SEARCH

SEARCH BY CITATION

Table S1. Results of the Kolmogorov-Smirnov tests. H indicates the result of the test: H=0 (Do not reject the null hypothesis), H=1 (Reject the null hypothesis at significance level of alpha). The null hypothesis is that empirical cumulative distribution functions (CDF) are equal, while the alternative hypothesis is ‘unequal’ CDFs. Additionally, whether RW data after 1950 or after 1970 has a ‘larger’ or ‘smaller’ CDF than before 1950 were also tested. Note that a ‘larger’ CDF corresponds to a major number of narrow rings, whereas ‘smaller’ corresponds to a high number of wide rings.

Figure S1. δ13Craw (above), δ13Cc (middle), δ13Cpin (below) data vs. June to August (JJA) temperature (T), precipitation (P) and vapor pressure deficit (VPD) from 1901 to 1999.

Figure S2. Yearly values of 13C discrimination ratios (Δ≅ci/ca), leaf intercellular CO2 concentration (ci) and intrinsic water-use efficiency (iWUE) inferred from δ13Craw tree-ring data at all the studied stands. The δ13Cc chronologies (grey line) and the δ13Cpin chronologies (yellow line) are also depicted. The limits defined by the two physiological tree response constraints are shaded: upper boundary (constant ci/ca, active response to ca) and lower boundary (constant ca-ci, passive response to ca).

Figure S3. Trends in the δ13Craw chronologies vs. the atmospheric CO2 concentration (ca). Linear regressions plotted since 1850 (green), 1901 (grey), 1950 (black) and 1970 (red) only when significant (P<0.05).

Figure S4. Trends in the δ13Cpin chronologies vs. the atmospheric CO2 concentration (ca). Linear regressions plotted since 1850 (green), 1901 (grey), 1950 (black) and 1970 (red) only when significant (P<0.05).

Figure S5. Left: Ring-width frequencies from 1970 to 1999 in comparison with the frequencies prior to 1950, since 1600 for pnCaz, psLil, puPed, and since 1800 and 1900 for psUrb and puUrb, respectively (n indicates the number of trees for each group). Right: Empirical cumulative distribution function (CDF) for each group of ring-width data at each site.

Figure S6. Mean of the ring-width (RW) raw data for all samples and mean after removing the first 150 years from all the series (selected samples). The first number after n indicates the total amount of samples, whereas the second number shows the amount of the selected samples.

Figure S7. Ring-width frequencies for the periods 1850–1900, 1901–49, 1950–69 and 1970–99 at all the studied forests, except for puUrb where the former period is not shown (n indicates the number of trees for each group).

Please note: Wiley-Blackwell is 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_2373_sm_supplinfo.pdf571KSupporting 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.