Tree Physiology 27, 537–548
© 2007 Heron Publishing—Victoria, Canada
Variation in the radial patterns of sap flux density in pubescent oak
(Quercus pubescens) and its implications for tree and stand
transpiration measurements
RAFAEL POYATOS,1,2 JAN ÈERMÁK3 and PILAR LLORENS1
1
Institute of Earth Sciences ‘Jaume Almera’ (CSIC), Lluís Solé Sabarís s/n, 08208 Barcelona, Spain
2
Corresponding author ([email protected])
3
Institute of Forest Ecology, Mendel University of Agriculture and Forestry, Zemìdìlská 3 , 61300 Brno, Czech Republic
Summary Radial variation in sap flux density across the sapwood was assessed by the heat field deformation method in
several trees of Quercus pubescens Wild., a ring-porous species. Sapwood depths were delimited by identifying the point
of zero flow in radial patterns of sap flow, yielding tree sapwood areas that were 1.5–2 times larger than assumed based on
visual examinations of wood cores. The patterns of sap flow
varied both among trees and diurnally. Rates of sap flow were
higher close to the cambium, although there was a significant
contribution from the inner sapwood, which was greater (up to
60% of total flow) during the early morning and late in the day.
Accordingly, the normalized difference between outer and inner sapwood flow was stable during the middle of the day, but
showed a general decline in the afternoon. The distribution of
sap flux density across the sapwood allowed us to derive correction coefficients for single-point heat dissipation sap flow
measurements. We used daytime-averaged coefficients that depended on the particular shape of the radial profile and ranged
between 0.45 and 1.28. Stand transpiration calculated using the
new method of estimating sapwood areas and the radial correction coefficients was similar to (Year 2003), or about 25%
higher than (Year 2004), previous uncorrected values, and was
20–30% of reference evapotranspiration. We demonstrated
how inaccuracies in determining sapwood depths and mean sap
flux density across the sapwood of ring-porous species could
affect tree and stand transpiration estimates.
Keywords: diurnal variability, radial patterns, ring-porous
species, scaling, transpiration.
Introduction
Sap flow measurements in ring-porous trees present several
technical difficulties that cause significant uncertainties in the
estimation of tree transpiration and subsequent extrapolations
to tree and stand water use. The greatest uncertainty in the application of the heat dissipation (HD) technique (Granier
1985, Granier 1987) in ring-porous species is associated with
the non-uniformity of xylem sap flow (Clearwater et al. 1999).
In ring-porous species, wide earlywood vessels in the most recent annual rings are responsible for most of the long distance
water transport (Miller et al. 1980, Granier et al. 1994). Because of this feature of xylem anatomy, oak trees usually have
only a narrow sapwood band that is difficult to identify either
visually or from changes in relative water content (Èermák
and Nadezhdina 1998, Gartner and Meinzer 2005). Therefore,
determining the depth within the xylem at which sap flow
reaches zero (Èermák and Nadezhdina 1998) is a critical step
in extrapolating from sap flux density measurements to estimates of tree and stand sap flux.
Deciduous Quercus spp. generally show a peak in sap flow
close to the cambium and a sharp decrease with depth (Granier
et al. 1994, Èermák and Nadezhdina 1998, Èermák et al.
1998). These large gradients in sap flow can result in a serious
underestimation of mean sap flow along the measuring length
when using HD probes (Clearwater et al. 1999). Alternatively,
assuming that sap flow measured in the outer xylem is uniform
throughout the sapwood will likely result in a significant overestimation of the true flow, depending on the shape of the flow
profile (Nadezhdina et al. 2002).
Measuring sap flow at different depths or measuring radial
patterns by HD (Phillips et al. 1996, James et al. 2002, Delzon
et al. 2004), heat field deformation (HFD) (Jiménez et al.
2000, Nadezhdina et al. 2002) or heat pulse velocity (HPV)
methods (Hatton et al. 1990) and then relating this variability
to a single-point reference probe in the outer sapwood (Lu et
al. 2000), is a widely used procedure. However, these approaches do not incorporate temporal variations in sap flux
density radial profiles, resulting from changes in meteorological conditions or soil water content or with time of day
(Phillips et al. 1996, Èermák and Nadezhdina 1998), although,
according to Ford et al. (2004a), temporal changes in the radial
profile of sap flux density do not significantly affect daily
sums of sap flow calculated from maximum flow rates derived
from measurements of radial profiles.
In this study we used an HFD multi-point sensor to measure
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
Received January 26, 2006; accepted June 27, 2006; published online January 2, 2007
538
POYATOS, ÈERMÁK AND LLORENS
radial variations in sap flow in mature Quercus pubescens
Willd. trees growing in NE Spain. Long-term stand transpiration has been monitored with HD sensors in this stand since
2003. In a first approach to compute stand transpiration for the
2003 growing season, Poyatos et al. (2005) determined
hydroactive sapwood by visual inspection and assumed a uniform sap flux density across the sapwood. The objectives of
our study were to (1) assess radial variation in pubescent oak
sap flow and its diurnal dynamics; (2) use radial sap flow patterns to delimit sapwood depths; (3) obtain correction coefficients to account for radial variation in sap flow and apply
them to single-point heat dissipation measurements; and finally, (4) apply the correction factors obtained to estimate
stand transpiration during the 2003 and 2004 growing seasons.
Heat dissipation (HD) sap flow measurements
Materials and methods
Heat field deformation (HFD) sap flow measurements
General site description and biometric characteristics of the
stand
The heat field deformation method (HFD) was used to measure
the radial patterns of sap flow for 2–7 days between July and
October 2004 in nine trees that were also monitored with HD
probes (Table 1). An additional tree was measured to cover a
wider range of tree sizes for the derivation of the relationship between sapwood and basal area (Table 1). The HFD method is
based on changes in the spatial variation of the heat field around a
linear heater placed in the tree’s sapwood. Two temperature gradients measured around the heater have been found to describe
this heat field, dTsym and dTasym, and allow for the calculation of
sap flow (Nadezhdina et al. 1998). The technique has been used
in studies on radial patterns of sap flow (Jiménez et al. 2000,
Nadezhdina et al. 2002) and long-term stand transpiration
(Meiresonne et al. 2003). An infrared imagery study has shown
the actual deformation of the heat field around the heater by moving sap (Nadezhdina et al. 2004), and an extensive validation
against other estimates of transpiration is currently under way
(N. Nadezhdina, Mendel University, Brno, Czech Republic,
pers. comm.).
A multi-point HFD sensor (Dendronet, Brno, Czech Republic) with six measuring points with 8 mm spacing between them
was inserted in the trees at breast height, on the same side as the
Granier probes, but far enough from them (about 10 cm) to prevent interferences between the measuring systems. Sensor readings were performed every 20 s and stored as 5-min means in a
data logger (DT50, DataTaker, Rowville, Australia); however,
only five points could be recorded because of technical difficulties. The first measuring point was located 3 mm inward from
cambium, hence, we had five estimates of sap flow at 3, 11, 19, 27
and 35 mm inside the sapwood.
Sap flow per section (q; mm3 mm –1 s –1) was calculated as
(Nadezhdina et al. 2006):
q = Dw
(K + dTs − a )zax
dTasym ztg
(1)
where
dTs – a = dTsym − dTasym
TREE PHYSIOLOGY VOLUME 27, 2007
(2)
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
Field measurements were carried out in the Cal Barrol experimental plot, located in the Vallcebre experimental area (42°12′
N, 1°49′ E, 1100 m a.s.l.) in the eastern Pyrenees in northeast
Spain. Climate is sub-Mediterranean, with a mean air temperature of 7.3 °C (measured at 1440 m a.s.l.) and an annual rainfall of 924 mm (Gallart et al. 2002). The plot is characteristic
of the climatic vegetation type in these Mediterranean
montane areas (Buxo-sempervirentis–Quercetum pubescentis). Quercus pubescens is a winter deciduous or marcescent
species that tends to hybridize with other Quercus species, resulting in a high variability in several morphological and
ecophysiological traits (Himrane et al. 2004). Other woody
species in the plot are Prunus avium L., Fraxinus excelsior
L. and a dense understory mainly composed of Acer campestre
L., Buxus sempervirens L., Prunus spinosa L., Rubus spp. and
Rosa spp. The upper soil, which is formed by a loamy matrix
and limestone boulders, is about 50 cm deep (Rubio 2005).
The experimental plot was located in a stand of about 1.2 ha.
We selected a 10-m radius circular plot with a stand density of
987 trees ha – 1 and a mean DBH of 21.1 cm. The density of pubescent oak trees was 828 trees ha – 1 and basal area was
32.5 m2 ha – 1 (99% of the total). Mean height of the stand was
10.6 m and maximum leaf area index (LAImax) was 2.1, obtained by allometry (Poyatos et al. 2005). Maximum age of the
trees was about 70 years.
Net radiation (NR-Lite, Kipp & Zonen, The Netherlands),
air temperature and relative humidity (HMP45C, Vaisala, Finland), and wind speed (A100R, Vector Instruments, U.K.)
were measured about 2 m above the canopy. Soil water
(0–30 cm) was monitored continuously with a reflectometer
(CS616, Campbell Scientific, U.K.) beginning in May 2004.
Weekly measurements of soil water (0–30 cm) were made by
time domain reflectometry (TDR) (Tektronix 1502C, USA)
to assess spatial variability of surface soil water (0–30 cm) at
six locations in the plot and in deeper soil layers (30–50 cm) at
two locations. One of the probes was used to obtain a site-specific calibration for the reflectometer.
Since May 2003, sap flow has been continuously monitored in
12 trees with heat dissipation probes (Granier 1985) installed
at breast height on the north side of each stem. Needles were
inserted with a vertical separation of about 12 cm and covered
with reflective material. Measurements were performed every
20 s and stored as 15-min means in a data logger (DT500,
DataTaker, Rowville, Australia), which also recorded the meteorological data. Probes were 10 mm long—such short
probes minimize the errors in estimating mean sap flow along
the sensor, as demonstrated by Clearwater et al. (1999), and
have been used in other studies of ring-porous oaks (Phillips et
al. 2003). The original calibration (Granier 1985) was used to
obtain the mean sap flux density in the outer 10 mm of sapwood (vHD).
RADIAL PATTERNS OF SAP FLOW IN RING-POROUS OAKS
539
Table 1. Tree coding, diameter at breast height (DBH) and period of measurement of sap flow radial patterns, showing daytime means and minimum and maximum values (in brackets) of volumetric soil water content (θ) and vapor pressure deficit (D).
Tree
DBH
(cm)
Measurement period
(day of the year 2004)
θ
D
(kPa)
1
2
3
4
5
6
7
8
9
101
12.6
17.9
20.9
21.1
21.5
26.7
26.2
34.5
38.2
49.2
197–199
203–206
279–281
265–267
272–275
233–235
277–279
269–270
200–202
281–286
0.27 (0.27–0.28)
0.25 (0.23–0.25)
0.35 (0.34–0.36)
0.23 (0.22–0.23)
0.24 (0.22–0.40)
0.25 (0.24–0.25)
0.31 (0.29–0.35)
0.30 (0.29–0.30)
0.35 (0.32–0.43)
0.29 (0.27–0.30)
1.05 (0.69–1.40)
1.01 (0.21–1.68)
1.03 (0.38–2.02)
1.23 (0.61–1.86)
0.89 (0.16–1.65)
0.62 (0.04–1.41)
1.43 (0.52–3.82)
0.65 (0.07–1.50)
0.73 (0.12–1.45)
0.70 (0.11–1.63)
1
This tree was not monitored with HD sensors, but its radial pattern and sapwood area were measured to cover a wider range of tree sizes.
vi =
q
ls
(3)
We assumed that each thermocouple (TC) junction sensed an
8-mm length interval, centered around the measuring point
(Hatton et al. 1990), except for the shallowest point (which was
only 3 mm inside cambium), and the deepest one, which we assumed representative of the sap flux density in the rest of the inner sapwood. Sap flow for the whole tree, Q, was integrated as:
Q = v 3 As,3 + v11 As,11 + v19 As,19 + v 27 As,27
+ v 35 As,35 + v 35 As,in
(4)
Analysis of diurnal variation in measured radial patterns of
sap flow
The difference between outer flow, Qo (represented by measuring
points at 3 and 11 mm, i.e., the first 15 mm of sapwood) and inner
flow Qi (the remaining conductive sapwood) was normalized by
total stem sap flow and expressed as a percentage (Qdiff) to investigate diurnal variations in the radial profile of sap flow:
Qdiff =
(5)
We calculated a reference sap flux density (vref) in the outer
10 mm of sapwood from the weighted average of the two shallower junctions of the HFD sensor (3 and 11 mm inside the
cambium). According to the measuring length around each
point (4 mm), the measuring point at 3 mm would be representative of the first 7 mm and the junction at 11 mm would account for the remaining 3 mm to cover the entire length of the
Granier probe. Therefore, we applied a weighting factor of 0.7
for v3 and a weighting factor of 0.3 for v11 to obtain the average
mean flux density—sap flux density obtained by the HD sensors (vHD) could be equivalent to the weighted-mean sap flow
in the outer sapwood obtained from the HFD measurements
(vref). We were able to compare simultaneous values of vref and
vHD throughout the measuring period for only two trees because of malfunctioning of, or erroneous readings from, the
Granier probes.
To account for the particular sap flux density profile in each
tree, we related sap flux density (vi) at the five measuring
points to vref, obtaining a radial correction coefficient for each
depth, Crad,i in which the subscript i indicates the corresponding depth (mm):
C rad,i =
where vi is sap flux density at each depth, As,i is the corresponding annulus area with As,in representing the conducting area beyond the influence of the last measuring point (deeper than
39 mm).
Q0 − Qi
100
Q
vi
v ref
(6)
The radial correction factor can also be used to obtain an estimate
of sap flow density at different depths (vrad,i) when only single-point HD measurements are available:
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
and dTsym is symmetrical temperature difference (°C), dTasym is
asymmetrical temperature difference (°C), dTs-a is temperature
difference (°C) between the upper points of the symmetrical and
asymmetrical thermocouple pairs (Equation 2), K is the value of
dTs-a under conditions of zero-flow, zax is axial distance between
the symmetrical pair of thermocouples (30 mm), ztg is tangential
distance between the heater and the upper end of the asymmetrical pair of thermocouples (50 mm), and Dw is thermal diffusivity
of fresh wood (mm2 s –1), assumed to have a value of 2.25 ×
10 –1 mm2 s –1.
Sapwood depth (ls) was then estimated by extrapolating the
observed radial patterns of sap flow per section to zero flow
(Èermák and Nadezhdina 1998). Third- or fourth-order polynomials were fitted to the sap flux density radial patterns and
used to extrapolate the point of zero flow for each tree. Sap
flux density (vi; mm3 mm – 2 sapwood area s – 1, or mm s – 1 for
brevity) for each measurement depth (mm), denoted by the
subscript i, was calculated as:
540
POYATOS, ÈERMÁK AND LLORENS
v rad,i = C rad,i v HD
(7)
Finally, whole-tree sap flow was integrated as in Equation 4,
replacing vi with vrad,i.
1i) to more gradual decreases (Figures 1d, 1g, 1h and 1j). In
one case (Figure 1c), the pattern was comparatively flat, and
sap flux density at a depth of 40 mm was still 40–60% of the
maximum.
Stand transpiration
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
After correcting for sapwood area and radial variation of sap
flow, the ratio between corrected sap flow on a sapwood area
basis and sap flux density from Granier’s equation were calculated for each tree. We used this ratio as a coefficient (C ) to account for the radial variation of sap flow by applying it directly
to the sap flux density estimates from the Granier sensors.
Stand transpiration (Ec) for the 2003 and 2004 growing seasons was calculated by assigning a different sap flow density
to each diametric class (5 cm increment), which was obtained
by averaging all the measured trees within that class. Transpiration of each class relative to total stand transpiration, calculated during periods when all diametric classes were measured, was used to fill the gaps whenever a diametric class was
missing (i.e., because of sensor failure). Finally, stand transpiration was compared with the previous estimate without the
corrections presented in this study (Poyatos et al. 2005) and to
the reference evapotranspiration (ET0) obtained by the Penman-Monteith equation (Monteith 1965) with fixed surface
resistance (Allen et al. 1998). Missing values of Ec due to failure of sap flow sensors (one missing day in 2003 and two missing in 2004), were estimated by averaging the relative transpiration (Ec /ET0) of the previous and following day and applying the value obtained to measured values of ET0.
Results
Meteorology during the measurement period
Meteorological conditions during the measurement period were
mild (Table 1) compared with the extreme summer drought in
2003 (Poyatos et al. 2005). Although volumetric soil water content (θ) in the first 30 cm of the soil reached 0.22, equivalent to a
soil water deficit (SMD, Granier and Loustau, (1994)) of about
0.85, θ in the deeper horizons (30–50 cm) was always below
0.30 (SMD less than 0.40).
Tree-to-tree variations in radial patterns
Sap flux densities were considerably higher in the outer xylem
than in the inner xylem of Q. pubescens trees. Maximum v was
0.069 mm s – 1 and 0.036 mm s – 1 at the measurement points located 3 and 11 mm beyond the cambium, respectively. The two
innermost points, at depths of 27 and 35 mm, showed maximum values of v of 0.010 mm s – 1, and v at a depth of 19 mm
peaked at 0.019 mm s – 1.
Radial patterns of sap flow showed great variability among
trees (Figure 1), with peak flows most commonly recorded at
the outer measuring point (Figures 1b, 1e, 1f and 1i) or at a
depth of 11 mm (Figures 1c, 1d, 1g, 1h and 1j). Only in the
case of Tree 1, the smallest tree, did sap flow peak at a depth of
19 mm (Figure 1a). The rate of decline in sap flux density with
depth also varied, from steep declines (Figures 1b, 1e, 1f and
Figure 1. Diurnal variation (0800–2000 h solar time) of representative radial patterns of sap flow in Quercus pubescens. The graphs
show the fitted surface (distance-weighted least squares) to 15-min
data throughout the entire measurement period for each tree (Table 1).
Tree order as in Table 1.
TREE PHYSIOLOGY VOLUME 27, 2007
RADIAL PATTERNS OF SAP FLOW IN RING-POROUS OAKS
Estimation of sapwood depth
Sapwood depths of pubescent oak trees estimated by extrapolation of radial patterns of sap flow measured at midday (Table 1) ranged between 35 and 49 mm (20–65% of total xylem
radius beginning at the cambium) (Figure 2). Sapwood depth
significantly increased with tree DBH, following a power relationship (P = 0.040). A power function between basal area (Ab)
541
and sapwood area (As) of each tree was also established (As =
4.16Ab0.66, n = 10, P = 0.000) and used to scale tree-level sap
flow to stand transpiration.
Sapwood areas obtained from radial patterns of sap flow
were 1.5–2 times larger than estimates made in the field based
on visual inspection of wood cores (Table 1). The ratio between sapwood and ground area for the stand (AS:AG) almost
doubled using the new allometric relationship instead of the
one used by Poyatos et al (2005).
Diurnal variations in radial patterns of sap flow
Figure 3. Diurnal variation in
the difference in sap flow between the outer and inner sapwood normalized by total sap
flow (Qdiff), in eight Quercus
pubescens trees. Each value is
the mean (± SE) of measurements taken within each
hourly time step.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
Figure 2. Sapwood depth estimation in two Quercus pubescens trees
from radial patterns of sap flow. Data points denote HFD measured
values of sap flux density and lines are polynomial fits, used to extrapolate to the point of zero flow.
Diurnal variation was also observed in the radial profiles of
sap flow. Sap flux density in the inner xylem, expressed as a
percentage of its maximum, was generally higher during the
afternoon (Figure 1). This trend, which could be inferred by
the upward folding of the fitted surface, especially between
1600 and 2000 h solar time, was observed for most, but not all,
trees (Figure 1). In the outer xylem, sap flux density relative to
its maximum value did not vary much during the day (Figure
1).
The normalized difference between outer and inner sap flow
(Qdiff) was rather stable during the middle part of the day, particularly in some trees (Figures 3b, 3d and 3j). Most of the
trees, though, showed a slightly decreasing trend during the
day (Figures 3f, 3g and 3h), and others presented a steeper decline in Qdiff from 1600 h onward (Figures 3c, 3e and 3i). For
trees with a pronounced peak profile (Figures 1a, 1e and 1f),
the absolute contribution to total flow of the internal xylem
was greater than that of outer xylem (negative Qdiff) during
most of the day.
In general, Qdiff was positively correlated with D and Rn and
negatively correlated with time of day (Table 2) for all the
542
POYATOS, ÈERMÁK AND LLORENS
trees except the largest (Tree 10). Some significant correlations were found between Qdiff and θ, but a consistent pattern
did not emerge.
Comparison of the diurnal variation in Qdiff with the magnitude of total stem flow revealed a clockwise pattern of hysteresis for most trees, indicating that, for a given rate of total flow,
the relative contribution of the outer xylem was higher during
the morning (Figure 4).
Radial correction coefficients
Tree
D
Rn
1
2
3
4
5
6
7
8
9
10
0.827 **
0.213 **
0.796 **
0.536 **
0.263 **
0.056 ns
0.652 **
0.071 ns
0.345 **
–0.372 **
0.621 **
0.471 **
0.657 **
0.849 **
0.398 **
0.547 **
0.683 **
0.438 **
0.627 **
–0.267 **
Time
–0.674 **
0.093 ns
–0.380 **
0.177 ns
–0.061 ns
0.124 ns
0.171 ns
0.675 **
–0.306 **
–0.020 ns
–0.564 **
–0.154 *
–0.456 **
–0.632 **
–0.444 **
–0.796 **
–0.822 **
–0.594 **
–0.547 **
0.068 ns
Figure 4. Difference between outer and
inner sapwood sap flow normalized by
total sap flow (Qdiff) as a function of
total stem flow (Q) and time. Lines
show 15-min data with data points
drawn every 30-min. Numbers in the
plots indicate solar time.
TREE PHYSIOLOGY VOLUME 27, 2007
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
For the two trees (Trees 5 and 9, Figure 5) in which we could
compare sap flux density in the outer 10 mm of sapwood as
measured by the Granier probe (vHD) with values estimated
from v3 and v11 (vref), very different results were obtained. Values of vHD and vref were rather similar for Tree 5 (Figure 5c),
whereas for Tree 9, vHD and vref differed greatly in daily dynamics and absolute value (Figure 5d), and the differences in-
Table 2. Correlation coefficients between Qdiff (the difference between outer and inner sap flow normalized by total flow) and vapor
pressure deficit (D), net radiation (Rn), volumetric soil water content
(θ) and solar time. Statistical significance: *, P < 0.05; **, P < 0.01;
and ns = not significant.
RADIAL PATTERNS OF SAP FLOW IN RING-POROUS OAKS
creased with increasing D (Figures 5b and 5d). For Tree 9, vHD
was more similar to v11 than to the weighted average vref (Figure 5d).
Because the radial patterns of sap flow were rather stable
during that part of the day when most sap flow occurred (Figure 3), we used a mean value of the radial correction coefficients (Crad,i) under conditions of high irradiation (Rn > 200
Wm – 2) to correct sap flux density measured by the Granier
probes. The values of Crad,i were lower and more variable with
increasing sapwood depth (Figure 6).
Sap flow was then calculated with the As obtained from the
radial patterns (Table 1) and applying the coefficients Crad,i. In
two of the trees equipped with Granier probes, radial patterns
were not measured and the mean values for all trees, except
Tree 1, which showed a very different pattern (Figure 1b),
were used instead. The correction yielded different results depending on the shape of the radial profile. Radially corrected
sap flow could be lower, (Figure 7a) similar to (Figure 7b) or
higher (Figure 7c) than the original uncorrected values. These
situations could be related, respectively, to three different radial patterns of sap flow: a steep decline in sap flow with depth
(Figure 1i), a gradually decreasing decline in sap flow with
depth (Figure 1b), or an approximately flat profile (Figure
1h). Not accounting for radial variation of sap flow led to inflated sap flow estimates (Figure 7).
We assessed the effects on daily transpiration of assuming
an average coefficient for daytime conditions by comparing
the use of the mean daytime coefficients with the actual coefficients obtained when concurrent measurements of sap flow,
determined by HD and radial patterns, were available. At the
15-min time scale, the use of a daytime-averaged coefficient
introduced errors of about ± 20%, but the overall impact of
Figure 6. Mean (± SE) and coefficient of variation (CV) of sap flux
density correction coefficients (Crad,i) at different depths in the xylem.
Figure 7. Three diurnal courses of stem sap flow (Q) calculated with
raw HD data and visually estimated sapwood (+), from raw HD data
and sapwood areas obtained from radial patterns (䊊), and HD data
corrected according to radial variation of sap flow and sapwood areas
obtained from radial patterns (䊉). Data for three trees are shown.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
Figure 5. Sap flux density measured at 3 and 11 mm beyond the cambium (v3 and v11, respectively), compared with the reference sap flux
density (vref, estimated as a weighted average of v3 and v11) and sap
flux density obtained by the HD probes (vHD).
543
544
POYATOS, ÈERMÁK AND LLORENS
these errors on daily transpiration was low (2–4%).
We related corrected values of sap flow on a sapwood area
basis to the raw data obtained from HD sensors to obtain a
whole-tree correction coefficient to account for radial variation of sap flow within the stem (C) (Table 3). The correlation
between coefficient C and DBH was not significant (P = 0.25).
Stand transpiration
Discussion
Radial patterns of sap flow in ring-porous oaks
Sap flow in ring-porous trees occurs predominantly in the
outer xylem, because earlywood vessels, which are highly efficient in water transport, are only functional in the outermost
growth ring (Ellmore and Ewers 1986). However, sap flow can
also occur in latewood vessels of older rings (Aloni et al.
1997). Our results on pubescent oak indicate a significant contribution from deeper xylem layers (Figures 1 and 3) as previously reported for this species (Èermák and Nadezhdina 1998)
and other ring-porous deciduous oaks (Èermák et al. 1992,
Granier et al. 1994). We have found that up to 60% of the total
flow can occur in the inner xylem (defined in this study as the
portion of sapwood deeper than 15 mm from the cambium),
Table 3. Sapwood areas obtained from visual estimations in the field and from radial patterns of sap flow for Quercus pubescens trees in the Cal
Barrol plot. The ratio between sapwood and ground area for the stand (AS:AG) was 9.7 m2 ha – 1 obtained from visual estimation and 17.3 m2 ha – 1
obtained from radial sap flow patterns. The whole-tree radial correction coefficients (C) are shown.
Tree
1
2
3
4
5
6
7
8
9
101
Sapwood area (cm2)
Visual estimation
Radial pattern of sap flow
48
84
108
110
113
161
156
243
286
–
94
159
223
209
215
285
278
391
441
590
Mean ± SE
1
Ratio
radial pattern:visual
C
1.96
1.89
2.06
1.90
1.90
1.77
1.78
1.61
1.54
–
1.28
0.58
0.87
0.64
0.51
0.49
0.56
0.94
0.45
–
1.82 ± 0.06
0.70 ± 0.09
This tree was not monitored with HD sensors, but its radial pattern and sapwood area were measured to cover a wider range of tree sizes.
TREE PHYSIOLOGY VOLUME 27, 2007
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
We applied coefficient C to the single-point measurements obtained with Granier sensors for the Years 2003 and 2004 (Figure 8). Results from two growing seasons showed that Ec was
20–30% of reference evapotranspiration in the studied oak
stand (Table 4). Considerable differences in Ec between years
were observed during the last part of the summer (Figure 8).
Neglecting radial pattern coefficients and using visually determined sapwood areas resulted in estimates of Ec that were 75%
lower than the corrected value for 2004, but no significant difference between corrected and uncorrected values was found
for Year 2003 (Table 4).
whereas Granier et al. (1994) found that up to 30% of the total
flow occurred deeper than 11 mm in a Quercus petraea (Matt.)
Liebl. tree. The maximum values of sap flux density across the
first 11 mm of sapwood in the study of Granier et al. (1994)
(0.065 mm s – 1) are comparable with the maximum values of
sap flux density we recorded at the shallowest measuring point
(0.069 mm s – 1).
The inner xylem in ring-porous oaks may act as a backup to
the hydraulic system when a disruption in the functionality of
the current-year earlywood vessels occurs (Granier et al.
1994). This could explain the significant contribution from
deeper latewood vessels when drought-induced embolism occurs in the outer xylem. Measurements of hydraulic conductance in the same species and growing in a similar climate
show that, from July to November, 40% of total hydraulic conductance is lost as a result of drought-induced embolism
(Nardini and Pitt 1999).
Sap flux density at a given point in the xylem depends on the
specific conductivity of the sapwood, the leaf area attached hydraulically to that portion of sapwood and the rate at which the
attached foliage is transpiring (Spicer and Gartner 2001). In
deciduous species, there is no direct connection between
leaves and the inner, older, xylem (Gartner and Meinzer 2005);
therefore, to explain the significant flow in this portion of the
sapwood, water transport in the radial direction must be assumed to take place. Kitin et al. (2004) have shown that earlywood vessels of the ring-porous Fraxinus lanuginosa Koidz.
are efficiently connected through bordered pits with the latewood vessels of the previous growth ring, confirming the existence of a pathway for radial movement of sap. Other findings
suggest that radial gradients of water potential also play a role
in determining the spatial patterns in stem water transport
(James et al. 2003, Domec et al. 2006). Combining radial sap
flow measurements and sapwood conductivity profiles (Spicer
and Gartner 2001) in ring-porous species, as recently done for
a radial-porous evergreen oak (Hirose et al. 2005), should pro-
RADIAL PATTERNS OF SAP FLOW IN RING-POROUS OAKS
545
Diurnal and tree-to-tree variations in radial patterns of sap
flow
vide increased insight on the ultimate causes of radial
variations in sap flow in ring-porous species.
Estimating sapwood depths from radial patterns of sap flow
Delimitation of sapwood depth based on radial profiles of sap
flow (Figure 1) is recommended for ring-porous species
(Èermák and Nadezhdina 1998) in which wood relative water
content may not be indicative of the true hydroactive xylem
area (Gartner and Meinzer 2005). Nadezhdina et al. (2002)
found that sapwood depths identified by radial patterns of sap
flow corresponded well with the limit estimated by changes in
xylem color, but no ring-porous species were represented in
their study. Poyatos et al. (2005) reported that visual identification of the inner sapwood boundary significantly underestimated the amount of hydroactive xylem area. The relationship
we obtained between tree basal area and sapwood area, determined from radial profiles of sap flow, was comparable with
the relationship found in a Quercus petraea stand in montane
areas of central Spain under similar climatic conditions
(Aranda et al. 2005).
Table 4. Reference evapotranspiration (ET0) and stand transpiration (Ec) for representative periods of the 2003 and 2004 growing seasons (Days
144–304). Uncorrected transpiration is calculated based on visually estimated sapwood area without accounting for radial variation in sap flow.
All values are in mm per growing season.
Year
Precipitation
ET0
Ec (uncorrected)
Ec (corrected)
Ec / ET0
2003
2004
363.4
483.3
584.7
553.5
115.2
131.3
118.2
164.1
0.20
0.30
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
Figure 8. Seasonal pattern of reference evapotranspiration (ET0) and
stand transpiration (Ec) in the pubescent oak stand during two growing seasons (Days 144–304, Years 2003 and 2004).
The general shape of the radial patterns in our study was similar to previous observations of the same species (Èermák and
Nadezhdina 1998, Èermák et al. 1998), but the observed
tree-to-tree variability suggests that other factors may have an
influence. Although there is relatively little information on
factors influencing spatial variation of sap flow in ring-porous
species, three factors have been identified. First, sap flow can
vary azimuthally around the trunk, and we only measured the
radial variation in one location (Loustau et al. 1998); however,
radial patterns of sap flux density in Quercus robur L. showed
little circumferential variation (Èermák et al. 1992). Second,
radial pattern measurements were not taken concurrently in all
the trees so variability in the shape of the sap flux density
profile within the trunk may have varied seasonally as a result
of changes in soil content or atmospheric conditions (Phillips
et al. 1996, Nadezhdina et al. 2002). Third, unrepaired embolisms during the season may have changed the spatial patterns
of xylem conductivity, affecting, in turn, the radial profile of
sap flow (Ford et al. 2004b).
The radial profile of sap flow varies diurnally in pubescent
oak. Although there is increasing evidence of such diurnal
variation in conifers and diffuse-porous species (Nadezhdina
et al. 2002, Ford et al. 2004a, 2004b), diurnal changes in radial
patterns of sap flow in ring-porous trees are still poorly documented. We found a proportionally greater contribution to total stem flow from the outer xylem, as defined in our study,
during conditions of high evaporative demand, whereas Ford
et al. (2004b), who studied several Pinus species, found that
conditions of high evaporative forcing are responsible for the
mobilization of water in the inner sapwood of pines. Our results are thus more consistent with those showing that sap flow
becomes intensified toward the cambium with increasing
evaporative demand (Nadezhdina et al. 2002). This pattern has
been related to hydraulic connections between the outer xylem
and sun-exposed foliage (Jiménez et al. 2000, Nadezhdina et
al. 2002). Studies on the connectivity between leaves and xylem in ring-porous species, as has been done for gymnosperms
(Maton and Gartner 2005), should help explain the diurnal
changes in radial patterns of sap flow.
The occurrence of drought-induced embolisms in earlywood vessels under conditions of high evaporative demand are
inconsistent with our observation of a proportionally greater
contribution to total stem flow from the outer xylem during
conditions of high evaporative demand (Granier et al. 1994).
546
POYATOS, ÈERMÁK AND LLORENS
However, the increase in the relative contribution of the inner
sapwood to total sap flow during the afternoon could be related
to the refilling of sapwood tissue. Embolism production and
repair are increasingly regarded as highly dynamic processes
even at the diurnal time scale (Clearwater and Goldstein
2005), and we predict that they play a role in determining the
spatial distribution of sap flux density.
Radial correction coefficients
Stand transpiration
After correcting for radial variation in sap flow and sapwood
depths, estimates of stand transpiration were up to 25% higher
than the results obtained based on visually determined sapwood and assuming a uniform distribution of sap flux density
along the sapwood (Poyatos et al. 2005). Corrected values of
Ec showed that growing season Ec, normalized by stand basal
area, was 3.6 and 5.1 m3 m – 2 for 2003 and 2004 respectively.
These values are lower than those found in a similar forest of
the same species in Italy (about 7 m3 m – 2 basal area per growing season) (Èermák et al. 1998). In Quercus robur stands of
different densities, growing season Ec expressed on a stand
basal area basis was between 2.1 and 5.8 m3 m – 2, showing
high interannual variation (Vincke et al. 2005). The mild me-
Acknowledgments
This research was supported by the projects PIRIHEROS
(REN2003-08768/HID) and CANOA (CGL2004-04919-C02-01),
funded by the Spanish Ministerio de Ciencia y Tecnología
(MCYT)/Ministerio de Educación y Ciencia (MEC). The Vallcebre
research area also operates with support from the RESEL network
through an agreement between the CSIC and DGCONA. The first author benefited from a predoctoral FPI grant by the MCYT-MEC and
additional funding by the Spanish MEC for a research stay at the Mendel University in Brno. The authors are indebted to Martin Èermák,
TREE PHYSIOLOGY VOLUME 27, 2007
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
We demonstrated that the relationship between the weighted
mean of the sap flux density of the two shallowest measurement points of the HFD sensor and mean sap flux measured by
HD probes differed between two trees in which both HFD and
HD probes were installed. Large radial gradients in sap flux
density along HD probes pose serious difficulties for estimating mean values of sap flux (Clearwater et al. 1999). In addition, errors associated with the unavoidable inaccuracies in the
positioning of sensors may have led to these discrepancies,
which are more critical when gradients of sap flux density are
steep (Nadezhdina et al. 2002). Because the HD and HFD
probes were placed about 10 cm apart circumferentially, azimuthal variability of sap flow could also have been responsible
for the lack of correspondence between sensors.
Approaches for correcting single-point sap flow measurements according to radial variation of sap flow, similar to the
approach we applied, have been described (Köstner et al.
1996, Zang et al. 1996, Wullschleger and King 2000, Delzon
et al. 2004). We applied whole-tree correction coefficients directly to sap flux density measurements obtained by Granier
probes in the outer 10 mm of sapwood. However, we found no
relationship between the whole-tree correction coefficients
and DBH or other biometric parameters as reported in other
studies (Delzon et al. 2004).
Because of the significant variability observed among our
study trees, we did not attempt to generalize the observed radial patterns based on a single equation as has been done for
conifers using a Gaussian-like curve (Ford et al. 2004b) and
for a Fagus species using a Weibull function (Kubota et al.
2005). However, double-Gaussian curves (Nadezhdina et al.
2002) may describe the variability in radial profiles that we observed, but more measuring points along the xylem would be
necessary to fit these curves.
teorological conditions during summer 2004, compared with
the extremely dry period in summer 2003 (rainfall was 42%
higher in 2004, Table 4), resulted in a parallel increase in Ec
(39% higher during 2004). Recent experiments have shown
that Ec of a deciduous Quercus spp. forest was reduced by
23–32% when rainfall inputs were reduced by 33%
(Wullschleger and Hanson 2006).
Although we could not validate our corrected estimates of
Ec, we showed that once sapwood depth is accurately known,
ignoring radial variation in stem sap flow would lead to significant overestimations of sap flow rates (Figure 7). In a recent
study, neglecting the radial pattern of sap flow was identified
as the cause of previous overestimations of stand transpiration
in maritime pine (Delzon and Loustau 2005).
In conclusion, the general shape of the radial patterns of sap
flow found in this study agrees with previous findings on the
same species, although we observed higher variability among
trees. These differences may be attributable to the interference
of azimuthal variability of sap flow due to crown irregularity
or to seasonal variation in radial distribution of sap flow. Sapwood depths estimated by extrapolation of radial patterns were
higher than previous estimates based on visual inspection,
confirming the discrepancies between estimates based on the
water content of the hydroactive area of xylem and the true
conducting area in ring-porous trees. We found no clear correspondence between sap flux density measured by the HD
probes and the calculated mean of HFD measurements, which
may be the result of uncertainties in the averaging of sap flux
density by HD probes, azimuthal variation in sap flow or positioning errors when installing the probes.
Sap flow in the inner xylem relative to total stem flow
tended to be higher during the early morning and late in the
day, when evaporative demand was lower. This observation
seems to contradict recent findings suggesting that high evaporative demand should mobilize water in the inner xylem in conifers; however, it supports the results of other studies indicating the relevance of hydraulic connections between more exposed foliage and outer xylem.
The use of sapwood areas determined from radial patterns
and the correction of HD sap flux density measurements to account for radial variation yielded stand transpiration estimates
that were up to 25% higher than previous uncorrected estimates. Given the high variability in radial patterns found
among trees growing in the same plot, the assessment of these
patterns is recommended when scaling-up single-point sap
flow measurements in ring-porous species.
RADIAL PATTERNS OF SAP FLOW IN RING-POROUS OAKS
Juliana Delgado, Núria Martínez-Carreras, Xavier Huguet, C.Rubio,
F.Gallart, J.Latron and M.Soler for their help with field work, and
Nadezhda Nadezhdina for her support in applying the HFD method.
References
Granier, A. 1985. Une nouvelle méthode pur la mesure du flux de sève
brute dans le tronc des arbres. Ann. Sci. For. 42:193–200.
Granier, A. 1987. Evaluation of transpiration in a Douglas-fir stand by
means of sap flow measurements. Tree Physiol. 3:309–320.
Granier, A. and D. Loustau. 1994. Measuring and modeling the transpiration of a maritime pine canopy from sap-flow data. Agric. For.
Meteorol. 71:61–81.
Granier, A., T. Anfodillo, M. Sabatti, H. Cochard, E. Dreyer,
M. Tomasi, R. Valentini and N. Bréda. 1994. Axial and radial water
flow in the trunk of oak trees: a quantitative and qualitative analysis. Tree Physiol. 14:1383–1396.
Hatton, T.J., E.A. Catchpole and R.A. Vertessy. 1990. Integration of
sapflow velocity to estimate plant water-use. Tree Physiol.
6:201–209.
Himrane, H., J.J. Camarero and E. Gil-Pelegrín. 2004. Morphological
and ecophysiological variation of the hybrid oak Quercus
subpyrenaica (Q. faginea × Q. pubescens). Trees 18:566–575.
Hirose, S., A. Kume, S. Takeuchi, Y. Utsumi, K. Otsuki and
S. Ogawa. 2005. Stem water transport of Lithocarpus edulis, an evergreen oak with radial-porous wood. Tree Physiol. 25:221–228.
James, S.A., M.J. Clearwater, F.C. Meinzer and G. Goldstein. 2002.
Heat dissipation sensors of variable length for the measurement of
sap flow in trees with deep sapwood. Tree Physiol. 22:277–283.
James, S., F.C. Meinzer, G. Goldstein, D. Woodruff, T. Jones,
T. Restom, M. Mejia, M.J. Clearwater and P. Campanello. 2003.
Axial and radial water transport and internal water storage in tropical forest canopy trees. Oecologia 134:37–45.
Jiménez, M.S., N. Nadezhdina, J. Èermák and D. Morales. 2000. Radial variation in sap flow in five laurel forest tree species in
Tenerife, Canary Islands. Tree Physiol. 20:1149–1156.
Kitin, P.B., T. Fujii, H. Abe and R. Funada. 2004. Anatomy of the vessel network within and between tree rings of Fraxinus lanuginosa
(Oleaceae). Am. J. Bot. 91:779–788.
Köstner, B., P. Biron, R. Siegwolf and A. Granier. 1996. Estimates of
water vapour flux and canopy conductance of Scots Pine at the tree
level utilizing different xylem sap flow methods. Theor. Appl.
Clim. 53:105–113.
Kubota, M., J. Tenhunen, R. Zimmermann, M. Schmidt and
Y. Kakubari. 2005. Influence of environmental conditions on radial
patterns of sap flux density of a 70-year Fagus crenata trees in the
Naeba Mountains, Japan. Ann. Sci. For. 62:289–296.
Loustau, D., J.C. Domec and A. Bosc. 1998. Interpreting the variations in xylem sap flux density within the trunk of maritime pine
(Pinus pinaster Ait.): application of a model for calculating water
flows at tree and stand levels. Ann. Sci. For. 55:29–46.
Lu, P., W.J. Muller and E.K. Chacko. 2000. Spatial variations in xylem sap flux density in the trunk of orchard-grown, mature mango
trees under changing soil water conditions. Tree Physiol.
20:683–692.
Meiresonne, L., D.A. Sampson, A.S. Kowalski, I.A. Janssens,
N. Nadezhdina, J. Èermák, J. Van Slycken and R. Ceulemans.
2003. Water flux estimates from a Belgian Scots pine stand: a comparison of different approaches. J. Hydrol. 270:230–252.
Miller, D.R., C.A. Vavrina and T.W. Christensen. 1980. Measurement
of sap flow and transpiration in ring-porous oaks using a heat pulse
velocity technique. For. Sci. 26:485–494.
Monteith, J.L. 1965. Evaporation and environment. Symp. Soc. Exp.
Biol. 19:205–234.
Nadezhdina, N., J. Èermák and V. Nadezhdin. 1998. Heat field deformation method for sap flow measurements. In 4th Int. Workshop on
measuring sap flow in intact plants. Eds. J. Èermák and
N. Nadezhdina. IUFRO Publications, Zidlochovice, Czech Republic, pp 72–92.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
Aloni, R., J.D. Alexander and M.T. Tyree. 1997. Natural and experimentally altered hydraulic architecture of branch junctions in Acer
saccharum Marsh. and Quercus velutina Lam. trees. Trees
11:255–264.
Allen, R.G., L.S. Pereira, D. Raes and M. Smith. 1998. Crop
evapotranspiration. Guidelines for computing crop water requirements. In FAO Irrigation and Drainage, Paper No.56 Ed. FAO,
Rome, 290 p.
Aranda, I., L. Gil and J.A. Pardos. 2005. Seasonal changes in apparent
hydraulic conductance and their implications for water use of European beech (Fagus sylvatica L.) and sessile oak [Quercus petraea
(Matt.) Liebl.] in South Europe. Plant Ecol. 179:155–167.
Èermák, J. and N. Nadezhdina. 1998. Sapwood as the scaling parameter-defining according to xylem water content or radial pattern of
sap flow? Ann Sci For. 55:509–521.
Èermák, J., E. Cienciala, J. Kuèera and J.E. Hallgren. 1992. Radial
velocity profiles of water flow in trunks of Norway spruce and oak
and the response of spruce to severing. Tree Physiol. 10:367–380.
Èermák, J., N. Nadezhdina, A. Raschi and R. Tognetti.1998. Sap flow
in Quercus pubescens and Quercus cerris stands in Itay. In 4th Int.
Workshop on Measuring Sap Flow in Intact Plants. IUFRO Publications-Publishing House of the Mendel University, Zidlochovice,
Czech Republic, 149 p.
Clearwater, M.J. and A.H. Goldstein. 2005. Embolism repair and long
distance water transport. In Vascular Transport in Plants. Eds.
N.M. Holbrook and M.A. Zwieniecki. Elsevier/Academic Press,
Oxford, U.K., pp 307–318.
Clearwater, M.J., F.C. Meinzer, J.L. Andrade, G. Goldstein and
N.M. Holbrook. 1999. Potential erros in measurement of nonuniform sap flow using heat dissipation probes. Tree Physiol.
19:681–687.
Delzon, S. and D. Loustau. 2005. Age-related decline in stand water
use: sap flow and transpiration in a pine forest chronosequence.
Agric For Meteorol. 129:105–119.
Delzon, S., M. Sartore, A. Granier and D. Loustau. 2004. Radial profiles of sap flow with increasing tree size in maritime pine. Tree
Physiol. 24:1285–1293.
Domec, J.C., F.C. Meinzer, B.L. Gartner and D. Woodruff. 2006.
Transpiration-induced axial and radial tension gradients in trunks
of Douglas-fir trees. Tree Physiol. 26:275–284.
Ellmore, G.S. and F.W. Ewers. 1986. Fluid flow in the outermost xylem increment of a ring-porous tree, Ulmus americana. Am. J. Bot.
73:1771–1774.
Ford, C.R., C.E. Goranson, R.J. Mitchell, R.E. Will and R.O. Teskey.
2004a. Diurnal and seasonal variability in the radial distribution of
sap flow: predicting total stem flow in Pinus taeda trees. Tree
Physiol. 24:941–950.
Ford, C.R., M.A. McGuire, R.J. Mitchell and R.O. Teskey. 2004b. Assessing variation in the radial profile of sap flux density in Pinus
species and its effect on daily water use. Tree Physiol. 24:241–249.
Gallart, F., P. Llorens, J. Latron and D. Regüés. 2002. Hydrological
processes and their seasonal controls in a small Mediterranean
mountain catchment in the Pyrenees. Hydrol. Earth Syst. Sci.
6:527–537.
Gartner, B.L. and F.C. Meinzer. 2005. Structure-function relationships in sapwood water transport and storage. In Vascular Transport in Plants Eds. N.M. Holbrook and M.A. Zwieniecki.
Elsevier/Academic Press, Oxford, U.K., pp 307–318.
547
548
POYATOS, ÈERMÁK AND LLORENS
Rubio, C. 2005. Hidrodinámica de los suelos de un área de montaña
media mediterránea sometida a cambios de uso y cubierta. In
Departament de Biologia Animal, Vegetal i Ecologia. Universitat
Autònoma de Barcelona, Spain, 194 p.
Spicer, R. and B.L. Gartner. 2001. The effects of cambial age and position within the stem on specific conductivity in Douglas-fir
(Pseudotsuga menziesii) sapwood. Trees 15:222–229.
Vincke, C., N. Bréda, A. Granier and F. Devillez. 2005.
Evapotranspiration of a declining Quercus robur (L.) stand from
1999 to 2001. II. Daily actual evapotranspiration and soil water reserve. Ann. Sci. For. 62:615–623.
Wullschleger, S.D. and A.W. King. 2000. Radial variation in sap velocity as a function of stem diameter and sapwood thickness in yellow poplar trees. Tree Physiol. 20:511–518.
Wullschleger, S.D. and P.J. Hanson. 2006. Sensitivity of canopy transpiration to altered precipitation in an upland oak forest: evidence
from a long-term field manipulation study. Global Change Biol.
12:97–109.
Zang, D., C.L. Beadle and D.A. White. 1996. Variation in sapflow velocity in Eucalyptus globulus with position in sapwood and use of a
correction coefficient. Tree Physiol. 16:697–703.
TREE PHYSIOLOGY VOLUME 27, 2007
Downloaded from http://treephys.oxfordjournals.org/ by guest on August 28, 2014
Nadezhdina, N., J. Èermák and R. Ceulemans. 2002. Radial patterns
of sap flow in woody stems of dominant and understory species:
scaling errors associated with positioning of sensors. Tree Physiol.
22:907–918.
Nadezhdina, N., H. Tributsch and J. Èermák. 2004. Infra-red images
of heat filed around a linear heater and sap flow in stems of lime
trees under natural and experimental conditions. Ann. Sci. For.
61:203–213.
Nadezhdina, N., J. Èermák, J. Neruda, A. Prax, R. Ulrich,
V. Nadezhdin, J. Gašpárek and E. Pokorny. 2006. Roots under the
load of heavy machinery in spruce trees. Eur. J. For. Res.
125:111–128.
Nardini, A. and F. Pitt. 1999. Drought resistance of Quercus
pubescens as a function of root hydraulic conductance, xylem embolism and hydraulic architecture. New Phytol. 143:485–493.
Phillips, N., R. Oren and R. Zimmerman. 1996. Radial patterns of
sylem sap flow in non-, diffuse- and ring-porous tree species. Plant
Cell Environ. 19:983–990.
Phillips, N., B.J. Bond, N.G. McDowell, M.G. Ryan and A. Schauer.
2003. Leaf area compounds height-related hydraulic costs of water
transport in Oregon White Oak trees. Funct Ecol. 17:832–840.
Poyatos, R., P. Llorens and F. Gallart. 2005. Transpiration of montane
Pinus sylvestris L. and Quercus pubescens Willd. forest stands
measured with sap flow sensors in NE Spain. Hydrol. Earth Syst.
Sci. 9:493–505.