higher tensions there (Figure 8). Changes in stem volume are
caused by transpirational extraction of water from tissues and
when taking into account the time shift (larger but almost con-
stant for the whole tree, smaller but gradually increasing in the
upper crown) both processes are linearly related. Soft tissues
account for only a small proportion of stored water, most of
which is in the sapwood.
Seasonal changes in daily water storage and stem volume
The amount of water withdrawn from storage on clear days
was either relatively stable (upper crown and stem; Figure 8A:
open rectangles) or increased as the season progressed (whole
tree; Figure 8A: solid rectangles). For the whole tree, the
amount of water withdrawn from storage each day increased
from about 40 dm
3
in early August to 50 dm
3
in late September
and ranged from 20 to 30% of daily sap flow. For the upper
stem, water withdrawn from storage (~10 dm
3
day
– 1
) averaged
about 10% of the water lost from the upper stem (96 to
128 dm
3
; Figure 8B) and was relatively stable. The amount of
water withdrawn daily from the whole stem represented 2 to
5% of the free stored water (again increasing from August to
October). For the same period, but considering only the upper
part of the stem, 20 to 25% of free water was used during the
day (Figure 8C). There was a disproportionate use of stored
water from the top of the tree.
Earlier in the season when about 40 dm
3
of stored water was
used by the entire tree (Figure 8A), about 6 dm
3
came from
changes in elastic tissues or about 15% of the total. This per-
centage decreased over the season. Daily volume changes in
elastic tissues (averaged 6 and 0.4 dm
3
day
– 1
for the whole tree
and upper crown, respectively; Table 2) were small compared
with the total quantity of free water used from storage. If ex-
pressed as a percentage of free water, diurnal water used from
elastic tissues of the entire tree never exceeded 0.5%. For the
upper stem, it increased from 0.7 to 1.0% over the study pe-
riod. Water from elastic tissue was never a substantial percent-
age of the total, about 14% of daily water used for transpira-
tion derived from storage or about 1% of the free water. On
average about 45 and 10 liters (about 23 and 10% of transpired
water) were taken from storage when considering the whole
tree and its upper part, respectively. This percentage did not
change much for the whole tree during the study (it increased
from about 20% in the fall to 31% in midsummer), but the per-
centage increased substantially in the upper part of the tree
(the upper stem supplied twice as much in October, up to 27%,
compared with earlier in the summer), even though tree water
loss was about 33% lower at this time of the year.
Water turnover rate
When considering just free stored water from the upper crown
and from the whole tree and their corresponding sap flows,
stored water could meet transpirational needs for a little more
than a third of a day (0.32 to 0.42) and for about a week (6.3 to
8.4), respectively (Table 3), with no clear seasonal variation.
Transpiration and water potential relationship
Figure 9 illustrates a constant decline in water potential as
transpiration increased at the tops of the three study trees. The
two slopes, short- and long-distance hydraulic conductivity,
were similar and linear. The water potential difference be-
tween the two lines at a given transpiration rate corresponds to
the frictional potential. Frictional potential is another negative
constraint, in addition to the gravitational potential, that can
decrease leaf water potential in these tall trees.
Discussion
Tree water storage in stems and branches
Sapwood and heartwood water contents at the stem base of the
studied trees were similar to those observed in other conifer-
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
DYNAMICS OF TREE WATER STORAGE AND STEM DIAMETER CHANGE
189
Figure 5. Diurnal measures of stem radius (
R) at the height of 4 m
(left) and 46 m (right) in the old-growth Douglas-fir sample tree
(Psme 1373) during selected days with fine weather (August 1 and 24,
September 10 and October 8). The radius displayed is real, but is rela-
tive to the measuring device—the actual radius would include the dis-
tance from the center of the stem to the measuring point (from the
position at 4 m, this was about 500 mm).
Downloaded from https://academic.oup.com/treephys/article-abstract/27/2/181/1664618
by guest
on 25 July 2018
ous trees (Waring and Running 1978, Sellin 1991
b, Èermák
and Nadezhdina 1998, Kravka et al. 1999), which also fit for
needle water content (Èermák et al. 1983). The extremely low
heartwood water content in the upper stem (46 m) of Psme
1373 was probably indicative of long-term desiccation. Unfor-
tunately, additional and more extensive sampling was not per-
mitted at the protected site.
Based on tissue volumes, water contents and assumptions
about free water as a proportion of total water, we estimated
that Psme 1373 contained 1426 dm
3
of free water (Table 1).
Most of this water was in the stem sapwood below a height of
46 m and represented the greatest source of stored water used
daily in transpiration. Water stored in tissues above 46 m and
in elastic tissues was also used, but these sources were less im-
portant. Similar results were noted by Phillips (unpublished
data) for a nearby 65-m-tall Douglas-fir tree. In contrast, water
storage in the upper half of the stem of Pinus pinaster Ait. was
reported to be more significant than in the lower portion of the
stem (Loustau et al. 1996); however, the pine trees studied by
Loustau et al. (1996) were considerably smaller than Psme
1373.
Water can be stored extracellularly or intracellularly in plant
tissues (Arcikhovskiy 1931, Holbrook 1995). In contrast to
extracellular stem water, intracellular water in trees is mostly
confined to living tissues between the bark and the newly de-
rived xylem. These tissues are highly elastic and may undergo
considerable dimensional changes during the day (Dobbs and
Scott 1971, Klepper et al. 1971, Goldstein et al. 1984, Milne
1989, Franco-Vizcaino et al. 1990, Holbrook and Sinclair
1992a, 1992b). For this large tree (Psme 1373), these intra-
cellular stores represent less than 1% of daily water use.
Extracellular water storage can be substantial in trees (Waring
and Running 1978, Kravka et al. 1999) and involves water held
by capillary forces in the sapwood as well as water released as
a result of cavitation (Zimmermann 1983, Tyree and Yang
1990). Easily available (“free”) extracellular or capillary water
could represent a substantial fraction of water (Holbrook
1995). Under severe drought, water released by cavitation may
support survival by preventing desiccation of fast-growing tis-
sues (Dixon et al. 1984). Tyree and Yang (1990) concluded
that water stored by small trees is mainly capillary water or
water released by cavitation and may comprise a large volume;
however, they also stated that the stored water in small trees is
typically released at either very high or very low water poten-
tials and thus has no significant role under most conditions, al-
though Zweifel et al (2001) reported that sap flow is buffered
by storage in small trees.
Our results for the large Psme 1373 tree suggest otherwise.
As water is transpired from the foliage, water is withdrawn
from needle tissue reserves (small), tensions develop in the
190
ÈERMÁK, KUÈERA, BAUERLE, PHILLIPS AND HINCKLEY
TREE PHYSIOLOGY VOLUME 27, 2007
Figure 6. The change in stored water at
any time (Cum
∆Q) and stem volume
(V) calculated for the whole tree (left)
and for the upper crown (above 51 m;
right) in the old-growth Douglas-fir
sample tree (Psme 1373) during se-
lected days with fine weather (August
1 and 24, September 10 and October
8). Scales differ because of the large
range of values.
Downloaded from https://academic.oup.com/treephys/article-abstract/27/2/181/1664618
by guest
on 25 July 2018