pleted mostly during morning hours and replenished during
the afternoon, particular changes were dependent on weather
conditions.
On a daily basis, water withdrawn from storage equaled wa-
ter returned to storage (consequently Q
t
= Q
crown
, which fits for
day-to-day changes, but not for a growing season). The results
presented in Figure 3 illustrate the within-day behavior of
stored water, whereas data in Figure 4 represent the total water
used from storage. The mean daily total quantity of water
withdrawn from storage for the whole tree Psme 1373 was
about 45 dm
3
(varying between 34 and 53 dm
3
) and for its up-
per crown it was about 13 dm
3
(9–27 dm
3
). The relative quan-
tity of stored water used from the whole tree was appreciable
and represented about 23% (20–31%) of daily total sap flow,
compared with only about 7% (5–16%) of daily sap flow used
from the upper crown. When expressed as a percentage of total
free water, total stored water used on clear days was 3.0%
(2.3–3.6%) and that from the upper crown was 0.9%
(0.6–1.8%): therefore, it was a relatively small fraction of total
free water, likely reflecting our definition of free water (see
Figure 1). However, as a percentage of total water lost by tran-
spiration, the daily use of stored water (~23%) represented a
biologically significant quantity.
Water withdrawn from storage came from both the upper
stem (> 51m) and the lower stem (< 51m); however, the quan-
tity coming from the lower stem was 3 to 5 times that from the
upper stem. Most of the stored water (75%) came from the
zone between 4 and 51 m (see Figure 3 and 4). The volume of
free water in the tree was an order of magnitude larger below
46 m than above 46 m (Table 1). For the upper part of the tree
(i.e., above 51 m), water used from storage was about 10 dm
3
.
Both elastic (phloem, needle, etc.) and inelastic tissue volumes
were small in the top of the tree (see Figure 1) and between 38
and 65% of these volumes could be withdrawn. There was
considerable temporal variability in water removed from or re-
turned to storage in the upper part of Psme 1373 (Figure 4).
Although there was a net depletion of water from storage in
the morning and early afternoon, there were short periods of
recharge. The pattern for the tree below 51 m was easily di-
vided into distinct phases of depletion and recovery (Figure 4).
As shown in Table 1, the amount of free water in the upper
stem was considerably less than in the lower stem (140 versus
5969 dm
3
, or 2.3% of the total). Thus, the percentages of free
water observed in the two regions differed (~24 versus ~4%
for the upper stem and whole tree, respectively) even though
the absolute amounts were in the opposite direction (~10 ver-
sus ~50 dm
3
, respectively; see Figure 4). From the standpoint
of proximity (when compared with soil water) and volume, the
sapwood of the lower stem was most important. This was evi-
dent in the distribution of free water in different tissues (stem
sapwood and phloem, branch sapwood and phloem and need-
les) and at different heights (upper middle and lower crown
and bare stem below crown; see Figure 1).
Diurnal changes in stem volume and stored water
The radius of the stem (R) measured at 4 and 46 m changed ap-
preciably during a 24-h period (Figures 5 and 6). Values of
∆R
reported in this study (with observed daily amplitude of about
0.1 mm) largely reflect volumetric changes in elastic tissues
and associated changes in their water content as first suggested
by MacDougal (1925), Arcikhovskiy (1931), and Molz and
Klepper (1973). The maximum radius was noted between
0730 and 0800 h at both 4 and 46 m. At 46 m, a minimum oc-
curred around 1400 h; there were no further changes in stem
radius until 1900 h, when radius increased rapidly. In contrast,
stem radius was minimal at 1730 h at 4 m. Stem radius contin-
ued to increase through the night, but at a lower rate. The
steeper slope during recovery at night likely illustrates rehy-
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
DYNAMICS OF TREE WATER STORAGE AND STEM DIAMETER CHANGE
187
Figure 3. Diurnal measures of sap flow in the old-growth Douglas-fir
sample tree (Psme 1373) at the base of the stem and for the upper
crown on four selected days with fine weather (August 1 and 24, Sep-
tember 10, October 8). Crown totals represent sap flow measured in
branches distributed at six locations in the crown. Stem totals repre-
sent flow measured at the stem base (at a height of 4 m). Upper crown
represent flows measured in branches close to the tree top (above
51 m), upper stem represents flow measured in the stem at the height
of 51 m.
Downloaded from https://academic.oup.com/treephys/article-abstract/27/2/181/1664618
by guest
on 25 July 2018
dration, whereas the gentler slope late at night and early in the
morning suggests growth. Diurnal changes in stem volume
paralleled the cumulated difference in sap flows during day-
time and volume increases during night, when no storage wa-
ter is extracted from tissues. The time shift, however, was
larger for the whole tree than for the upper crown.
The slightly greater dimensional changes in the upper stem
radius than in the lower stem radius confirmed earlier findings
of Dobbs and Scott (1971). For Psme 1373, the diurnal pattern
of stem shrinkage (–
∆R and ∆Q), refilling (+∆R and +∆Q) and
growth (+
∆R) was similar at the base of the stem to that in the
upper crown. However, there were large differences in the tim-
ing of changes in stem volume and water storage in the upper
stem versus the lower stem, 15 min versus 3 h (Figure 6). The
delay in the upper stem increased substantially during the
growing season, whereas the delay in the lower stem remained
constant. If these time shifts are taken into account and the late
night and early morning increases due to growth are excluded,
then the relationship between a volume change and a change in
water storage was linear during most of the daytime—from
about 0800 to 2100 h (Figure 7).
Diurnal changes in stem volume (
∆V calculated from mea-
sured
∆
R with Equation 7, Figure 6) were strongly related to
changes in the quantity of water removed from storage (±
∆Q)
(cf. Figures 6 and 7). Stem volume decreased with increasing
transpiration (and water depletion from storage, –
∆Q) early in
the day and increased with decreasing transpiration (and grad-
ual refilling of storage, +
∆Q) later in the day. Despite these
changes, growth or a net day-to-day increase in volume oc-
curred only during the night when transpiration approached
zero and internal storage comparments had been largely re-
filled.
When measured volume changes were expressed as frac-
tions of free water for different tissues, they appeared highest
for needles of the whole crown, followed by phloem and other
wet tissues. The situation was similar when daily storage was
evaluated the same way, but the significantly higher percent-
age volume change occurring in the upper crown indicated
188
ÈERMÁK, KUÈERA, BAUERLE, PHILLIPS AND HINCKLEY
TREE PHYSIOLOGY VOLUME 27, 2007
Figure 4. Diurnal courses of differ-
ences between transpiration of crown
foliage (sap flow in branches) and sap
flow in the stem for the whole tree
(whole crown and stem base; left) and
upper crown (above height of 51m;
right) in the old-growth Douglas-fir
sample tree (Psme 1373) for four se-
lected days with fine weather (Aug-
ust 1 and 24, September 10 and Octo-
ber 8).
Downloaded from https://academic.oup.com/treephys/article-abstract/27/2/181/1664618
by guest
on 25 July 2018