compared with those of tissues external to the sapwood.
Changes in stem dimensions are useful for modeling and re-
cording stem water potentials, particularly if lags are incorpo-
rated (e.g., Zaerr 1971, Molz and Klepper 1973, So 1979,
Herzog et al. 1995, Irvine and Grace 1997, Zweifel et al. 2000.
Intrigliolo and Castel 2006).
Changes in response to hydration are attributed to the lateral
transfer of water between these tissues and the conducting xy-
lem (Molz and Klepper 1973). Hinckley and Bruckerhoff
(1975) in white oak, Lassoie (1979) in Douglas-fir and Anto-
nova et al. (1995) in Scots pine noted three general patterns of
dimensional changes in trees during the growing season. First,
during periods of high soil water content and low evaporative
demand, stem diameter increased from one morning to the
next and often during the day there was either no decrease or
just a reduction in the rate of increase in diameter. This pattern
was largely characterized by growth. Second, during periods
of high soil water content and high evaporation demand, there
was an increase in stem diameter from one morning to the
next, but during the day there could be an appreciable de-
crease. A mixture of growth and tissue rehydration character-
ized this pattern. Third, during periods of low soil water
content and high evaporation demand, there was a daytime re-
duction in stem diameter with only partial recovery overnight.
Only changes in hydration characterized the last pattern. We
observed only Pattern 2 in Psme 1373. Recently, these three el-
ements have been integrated in a model (Steppe and Lemeur
2004).
Seasonal changes in daily water storage and stem volume
A critical assumption in our calculations of the total amount of
stored water was that there was no net change from day-to-day
(i.e., complete refilling occurred). Three lines of evidence sug-
gested that this assumption was likely justified: first, predawn
water potentials were consistently high in Psme 1373 (as well
as other Douglas-fir trees measured within the crane circle),
second, plotting the radial changes from day-to-day did not
demonstrate a progressive decline in radius, and third, on Au-
gust 8, we added over 800 liters (8.5 mm) of water to the soil
surrounding the study tree in an effort to reduce any water defi-
cits. Loustau et al. (1996) made a similar assumption in their
calculations for the water relations of Pinus maritima Poir.
During an extraordinarily dry summer, Hinckley and
Bruckerhoff (1975) noted a continuous loss of stem diameter
in a white oak tree. They assumed that day-to-day volume
changes in extensible tissues reflected a net loss in stem water
content—these day-to-day decreases were linearly related to
decreases in predawn water potential. Similarly, Waring and
Running (1978) observed a progressive decrease in sapwood
water content over the growing season in old-growth
Douglas-fir trees; maximum water contents of 100% satura-
tion were observed in late February and March and a minimum
of 50% was reached in mid-August. Similar conclusions were
drawn by Èermák and Nadezhdina (1998) for adult Norway
spruce trees, where sapwood was maximally hydrated in early
spring and dehydrated substantially, especially in the inner
sapwood, during a summer drought. Similar results were
found in broadleaf species (Tatarinov and Èermák 1999).
These and other studies indicated that our assumption of no net
change in tissue water content might have resulted in a slight
overestimation of the daily water use from storage: however,
predawn water potential did not change appreciably during the
summer in our study trees.
Although the total change in water volume (i.e., the amount
used) in elastic tissues of the stem was small and mostly con-
stant over the growing season (about 6 dm
3
for the whole trees
and 0.3 dm
3
for the treetop; see Figures 6 and 8), daily total
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
DYNAMICS OF TREE WATER STORAGE AND STEM DIAMETER CHANGE
193
Figure 9. Relationship between transpiration (mmol m
– 2
s
– 1
) and wa-
ter potential (
Ψ, MPa) in the exposed upper crown of old-growth
Douglas-fir trees. Each value is the mean of two readings on each of
three trees and three measurement days. Short distance hydraulic re-
sistance (
᭹) is the slope of the regression for Ψ of uncovered leaves at
solar noon minus
Ψ of aluminum-foil-covered leaves Ψ versus the un-
covered foliage’s transpiration rate. Long distance hydraulic resis-
tance (
) is similarly determined, except the values shown are the Ψ
of uncovered leaves at solar noon minus the
Ψ of covered leaves at
predawn versus the corresponding solar noon values of transpiration
in the uncovered foliage (it was assumed that transpiration in the cov-
ered foliage was zero).
Table 3. Water turnover rate, i.e., theoretical mean time during which transpiration can be supported by free water storage in different tree parts and
tissues of Douglas-fir 1373 (values of total water storage for corresponding tree parts were taken from Table 1).
Tree part
Time interval Stem (xyl + phl) Branch (xyl + phl) Sapwood (stem + bran) Phloem (stem + bran)
Needles
Tree total
Upper crown
Days
0.228
0.073
0.255
0.055
0.064
0.374
Hours
5.5
1.8
6.1
1.3
1.5
9.0
Entire tree
Days
6.57
0.36
6.12
0.81
0.236
7.17
Hours
158
9
147
19
6
172
Downloaded from https://academic.oup.com/treephys/article-abstract/27/2/181/1664618
by guest
on 25 July 2018