Effect of Water Stress on Stem Diameter Changes of Peach Trees Growing in the Field

Author(s): E. Garnier and A. Berger

Source: Journal of Applied Ecology , Apr., 1986, Vol. 23, No. 1 (Apr., 1986), pp. 193-209
Published by: British Ecological Society

Stable URL: https://www.jstor.org/stable/2403091

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

British Ecological Society is collaborating with JSTOR to digitize, preserve and extend access to
Journal of Applied Ecology

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
Journal of Applied Ecology (1986) 23, 193-209

EFFECT OF WATER STRESS ON STEM DIAMETER

CHANGES OF PEACH TREES GROWING IN THE FIELD

BY E. GARNIER AND A. BERGER*

Laboratoire d'Ecophysiologie, Centre Emberger (C.N.R.S.), Route de Mende,

BP 5051, 34033 Montpellier-Cedex, France

SUMMARY

(1) Daily stem diameter changes were recorded continuously from June to September
on peach trees irrigated at 50 and 100% of calculated maximum evapotranspiration.
In addition, the stem water potential of these trees was measured on several days
of the season.
(2) The maximum daily shrinkage (MDS) is shown to be dependent on environmental
factors affecting plant transpiration, i.e. the soil water content-the single factor best
explaining MDS variance-and the potential evapotranspiration. A multiple linear
regression involving soil water content and potential evapotranspiration accounted for
66% of the MDS variance.
(3) The relationship between stem water potential and stem diameter changes during
the course of the day showed a marked hysteresis, with, for a given water potential,
a greater diameter in the morning than in the afternoon. It is shown that a minimum
potential drop of 0-4 MPa compared to predawn value was necessary to induce onset
of stem shrinkage. The reasons for this time lag between these two parameters is
discussed.
(4) A good linear relationship was found (r2 = 0.83) between MDS and the minimum
daily stem water potential of the day, suggesting that environmental conditions affect
MDS through the mediation of plant water potential.
(5) Finally, the possibility of using stem diameter changes for irrigation purposes
is briefly discussed.

INTRODUCTION

The water stored in the different parts of a tree can, to a certain extent, feed the
transpiration stream flowing in the conducting units (Chaney 1981). In hardwood plants,
the quantity of water stored in the stem alone has been evaluated to represent about
19 h of potential transpiration distributed in the sapwood and in the extensible tissues
(Chaney 1981). The availability of this stored water is a matter of discussion (Zimmermann
& Milburn 1982), since Powell & Thorpe (1977) showed from different tree capacitance
calculations that nearly all the water available for transpiration was in the extensible tissues
in apple trees, whereas in conifers, the sapwood can also substantially feed the transpiration
stream (see, for example, Jarvis 1975).
The shrinking and swelling that take place in the extensible plant tissues during
daylight periods provide an indirect measure of the quantity of water released in the
transpiration stream from these tissues, and are related to the changes in the water content
and turgor potential of their cells (reviewed by Kozlowski 1972). In the stem, these
extensible tissues consist essentially of a narrow ring located outside the dead xylem
vessels, and include the phloem, the cambium and the immature cells of the protoxylem

* To whom offprint requests should be sent.

193

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 194 Stem diameter changes in peach trees

(this group of tissues will be referred to as 'phloem') the xylem tissue being almost totally
rigid (Dobbs & Scott 1971; Molz & Klepper 1973; Jarvis 1975).
It is generally assumed that shrinking and swelling are induced by the development of
water potential gradients between the phloem and the conducting xylem, producing
water fluxes between these two tissues (Dobbs & Scott 1971; Molz & Klepper 1972).
The link between these stem deformations and the water status of the plant remains
unclear, especially in the explanation of the phase lags observed between stem deformations
and water potential (see the discrepancy between measured and predicted values in the
'dynamic 91ux method' presented by Huck & Klepper 1977). However, numerous attempts
have been made to use stem diameter variations as an integrated indicator of plant
water stress since measurement is non-destructive and can be recorded continuously
(e.g. Namken, Bartholic & Runkles 1969, 1971; So 1979; Meyer 1980; Gensler &
Diaz-Munoz 1983).
In a study concerning peach-tree irrigation, attempts were made to quantify the influence
of varying soil and atmospheric water conditions on stem diameter variations to test
whether they could be used as a sensitive indicator of plant water stress for irrigation
scheduling. Stem water potential was also measured on several days, and the relation
between these two parameters of plant water status was examined.

MATERIAL AND METHODS

Experimental site and plant material

This research was conducted in an experimental orchard of the Ecole Nationale

Sup&rieure Agronomique de Montpellier, 8 km south-east of Montpellier (France;
43034'N, 3057'E) from June to September 1984. Peach trees (Prunuspersica (L.) Batsch
cv. Springcrest on GF 305 rootstock) were planted in January 1981 at 4.5 m spacing
in rows 6 m apart. The soil is alluvial with a gravelly layer at 50 cm depth. It is classified
as calcic-luvisol (F.A.O. classification). Main roots are located in the upper 0-50 cm
clay-silt layer with some minor ones going down to 60 cm.

Meteorological data
Precipitation, short-wave radiation (Rg), Piche evaporation (Ep), air temperature and
vapour pressure, windspeed and relative duration of sunshine were kindly provided by
Montpellier-Frejorgues airport meteorological station, 1.7 km south-east of the orchard.
Potential evapotranspiration (PET) was calculated as:

PET=m-Rg + nEp, (1)

where Rg is in cal* cm-2, PET and

ing on attitude and time of the year (Brochet & Gerbier 1972). This was used to time the
irrigations (see below). The validity of this formula has been demonstrated for periods
of 10 days or more (Brochet & Gerbier 1972; Seguin 1975). Penman potential evapo-
transpiration (see Seguin 1975) was also calculated and used when daily PET values
were necessary.

Soil moisture

Soil water content data presented in this paper were obtained with a neutron probe
(Pitman Instruments, Wallingford, model 225) used in access tubes placed at 50 cm

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 195

from the tree trunks within a row (one tube for each treatment). Complete description
of the device and the method used can be found in Garnier, Berger & Rambal (1986).

Irrigation

The trees were sprinkler irrigated and two treatments were applied:
'Wet treatment': eight trees were irrigated at 100% of maximum evapotranspiration
(MET) when half of the estimated readily available soil water between 0 and 70 cm
was depleted. The MET was calculated from Brochet-Gerbier equation, corrected by a
crop coefficient of 0.9 until harvest (23 June) and 0.6 thereafter. These are the coefficients
recommended for peach-tree irrigation in the south of France.
'Dry treatment': six trees were irrigated on every other occasion when wet treatment
was applied, with the same amount of water per irrigation as in the wet treatment.
Integrated over a complete cycle of drying and irrigation, this represents 50% MET.

Stem diameter changes

Stem diameter changes were measured with four linear variable displacement
transducers (LVDT; If6lec, Paris, model L5R), each one mounted in a metal frame
(Fig. 1); a release spring allowed the plunger of the LVDT to be permanently in contact
with the outer bark of the trunk. Output of the LVDT was 4 mV (10 Alm)-', with a f
range of 2.5 mm and 1 V output. The whole apparatus-LVDT plus frame-was
calibrated in the range 10-35 0C to account for temperature drift, and temperature in
the immediate proximity of the apparatus was continuously recorded during operation
in the field.

FIG. 1. Frame for attaching the LVDT (linear variable displacement transducer) to peach-tree
stem. P, plunger of the LVDT; G, guide for the LVDT; LR, lateral rods in INVAR adjustable
between 7 and 15 cm; RF, rubber foam layer; J, jaws of the frame between which the tree
trunk is inserted.

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 196 Stem diameter changes in peach trees

The LVDT was placed on the main trunk, below the first branch of the tree, 10-30 cm
above ground. This is the best way to integrate all that happens in the tree canopy.
During operation, a plastic cylinder painted in white surrounding the device was set on
the ground. An aluminium roof at the top of the cylinder completed the protection of
the device from both direct sunlight and rain.
Four trees-two per treatment-were equipped continuously with the devices from
1 June to 24 September except for 10 days in July, and when irrigation water was applied.
The numeric outputs of the four LVDTs and sensor temperature were recorded every
32 min on the tape of a data logger (Alep Electronique, Grenoble, model ED 7210)
which had an autonomy of about 6 days. Once removed from the data logger, the tape
was read and the data transferred to a microcomputer floppy disc and then to a larger
computer.
The LVDTs, temperature sensor and data logger were powered by a 12-V car battery
located with the data logger in a wooden shelter painted in white placed under a tree
canopy. Temperature in the shelter was always maintained below 30 IC, which was below
the maximum working temperature of the data logger.
Stem diameter changes can be divided into an irreversible component due to the
growth of tissues and a reversible component due to the varying hydration of the extensible
tissues (Kozlowski 1972). It is possible, therefore, to define a daily growth, corresponding
to the difference between two successive daily maximum values of the stem diameter
(before shrinkage), and a daily maximum shrinkage corresponding to the difference
between the maximum (before onset of shrinkage) and the minimum values of the
diameter attained during this day.

Stem water potential

The xylem water potential of the stem (ts) was measured with a pressure chamber
(PMS Instrument Co.) (Scholander et al. 1965) in the following way: a sunlit mature leaf
was selected in the middle of a shoot and enclosed 2 h prior to measurement in a
cellophane bag covered with aluminium foil. This stops leaf transpiration and enables the
potential in the xylem of the leaf to come to equilibrium with the potential in the xylem of
the stem at the point of attachment of the petiole (Begg & Turner 1970; Powell 1974;
Passioura 1982, p. 16). This assumption has been confirmed by the work of McBurney &
Costigan (1982), who showed on young cabbage plants, that the water potential measured
with a hygrometer attached to the stem was the same as the one measured with a pressure
chamber on leaves previously covered with aluminium foil. The As values presented
are the means of three measurements taken at a given time and were measured on the same
trees (one per day) as the stem diameter changes.

RESULTS

Seasonal course of daily growth and maximum daily shrinkage

Cumulative trunk growth of two trees is presented at the top of Fig. 2. The 0 reference
values of the stem diameters on 1 June correspond to real diameters of 10-52 cm for the
50% MET tree and 11-54 cm for the 100% MET tree. In spite of somewhat different
behaviour of the two trees until the end of June, growth did not seem to be much
affected in the dry treatment until mid-July, when daily growth began to decrease
compared to the wet treatment. Stem diameter increase stopped in both treatments at

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 197

E 8 -
E

2 6

E 4 -

E/

CX) 2 -

100

E 0 r -' -Tlrm
E >3mZ 150-
05 100 I
200
100

oO 1 50 -

20 -I

June July August September

FIG. 2. TOP: seasonal course of stem diameter growth (morning diameters, before onset of stem
shrinkage --:50% MET -: 100% MET). Middle: evolution of maximum daily shrinkage
(MDS) represented by a vertical bar, for 100% MET (top) and 50% MET (bottom) treatments.
The bottom of the figure shows natural rainfall (--), irrigation on both treatments (---) and
on wet treatment only (---)

the beginning of September. The cumulative growth over the season was 6.6 and 7.9 mm
in dry and wet treatments, respectively.
Individual values of maximum daily shrinkage of the stem (MDS) in a given treatment
varied widely from day to day (Fig. 2, middle), depending partly on environmental
conditions (see below). For example, MDS values can be very low during rainy days:
9 and 11 ,um for dry and wet treatments respectively on 24 August, a day with greatly
overcast sky and heavy thunderstorms. The maximum MDS value (257 yin) was obtained
on 7 August for the dry treatment.
In both treatments, MDS tended to increase as the season progressed: Table 1
shows that the mean MDS value averaged over 5-7-day periods is greater in August
and September compared to June, which was associated with a marked decrease in
soil water content (Table 1) and to the end of the growth period (Fig. 2, top).
The MDS was greater in the dry treatment than in the wet one throughout the season
(Fig. 2). The mean average ratio estimated over the season between MDS in the two
treatments was roughly 0 5 (Fig. 3).

Influence of environmental conditions on stem shrinkage

Soil water content

In Fig. 4, MDS is plotted against the soil water content averaged over 60 cm, for
the days when soil moisture was measured (i.e. 24 points). For these days, PET varied

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 198 Stem diameter changes in peach trees

TABLE 1. Mean daily values of the maximum daily shrinkage (MDS, with the
range of variation between brackets) and potential evapotranspiration (PET)
calculated for the given time period, and soil content (SWC) averaged over the
upper 60 cm of soil for that period
50% MET 100% MET
PET MDS SWC MDS SWC
Time period (mm) (gim) (cm3. cm-3) (am) (cm3. cm-3)
6-10June 4.8 66 (29-116) 0.197 18 (3-43) 0 217
15-21 June 5.4 75 (58-107) 0.167 51 (34-76) 0 210
27June-l July 5.1 37 (3-100) 0.188 14 (3-44) 0 212
2-8July 6.5 87 (5-129) 0.161 12 (0-39) 0.190
24-30July 7.1 102 (34-174) 0.190 56 (14-76) 0.214
6-12August 5.2 177 (50-257) 0.160 61 (7-107) 0 203
19-25 August 3.6 49 (9-85) 0.166 32 (0-60) 0.199
26 Aug-1 Sept 3.9 110 (5-173) 0.168 65 (1-90) 0.200
4-10September 4.2 165 (95-206) 0.162 70 (29-112) 0.189
11-15September 3.7 172(117-217) 0.161 103 (72-147) 0.178
17-23 September 3-6 183 (117-226) 0.155 119 (35-172) 0.172

E~ 120
E120 I~~~~~~~~~ 0 0 0

Q~o 0 0 ?0
~~~ 80 ~~~~~~~~0
0)~~~~~~~~~~~~

Z 100 - 0
0- - ?

C~ 60 0 00X

0
Ec4 00
00 0
0 0

E 20 O O00 0

20 60 100 140 180 220

Maximum daily shrinkage, 50% MET (pum)

FIG. 3. Relation between MDS in the 50% MET and in the 100% MET treatments. The 0.5 line
has also been drawn.

between 3 and 7.7 mm. In spite of this wide range of PET values, there was a strong
influence of soil water content on stem shrinkage: for example, on 29 June (PET =
6.2 mm), the soil water content was 0*19 and 0.21 cm3 * cm-3 in dry and wet treatments
respectively, and MDS was respectively 119 and 51 sum. The data indicates that a
decrease in soil water content from 0.21 to 0. 15 cm3*cm-3 has led to an increase in
MDS from 34 to 152 tim.
This type of reaction contrasts markedly with the response of predawn water potential
to soil water content: this latter parameter is insensitive to the decrease in soil water
content from 0.22 to 0.17 cm3*cm-3, and decreases sharply thereafter (B. Gamnier &
A. Berger, unpublished data).
The circled point on this figure (not taken into account in the regression analysis)
corresponds to a day following an irrigation on the wet treatment. On several occasions
(see Fig. 2), we noticed that just after the irrigations, when the soil was flooded, daily

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 199

180 0
0

1 60
10 \ o A
3 140

V 120 \ 0
C ~~~~0
100 0

- 80 0

E ZsA
x~~~~~~~
co 0~~~A
-40 0 A

20A

0
0 15 0 17 0 19 0 21 0 23
Average soil water content (0-60 cm)(cm3-cm-3)

FIG. 4. Relationship between the soil water content averaged over the upper 60 cm of soil and
MDS (mean of the two trees in each treatment). The linear regression is MDS = -1963 SWC
+ 447 (r=0.79,P < 0.001). (0), 50% MET; (A), 100% MET.

shrinkage was very pronounced. This may be due to a poor aeration of the roots just
after water application.

Potential evapotranspiration
Figure 5 shows the influence of PET on MDS for the dry treatment only (for the
whole season, i.e. 73 points). The regression analysis shows only a fair correlation between
these two variables (r2 = 0.38 and 0. 15 (P < 0.001) for the wet treatment (not shown)),
probably reflecting the pronounced influence of soil water content which is not taken into
account in this simple linear regression. As the correlation between soil water content and

220 0
w 200 -
o0

180 -0000 0

? 160 - Cb X
~140 -0 00
C2120 ? o0 0
xo 20 0 0 00 0
00 %Of 0
?80 -8 o
-0 60 0 0 0%
E
~ 40 0 ~ 0 00
E 0 - 0 0 000
x 20 0 0 0
co 000 0
75 0 0
1 3 5 7 9
Potential evapotranspiration (mm)

FIG. 5. Relationship between the daily potential evapotranspiration (PET) and MDS for the
50% MET treatment (mean of two trees). The regression line is: MDS = 19. 1 PET - 6.6
(r=0.62,P <0.001).

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
200 Stem diameter changes in peach trees

TABLE 2. Correlation matrix for the simple linear relationships between the different
variables (see text). Each point for MDS is the mean of the two trees in each
treatment. N= 24

MDS SWC PET

MDS 1
SWC -0-79 1
PET 0 04 0-19 1

MDS is the maximum daily shrinkage, SWC is

the soil water content and PET is the potential
evapotranspiration.

PET is low (r = 0. 19, see Table 2), this data shows that an increase in PET leads to an
increase in the stem shrinkage of the trees.

Combined influence of the different factors

A step-wise multiple regression was used to investigate the influence of the different
variables on MDS. Table 3 shows that soil water content alone explains 62% of the
MDS variance. Introducing the other environmental variable, PET, increases to 66%
the part of the explained variance. However, this improvement is significant at the
0.15 level only. This contrasts with the significance of the simple linear fits when
treatments are separated (P < 0.001 for both treatments, see above). Such a difference
may be due to (i) a difference in the number of points used in the regressions: 24 for the
multiple one (days when the soil water content was measured) and 73 for the simple
ones (whole season), and (ii) the range of PET values, which was 3-7.7 mm in the
multiple regression and 1X5-9 mm in the simple ones. Indeed, it seems that extreme PET
values (below 3 mm for the low values and above 7.7 mm for the high ones) stabilize the
regression line.

TABLE 3. Results of a step-wise multiple linear regression introducing first soil water
content (SWC) and then the potential evapotranspiration (PET). MDS is in gm,
SWC is in cm3. cm-3 and PET is in mm.

% variance
Regression equation accounted for

MDS=-1963 SWC + 447 62

MDS =-2054 SWC + 6.6 PET + 430 66

Stem shrinkage in relation to stem water potential

Diurnal course of plant parameters: clear days

Figure 6 shows the diurnal course of stem diameter changes, stem water potential and
short-wave radiation for two 24-h periods with clear daylight periods.
On 18 July (2 days before a 50-mm irrigation on both treatments), the soil water
content averaged over the upper 60 cm of soil was 0. 15 and 0. 19 cm3 cm-3 in dry and
wet treatments respectively.
Figure 6 (left) shows that stem water potential decreased in both treatments as soon as
the sun rose (between 04.30 and 05.00 h T.S.T.) until 10.00 h in the wet treatment and
11.30 h in the dry one. Between 10.00 and 15.00 h, Vs values fluctuated between 1.06
and 119 MPa (minimum daily value) in the wet treatment and between 1.47 and

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 201

900 900

l, E 700 - 700
? 500 5 A00

C/O 300 -300

100 r PET= 7 4 mm PET= 5 4 mm 100

02 81 2

0~

( u0 1 4 t w-1 4

I,~~ ~ I , ,I ,I I I, Ii ,1 ,I I

(predawn, 8 J
FIG.
E 6.
_ Dail1-
6oreo 00
hr-aerdain(o)
--'00 tmwtrptnil(iden
E co~~~~~~~~~~c

200 NO-Ce I ~~~~~~~~~~~~-100

-0

1g5 Q)a (mniu dalCau)i(h r nfrmn suopaeufr5h

4 8 1 2 1 6 20 24 4 8 4 8 12 16 20 24 4 8
1 8 July 1 6 August
True solar time

FIG. 6. Daily course of short-wave radiation (top), stem water potential (middle) and
stem diameter changes (bottom) for 2 clear days: 18 July (left) and 16 August (right).
(0----0), 50% MET; (A ~A), 100% MET.

1p56 MPa (minimum daily value) in the dry one, forming a pseudo-plateau for 5 h.
Stem water potential values increased thereafter, in two phases. A rapid increase
(around 01 MPa h1 ) until 20.30 0 h, was followed by a slower one (0.02 MPa h ) lasting
until sunrise the next day. One interesting point to notice is that the onset of water
potential decrease on 19 July occurred when water potential values were still increasing.
This seems to indicate that the equilibrium between the soil and the plant was not reached
at that time. Differences in s between the two treatments ranged from 0J28 MPa
(predawn, 18 July) to 0.45 MPa (16.05 h), which shows the pronounced effect of different
soil water contents on this parameter (see also Gamnier and Berger 1985).
Stenfi diameter changes (Fig. 6, bottom) followed the same general trend as water
potential with some different features: decrease of stem diameter started around 06.00 h,
at least 1 h after the onset of decrease of water potential. Minimum diameters were
reached at 13.00 and 14.07 h in wet and dry treatments respectively, and started to
increase thereafter, with no pseudo-plateau. The two-phase kinetics of the diameter
increase is particularly clear in the dry treatment: from 14.00 to 20.00 h, the rate of
recovery was 23 um h-' whereas from 20.00 to 06.00 h, it was only 13 pUm h-'. This
phenomenon has been noticed on both treatments, particularly from mid-July to
September, and has also been reported by Namken, Bartholic & Runkles (1969) and
Jordan & Ritchie (1971) for the stems of cotton plants and by Elfving & Kaufmann
(1972) for citrus fruits.

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
202 Stem diameter changes in peach trees

Maximum daily shrinkage was 213 and 70,um in dry and we

and daily growth was respectively 28 and 67 um. Stem diameter recovered its morning
value much earlier in the wet treatment (21.00 h) than in the dry one (03.30 h), which
was also commonly observed throughout the season, except after the irrigations on
both treatments (see below).
Figure 6 (right) shows the diurnal pattern of the same parameters on 16 August,
2 days after a 32-mm irrigation on both treatments. Soil water content averaged over
0-60 cm was 0 19 and 0.23 cm3 cm-3 on dry and wet treatments respectively. However,
the upper part of the profile was near field capacity in both treatments, and therefore,
some of the roots could extract water in soil zones with very similar soil water contents.
Predawn water potential was 0.20 MPa and minimum daily Vs value 1.08 MPa in wet
treatment, and 0.28 and 1. 19 MPa in the dry one. Differences between the two treatments
were much less pronounced than on 18 July: they range from 0.02 MPa at 06.40 h to
a maximum of 0.15 MPa at 10.15 h. No real plateau could be observed on that day,
which was rather exceptional.
Stem diameter changes were alike in the two treatments, with a phase lag between
water potential and diameter changes of about 01.30 h in the dry treatment and 03.30 h
in the wet one (these values of the time lags are for the beginning of the day; indeed,
phase lags are difficult to observe for the minimum daily values, because the water
potential curve usually forms a pseudo-plateau at that time of the day). Minimum
diameter was reached at 14.00 h in both treatments, with nearly the same MDS:
108,um for the dry treatment, and 113 um in the wet one.
The rate of diameter increase was 14 um h-' in the rapid phase (from 14.00 to
19.00 h), and about 7 um h-' (from 19.00 to 06.00 h) in the slow one. Daily growth
was 27,um in the dry treatment and 41 um in the wet one. Stem diameters recovered thei
morning values at the same time in the two treatments (21.30 h).

Diurnal pattern of stem diameter changes: cloudy days

Figure 7 shows the stem diameter changes of two trees for 2 days when the weather
was either variable (9 August) or very cloudy (10 August). On 9 August, the small
increase in radiation between 15.00 and 16.00 h occurring when the diameters were
increasing, caused the stem to shrink much more rapidly (within 30 min) than in th
morning when the sun rose. This very rapid response to the rapidly changing
environment has been described by Stansell et al. (1973). However, the scanning interva
of our data logger did not allow a precise analysis of the phenomenon.
On 10 August, diurnal shrinkage was hardly apparent, especially in the wet treatment,
due to the very cloudy weather (Rg = 429 J. cm-2), and low evaporative demand.
During this 2-day period, cumulative growth was important in the two treatments:
90,um day-' in the dry one and 123 um day-' in the wet one, which can be compared
to the values for clear days presented above.

Link between stem diameter changes and water potential

Figure 8 shows stem diameter changes during the course of 2 days plotted against
the difference between stem water potential at a given time and the corresponding
predawn (or base) water potential of each day.
There was a marked hysteresis in this relation on both days, with a larger diameter
for a given water potential, in the morning than in the afternoon. This reflects the fact
that stem diameter changes lag behind the variations of stem water potential (Fig. 6). On

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 203

800
as E t~PET= 4 7 mm PET= 1 5 mm
co 600 -

O.90 400-

co 200-

200 -

160 - 0

120

-80 -

C) 0

40 -

E 0 - -

4 8 12 16 20 24 4 8 12 16 20 24
9 August 1 0 August
True solar time

FIG. 7. Daily course of short-wave radiation (top) and stem diameter changes (bottom) for
2 cloudy days: 9 and 10 August. Same symbols as Fig. 6.

both treatments for these 2 days, a fall of about 0.4-0.6 MPa of stem water potential
values compared to predawn values was necessary before the stem began to shrink.
On 18 July, shrinkage was much more pronounced in the dry treatment than in the
wet one, for a slightly greater (1.0 MPa instead of 0.92 MPa) potential drop. On
16 August, after the irrigation, the curves for the two trees were very similar, indicating
a good recovery of the droughted tree after water stress.
If we plot the stem diameter changes during the course of a day against Vs - Vb for
the desorption phase only (after Namken, Bartholic & Runkles 1971), i.e. when water
potential and stem diameter both decrease, we obtain the graph on Fig. 9. The
points are for 7 days of measurement, with varying values of predawn water potential
(from 0.18 to 0.82 MPa) and of minimum daily stem water potential (from 0.73 to
1-.88 MPa). For all these days, we can see that a potential drop of about 0.4 MPa
compared with the predawn values was necessary to cause stem shrinkage. A further
0.4-MPa drop from 0.4 MPa to 0.8 MPa caused the stem to shrink between 20 and
100 pIm, and a further 0.2-MPa decrease to 1.0 MPa caused major shrinkage (up to
180 ,um).
Finally, MDS values from both treatments were plotted against the minimum daily
values of stem water potential (Fig. 10), producing a good linear relationship (r2 = 0.83,
P < 0.001): a drop of minimum daily stem water potential from -0.8 to -1.8 MPa
caused an increase in MDS from 40 to 200 um. The two extreme points on this figure

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
204 Stem diameter changes in peach trees

18 July 16 August

40 X 40

20 20

0-~~~~~~~~~~~~~~~~~
E' --20 -20

-l 40 # ,8 _ _ _ - 40

c-10 _ - 60
Co10
o60P / ,' 0 1
-80 / / | - -80
E /

-1 ~~-00 0 0 -1 00
E

- - b -1 20
0 20
-1 40 -1 40

FIG. 8. Relationship between stem diameter changes and the calculated difference between
stem and base (== predawn) water potentials during the course of a day for the same days as
Fig. 6: 18 July (left) and 16 August (right). Arrows indicate the daily course. (O --- 0),
50% MET; (/ \,100% MET.

0~~~~~~~~

-160' ' 'f - 020

;_____________EX__D OO- A

_7- -20
, I , , , I I
- ~ 50 ET A A),100% MET. - 6
'P~tem-Yf'bae (M~a)

FIG. 8. Relationship between stem diameter changes and the calculated difference between
0
stem and base (=pean water potentials a0
during the courtonpas see oft) aTa o he ponsamefo7 days a
0~~~ o _ ~~~~~~~-140
Fig.~~ ~o 6:suemn 18om2 July (lf)on 16 August rgt).Arosidctthdalcuse

- ~ o~~~~ 50%ME
~~~A a-200
- 8-106-048-02 6 0 -- 0-02 0
-10 _~~~
o - A -40 - -180
4 -02E

P -60 Pbae M a
_ -200
FI.9.Rlainsi eten tm imee hags n tecaclte ifeenebewe
stm n bsewte ptntal urngte esrtin hse(setet) hepons r80 r dy
ofmaueet(rm2 uet 6Ags)

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 205

- 200

_180

- 160 E

- 140 C,

X 120'

\
0~~~~~~~~~~~~n
100>

\ 80 E
A :3
E
A W -60 x
o50% MET co
A100% MET A 40

_ - 20

-1 8 -1 6 -1 4 -1 2 -1 0 -0 8
Minimum daily stem water potential (MPa)

FIG. 10. Relationship between MDS and the minimum stem water potential value obtained
during the day. The regression line is: MDS = -177 S - 104 (r = 0-91, P < 0-00 1).

are for 18 July and 8 August, 2 days ending drying periods on the 50% MET treatment,
and thus corresponding to very dry soil conditions.

DISCUSSION

As an indication of the amount of water stored released in the transpiration stream, it is

reasonable to expect that MDS depends on environmental factors affecting the water
balance of the plant.

Influence of atmospheric and soil water conditions on daily shrinkage

Although many workers have shown that MDS increases with increasing evaporative
demand (e.g. Impens & Schalk 1965; Holmes & Shim 1968; Waggoner & Turner 1971
for stems, and Higgs & Jones 1984 for fruits), we had to separate the two treatments to
show a relationship between MDS and potential evapotranspiration (Fig. 5). This is due
to the very strong influence of soil water content on daily stem shrinkage (Fig. 4). Table 2
shows that it is the single factor explaining best the daily stem shrinkage of our peach
trees. Such an influence of soil water content has been shown by Kozlowski & Winget
(1964), Holmes & Shim (1968), Hinckley & Bruckerhoff (1975) and Pereira & Kozlowski
(1978). These authors also show that after a first stage when drought causes an increase
in stem shrinkage (as in Fig. 4), a sustained period of drought could induce a decrease in
daily stem shrinkage. This can be explained by stomatal closure limiting water loss
(Hinckley & Bruckerhoff 1975), but also by a modification of the elasticity modulus of th
cells, as shown in the short term by Cruiziat et al. (1980) on detached sunflower leaves.
The frequency of the irrigations in our study probably prevented the trees from reaching
such a droughted stage.
Figure 10 suggests that the action of environmental factors on the stem shrinkage of the
trees is mediated by the water potential of the plant.

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
206 Stem diameter changes in peach trees

Although non-environmental factors may have played a role in the amount of daily
shrinkage (e.g. development of the leaf area, competition for water between stems and
fruits: Kozlowski 1968; Chalmers & Wilson 1978) especially in June (complete canopy
development in early July: see Chalmers, Olsson & Jones 1983; harvest on 22 June),
the importance of these factors has not been studied in our trees.

Link with the stem water potential

The hysteresis loops shown in Fig. 8 caused by the delayed response of stem diameter
changes compared to water potential are quite common (e.g. Namken, Bartholic &
Runkles 1969; Hinckley & Bruckerhoff 1975; Lassoie 1979). It is generally assumed
that this delay is caused by a large radial resistance to water flow between xylem and
phloem, since axial resistance in the xylem is supposed to be low (Klepper, Browning &
Taylor 1971). However, it seems that the vertical propagation of water potential in a
tree can be progressive (Jarvis 1975), as also suggested by the data of Dobbs & Scott
(1971) and Chalmers & Wilson (1978): these authors showed that the stem began to
shrink earlier in the day in the upmost parts of the crown, supporting the view that
water is first withdrawn near the evaporation sites of the plant (Jarvis 1975).
Furthermore, Worrall (1966) working on potted trees, found a linear relationship
(and then, no phase lag) between stem diameter and water potential, and Hinckley et al.
(1974) noticed that the phase lags were much less important in white-oak saplings than
in adult trees. Results of McBurney & Costigan (1984) showed that when stem water
potential (measured with a hygrometer attached to the stem) is measured in the immediate
proximity of the diameter changes, these two parameters are linearly related, with no
phase lag between them. All these results suggest that when the distance between the points
of measurement of water potential and stem diameter changes is low, the phase lags
are reduced or cancelled.
In this study, the distance between the leaf-where stem water potential was measured
and the location of the LVDT on the trunk was between 2 0 and 2.5 m. Therefore, the
water potential drop necessary to induce stem shrinkage (Fig. 9) may be partly interpreted
as the value necessary to create a sufficient axial water potential drop along the con-
ducting units of the xylem between the leaf and the point of diameter measurement.
As stated by Lassoie (1979), the phase lag observed would be the time necessary for the
transmission of water potential over that distance.
The water potential gradient between a covered leaf located at the periphery of the
canopy and a covered leaf near the LVDT (inside the canopy) could reach 0.3 MPa
when transpiration was high (unpublished results). Under these conditions, a water
potential gradient of around 0.15 MPa m-1 developed (a value comparable to those of
Jarvis 19-75), which shows that axial flow resistance is not negligeable. As a pathway
composed of resistances alone cannot explain the phase lags (Jarvis 1975; Weatherley
1976), it must be the case that before any detectable change in stem diameter is recorded
at the LVDT level, water is withdrawn from other compartments located above, i.e.
the leaves and the upper phloem.
Therefore, the modelling of the relation between stem diameter changes and water
potential has to take into account this progressive vertical propagation of water potential
down the xylem tissue.
This phase lag between water potential and stem diameter changes is much less
pronounced when climatic conditions vary rapidly (Fig. 7). This phenomenon has also
been noticed by Stansell et al. (1973) on cotton and by Hinckley et al. (1974) on a

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 207

white-oak sapling. The reasons for this particular behaviour remain unclear, but may be
due to the reduced capacitance of the tissues higher up the stem.
There was good correlation between MDS and the minimum daily stem water potential
(Fig. 10), indicating that the more the water stress is pronounced, the more the internal
water reservoirs of the plant are depleted during the day. Lassoie (1979) found such
relations between minimum leaf water potential and maximum daily shrinkage (with
r2 between 0.57 and 0.83). Extrapolation of the regression line presented on Fig. 10
indicates that when the minimum stem water potential of the day is greater (less negative)
than -0.55 MPa, the stem does not shrink during the day. Tromp (1984) found a similar
result on apple trees where a leaf water potential value of -0.8 MPa was necessary for
the onset of fruit shrinkage.

Application of stem diameter changes for the timing of irrigation

Stem diameter changes appear to be a sensitive physiological indicator of plant water

stress, since differences in stem shrinkage between the two treatments were detectable as
soon as early June (Fig. 2). For clear interpretation, however, this measure must be
coupled with microclimatic measurements (Fig. 5). The growth pattern of the trees
seems to be influenced much later in the season (mid-July, Fig. 2). For appropriate timing
of irrigation, the threshold values of stem shrinkage beyond which production might
be affected must still be defined.
Furthermore, much work remains to be done before a clear link between the stem
diameter changes and the water status of the plant can be established.

ACKNOWLEDGMENTS

The authors wish to thank J. Godefroy for his constant support during this work,
J. Hugard for facilities to work in the E.N.S.A.M. experimental orchard, J. Fabreguettes
for the arrangement of electrical and electronical devices, F. Jardon for the construction
of the frames, J. L. Salager for the data processing step, and E.N.S.A.M. staff for help
in the field.

REFERENCES

Begg, J. E. & Turner, N. C. (1970). Water potential gradients in field Tobacco. Plant Physiology, 46,343-346.
Brochet, P. & Gerbier, N. (1972). Une methode pratique de calcul de l'evapotranspiration potentielle.
Annales Agronomiques, 23,31-49.
Chalmers, D. J., Olsson, K. A. & Jones, T. R. (1983). Water relations of peach trees and orchards. Water
Deficits and Plant Growth, Vol. VII (Ed. by T. T. Kozlowski), pp. 197-232. Academic Press, New York
Chalmers, D. J. & Wilson, I. B. (1978). Productivity of peach trees: tree growth and water stress in relation
to fruit growth and assimilate demand. Annals of Botany, 42, 285-294.
Chaney, W. R. (1981). Sources of water. Water Deficits and Plant Growth, Vol. VI (Ed. by T. T.
Kozlowski), pp. 1-47. Academic Press, New York.
Cruiziat, P., Tyree, M. T., Bodet, C. & LoGullo, M. A. (1980). The kinetics of rehydration of detached
sunflower leaves following substantial water loss. New Phytologist, 84, 293-306.
Dobbs, R. C. & Scott, D. R. M. (1971). Distribution of diurnal fluctuations in stem circumference of
Douglas-fir. Canadian Journal of Forest Research, 1, 80-83.
Elfving, D. C. & Kaufmann, M. R. (1972). Diurnal and seasonal effects of environment on plant water
relations and fruit diameter of Citrus. Journal of the American Society for Horticultural Science,
97, 566-570.
Garnier, E. & Berger, A. (1985). Testing water potential in peach trees as an indicator of water stress.
Journal of Horticultural Science, 60, 47-56.
Garnier, E., Berger, A. & Rambal, S. (1986). Water balance and pattern of soil water uptake in a peach
orchard. Agricultural Water Management (in press).

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
208 Stem diameter changes in peach trees

Gensler, W. & Diaz-Munoz, F. (1983). Stem diameter variations in cotton under field conditions. Crop
Science, 23, 907-912.
Higgs, K. H. & Jones, H. G. (1984). A microcomputer-based system for continuous measurement and
recording fruit diameter in relation to environmental factors. Journal of Experimental Botany, 35,
1646-1655.
Hinckley, T. M. & Bruckerhoff, D. N. (1975). The effects of drought on water relations and stem shrinkage
of Quercus alba. Canadian Journal of Botany, 53, 62-72.
Hinckley, T. M., Chambers, J. L., Bruckerhoff, D. N., Roberts, J . E. & Turner, J. (1974). Effect of mid-day
shading on stem diameter, xylem pressure potential, leaf surface resistance, and net assimilation rate
in a white oak sapling. Canadian Journal of Forest Research, 4, 296-300.
Holmes, J. W. & Shim, S. Y. (1968). Diurnal changes in stem diameter of Canary Island pine trees (Pinus
canariensis, C. Smith) caused by soil water stress and varying microclimate. Journal of Experimental
Botany, 19, 219-232.
Huck, M. G. & Klepper, B. (1977). Water relations of cotton. II. Continuous estimates of plant water potential
from stem diameter measurements. Agronomy Journal, 69, 593-597.
Impens, I. I. & Schalk, J. M. (1965). A very sensitive electric dendrograph for recording radial changes of
a tree. Ecology, 46, 183-184.
Jarvis, P. G. (1975). Water transfer in plants. Heat and Mass Transfer in the Biosphere. I. Transfer
Processes in Plant Environment (Ed. by D. A. de Vries & N. H. Afgan), pp. 369-394. John Wiley &
Sons, New York.
Jordan, W. R. & Ritchie, J. T. (1971). Influence of soil water stress on evaporation, root absorption, and
internal water status of cotton. Plant Physiology, 48, 783-788.
Klepper, B., Browning, V. D. & Taylor, H. M. (1971). Stem diameter in relation to plant water status.
Plant Physiology, 48, 683-685.
Kozlowski, T. T. (1968). Diurnal changes in diameters of fruits and tree stems of Montmorency cherry.
Journal of Horticultural Science, 43, 1-15.
Kozlowski, T. T. (1972). Shrinking and swelling of plant tissues. Water Deficits and Plant Growth, Vol. III
(Ed. by T. T. Kozlowski), pp. 1-64. Academic Press, New York.
Kozlowski, T. T. & Winget, C. H. (1964). Diurnal and seasonal variation in radii of tree stems. Ecology,
45, 149-155.
Lassoie, J. P. (1979). Stem dimensional fluctuations in Douglas-fir of different crown classes. Forest Science,
25, 132-144.
McBurney, T. & Costigan, P. A. (1982). Measurement of stem water potential of young plants using a
hygrometer attached to the stem. Journal of Experimental Botany, 33, 426-431.
McBurney, T. & Costigan, P. A. (1984). The relationship between stem diameter and water potentials in
stems of young cabbage plants. Journal of Experimental Botany, 35, 1787-1793.
Meyer, J. P. (1980). Mesure des variations journalieres de tronc d'avocatier comme indicateur de stress
hydrique. Note Interne de 1'.R.F.A., R.A. 1980, Document n0 F.D. 21, 5 pp.
Molz, F. J. & Klepper, B. (1972). Radial propagation of water potential in stems. Agronomy Journal,
64,469-473.
Molz, F. J. & Klepper, B. (1973). On the mechanism of water-stress induced stem deformation. Agronomy
Journal, 65, 304-306.
Namken, L. N., Bartholic, J. F. & Runkles, J. R. (1969). Monitoring cotton plant stem radius as an indication
of water stress. Agronomy Journal, 61,891-893.
Namken, L. N., Bartholic, J. F. & Runkles, J. R. (1971). Water stress and stem radial contraction of cotton
plants (Gossypium hirsutum L.) under field conditions. Agronomy Journal, 63, 623-627.
Passioura, J. B. (1982). Water in the soil-plant-atmosphere continuum. Encyclopedia of Plant Physiology,
New Series, Vol. 12 B. Physiological Plant Ecology. II. Water relations and Carbon Assimilation (Ed. by
0. L. Lange, P. S. Nobel, C. B. Osmond & H. Ziegler). pp. 5-33. Springer-Verlag, Berlin.
Pereira, J. S. & Kozlowski, T. T. (1978). Diurnal and seasonal changes in water balance of Acer saccharum
and Betula papyrifera. Physiologia Plantarum, 43, 19-30.
Powell, D. B. B. (1974). Some effects of water stress in late spring on apple trees. Journal of Horticultural
Science, 49, 257-272.
Powell, D. B. B. & Thorpe, M. R. (1977). Dynamic aspects of plant-water relations. Environmental Effects
on Crop Physiology (Ed. by J. J. Landsberg & C. V. Cutting), pp. 259-285. Academic Press, London.
Scholander, P. F., Hammel, H. T., Bradsteet, E. D. & Hemmingsen, E. A. (1965). Sap Pressure in vascular
plants. Science, 148, 339-346.
Seguin, B. (1975). Etude comparee des m6thodes d'estimation d'ETP en climat m6diterran6en du sud de la
France (region d'Avignon). Annales Agronomiques, 26, 671-69 1.
So, H. B. (1979). An analysis of the relationship between stem diameter and leaf water potential. Agronomy
Journal, 71, 675-679.
Stansell, J. R., Klepper, B., Browning, D. V. & Taylor, H. M. (1973). Plant water status in relation to
clouds. Agronomy Journal, 65, 6 77-678.

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms
 E. GARNIER AND A. BERGER 209

Tromp, J. (1984). Diurnal fruit shrinkage in apple tree as affected by leaf water potential and vapour pressure
deficit of the air. Scientia Horticulturae, 22, 81-87.
Waggoner, P. E. & Turner, N. C. (1971). Transpiration and its control by stomata in a pine forest.
Bulletin of the Connecticut Agricultural Experiment Station, New Haven, No. 726, 87 pp.
Weatherley, P. E. (1976). Introduction: water movement through plants. Philosophical Transactions of the
Royal Society of London, B273, 43 5-444.
Worrall, J. (1966). A method for correcting dendrometer measures of tree diameter for variations induced
by moisture stress change. Forest Science, 12,427-429.
Zimmermann, M. H. & Milburn, J. A. (1982). Transport and storage of water. Encyclopedia of Plant
Physiology, New Series, Vol. 12 B. Physiological Plant Ecology. II. Water Relations and Carbon
Assimilation (Ed. by 0. L. Lange, P. S. Nobel, C. B. Osmond & H. Ziegler), pp. 135-151. Springer-
Verlag, Berlin.

(Received 2 July 1985; revision received 17 September 1985)

This content downloaded from

200.23.135.22 on Mon, 21 Nov 2022 16:02:45 UTC
All use subject to https://about.jstor.org/terms