appendices

Appendix 1. Description of the irrigation experiment

Table A1. May, June and annual [median ± SE (minimum-maximum)] rainfall in mm for the 1951-2007 period. Plants in the dry treatment received ambient rainfall and plants in the mesic treatment received ambient plus added rainfall. Water was added simulating 4 (May) and 6 (June) mm m-2 precipitation events.

1951-2007 / 2009
Ambient rainfall / Added rainfall / Total rainfall
mesic treatment
May Rainfall / 66.1 ± 4.8 (12.6-176.8) / 21 / 20
(5 x 4 mm events) / 41
June Rainfall / 53.4 ± 4.3 (9-164.5) / 8 / 42
(7 x 6 mm events) / 50
Annual Rainfall / 496.1 ± 14.7 (270.8-769) / 299.2*

* No climatic data available for December 2009, total 2009 rainfall estimated with long-term mean December rain.

Appendix 2. Differences in abiotic characteristics

Table A2. Results of the statistical analyses of the effects of different study sites, ecosystem types, microhabitats and the interaction between the latter two. Analysed variables were different soil nutrient concentrations, percentages of direct (direct site factor, DSF) and indirect (indirect site factor, ISF) radiation; and total [photosynthetically active radiation (PAR) integrated over the day]. Significant effects (P < 0.05) are indicated in bold.

Variable / Site / Ecosystem Type (EC) / Microhabitat(MH) / ECx MH
Na / F / 29.74 / 30.85 / 42.06 / 2.84
P / < 0.001 / < 0.001 / < 0.001 / 0.062
K / F / 12.96 / 19.60 / 54.81 / 3.89
P / < 0.001 / < 0.001 / < 0.001 / 0.023
Mg / F / 48.94 / 49.98 / 42.97 / 10.95
P / < 0.001 / < 0.001 / < 0.001 / < 0.001
Ca / F / 39.03 / 17.53 / 27.48 / 1.18
P / < 0.001 / < 0.001 / < 0.001 / 0.311
N / F / 8.98 / 10.44 / 73.01 / 0.65
P / < 0.001 / < 0.001 / < 0.001 / 0.523
P / F / 34.99 / 31.62 / 7.45 / 0.26
P / < 0.001 / < 0.001 / 0.007 / 0.769
Corg / F / 10.90 / 15.94 / 81.86 / 1.87
P / < 0.001 / < 0.001 / < 0.001 / 0.158
Corg /N / F / 2.59 / 1.43 / 3.79 / 0.15
P / < 0.001 / 0.242 / 0.054 / 0.858
ISF / F / 0.42 / 0.27 / 2225.63 / 7.78
P / 0.797 / 0.766 / < 0.001 / 0.001
DSF / F / 1.02 / 0.25 / 3011.89 / 11.03
P / 0.401 / 0.780 / < 0.001 / < 0.001
PAR / F / 0.83 / 0.26 / 3103.40 / 11.27
P / 0.509 / 0.769 / < 0.001 / < 0.001

Table A3. Mean (SE, n is the number of plots for each combination of woodland type, site and microhabitat indicated in Table 1) soil nutrient concentrations, in mg per soil kg (Na, K and Mg) and in mg per soil g [Ca, N, P and organic C (Corg)] in different woodland types, microhabitats (under the canopy of adult trees and in the open) and sites (codes as in table 1). Significant differences (P < 0.05) are indicated with capital letters (among woodland types), † and ‡ symbols (between microhabitats) and lower-case letters (among sites within woodland types).

WOODLAND TYPE / Microhabitat / Site
Canopy / Open / ABW / RRW / RSW
Mature
Woodland / Na / 7.1 (0.4)A / 8 (0.5)† / 5.9 (0.4)‡ / 7.1 (0.3)a / 9.1 (0.7)a / 4.5 (0.3)b
K / 448 (18.8)A / 530 (23.2) † / 331 (14.9) ‡ / 478 (23.6)a / 519 (31.9)a / 314 (26.5)b
Mg / 170 (9.5)A / 197 (12.4) † / 133 (11.9) ‡ / 153 (11.9)a / 130 (9.8)a / 248 (19.2)b
Ca / 5.3 (0.1)A / 5.7 (0.2) † / 4.8 (0.2) ‡ / 6.5 (0.2)a / 4.9 (0.1)b / 4.3 (0.2)c
N / 4 (0.2)A / 4.8 (0.3) † / 2.9 (0.2) ‡ / 5.1 (0.4)a / 3.6 (0.3)b / 3.2 (0.2)b
P / 0.7 (0.04)A / 0.8 (0.05) † / 0.6 (0.04) ‡ / 1 (0.06)a / 0.6 (0.03)b / 0.5 (0.01)b
Corg / 74.1 (3.4)A / 88.5 (4.3) † / 53.5 (2.6) ‡ / 93.3 (5.7)a / 65.9 (4.3)b / 60.3 (5.7)b
C/N / 21.8 (2.9) / 18.9 (0.4) / 26 (7) / 19 (0.6) / 27.2 (7.8) / 18.3 (0.6)
Canopy / Open / ABA / LRA
Former
Agricultural
Fields / Na / 9.7 (0.6)B / 10.4 (0.7) / 8.2 (0.7) / 7.7 (0.8)a / 11.5 (0.5)b
K / 347 (17)B / 366 (23.3) / 306 (15) / 360 (29.2) / 335 (19)
Mg / 257 (17.1)B / 261 (23.3) / 249 (21.4) / 347 (18.4)a / 170 (10.2)b
Ca / 4.8 (0.1)B / 4.9 (0.1) † / 4.5 (0.1) ‡ / 4.8 (0.1) / 4.8 (0.1)
N / 3.3 (0.3)B / 3.9 (0.3) † / 2 (0.2) ‡ / 2.8 (0.3) / 3.9 (0.4)
P / 0.5 (0.01)B / 0.5 (0.02) / 0.5 (0.03) / 0.5 (0.02) / 0.5 (0.02)
Corg / 57.2 (3.6)B / 66.1 (4.3) † / 38.7 (3.2) ‡ / 58.1 (5.7) / 56.3 (4.7)
C/N / 19.2 (0.7) / 18.8 (1) / 20 (1) / 21.7 (1.1) / 16.9 (0.7)
Canopy / Open / HHL / RRL
Former
Livestock
Pastures / Na / 10.5 (0.7)B / 12.4 (0.9) † / 7.9 (0.9) ‡ / 13.3 (1)a / 7.9 (0.7)b
K / 493 (23.5)C / 559 (28.5) † / 397 (25.8) ‡ / 512 (22.6) / 475 (40.8)
Mg / 146 (7)A / 167 (7.8) † / 116 (8.2) ‡ / 169 (8.1) / 125 (9)
Ca / 4.5 (0.1)C / 4.7 (0.1) † / 4.3 (0.1) ‡ / 4.4. (0.1) / 4.7 (0.1)
N / 3.9 (0.2)B / 4.5 (0.3) † / 3.1 (0.3) ‡ / 4.5 (0.4) / 3.4 (0.2)
P / 0.8 (0.03)A / 0.8 (0.03) / 0.7 (0.06) / 0.9 (0.05)a / 0.7 (0.02)b
Corg / 65.8 (3)A / 73.3 (3.9) † / 55.1 (3.2) ‡ / 73.9 (4.1) / 58.1 (3.6)
C/N / 18.8 (2.2) / 16.5 (0.7) / 22.1 (5.2) / 20.4 (4.5) / 17.3 (0.7)

Appendix 3. Climatic conditions at the study area

Figure A1. Climatic diagram of the meteorological station closest to the study sites (Molina de Aragón, 40º 50’ 40’’ N, 1º 53’ 07’’ W, 1063 m a.s.l.). Bars and lines represent mean monthly rainfall and temperature, respectively, during the period 1951-2009 and in the two years studied: 2008 and 2009 (no climatic data available for December 2009). Filled and dashed portions of the horizontal lower bar indicate the periods with certain and possible frost, respectively.

Figure A2. Climatic conditions from 2008 to 2009 at the study area. (a) daily photosynthetically active radiation (PAR), numbers are the total annual PAR; (b) mean (continuous line), maximum and minimum (dotted lines) daily temperatures, numbers are the mean temperature of the coldest (January, min) and hottest month (July, max), and the annual mean; and (c) daily rainfall (bars) and relative soil humidity (line, Soil RH), numbers are the total annual and spring rainfall. Arrows indicate the sampling dates of the fluorescence measurements. Climatic conditions were monitored with a HOBO ® weather station close to the study sites (40º 49’ 08’’ N, 2º 12’ 47’’ W, 1200 m a.s.l.). The station incorporated a rain gauge (RGA-M0XX), a photosynthetically active radiation (PAR) sensor (S-LIA-M003), a temperature sensor (12-Bit Temperature Smart Sensor) and a soil moisture sensor (S-SMA-M003), placed at 30 cm depth to measure relative soil volumetric water content. Readings of each sensor were recorded every 30 minutes with a data logger (HOBO ® H21-001; all components from HOBO ® Onset Computer Corporation, Bourne, MA, USA).

Appendix 4. Total growth in height

Figure A3. Mean (SE, n varied upon survival and treatment combination) total growth in height: (a) in three different woodland types (mature woodlands, former agricultural fields and livestock pastures) and microhabitats (only non-watered plants) and (b) total growth of plants in two different woodland types, microhabitats and different watering treatments (M: mesic and D: dry). Significant differences (P < 0.05) are indicated with letters.

Appendix 5.Principal component analysis of soil nutrient concentrations.

Material and methods

We collected soil samplesfrom each plot, to measure seven soil nutrient concentrations (N, P, K, Corg, Na, Ca and Mg), relevant for plant mineral nutrition (Lambers and others 2000). To characterize the main trend of variation of the response variables with respect to all nutrient concentrations we performed a principal components analysis (PCA). PCA generates new independent variables (named factors or principal axes), each accounting for a certain amount of the total variability. PCA can summarize in a few dimensions (axes) most of the variability of a dispersion matrix from a large number of descriptors and provide the amount of variance explained by each of these few independent axes (Legendre and Legendre). The explanatory variables were the seven nutrient concentrations and PCA was computed for all 165 micro-sites. Previous to analysis all nutrient concentrations were log-transformed to avoid overdispersion of highest values (Quinn and Keough2002). This analysis was performed in Statistica 6.0 (StatSoft, Tulsa, OK, US).

Results

The PCA of the seven soil nutrient concentrations produced one axis accounting for 53% of the variance (Fig. A4), a second and third axes accounting for 15% and 14%, respectively, and four more axes accounting for 6% or less of the variance each (Table A4). For the SEM we used the coordinates related to the first axis (multiplied by -1), for each plot, as exogenous variable indicator of soil fertility.

References

Legendre P, Legendre L. 1998. Numerical ecology. Amsterdam: Elsevier.

Quinn G, Keough M. 2002. Experimental design and data analysis for biologist. Cambridge: Cambridge University Press.

Table A4. Results of the PCA of soil nutrient concentrations. Percentage of variance explained by each of the seven axes and relative contribution (in %) of each analysed variable to each axis. In bold values for the variable/s contributing the most to each axis.

Axis 1 / Axis 2 / Axis 3 / Axis 4 / Axis 5 / Axis 6 / Axis 7
%Variance explained / 53.3 / 15.5 / 13.8 / 6.2 / 5.0 / 4.8 / 1.4
Variable
contribution / N / 21.7 / 0.6 / 1.1 / 2.4 / 0.2 / 38.4 / 35.8
P / 13.3 / 23.7 / 0.1 / 44.5 / 4.4 / 10.1 / 3.8
K / 16.6 / 1.5 / 14.6 / 0.2 / 49.9 / 13.2 / 4.0
Corg / 22.2 / 0.0 / 3.9 / 4.9 / 4.6 / 13.6 / 50.8
Ca / 16.4 / 2.3 / 10.0 / 34.2 / 13.1 / 20.0 / 4.0
Mg / 2.6 / 54.0 / 25.4 / 13.6 / 0.1 / 4.3 / 0.0
Na / 7.3 / 17.9 / 44.9 / 0.2 / 27.7 / 0.4 / 1.6

Figure A4. Correlation biplot of the first and second axes of the PCA for nutrient concentrations from all plots from different woodland types and microhabitats (in the open, open symbols and under the canopy, closed symbols). Coordinates of the first and second axes have been multiplied by -1.

1