulated from climate-only (dotted lines) and climate-plus-edaphic (uninterrupted lines) SDMs between 2010 and 2100 over the French forest territory. (a) Number of patches. (b) Mean area of patches. Lines with and without hollow circles displayed trends computed under the HadCM3 A2 and PCM B2 climate scenarios respectively.
gcb2679-sup-0008-TableS1.docWord document92KTable S1. Results of the environmental variables' selection in the SDM of Quercus pubescens. The best model is composed of climatic water balance of July (CWB7), soil pH inferred from plant assemblage (pH) and mean temperature of January (Tm1) (D² = 48.09 and AUC = 0.929; value computed from the calibration dataset, n = 111 013 plots).
gcb2679-sup-0009-TableS2.docWord document40KTable S2. Environmental variable performance in SDM of Quercus pubescens according to the spatial resolution. The combination of climatic water balance of July (CWB7), soil pH (pH) and mean temperature of January (Tm1) explained significantly the distribution of Q. pubescens whatever the spatial resolution (1 × 1 km to 50 × 50 km).

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.