Tree Physiology 27, 621–629
© 2007 Heron Publishing—Victoria, Canada
Nocturnal transpiration causing disequilibrium between soil and stem
predawn water potential in mixed conifer forests of Idaho
KATHLEEN L. KAVANAGH,1,2 ROBERT PANGLE1 and ALISA D. SCHOTZKO1
1
Department of Forest Resources, University of Idaho, P.O. Box 441133, Moscow, ID 83844-1133, USA
2
Corresponding author ([email protected])
Received March 16, 2006; accepted August 31, 2006; published online January 2, 2007
Summary: Soil water potential (Ψs) is often estimated by
measuring leaf water potential before dawn (Ψpd), based on the
assumption that the plant water status has come into equilibrium with that of the soil. However, it has been documented for
a number of plant species that stomata do not close completely
at night, allowing for nocturnal transpiration and thus preventing nocturnal soil–plant water potential equilibration. The potential for nighttime transpiration necessitates testing the assumption of nocturnal equilibration before accepting Ψpd as a
valid estimate of Ψs. We determined the magnitude of disequilibrium between Ψpd and Ψs in four temperate conifer species
across three height classes through a replicated study in northern Idaho. Based on both stomatal conductance and sap flux
measurements, we confirmed that the combination of open
stomata and high nocturnal atmospheric vapor pressure deficit
(D) resulted in nocturnal transpiration in all four species. Nocturnal stomatal conductance (gs-noc) averaged about 33% of
mid-morning conductance values. We used species-specific
estimates of gs-noc and leaf specific conductance to correct Ψpd
values for nocturnal transpiration at the time the samples were
collected. Compared with the unadjusted values, corrected values reflected a significantly higher Ψpd (when D > 0.12 kPa).
These results demonstrate that comparisons of Ψpd among species, canopy height classes and sites, and across growing seasons can be influenced by differential amounts of nocturnal
transpiration, leading to flawed results. Consequently, it is important to account for the presence of nocturnal transpiration,
either through a properly parameterized model or by making
Ψpd measurements when D is sufficiently low that it cannot
drive nocturnal transpiration. Violating these conditions will
likely result in underestimation of Ψs.
Keywords: Larix occidentalis, nocturnal transpiration,
Pseudotsuga menzesii, sapflux, soil water, Thuja plicata,
Tsuga heterophylla.
Introduction
A common assumption is that in trees and other C3 plants,
stomata close at night, thus preventing nocturnal transpiration.
In the absence of transpiration, soil and stem water generally
come into equilibrium, therefore it has been assumed that soil
water potential (Ψs) is equivalent to leaf water potential before
dawn (Ψpd) (Waring and Cleary 1967). Predawn leaf water potential is considered a particularly robust measure of soil water
availability because it integrates soil water throughout the soil
volume occupied by plant roots. However, there is considerable evidence that stomata do not close completely during the
night in some species (Benyon 1999, Wheeler et al. 1999,
Donovan et al. 1999, 2001, 2003, Snyder et al. 2003, Bucci et
al. 2004, Grulke et al. 2004, Daley and Phillips 2006). Incomplete stomatal closure resulting in nocturnal stomatal conductance (gs-noc), if unaccounted for, will impair the accuracy of Ψs
estimated from Ψpd. The occurrence of gs-noc has been shown to
prevent plant–soil water potential equilibration, causing leaf
Ψpd to become more negative than that of the soil (Donovan et
al. 2003). Furthermore, calculation of hydrologic budgets for
ecosystems commonly include estimation of canopy-level water loss and if these calculations do not account for nocturnal
water loss, there exists the potential for significant error. For
example, Benyon (1999) suggests that models used to determine watering regimes for plantations need to account for this
phenomenon. The potential occurrence of nocturnal transpiration, as a result of incomplete stomatal closure, makes it important to reassess the use of Ψpd as an estimate of Ψs.
Nocturnal transpiration is not the only cause of disequilibrium between Ψpd and Ψs. During short nights at northern latitudes there may be insufficient time for complete refilling of
the tree sapwood before dawn (Sellin 1999). In desert ecosystems, high concentrations of leaf apoplastic solutes common
in halophyte species can cause a more negative leaf water potential relative to soil water even in the absence of nocturnal
transpiration (Donovan et al. 2001). Finally, disruptions in the
soil to leaf hydraulic pathway through catastrophic xylem cavitation, or decreased root to soil hydraulic conductivity, or
both, can impede water flow and nocturnal tissue rehydration
(Kavanagh and Zaerr 1997, Sperry et al. 2002).
In the absence of complete nocturnal impediments to tissue
rehydration, the relationship between Ψpd and Ψs can be described mathematically (Jones and Sutherland 1991, Dewar
1995, Whitehead 1998, Bond and Kavanagh 1999) by illustrating the effect of stomatal conductance and leaf-to-air vapor
622
KAVANAGH, PANGLE AND SCHOTZKO
pressure gradients on the predawn plant-to-soil water potential
equilibrium:
Ψpd =
Ψs − ( gs -noc D )
KL
(1)
where gs-noc is nocturnal stomatal conductance, D is the vapor
pressure deficit representing the leaf-to-air vapor pressure gradient, and KL is leaf specific conductance. This equation predicts equilibrium between Ψpd and Ψs if stomata are closed
(i.e., gs-noc = 0), or the vapor pressure gradient is zero. If
stomata are open at night, then nocturnal D will determine
whether predawn equilibration between soil and plant water
potential occurs.
Stomatal conductance (gs) is controlled by a complex combination of physiological and environmental variables, which
tend to minimize water loss per unit of carbon uptake (Cowan
1977). Stomatal control may involve a chemical signal, such as
abscisic acid, from roots exposed to decreasing soil water content (Wilkinson and Davies 2002). Stomatal conductance is
also indirectly responsive to changes in the atmospheric vapor
pressure deficit (Mott and Parkhurst 1991, Oren et al. 1999,
Addington et al. 2004, Bucci et al. 2004), as well as to daily
and seasonal variation in plant hydraulic conductance (Fuchs
and Livingston 1996, Bond and Kavanagh 1999).
Stomatal responses to environmental conditions appear to
be largely species specific (Bond and Kavanagh 1999, Oren et
al. 1999, Snyder et al. 2003), however, and incomplete nighttime stomatal closure is not found in all species (Donovan et
al. 1999, 2003, Grulke et al. 2004), indicating that the occurrence of nighttime transpiration cannot be assumed.
The objective of this study was to derive leaf-level parameter estimates for Equation 1 to determine the likely magnitude
of disequilibrium between Ψpd and Ψs in four temperate conifer species across a range of sites and environmental conditions. Additionally, we explored the presence of nocturnal
canopy-level transpiration based on sap flux measurements
and examined the accuracy of our previous season’s Ψpd estimates of Ψs with, and without, consideration of gs-noc.
Materials and methods
Study site and species
In the summers of 2003, 2004 and 2005 we measured diurnal
leaf and canopy gas exchange patterns in selected trees from
nine coniferous stands located in the Priest River Experimental Forest of northern Idaho, USA. The Priest River Experimental Forest (48°45′ N, 116°49′ W) is located in complex
mountainous terrain, with elevations ranging from 675 to 1825
m. Located in the northern Rocky Mountains, the forest experiences wet cold winters that are heavily influenced by Pacific
maritime air masses. Mean annual precipitation is about 812
mm at lower elevations with high elevations receiving up to
1270 mm (Finklin 1983). The majority of precipitation occurs
as winter snow and spring rain, whereas summers are typically
warm and dry with an extended drought period. The historical
annual variation in temperature ranges from a monthly mean
of –4.5 °C during January to 18.2 °C in July. Soil texture in the
upper 60 cm ranges from silt loam to loam depending on site,
and the soils are classified as either Typic Vitrixerands or
Andic Haploxerepts depending on the percentage of volcanic
ash parent material in the upper profile. Soil textures deeper
than 60 cm vary across sites, generally consisting of a loam,
gravelly loam, or very gravelly loam texture down to 100 cm.
Our experimental design comprised three sites within the
Benton Creek watershed of the experimental forest. Within
each site, select co-dominant and dominant trees were measured in adjacent stands that differed in height and age, but
were similar in species composition, habitat type, aspect and
slope. The selected trees ranged in height from 6 to 43 m, and
are located in 15- to 130-year-old stands (Table 1). We measured canopy gas exchange rates in a total of 44 trees including; Douglas-fir (Pseudotsuga menziesii var. glauca (Beissn.)
Franco), western hemlock (Tsuga heterophylla (Raf.) Sarg.),
western larch (Larix occidentalis Nutt.), and western red cedar
(Thuja plicata Donn ex D. Don). At each site, trees from adjacent stands were classified into short, intermediate and tall
height classes (Table 1). The Lower Benton (LB) site is located
on a hill slope position (20° slope, N aspect), at an elevation of
775 m. The Upper Benton (UB) site is located on a hill slope
position (20° slope, N aspect), at an elevation of 1060 m. Both
the LB and UB sites primarily comprised Thuja plicata and
Tsuga heterophylla habitat types. The Center Ridge (CR) site
is located on a shoulder-hill slope position (15° slope, S aspect), at an elevation of 1160 m, with an Abies grandis series
habitat type (Cooper et al. 1991).
Environmental variables
At each site, air temperature (Ta ) and relative humidity (RH)
were measured at the top of the canopy in the short and tall
stands with Hobo Pro® RH/Temp sensors (Onset Computer
Corporation, Pocasset, MA), and wind speed was measured
with Model 014A Met-One wind speed sensors (Met-One Instruments, Inc., Grant Pass, OR). Photosynthetic photon flux
(PPF) was measured with Model 3668 Spectrum Quantum
light sensors (Spectrum Technologies, Inc., Plainfield, IL) in
the short canopies only. Hourly measurements of soil volumetric water content (10–30 cm soil depth) were recorded in each
of the nine stands with ECH2O soil moisture probes (Decagon
Devices, Inc., Pullman, WA) that integrate the soil water content across the soil profile with which they are in contact. With
the exception of the soil water content measurements, all environmental variables were recorded as 10-min means.
Leaf gas exchange
Leaf-level stomatal conductance (gs) and transpiration were
measured with a Li-Cor LI-1600 steady-state porometer
(Li-Cor, Inc., Lincoln, NE). Samples of sun foliage were obtained from a zone between the canopy midpoint and the base
TREE PHYSIOLOGY VOLUME 27, 2007
NOCTURNAL TRANSPIRATION AND ESTIMATE OF SOIL WATER POTENTIAL
623
Table 1. Descriptive statistics for sap flux trees measured during the 2005 field season at Priest River Experimental Forest, Idaho, USA. Values are
means (± SE). Abbreviations: LB = Lower Benton (25- to 130-year-old stands); UB = Upper Benton (20- to 125-year-old stands); CR = Center
Ridge (15- to 125-year-old stands); and DBH = diameter at breast height.
Species
n
Height class
Site
Height (m)
DBH (cm)
Sapwood area (cm2)
Sapwood depth (cm)
Douglas-fir
Douglas-fir
Douglas-fir
Western hemlock
Western hemlock
Western hemlock
Western larch
Western larch
Western larch
Western red cedar
Western red cedar
Western red cedar
4
4
4
4
4
4
4
4
6
2
2
2
Short
Intermediate
Tall
Short
Intermediate
Tall
Short
Intermediate
Tall
Short
Intermediate
Tall
UB, CR
UB, CR
UB, CR
LB, UB
LB, UB
LB, UB
UB, CR
UB, CR
LB, UB, CR
UB
UB
LB
6.4 (0.30)
20.8 (1.34)
30.8 (0.71)
9.1 (0.75)
18.5 (0.58)
29.0 (0.77)
8.7 (0.81)
20.8 (0.43)
34.2 (2.30)
6.7 (0.59)
16.5 (0.90)
24.8 (0.35)
10.6 (0.56)
34.3 (3.57)
52.3 (5.55)
12.2 (0.90)
28.0 (0.84)
42.0 (1.57)
11.6 (0.40)
28.5 (2.64)
46.8 (3.29)
8.9 (0.38)
26.2 (0.75)
42.4 (6.54)
54.8 (4.87)
314.0 (47.7)
517.7 (76.3)
99.0 (15.3)
351.0 (22.6)
661.0 (22.3)
75.8 (4.02)
169.9 (18.9)
279.4 (28.9)
40.3 (2.68)
148.8 (7.53)
296.6 (64.7)
2.1 (0.07)
3.8 (0.38)
4.0 (0.36)
4.0 (0.34)
5.5 (0.18)
6.6 (0.32)
2.9 (0.08)
2.5 (0.22)
2.4 (0.24)
1.9 (0.03)
2.1 (0.04)
2.6 (0.12)
of the upper third of the live crown of each tree with a shotgun
or pole pruner. Samples were immediately placed in a plastic
bag and measured within 5 min of collection to minimize
changes in sample stomatal conductance and water content
(Kaufmann and Thor 1982, Meng and Arp 1993). Porometer
measurements were performed under ambient light, temperature and humidity at ground level, and the measurements were
subsequently recalculated for upper-canopy temperature and
RH at the time of sampling based on simultaneous data from
the canopy temperature and RH sensors and porometer estimates of gs. Early summer diurnal porometer measurements
were performed on (day of year in parentheses) July 7 (Day
188), July 8 (Day 189) and July 13 (Day 194), 2005 at the
Lower Benton (LB), Upper Benton (UB) and South Center
Ridge (SR) sites, respectively. A second set of summer measurements was made on August 19 (Day 231) and August 20
(Day 232), 2005 at the UB and CR sites. Nighttime stomatal
conductance and transpiration were assessed each day during a
predawn measurement period that ranged from 60 to 90 minutes before canopy illumination (PPF < 20 µmol m– 2 s – 1). On
each day, porometer measurements were also made at midmorning (0900 to 1145 h) and midday (1300 to 1530 h). To
correct for leaf area, sample foliage was scanned and projected
leaf area was determined with ImageJ analysis software
(ImageJ, Version 1.33u). For each sampling period, water potential of the leaves used in the porometer measurements was
determined with a pressure chamber (PMS Instrument Company, Corvallis, OR). To obtain initial estimates of soil water
potential, predawn foliar values were adjusted for sample
height (0.01 MPa m –1) to account for the gravitational influence on foliar water potential. Sample height was measured
with an Impulse 200 laser range finder (Laser Tech Inc., Centennial, CO).
KL =
gs -mm D
(Ψpd − Ψmm )
(2)
where gs-mm is midmorning stomatal conductance, D is the canopy vapor pressure deficit when gs-mm was measured, Ψpd is
predawn leaf water potential measured in the absence of nocturnal transpiration, and Ψmm is mid-morning leaf water potential. Mid-morning stomatal conductance was chosen because
it was commonly higher relative to mid-afternoon conductance when increasing D causes stomatal closure (data not
shown). Predawn leaf water potential was measured when
gs-noc or canopy nocturnal D was zero, so it was assumed to be
an accurate estimate of soil water potential. Thus, for species
that have significant gs-noc, it was necessary to sample for Ψpd
during a period of low nocturnal D, followed by clear morning
skies for measurement of foliar gas exchange and Ψmm. These
conditions occurred on August 18, 2005 (Day 230) for sites
UB and CR, when predawn canopy D averaged 0.008 kPa followed by morning hours with clear skies. Similar conditions
did not occur at LB, so the KL calculations were limited to species or individuals that exhibited no gs-noc on July 7, 2005 (Day
188).
Nocturnal tree flux measurements
Diurnal patterns of whole-tree water use were measured from
May to October 2005 with Granier-type sap flux sensors installed in the sapwood at a height of 1.3 m (Granier 1987). Sap
flux density (Js; g m – 2 s – 1) was calculated as (Granier 1987):
Js =119K 1.231
(3)
where K is given by:
Leaf specific conductance
Estimates of leaf-level KL were calculated by rearranging
Equation 1:
K=
( ∆ Tb - ∆ Tf )
∆ Tf
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
(4)
624
KAVANAGH, PANGLE AND SCHOTZKO
∆Tb is the temperature difference between a heated and an unheated probe when there is zero xylem sap flow, and ∆Tf is the
temperature difference between the probes when xylem sap is
flowing (Granier 1987). Each tree had outer sensors installed
at opposite sides of the bole, and select trees with deep sapwood had a deep sensor installed to provide information on the
radial change in Js. For trees with inner sapwood sensors, the
reported tree Js values are a weighted mean Js calculated from
the proportion of outer and inner sapwood to total sapwood
area. All sap flux data were recorded as 10-min means by
CR10X data loggers (Campbell Scientific, Inc.).
A key assumption in the use of the Granier sap flux method
is the accurate determination of a stable baseline ∆Tb during
periods of zero flow (Ewers and Oren 2000). Stable ∆Tb is
reached on nights with low D after complete refilling of tree
water stores (capacitance). However, for trees with incomplete
stomatal closure and high nocturnal D, a stable baseline ∆Tb
may not be achieved. For these days, the apparent (∆Tb) value
will be underestimated, and subsequently the daily Js values
will also be underestimated. To avoid these errors, we obtained
a mean stable ∆Tb value for each sensor on nights with low D
(< 0.1 kPa). The mean stable ∆Tb value for each sensor was
then used to calculate Js during periods when nocturnal D created conditions that were favorable to nighttime canopy water
loss (D > 0.1 kPa). The use of a more accurate ∆Tb value allowed for the correct calculation of daytime Js, and the estimation of nighttime Js. Nighttime Js values can be indicative of
storage tissue refilling or canopy water loss, or both. Because
we did not quantify nighttime refilling, we made no attempt to
partition nighttime sap flux into these two components. Quantification of nighttime sapwood refilling requires the installation of multiple sensors both at the base of the live crown and
in the lower bole, and this type of sampling protocol was beyond the original scope of the project.
Table 2. Mean leaf nocturnal stomatal conductance (gs-noc), midmorning stomatal conductance (gs-mm), and percent of gs-noc relative to
gs-mm (%) for several conifer species at Priest River Experimental Forest. Data were collected in July and August 2005. Numbers in parentheses are standard errors. Within a column, different letters indicate
significant differences at P < 0.05.
Species
gs-noc
(mmol m – 2 s – 1)
gs-mm
%
(mmol m – 2 s – 1)
Western larch
Douglas-fir
Western hemlock
Western red cedar
42.3 (4.1)a
29.1 (4.2)b
19.8 (4.4)bc
10.4 (5.5)c
116.6 (8.9)a
86.2 (9.1)b
47.4 (9.6)c
48.4 (5.2)c
36
34
42
21
n
21
21
15
8
ern red cedar and western hemlock the lowest value. There
were no significant differences (P > 0.05) in gs-noc or gs-mm between sites, height classes and sampling periods (time), with
the exception of higher gs-mm at UB relative to LB (P = 0.0024).
Minimum nocturnal D values at UB and CR often exceeded
0.1 kPa throughout the growing season, although values were
uniformly lower in the short stands than in the tall stands (Fig-
Statistical analysis
Data were analyzed and regression trends fitted using
Origin 7.0 graphing and analysis software (Origin Lab Corporation, Northampton, MA). Models for repeated measures statistical analyses and means tests were constructed with a
mixed-effects repeated measures ANOVA (PROC MIXED),
SAS v9 software, SAS Institute, Cary, NC. Within subject
covariance structures were compared based on Akaike’s Information Criteria (AIC) to select the best model. The model that
received a lower AIC rating was selected. All statistical results
were deemed significant or otherwise based on a P-value of
0.05.
Results
Nocturnal stomatal conductance at the leaf level was evident in
all the species sampled averaging about 33% of gs-mm (Table 2).
Western larch had significantly higher gs-noc than the other species (P < 0.05), with western red cedar having the lowest gs-noc.
Mid-morning stomatal conductance followed a similar pattern
to gs-noc with western larch having the highest value and west-
Figure 1. Daily minimum nocturnal atmospheric vapor pressure deficit (D) at Lower Benton (LB), Upper Benton (UB) and Center Ridge
(CR) at Priest River Experimental Forest during the 2005 growing
season. Dashed lines represent values at the canopy top of tall trees
and solid black lines represent values at the canopy top of the short
trees. The lines have been smoothed for clarity.
TREE PHYSIOLOGY VOLUME 27, 2007
NOCTURNAL TRANSPIRATION AND ESTIMATE OF SOIL WATER POTENTIAL
ure 1). In addition, nocturnal D decreased as elevation declined, with D at the short stands at LB rarely exceeding
0.2 kPa at predawn (Figure 1).
The combination of open stomata and high nocturnal D resulted in nighttime canopy transpiration, as indicated by the
magnitude of the Js values, and the sensitivity of sap flux to
changing D in western larch and Douglas-fir (Figures 2 and 3).
Western hemlock and western red cedar exhibited a lower nocturnal sap flux and lower sap flux sensitivity to D, which is
consistent with a reduced gs-noc relative to the other species.
The amount of water that is transpired nocturnally could not be
calculated from sap flux measurements because we were unable to partition bole refilling and losses to the atmosphere.
This prevented us from performing an in-depth analysis of the
sensitivity of nocturnal canopy conductance (Gs) to changes in
atmospheric D. The strong hysteresis pattern (data not shown)
observed in the diurnal sap flux response to changing D indicated that refilling occurred (Meinzer et al. 1997, Oren et al.
2001).
Leaf specific conductance varied by species and was significantly (P < 0.05) higher in western larch than in the other species (Table 3). There were no within-species differences in KL
625
among sites or canopy height classes.
Using species-specific estimates of gs-noc and KL, Equation 1
was applied to correct Ψpd values for nocturnal transpiration at
the time predawn samples were collected on the same trees in
2003, 2004 and 2005 (Figure 4). Compared with the unadjusted Ψpd values, the corrected values were higher, despite a
declining soil volumetric water content at 10–30 cm soil
depth. The nonlinear curves were fitted with a least squares
method, and the adjusted curve reflected a typical soil water
retention curve (Figure 4). When soil water content was near
field capacity, large declines in soil water resulted in small
changes in Ψpd. However, once soil (from 10–30 cm depth)
reached a critical volumetric water content (about 7%), there
was a steep decline in Ψpd. The unadjusted curve in Figure 4
depicts an atypical response to decreasing soil water content,
where decreases in volumetric water content (10–30 cm
depth) resulted in apparent declines in Ψpd across the range of
observed volumetric soil water content values. Seasonal declines in volumetric soil water content were negatively correlated with D (data not shown) and disequilibrium was greatest
at high D (Figure 5). When D was greater than 0.6 kPa there
was a decline of about 0.2 MPa in Ψpd relative to the corrected
value, and the difference between corrected and uncorrected
Ψpd increased with increasing D. Thus, higher nocturnal transpiration during periods of increasing nocturnal D, and the
subsequent disequilibrium between Ψpd and Ψs (Figure 5), led
to the erroneous conclusion that tree access to soil water became limited as summer progressed (e.g., more negative unadjusted values in Figure 4). Furthermore, Figure 5 suggests that
the unadjusted Ψpd was more responsive to soil volumetric water content at 10–30 cm than it actually was. Below a D of
about 0.12 kPa, there were no differences between the measured and corrected Ψpd values (Figure 5).
The difference in nocturnal D with site and height class
(Figure 1) resulted in higher rates of nocturnal transpiration in
the intermediate and tall stands than in the short stands (e.g.,
smaller correction in the short trees in Figure 5). Thus, before
the correction for nocturnal transpiration was applied, it would
have appeared that the short stands had a higher Ψpd than the
intermediate and tall stands (Figure 6). Once the correction
was applied, there was no difference in Ψpd (P < 0.05) across
the height classes. In addition, the apparent variation in Ψpd
with species changed once the values were adjusted (Figure 7).
These species differences in Ψpd reflect interspecific differences in gs-noc (Table 2), and species locations on sites that have
a higher nocturnal D due to elevation (Figure 1).
Discussion
Figure 2. Sap flux (Js) at breast height and vapor pressure deficit (D,
gray line) during the nights of July 8, July 19, August 5 and August
19, 2005 (from left to right) in two western larch (WL), Douglas-fir
(DF), western hemlock (WH) and western red cedar (WRC) trees
from the medium height class at Upper Benton (UB). Night was defined as periods when photosynthetic photon flux was below 20 µmol
m–2 s–1 and, depending on the day, ranged from 2030 to 0610 h.
We found that gs-noc was sufficient to prevent equilibrium between Ψs and Ψpd in temperate coniferous forests of northern
Idaho. It has been recognized that Ψpd must be adjusted for
sample height (Scholander et al. 1965) and we have now demonstrated that Ψpd also needs to be adjusted for predawn D and
gs-noc. Measures of Ψpd are further confounded by changes in D
with canopy height, topographic position, season and species
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
626
KAVANAGH, PANGLE AND SCHOTZKO
Figure 3. Sensitivity of sapflux
(Js) to vapor pressure deficit (D)
at night (䊉) and during the day
(䊊). Sapflux (Js, hourly means)
was measured at breast height in
12 Douglas-fir trees (DF), 14
western larch trees (WL), 16
western hemlock trees (WH) and
six western red cedar trees (WRC)
on July 8, July 19, August 5 and
August 19, 2005. To limit the influence of refilling on sensitivity
to D, night was defined as 0001 to
0600 h and daytime as 0601 to
2000 h. The exponential relationship (y = a(1 – e – b x )c) is shown
for both night (gray line) and day
(black line).
differences in gs-noc. Therefore, when comparing Ψpd among
species, canopy heights, topographical positions, or dates, it is
necessary to account for nocturnal transpiration, or to make all
measurements when D is too low to drive significant nighttime
transpiration (Figure 5) (Sellin 1999).
The existence of gs-noc in four temperate conifer species provides further evidence that nocturnal stomatal conductance is a
common phenomenon in a large variety of plant functional
groups and species (Benyon 1999, Wheeler et al. 1999, Donovan et al. 1999, 2001, 2003, Bucci et al. 2004, Snyder et al.
2003, Grulke et al. 2004, Daley and Phillips 2006, Woodruff et
al. 2004). However, this is one of the first studies to explore
Table 3. Mean leaf specific conductance (KL) for several conifer species at Priest River Experimental Forest in northern Idaho. Data were
collected in July and August 2005 during periods when nocturnal
transpiration or nocturnal vapor pressure deficit was zero. Numbers in
parentheses are standard errors and different letters indicate significant differences at P < 0.05.
Species
KL (mmol m – 2 s –1MPa –1 )
n
Western larch
Douglas-fir
Western hemlock
Western redcedar
1.34 (0.14)a
0.93 (0.07)b
0.70 (0.08)b
0.99 (0.05)b
11
8
9
6
changes in gs-noc with changing D, and to explore the extent of
disequilibrium in co-occurring temperate conifer species
growing in differing landscape positions within a complex
mountainous setting.
Estimates of leaf-level and whole-tree KL require a precise
measure of Ψpd to accurately determine the leaf-to-soil water
potential gradient. Because this value is in turn used to calculate the adjustment factor for Ψpd, it is essential that KL be calculated from values of Ψpd determined when nocturnal
transpiration is negligible. This unique set of environmental
conditions includes a relatively cool moist night with low D,
followed by daytime conditions when foliage is dry, irradiances are sufficient for maximum gs, and D is adequate to drive
leaf to atmosphere water flux. This combination of moist cool
nights and warm dry days were rare at our northern Idaho sites
with only seven nights out of 70 having D below 0.12 kPa in all
height classes (Figure 1), and only a few of these nights were
followed by warm sunny days. The lack of appropriate climatic conditions limited the sampling dates for KL, and thus our
sample size. This small sample size may be the reason we were
unable to detect differences in KL across height classes.
Once Ψpd was adjusted for nocturnal conductance it became
evident that the trees at our study were not soil water limited in
2004 and 2005, because Ψpd rarely dropped below 0.7 MPa.
The sharp drop in Ψpd that occurred when volumetric soil water content at 10–30 cm depth declined below 6% (Figure 4)
was in 2003; a drier year relative to 2004 and 2005 (data not
TREE PHYSIOLOGY VOLUME 27, 2007
NOCTURNAL TRANSPIRATION AND ESTIMATE OF SOIL WATER POTENTIAL
Figure 4. Predawn leaf water potential (Ψpd) as a function of volumetric soil water content (volume % from 10–30 cm) for all the trees
measured in 2003, 2004 and 2005 across all three sites. A model
based on Equation 1 was applied to correct the water potential values
for nocturnal transpiration: corrected Ψpd values (䊊); and uncorrected
Ψpd values (䊉). The exponential relationship (y = a(1 – e– b x )c ) is
shown for both corrected (gray line) and uncorrected data (black line),
and there are no significant differences between the lines (P > 0.05).
shown). The lack of seasonal soil water limitations is most
likely because the tree had access to deep soil water. This conclusion is consistent with a comprehensive seasonal soil water
storage study conducted by Warren et al. (2005) showing that
Douglas-fir and ponderosa pine (Pinus ponderosa (Dougl. ex
Laws) forests access soil water below 30 cm as the typical
summer drought progresses. In addition, there is strong evidence from 18O isotopic studies that the tree species at our
study sites access soil water well below 30 cm (J.D. Marshall,
personal communication), thus mitigating the effects of a pro-
Figure 5. Difference in mean predawn water potential (Ψpd) in short
(䉭,䉱), intermediate (䊐,䊏) and tall (䊊,䊉) trees in response to vapor
pressure deficit (D) at the time the predawn measurement was made.
A model based on Equation 1 was applied to correct the Ψpd values for
nocturnal transpiration. Corrected Ψpd values are denoted by open
symbols and uncorrected values by filled symbols. Data includes all
trees sampled in 2003, 2004 and 2005 on all three sites. Bars represent
1 SE.
627
Figure 6. Mean predawn leaf water potential (Ψpd ) for the short- (S),
intermediate- (I) and tall- (T) height classes for all the trees measured
in 2003, 2004 and 2005 across all three sites. A model based on Equation 1 was applied to correct the water potential values for nocturnal
transpiration. Corrected Ψpd values are denoted by open bars and uncorrected Ψpd values by filled bars. The bars with the same letters are
not significantly different. Vertical bars represent 1 SE.
longed summer drought. These studies reinforce the concept
that tree roots are a good integrator of soil water availability
throughout the rooting zone. In contrast, measurement of volumetric soil water content within the 10–30 cm profile did not
accurately portray soil water availability to the plant, except
during the relatively dry year of 2003, when the entire soil profile accessed by the roots possibly dried out. Our research indicates that, with adjustments for nocturnal transpiration, Ψpd
can accurately assess seasonal, topographic, and species-specific patterns of soil water availability.
There were species-specific differences in gs-noc, which may
offer clues about the ecological advantages of incomplete
stomatal closure. In western hemlock and western red cedar
low gs-noc and limited sap flux response to nocturnal D suggests
that nocturnal transpiration is minimal in these species. The el-
Figure 7. Mean predawn leaf water potential (Ψpd ) for all western
hemlock (WH), western red cedar (WRC) western larch (WL), and
Douglas-fir (DF) trees measured in 2003, 2004 and 2005 across all
three sites. A model based on Equation 1 was applied to correct Ψpd
values for nocturnal transpiration. Corrected Ψpd values are denoted
by open bars and uncorrected values by filled bars. Bars with the same
letters are not significantly different (P < 0.05). Vertical bars represent
1 SE.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
628
KAVANAGH, PANGLE AND SCHOTZKO
evated nighttime sap flux from evening until midnight (Figure 2) most likely signifies refilling of the sapwood and
nighttime transpiration (Daley and Phillips 2006). Douglas-fir
nocturnal sap flux showed a moderate response to nocturnal D,
whereas sap flux in western larch was tightly coupled with
changes in D. The high gs-noc and rapid response to D indicate
that, during periods of high nighttime D, there are few opportunities for bole refilling in western larch. Our results are consistent with the findings of Daley and Phillips (2006) who
reported that the highest gs-noc and rates of nighttime transpiration were in the most shade-intolerant species sampled. Paper
birch (Betula papyrifera Marsh.), the most shade-intolerant
species Daley and Phillips (2006) sampled, had elevated gs-noc.
Of the species we sampled, western larch is the most shade-intolerant, as reflected by its requirement for full sun to maintain
a positive carbon balance (Dang et al. 1992, Gower and Richards 1990). We hypothesize that the partial nighttime stomatal
closure in western larch may allow higher stomatal conductance in the evening and early morning when D limitations to
stomatal aperture are less likely. Western larch had higher sap
flux rates early in the morning, responding earlier (~ 0530 versus ~ 0700 h) to increasing PPF relative to the more shade-tolerant co-occurring species examined (Figure 8). This pattern
indicates that the western larch trees, by keeping their stomata
open, were able to take advantage of the combination of low D
and increasing PPF early in the morning. The pattern of increasing sap flux with illumination occurred during a period
when D was constant (data not shown), which indicates that
the increase in sap flux was due to increasing stomatal aperture, not changing D. This pattern further supports the conclusion that these trees are not soil water limited. In a soil-water
limited system, it is expected that water is highly conserved
when PPF is below the light compensation point (Cowan
1977).
In conclusion, nocturnal conductance caused Ψpd to be more
negative than Ψs, and the difference between Ψpd and Ψs increased during the growing season, and with elevation, as D increased. The occurrence of nocturnal transpiration can lead to
the false conclusions concerning tree root access to soil water
on upland sites and late in the growing season. To assess Ψs on
our study sites, it is necessary to measure Ψpd when D is at or
below 0.12 kPa, however, if Ψpd is measured when D is above
this threshold, Ψs can be estimated with a properly parameterized model that includes gs-noc, D and KL.
Acknowledgments
We thank the Priest River Experimental Forest and the U.S. Forest
Service for research support and forest access. We are especially
grateful for the logistical support provided to us by our field and lab
crews including Howard Jennings, Benjamin Miller, Pete Gag, Matt
Thompson, Aki Koyama, Kirsten Stephan, Dana Townsend, Brian
Loyd, Jason Dungan and Jessica Schedlebauer. We also thank Nathan
Phillips for advice and technical expertise regarding sap flux measurement, and Colin Campbell for technical help with instrumentation
and data-logging. This project was supported in part by the NSFIdaho EPSCoR Program and by the National Science Foundation under Award No. EPS-0132626 and McIntire-Stennis Funds.
Figure 8. Mean sap flux (Js) at breast height during the early mornings
of July 7 and 8, July 19 and 20, August 5, and August 19 and 20, 2005
in two co-occurring western larch (WL), Douglas-fir (DF), western
hemlock (WH) and western red cedar (WRC) trees from the
intermediate height class at Upper Benton (UB). Bars represent the
standard deviation of the mean. Symbols: 䊉 = Tsuga heterophylla
(Raf.) Sarg.; × = Pseudotsuga menziesii var. glauca (Beissn.) Franco;
䉫 = Thuja plicata Donn ex D. Don; 䉱 = Larix occidentalis Nutt. and
䊐 = photosynthetic photon flux (PPF).
References:
Addington, R.N., R.J. Mitchell, R. Oren and L.A. Donovan. 2004.
Stomatal sensitivity to vapor pressure deficit and its relationship to
hydraulic conductance in Pinus palustris. Tree Physiol. 24:
561–569.
Benyon, R.G. 1999. Nighttime water use in an irrigated Eucalyptus
grandis plantation. Tree Physiol. 19:853–859.
Bond, B.J. and K.L. Kavanagh. 1999. Stomatal behavior of four
woody species in relation to leaf-specific hydraulic conductance
and threshold water potential. Tree Physiol. 19:503–510.
Bucci, S.A., F.G. Scholz, G. Goldstein, F.C. Meinzer, J.A. Hinojosa,
W.A. Hoffmann and A.C. Franco. 2004. Processes preventing nocturnal equilibration between leaf and soil water potential in tropical
savanna woody species. Tree Physiol. 24:1119–1127.
Cooper, S.V., K.E. Neiman and D.W Roberts. 1991. Forest habitat
types of northern Idaho: A second approximation. USDA, Forest
Service, Gen. Tech. Rep. INT-236. 143 p.
Cowan, L.R. 1977. Stomatal behavior and environment. In Advances
in Botanical Research. Eds. R.D. Preston and H.W. Woodhouse.
Academic Press, San Francisco, pp 117–228.
Daley, M.J. and N.G. Phillips. 2006. Interspecific variation in nighttime transpiration and stomatal conductance in a mixed New England deciduous forest. Tree Physiol. 26:411–419.
Dang, Q.L., V.J. Lieffers and R.L. Rothwell. 1992. Effects of summer
frost and subsequent shade on foliage and gas exchange in peatland
tamarack and black spruce. Can. J. For. Res. 22:973–979.
Dewar, R.C. 1995. Interpretation of an empirical model for stomatal
conductance in terms of guard cell function. Plant Cell Environ.
18:365–372.
Donovan, L.A., D.J. Grise, J.B. West, R.A. Papport, N.N. Alder and
J.H. Richards. 1999. Predawn disequilibrium between plant and
soil water potentials in two cold-desert shrubs. Oecologia 120:
209–217.
TREE PHYSIOLOGY VOLUME 27, 2007
NOCTURNAL TRANSPIRATION AND ESTIMATE OF SOIL WATER POTENTIAL
Donovan, L.A., M.J. Linton and J.H. Richards. 2001. Predawn plant
water potential does not necessarily equilibrate with soil water potential under well watered conditions. Oecologia 129:328–335.
Donovan, L.A., J.H. Richards and M.J. Linton. 2003. Magnitude and
mechanisms of disequilibrium between predawn plant and soil water potentials. Ecology 84:463–470.
Ewers, B.E. and R. Oren. 2000. Analyses of assumptions and errors in
the calculation of stomatal conductance from sap flux measurements. Tree Physiol. 20:579–589.
Finklin, A.I. 1983. Climate of the Priest River experimental forest,
northern Idaho. USDA, Forest Service, Gen. Tech. Rep. INT-159.
Fuchs, E.E. and N.J. Livingston. 1996. Hydraulic control of stomatal
conductance in Douglas-fir (Pseudotsuga menziesii (Mirb.)
Franco) and alder (Alnus rubra (Bong.)) seedlings. Plant Cell Environ. 19:1091–1098.
Granier, A. 1987. Evaluation of transpiration in a Douglas-fir stand by
means of sap flow measurements. Tree Physiol. 3:309–320.
Gower, S.T. and S.T. Richards. 1990. Larches: deciduous conifers in
an evergreen world. Bioscience 40:818–826.
Grulke, N.E., R. Alonso, T. Nguyen, C. Cascio and W. Dobrowolski.
2004. Stomata open at night in pole-sized and mature ponderosa
pine: implications for O3 exposure metrics. Tree Physiol. 24:
1001–1010.
Jones, H.G. and R.A. Sutherland. 1991. Stomatal control of xylem
embolism. Plant Cell Environ. 14:607–612.
Kaufmann, M.R. and G.L. Thor. 1982. Measurement of water stress
in subalpine trees: effects of temporary tissue storage methods and
needle age. Can. J. For. Res. 12:969–972.
Kavanagh, K.L. and J.B. Zaerr. 1997. Xylem cavitation and loss of
hydraulic conductance in western hemlock following planting.
Tree Physiol. 17:59–63.
Meinzer, F.C., T.M. Hinckley and R. Ceulemans. 1997. Apparent responses of stomata to transpiration and humidity in a hybrid poplar
canopy. Plant Cell Environ. 20:1301–1308.
Meng, F.R. and P.A. Arp. 1993. Net photosynthesis and stomatal conductance of red spruce twigs before and after twig detachment.
Can. J. For. Res. 23:716–721.
629
Mott, K.A. and D.F. Parkhurst. 1991. Stomatal responses to humidity
in air and helox. Plant Cell Environ. 14: 509–515.
Oren, R., J.S. Sperry, G.G. Katul, D.E. Pataki, B.E. Ewers, N. Phillips
and K.V. Schäfer. 1999. Survey and synthesis of intra- and interspecific variation in stomatal sensitivity to vapor pressure deficit.
Plant Cell Environ. 22:1515–1524.
Oren, R., J.S. Sperry, B. Ewers, D. Pataki, N. Phillips and J. Megonigal. 2001. Sensitivity of mean canopy stomatal conductance to vapor pressure deficit in a flooded Taxodium distichum L. forest:
hydraulic and non-hydraulic effects. Oecologia 126:21–29.
Scholander, P.F., H.T. Hammel, E.D. Bradstreet and E.A. Hemmingsen. 1965. Sap pressure in vascular plants. Science 148:339–346.
Sellin, A. 1999. Does predawn water potential reflect conditions of
equilibrium in plant and soil water status? Acta Oecol. 20:51–59.
Snyder, K.A., J.H. Richards and L.A. Donovan. 2003. Nighttime conductance in C3 and C4 species: do plants lose water at night?
J. Exp. Bot. 54:861–865.
Sperry, J.S., U.G. Hacke, R. Oren and J.P. Comstock. 2002. Water
deficits and hydraulic limits to leaf water supply. Plant Cell Environ. 25:251–263.
Waring, R.H. and B.D. Cleary. 1967. Plant moisture stress evaluation
by a pressure bomb. Science 155:1248–1254.
Warren, J.M., F.C. Meinzer, J.R. Brooks and J.C. Domec. 2005. Vertical stratification of soil water storage and release dynamics in Pacific Northwest coniferous forests. Agric. For. Meteorol. 130:
39–58.
Wheeler, R.M., C.L. MacKowiak, N.C. Yorio and J.C. Sager. 1999.
Effects of CO2 on stomatal conductance: Do stomata open at very
high CO2 concentrations? Ann. Bot. 83:243–251.
Whitehead, D. 1998. Regulation of stomatal conductance and transpiration in forest canopies. Tree Physiol. 18:633–644.
Wilkinson, S. and W.J. Davies. 2002. ABA-based chemical signaling:
the co-ordination of responses to stress in plants. Plant Cell Environ. 25:195–210.
Woodruff, D.R., B.J. Bond and F.C. Meinzer. 2004. Does turgor limit
growth in tall trees? Plant Cell Environ. 27:229–236.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com