located within the stand. Several topographical variables were obtained for each site
and tree. Elevation, aspect and slope steepness were measured at the tree level.
The diameter at 1.3 m (diameter at breast height, dbh) of each tree located within
the transect was also measured, and I assessed their vigour using a semi-quantitative
scale based on the percentage of crown defoliation (Müller and Stierlin 1990): class
0, 0–10% defoliation (healthy tree); (1) 11–25% (slight damage); (2) 26–50%
(moderate damage); (3) 51–75% (severe damage); (4) 76–90% (dying tree);
(5) dead trees with >91% defoliation or only retaining red needles. Since estimates
of percent crown defoliation may vary among observers and places, I used as a
reference a tree with the maximum amount of foliage at each site. Declining trees
were considered as those with crown defoliation greater than 50%, and declining
sites were regarded as those with more than 25% trees with such degree of defoliation. Dead trees were regarded as those whose crowns showed complete defoliation or only retained red needles and whose most recently formed rings
corresponded to years prior to the sampling year. Lastly, the number and dbh of all
neighbouring trees found within a circular plot of 7.62 m in radius placed around
each subject tree was measured to estimate the basal area (m
2 ha
−1 ) of the silver-fir
neighbourhood. Values are given as means ± standard errors throughout the text.
6.1.3 Tree-Ring Data
I followed established dendrochronological methods to analyse tree-ring data (Fritts
1976). Two or three cores were taken from each tree at breast height (1.3 m) using an
increment borer. In the field, sapwood length was estimated visually whenever
possible (n = 92 trees from 22 sites). The wood samples were air-dried and polished
with a series of successively finer sandpaper grits. Then, wood samples were visually
cross-dated. Tree rings were measured to the nearest 0.01 mm using a binocular
scope and an LINTAB measuring device (Rinntech, Heidelberg, Germany).
Cross-dating of the tree rings was checked using the program COFECHA (Holmes
1983). To calculate tree age at 1.3 m, in the case of cores without pith, a geometric
method based on the curvature of the innermost tree ring was used to estimate the
missing distance to the pith. Stem sections and cores with pith (n = 120) were used
to calculate regressions between the distance to the pith and the number of tree rings
(r > 0.98 and P < 0.05 in all cases).
The percentage growth change (GC) filter of Nowacki and Abrams (1997) was
applied to identify abrupt and sustained increases or decreases in radial growth (i.e.
releases or suppressions, respectively). First, I calculated the ring-width medians of
subsequent 10-year periods along all the growth series because medians are more
robust estimators of central tendency than means. The M1 and M2 values are
defined as the preceding and subsequent 10-year ring-width medians of a given
dated ring, respectively. For instance, the periods M1 = 1946–1955 and
138
J.J. Camarero
and tree. Elevation, aspect and slope steepness were measured at the tree level.
The diameter at 1.3 m (diameter at breast height, dbh) of each tree located within
the transect was also measured, and I assessed their vigour using a semi-quantitative
scale based on the percentage of crown defoliation (Müller and Stierlin 1990): class
0, 0–10% defoliation (healthy tree); (1) 11–25% (slight damage); (2) 26–50%
(moderate damage); (3) 51–75% (severe damage); (4) 76–90% (dying tree);
(5) dead trees with >91% defoliation or only retaining red needles. Since estimates
of percent crown defoliation may vary among observers and places, I used as a
reference a tree with the maximum amount of foliage at each site. Declining trees
were considered as those with crown defoliation greater than 50%, and declining
sites were regarded as those with more than 25% trees with such degree of defoliation. Dead trees were regarded as those whose crowns showed complete defoliation or only retained red needles and whose most recently formed rings
corresponded to years prior to the sampling year. Lastly, the number and dbh of all
neighbouring trees found within a circular plot of 7.62 m in radius placed around
each subject tree was measured to estimate the basal area (m
2 ha
−1 ) of the silver-fir
neighbourhood. Values are given as means ± standard errors throughout the text.
6.1.3 Tree-Ring Data
I followed established dendrochronological methods to analyse tree-ring data (Fritts
1976). Two or three cores were taken from each tree at breast height (1.3 m) using an
increment borer. In the field, sapwood length was estimated visually whenever
possible (n = 92 trees from 22 sites). The wood samples were air-dried and polished
with a series of successively finer sandpaper grits. Then, wood samples were visually
cross-dated. Tree rings were measured to the nearest 0.01 mm using a binocular
scope and an LINTAB measuring device (Rinntech, Heidelberg, Germany).
Cross-dating of the tree rings was checked using the program COFECHA (Holmes
1983). To calculate tree age at 1.3 m, in the case of cores without pith, a geometric
method based on the curvature of the innermost tree ring was used to estimate the
missing distance to the pith. Stem sections and cores with pith (n = 120) were used
to calculate regressions between the distance to the pith and the number of tree rings
(r > 0.98 and P < 0.05 in all cases).
The percentage growth change (GC) filter of Nowacki and Abrams (1997) was
applied to identify abrupt and sustained increases or decreases in radial growth (i.e.
releases or suppressions, respectively). First, I calculated the ring-width medians of
subsequent 10-year periods along all the growth series because medians are more
robust estimators of central tendency than means. The M1 and M2 values are
defined as the preceding and subsequent 10-year ring-width medians of a given
dated ring, respectively. For instance, the periods M1 = 1946–1955 and
138
J.J. Camarero