,I + Sa11 et al, I ..a ” “‘: /- ...
,I
+ Sa11 et al, I
..a
” “‘: /-
‘ .,
,’
.<
j ”
.
.
EFFECTS OF WATER SHORTAGE ON GROWTH OF
EUCALYPTUS CAMALDULENSIS DEHN UNDER SAHELO-
SUDANIAN CLIMATE IN SENEGAL
P . N . S A L L *, G . A U S S E N A C * * , E . D R E Y E R * * a n d A .
GRANIER**
*Centre national de recherches forestitres, ISRA, Parc Forestier de Hann-Dakar,
1
*
I I
BP 2312, DAKAR, Senegal
t
**Laboratoire de Bioclimatologie et Ecophysiologie Forestières, Station de
Sylviculture et Production, INRA-Nancy, Champenoux, F-54280 SEICHAMPS, France
. v Sali et al, 2
a
Summary.
Growth and water potentials of Eucalyptus camuldulensis Dehn. have been studied
under Sahelo-sudanian conditions in Senegal. Water stress adaptation of Young seedlings
has been tested under serni controlled conditions, revealing a strong growth limitation as
soon as predawn water potentials reached values around -1.0 MPa, and a quasi-cessation
below -2.OMPa. On an in situ plantation near Bandia, the same values of predawn water
potential caused growth cessation, and reversely, growth resuming when water status
irnproves above these levels. These features correspond to a poor adaptation to drought
conditions. The use of an experimentally established relation between leaf predawn water
potential and soi1 water reserve on Young trees in a plantation, allowed a survey of
calculated predawn potentials during four successive years. In fact, values below the
above mentioned thresholds were often reached, explaining pakially the poor growth and
3%
survival observed in the field. Years 1983 and 1984 were particularly severe in this
.
regard.
\\.
Adaptability of Eucalyptus camaldulensis to sahelo-sudanian irregular water supply
conditions appears very poor; this species should be reserved to regions with more
important and regular rainfall.
d .
.
Sa11 et al, 3
.
l.Introduction.
The fast dwindling of forested areas in arid and semi-arid tropical regions
endangers more and more the supply of fuel wood to local populations. An important
effort of reforestation is needed to meet the.continuously increasing demand for fuel
wood. Use of fast growing exotic tree species has been thought to constitute a good
solution to meet this challenge. Under Sahelo-sudanian conditions with poor soils, long
dry periods and irregular rainfall, a good adequacy between the minimal water
requirements of afforested species, and the hazardous water supply is of prime
importance to allow not only tree survival, but also a significant productivity.
Eucalyptus camaldulensis is frequently used by foresters for new stand inst’allation,
because of the short revolutions it may ensure, and of its high productivity and good
coppicing ability.
In Senegal, a wide ranged project has been designed near Dakar in order to replace
natural tree stands of Acacia s&al Del., showin’g very low productions (about 1 m3 year-
1), by intensively cultivated Erlcalyptus camaldulensis ( Cissokho,l9S3). A study on
Eucalyptus growth potential under limited water supply has therefore been initiated. In
fact, in its natural area in Australia, E. camaldulensis is rather located near river banks,
with some kind of groundwater supply; therefore its adaptation to semi-arid conditions
may be questioned.
E. camaldulensis has been intensively studied in respect to rooting ability
(Riedacker, 1973, Ame et al, 1976), high temperature resistance (Kreeb 1965),
and water consumption (Poupon,l968). Productivity of different origins has also been
assessed. But Quraishi and Kramer (1980) and Moreshet (1982) alone have
published informations on drought effects on growth in this species.
In order to determine growth potential of E. camaldulensis under limited water
‘I
supply, data from seedlings grown in pots with controlled irrigation have been compared
Y
u
.
*
Sa11 et al, 4
with those obtained on trees growing at Bandia, near Dakar, under natural conditions,
*
without any additional water supply.
‘” Sa11 et al, 5
2.Material and methods.
Two experiments have been designed, one under natural conditions in a newly
planted stand, and one under semi controlled conditions in a nursery.
2.l.Plant material.
In both experiments, seedlings were obtained from seed harvested on trees in
Ko&1 Verger (Kaolack, Senegal, annual rainfall: 790 mm ); the Australian origin was
Derby (W.A., long.:123 deg. 59’ E, lat.17 deg.19’ S, alt.12 m, annual rainfal1: 610
mm).
,
2.2.Nursery experiments.
Soi1 was taken from plot 2 in Bandia (see below). Extiactable soil water reserve
was estimated as the difference in water content between permanent wilting point (y~=-
1.6 MPa) and field capacity (Ys=-0.06MPa.). The relation between soi1 water potential
and soi1 water content is sketched on figure 11
They were conducted during two consecutive years, in spring and summer 1984
and 1985 (end of dry period and beginning of rain season). Seeds were germinated in
seed beds, and seedlings transplanted to plastic pots (vol: 10 1). They were watered twice
daily; as soon as they reached 35 cm height, they were submitted to a water shortage.
Intensities of imposed stress in the different treatments are indicatcd in table 1.
Transpiration and soi1 water reserves were measured by daily weighing of individua
pots.
As soon as the desired water content was attained, pots were covered with
sulfurized paper to avoid direct evaporation from the soil, Water content was maintained
by daily partial rewatering up to desired weight. Imposed drought levels were maintained
*
during 30 days.
k
Sali et al, 6
Seedlings’ height was measured daily, and predawn leaf water potential (Ypd) was
*
estimated with a pressure chamber on 5 and 12 seedlings per treatment respectively in
1984 and 1985. Height growth and water stress intensity relationships were assessed
through mean growth and mean Ypd on each treatment.
2.3.Plantation experiments.
The plantation was installed in the Bandia forest resort, about 70 km southeast
from Dakar. This region is submitted to a sahelo-sudanian climate, where annual rainfall
exhibits very important variations; for instance, during a six year period (1981-1986),
mean annual rainfall was limited to 3958 mm, with a minimal value of 246,6 mm during
1983 ( see table 2). In fact, rainfalls are decreasing in this area, from mean levels of about
600 mm during the fifties, two the low levels of about 400 mm in the eighties.
Annual evapotranspiration has been estimated to about 2000 mm, according to
Riou (1975).
Two plots have been defined,corresponding to two different reference soils of
Bandia: the fiist one (plot 1 ) is a haplaquent, and plot 2; a typic eutrochrept (according to
american classification, S-111, 1988).
In each plot, an experimental area of 28*28 m. was defined, with 49 trees spaced
4m*4m. Four months old seedlings were planted at the end of July 1981, after a
mechanical deforestation and ploughing. 42 trees survived the whole experiment on plot 1
and 39 on plot 2.
Soi1 humidity was measured from December 1981 to December 1986 with a
Campbell 501 B neutron probe. Five 3 m. long duralumin tubes were installed on each
plot. During dry season, water content was measured once monthly, and during the rain
period (from June to October), once weekly. Humidity measurements were made on 3 m.
profiles, but water stocks were calculated over 1.75 m only, as we use them exclusively
c.
c
.
.
;.
.:
r
. :
_.
. ,
.
.
Sa11 et al, 7
to assess the relationship between mean sojl water content and tree water status.
Furthermore, direct observations revealed that no roots grew deeper (Sall, 19SS).
Meanwhile, tree heights and circumferences were monitored monthly.
__.
During 1985 and 1986, predawn leaf water potential (Ypd) was monitored in
partiel with soi1 water content, and the daily minimum potential (ym) was measured at 1
~ - - -
PM UT. Measurements were made with a pressure chamber on 5 trees on plot 1 and 11
on plot 2.
:
Nine characteristic days were chosen during rain season and dry period, and water
potential evolution was followed hourly during these days.
The critical Ybp, as defined by Aussenac and Grnnier (1978) has been
’
established graphically, by plotting together Ybp and maximal’ daily variations of Y.(X),
and assuming that this threshold value of Ybp is reached when aY is lower than 0.3
*
MPa. This leaf water status level may involve a complete stomatal closure (Aussenac
and Granier, 1978).
*
Sall et al, 8
3.ResuIts.
3.1.Effects of drought on height gromth of Young seedlings.
Figure 2 sho\\i;s results of both experiments (1984 and 1985); a very close
relationship appears between mean cumulated elongation and mean Ypd for the different
treatments, in spite of some slight differences due to microclimate from one year to the
other. Elongation was in both cases drastically limited for Ypd values below -1.9 MPa.; a
predawn water potential threshold completly impeding elongation seemed to appear
between -1.9 and -2.1 MPa.
3.2.Water potential evoIution in the fieId.
Seasonal evolutions of Ypd and Ym have been followed during 1985 and 1986. A
very close relationship appears on each plot between predawn leaf water potential and
water stock in the profile (fig. 4). This closerelationship allowed an extrapolation of
predawn water potentials over the previous growth periods of 1983 and 1984, for each
plot. Differences appearing between plots on figure 4 may be attributed to variations in
soi1 texture and soi1 water characteristics between both plots.
Seasonal fluctuations of Ypd displayed very different pattems from one year to the
other, depending on the importance and distribution of rainfall (fig 3). During 1983, Ypb
reached
-2.35 MPa during dry season, and did not fully recover during min period (
maximal value of -1.6 MPa). Evolution was similar during 1984 with a minimal value of
-2.9 MPa, but a better recovery during tht rains (-1.4). Year 1985 was far more
favorable, as dry season minimum reached -3.0 MPa, but recovery values attained 0.3
MPa, due to important rainfalls. Following dry season was marked by a limited stress (-
1.95 MPa.); and rains induced a good recovery up to -0.6 MPa.
c
I
‘.
-
Sali et al, 9
Hourly variations of Ym are shown on figure 5. During rain season (July 23,
c
August 21 and September 27 1985) Yw showed important variations with time, reaching
.-
an amplitude of about 1.7 MPa. During dry season, in opposition (June, 5 1985), Yw
varied only slightly, staying between -2.9 and -3.4 MPa.
As shown on figure 6, critical Ypd may be located at a level of about -3.0 MPa, for
which daily variation (aY> remains below -0.3 MPa.
3.3.Effects of drought on growth in the plantation.
Mean yearly productivity, followed during 6 consecutive years, remained very
low, reaching only 1.1 m3 ha-l yr-1 on plot 1, and slightly higher values of 4.6 m3 ha-l
yr-1 on plot 2.
Figure 7 shows the parallel evolution of mean height and circumference growth and
of Ypd for both plots. From these data, we have dctermined the value of Ypd which was
accompanied by a stop of elongation and circumference growth. Table 3 indicates the
values obtained during successive years. They vary only slightly, with a mean of about -
c
2.0 MPa. for elongation and -1.9 MPa. for the diametér. Inversely, after the first rains,,
growth resumes immediately after the moment where Ypd reached the same values.
From these data, way may determine duration of periods favorable to growth,
marked by different threshold values (-1.0, -1.5 and -2.0 MPa., table 3). These periods
appear very short during 84 and 85, and far from optimum.
It is of great interest to observe that the Ypd thresholds for growth observed on
nees are very narrow to those estimated on Young seedlings .
.
.
.
.‘.
c Sa11 et al, 10
:.
4.Discussion and conclusion.
E. camaldulensis appeared in our experiments, to be submitted to very drastic
.
important water stresses in the sahelo-sudanian zone, which is characterized by low-. ’
annual rain falls, grouped over a very short period (mid June-mid September). In fact, the
predawn leaf water potential (Ypd) remained at levels of about -0.3 MPa. during the rain
season, but reached values as low as -3.2 MPa. during the dry period
The presented results show that a critical predawn water potential may appear at
levels of about -3.O.MPa. Although no published data are available about a critical Ybp in
Eucalyptus camaldulensis , some results may support our iden.bi Runwald and
Karshon (1982) observed on the same species that, as soon as Ypd reached -310 MPa,
Ym never fell below 3.4 MPa. In a other work, Quraishi and Kramer (1980)
observed a complete stomatal closure at water potentials below -2.8 MPa. Our results
therefore are in accordance with these observations. Critical Ypd levels have been
estimated to values between -3.0 to -3.5 for a set of species like Cedrru arlantica, Qrrercus
ilex, Quercus prrbescens (Aussenac and Valette, 1952).
This notion of critical Ypd may be questioned, as seasonal variations in
osmoregulation may influence the relations between water potential, turgor and therefore
stomatal conductance. Myers and Neales (l?SG), for instance, demonstrated the
importance of osmotlc adjustment after preconditioning by drought, on seedlings of
various Eucalyptus species (E. behriana, E. microcarpa, E. polyanthemos). But on the
other hand, Cuyon (1987) showed a good agreement between estimated critical Ypd,’
and Yw at loss of turgor, determined during pressure volume experiments.
Our measures allowed a relatively precise determination of a Ypd threshold value,
below which no more growth occured. This threshold is located near -2.0 MPa., as well
for shoot elongation as for diameter growth. Quraishi’and Kramer (19SO) observed
growth cessation in the same species at Yw below -2.4 MPa. We may therefore assume
. .
i Sa11 et al, 11
0
that growth is at least drastically limited below -2.0 MPa.. In this respect, E.
camaldulensis appears less adapted than other Eucalyptus, like E. polyantlzemos or E.
sideroxylon (Quraishi and Kramer, 1980). In related species as E. melliodora and
E. microcarpa, Claytgn-Greene (1983) observed loss of turgor at ‘i! values of about -
2,68 MPa., and -2.99 MPa., which seem very narrow to our critical Ypd values. Very
low Ypd values are also reported for E. globulus in Portugal (-4.0 MPa.,Pereira et al,
1986), for E. incrassata in Victoria, Australia (-3.5 MPa,, Wellington, 1983), and
for E. obliqua and E. fasciculosa in Western Australia (-4.0 MPa., Sinclair, 1980).
But growth and productivity are ver-y seldom reported together with water status, and no
final conclusion may be adopted.
Comparisons with some European trees show that growm cessation occurs at Ypd
values of about -0.8 MPa on Juglans regia (Dreyer, 1984), -l.lMPa. on Fraxinus
”
excelsior and Quercus robur (Aussenac and Levy, 1983), -1.5 MPa. on Pseudotsuga
c’
menziesii (Grieu, 1986), and -2.0 MPa on Cedrus atfantica. (Aussenac and
Finkelstein, 1983)
Therefore, even if E. camaldulensis may grow till quite low Ypd, it is not as
competitive as it could have been necessary for a realistic production under sahelo-
sudanian conditions. Would other provenances have displayed better productivities?
Moreshet (1981) showed that under the exueme conditions of Neguev desert (228 mm
annuial min), a provenance from southern Australia (347 mm) displayed better survival
and growth, than one from tropical Australia (958 mm). The provenance used here seems
intermediate between both of them, as in Derby annual rainfall reaches 610 mm.
These limited growth abilities of Eucalyptus camaldulensis at Ypd below -2.0
MPa, combined with the short period during which water availability is adequate for tree
3
life, explains the very low productions measured on the plots: after 6 years of plantation,
* Sali et al, I2
the mean productivity never exceeded 4.6 m-3 ha-l yr-1. Use of the same seed on other
r-
sites with same water supply, confirmed that production never exceeded this value.
At the same time, in Senegal, studies conducted with the same species, under
. .
sudanian climate (annÜa1 rainfall between 600 and 750 mm), showed that the production
could reach 7 to 8 m3 ha-l yr-1, and even 17 to 18 m3 ha-l yr-1 under guinean climate
(1200 mm rainfall) (Sall, 19SS).
These data confirm clearly that, for a significant wood production in the sahelian
zone, it is of no interest to use E. camaldulensis, Question arise to know which species,
from the local flora, or imported, may have a significant wood production potential with
less than 500 mm rainfall. Trials have to be undertaken in coming years with different
-
.
Acacia species(A. albida, A. sene@), which present the important advantage of being of
. .
great utility in aftican agroforestry systems.
Acknowledgements.
The authors thank particularly the Centre Technique Forestier Tropical, an
Institute of the Centre International de Recherche en Agronomie Tropicale
pour le Développement (CIRAD) for financial and technical help in the realization of
this research programme, and in particular M. Bnilly, who has initiated this work in
Senegal. They thank also the Institut SMgalais de Recherches Agronomiques
(ISRA) for facilities offered to P. Sa11 during his work.
.
.
.
. Sa11 et al, 13
REFERENCES.
c
Awe J.O., K.R. Shephered K.R., and Florence R.G., 1976. Roo t
development in provenances ‘of Eucalyptus camaldulensis Dehn. Austr. For., 39,201-
209.
Aussenac G and Granier A., 1978. Quelques resultats de cinétiques journalières du
potentiel de sève chez les arbres forestiers. Ann. Sci. for.,35,19-32.
Aussenac G. and Valette J. C.,1982. Comportements hydriques estivaux de
Cedrus atlantica, Quercus ilex et Quercus pubescens, et de divers pins dans le Mont
Ventoux. Ann. Sci. for.,39, 41-42.
Aussenac G. and Finkelstein D., 1983. Influence de la secheresse sur la
croissance et la photosynthèse du cèdre. Ann. Sci. for., 40,67-77.
Aussenac G. and Lévy G., 1953. Influence du déss&chement du sol sur le
comportement hydrique et la croissance du chêne pédonculé (Quercus pedunculata Ehrl.)
et du frêne (Fraxinus excelsior L.) cultivés en vases de vegétation. Ann. Sci. for.,40,251-
264.
Cissokho C.,1983. La place de 1’Eucalyptus dans la ‘politique forestière du Sénégal.
Ministère de la Protection de la Nature, Dakar.
Clayton-Greene K.A.,1983. The tissue water relationship of Callitris columellaris,
Eucalyptus melliodora and Eucalyptus microcarpa investigated using the pressure volume
technique. Oecologia,57,368-373.
Dreyer E., 1984. Comportement d’une plante perenne soumise Zt des contraintes
hydriques: réponses physiologiques de jeunes noyers soumis & des pt’riodes de
secheresse. Thése UniversitC de Clermont Ferrand, 160 pp.
Grieu P., 1956. Ecophysiologie du Douglas: contribution a 1’Ctude de l’influence des
d&ïcits hydriques sur les &Changes gazeux, la croissance et l’accumulation des metabolites
organiques. Thèse Université de Nancy, 169 pp.
Guyon J.P;,1987. Analyse des courbes pression volume de rameaux de trois espkes
forestières. Oecol. Applic.,8,363-370.
:
.
t.
..,‘.
._.
-
Sa11 eral, 14
Kreeb K., 1965. Untersuchungen über die Hitze und Trockenresistenz an
c
Australischen Immergriinen im Keimlingsstadium. Ber. dt. Bot. Ges.,78,90-98.
Moreshet S. 1981. Physiological activity in a serni arid environment of Eucalyptus
camaldulensis Dehn..from two provenances. Aust. J. Bot., 29,97-l 10. -
Myers B.A. and Neales T.F., 1986. Osmotic adjustment, induced by drought, in
seedlings of three Eucalyptus species. Aust. J. Plant Physiol.,13,597-603.
Pereira J.S,,Tenhunen J.D., Lange O.L. Beyschlag W., Meyer A. and
David M.M., 19% Seasonal and diumal patterns in leaf gas exchange of Eucalyptus
glubulus trees growing in Portugal. Can. J. Forest Res., 16, 177-184.
Poupon H., 1968. Premières études des caractéristiques hydriques des feuilles
d’Eucalyptus camaldulensis en Tunisie. Ann. INA Tunisie, 1, l-22.
Quraishi M.A., and Kramer J.T., 1970. Water stress in three species of .
.
Eucalyptus. For. Sci., 16,74-78.
Riedacker A., 1972. Physiologie des souches d’Eucalyptus camaldulensis traitées en
taillis au Maroc. Le problème de leur vieillissement. C.R. VIIIeme Congrès Forestier
Mondial, Buenos Aires.
Riou C., 1975. La determination pratique de l’evaporation. ORSTOM, 236~.
Runmald C. and Karschon A., 1982. Leaf xylem water potentials and water
saturation defïcits as related to seed origin of Eucalypnu camaldulensis. Aust. For. Res.,
12,175181.
Sa11 P.N., 198s. Etude écophysiologique de Eucalyptus camaldulensis Dehn. en zone
sahelo-soudanienne. Thèse Université de Nancy, 202 pp.
Scholander P.F., Hammel H.T., Bradstreet ED. and Hemmingsen E.A.,
1965. Sap pressure in vascular plants. Science, 148,339-346.
Sinclair R., 1980. Water potential and stomatal conductance of three Eucalyptus
species in the Mount Lofty Ranges, South Australia: responses to summer drought. Aust.
J. Bot., 18, 261-273.
Wellington A.B., 1954. Leaf water potentials, fire, and the regeneration of mallee
c
Eucalyptus in semi-arid, South-eastem Australia. Oecologia, 64, 360-362.
Sa11 et al, 15
a
TABLES.
Table 1. Characterization of drought treatments applied to Eucalyptus
camaldulensis seedlings cultivated in pots during 1984 and 1985.
YSOil
Nlean Ypd
(MPa.)
(MPa.)
-0.06
-0.7kO.17
-0.16
-0.9f0.13
-0.35
-1.lf0.18
-0.89
-1.87rt0.22
1985
10
-0.35
-0.92+0.05
10 .
;;
-0.45
-1.10+0.05
-0.56
-1.28+0 03
ii
:o
-0.79
-1.60+0’05
10
5
-0.89
-1.91+0:11
REW: relative extractable water content, expressed as fraction of total extractable
water (840 g per pot)
Ysoil: calculated mean soi1 water potential in the pots; these values are determined
according to potential-water content relation plotted on figure 1.
Ypd: mean predawn leaf water potential,measured with a pressure chamber during
the treatments, indicated values display confidence limits at 5%.
.
TabIe 2. Annual rainfall evolution at Bandia, during the experiment (mm) .
years
1981
1982
1983
1984
1985
lYS6
Mean
rainfall
403.0
445.5
246.6
327.9
474.2
478.9
395.8
:
.
.-
(
: 8
.
SaO et al, 16
Table 3. Predawn leaf water potential (Ypd, MPa.) levels for which growth
cessation and resuming occurred, during successive years, at Bandia.
Year
Height growtl
Diametix- growth
cessation
resuming
cessation
_ .
resuming
1983
plot 1
-1.7
-1.7
plot 2
-2.35
-2.15
-2.15
-2.15
1984
plot 1
-1.8
-1.75
-1.7
-1.75
plot 2
-2.35
-2.35
-2.0
-2.35
1985
plot 1
: -1.85
-1.85
-1.85
-1.85
plot 2
-1.85
-1.85
1986
plot 1
plot 2
-1.85
-1.85
.
mean
-2.01fO.3
-1.97kO.23 -1.1)7&0.17
-2.0-10.28
Table 4. Duration of periods (months) marked by predawn Ieaf water potentials
(Ypd) above threshold values of -1.0, -1.5 and -2.0 MPa. in the plantation at Bandia.
<,
I
Plot 1
Plot 2
I
Ypd
-1.0
-1.5
-2.0
-1.0
-1.5
-2.0
years
*-
1983
1984
0
01
ii
5 5
1985
1986
2.75
:
55
3::
5
12’
.
. .
Sa11 et al, 17
.z.,
n
FIGURES.
Figure 1. Relationship between soi1 water potential (MPa), and water content for
the soi1 used to grow potted Eucalyptus seedlings. Drawn lines represent the limits used
to define extractable soil water content.
Figure 2. Relationship between mean predawn leaf water potential (Ypd) and
mean elongation for potted Eucalyptus seedlings submitted to drought treatments as
described in table 1. Mean values are estimated for each treatment; bars represent
confidence intervals at 5%. Note that results of year 1985 (open symbols) and 1984
(closed symbols) display only a slight drift.
.
Figure 3. Relationship between measured predawn leaf water potential (Ypd) of
Eucalyptus grown in Bandia, and water content of soils, expiessed as mm in a 1.85 m
profile. Open symbols: plot 1, closed symbols: plot 2.; curve fitted by eye. On each plot,
the relation appears very well defined ail over measurement years 1985 and 1986. Plots
differ because of there different soi1 qualities: These relations were used to extrapolate
Ypd over two years previous to measurements.
’
Figure 4. Evolution over 4 years of monthly precipitations (bars), and measured
or estimated predawn leaf water potential (Upd) of Eucalyptus camaldulensis grown in
Bandia; 4a, plot 1; 4b plot 2. Notice the length of peri0d.s with very loti Ypd.
Figure 5. Daily variations of leaf water potential (Y) on Eucalyptus camaldulensis
grown in Bandia, during different days along rain and dry season.1985. Selected trees
were located on plot2.
Figure 6. Relation between leaf predawn water potential (Ypd) of Eucalyptus
camaldulensis grown in Bandia, and daily amplitude of variations (ay>. A critical Ypd
may be defined as the value below which JY remains below approximately 0.3 MPa.
:._
.
I.
. . .<
::.
,’
‘.
I
‘-_
.,.._
1’
Figure 7. Evolution of mean height (h), mean circumference (c) of Eucalyptus
.
camaZduZensis grown in Bandia on plot 1, during 1985 and 1986, and their mean predawn
leaf water potential (Ypd).Growth cessation and recovery occured at Ypd values of about
-1.9 MPa -
._<
_
.
- --
.
-
.._
. .
MPa
1.6
0.6
0 .1
0.06
0.032
0
5
10
15
20
Soi1 water content f%)
I
0 .
1.
I
1
I
0
I 0 . u1
I d . 0
I d . m
I N . 0
I rd3 . ul
I w . 0
a
.
l
I
1
1
t3
I
I
I
,
I
E; 0) cn
l
c-
v)
._
--.-
‘-..’
l,-:‘_ -. .
7
*
_.
.
_
,
_
-.
..-
,...
-
.,
---.
.
<’
.
.
_. Y.
c.
6 )::8 '10 12 li 16 .18 t
6 8 10 12 14 16 18 :/
a
1
1
1
1
I
1
r5.0 6.1985
. ..
- - - -. - -
.
.
-
.
_.
:.:
‘_
‘_ . -
m
.
0
-0.5 , -85
27.093985
- 1.0
I<
0
-0.5
17.12.1985
18.02.1985
- 1.0
-2.0
,
-3.0
.
.”
w
(MPa)
*
.
.
. .._.. -
:
:
1:: I. -7’ *-
.-.
-.
*
‘-
.:
:-
:
,. :
:
.
_
:-
‘.iom2
:.: -0.4 .‘- - 0 . 6
-4.8
-1.0
- 1 . 2
0
l
-0.5
/
l
l
__
4.0
- 1 . 5
r
.«
i.0
- 2 . 5
- 3 . 0
v PD
‘,
:JMPa)
‘:
-5c. -
t*c..
.
_ ” . _
..!’
.
.
.
.
. . .
.
.,
/
.
d
.
‘*I
+ Sa11 et al, I
..a
” “‘: /-
‘ .,
,’
.<
j ”
.
.
EFFECTS OF WATER SHORTAGE ON GROWTH OF
EUCALYPTUS CAMALDULENSIS DEHN UNDER SAHELO-
SUDANIAN CLIMATE IN SENEGAL
P . N . S A L L *, G . A U S S E N A C * * , E . D R E Y E R * * a n d A .
GRANIER**
*Centre national de recherches forestitres, ISRA, Parc Forestier de Hann-Dakar,
1
*
I I
BP 2312, DAKAR, Senegal
t
**Laboratoire de Bioclimatologie et Ecophysiologie Forestières, Station de
Sylviculture et Production, INRA-Nancy, Champenoux, F-54280 SEICHAMPS, France
. v Sali et al, 2
a
Summary.
Growth and water potentials of Eucalyptus camuldulensis Dehn. have been studied
under Sahelo-sudanian conditions in Senegal. Water stress adaptation of Young seedlings
has been tested under serni controlled conditions, revealing a strong growth limitation as
soon as predawn water potentials reached values around -1.0 MPa, and a quasi-cessation
below -2.OMPa. On an in situ plantation near Bandia, the same values of predawn water
potential caused growth cessation, and reversely, growth resuming when water status
irnproves above these levels. These features correspond to a poor adaptation to drought
conditions. The use of an experimentally established relation between leaf predawn water
potential and soi1 water reserve on Young trees in a plantation, allowed a survey of
calculated predawn potentials during four successive years. In fact, values below the
above mentioned thresholds were often reached, explaining pakially the poor growth and
3%
survival observed in the field. Years 1983 and 1984 were particularly severe in this
.
regard.
\\.
Adaptability of Eucalyptus camaldulensis to sahelo-sudanian irregular water supply
conditions appears very poor; this species should be reserved to regions with more
important and regular rainfall.
d .
.
Sa11 et al, 3
.
l.Introduction.
The fast dwindling of forested areas in arid and semi-arid tropical regions
endangers more and more the supply of fuel wood to local populations. An important
effort of reforestation is needed to meet the.continuously increasing demand for fuel
wood. Use of fast growing exotic tree species has been thought to constitute a good
solution to meet this challenge. Under Sahelo-sudanian conditions with poor soils, long
dry periods and irregular rainfall, a good adequacy between the minimal water
requirements of afforested species, and the hazardous water supply is of prime
importance to allow not only tree survival, but also a significant productivity.
Eucalyptus camaldulensis is frequently used by foresters for new stand inst’allation,
because of the short revolutions it may ensure, and of its high productivity and good
coppicing ability.
In Senegal, a wide ranged project has been designed near Dakar in order to replace
natural tree stands of Acacia s&al Del., showin’g very low productions (about 1 m3 year-
1), by intensively cultivated Erlcalyptus camaldulensis ( Cissokho,l9S3). A study on
Eucalyptus growth potential under limited water supply has therefore been initiated. In
fact, in its natural area in Australia, E. camaldulensis is rather located near river banks,
with some kind of groundwater supply; therefore its adaptation to semi-arid conditions
may be questioned.
E. camaldulensis has been intensively studied in respect to rooting ability
(Riedacker, 1973, Ame et al, 1976), high temperature resistance (Kreeb 1965),
and water consumption (Poupon,l968). Productivity of different origins has also been
assessed. But Quraishi and Kramer (1980) and Moreshet (1982) alone have
published informations on drought effects on growth in this species.
In order to determine growth potential of E. camaldulensis under limited water
‘I
supply, data from seedlings grown in pots with controlled irrigation have been compared
Y
u
.
*
Sa11 et al, 4
with those obtained on trees growing at Bandia, near Dakar, under natural conditions,
*
without any additional water supply.
‘” Sa11 et al, 5
2.Material and methods.
Two experiments have been designed, one under natural conditions in a newly
planted stand, and one under semi controlled conditions in a nursery.
2.l.Plant material.
In both experiments, seedlings were obtained from seed harvested on trees in
Ko&1 Verger (Kaolack, Senegal, annual rainfall: 790 mm ); the Australian origin was
Derby (W.A., long.:123 deg. 59’ E, lat.17 deg.19’ S, alt.12 m, annual rainfal1: 610
mm).
,
2.2.Nursery experiments.
Soi1 was taken from plot 2 in Bandia (see below). Extiactable soil water reserve
was estimated as the difference in water content between permanent wilting point (y~=-
1.6 MPa) and field capacity (Ys=-0.06MPa.). The relation between soi1 water potential
and soi1 water content is sketched on figure 11
They were conducted during two consecutive years, in spring and summer 1984
and 1985 (end of dry period and beginning of rain season). Seeds were germinated in
seed beds, and seedlings transplanted to plastic pots (vol: 10 1). They were watered twice
daily; as soon as they reached 35 cm height, they were submitted to a water shortage.
Intensities of imposed stress in the different treatments are indicatcd in table 1.
Transpiration and soi1 water reserves were measured by daily weighing of individua
pots.
As soon as the desired water content was attained, pots were covered with
sulfurized paper to avoid direct evaporation from the soil, Water content was maintained
by daily partial rewatering up to desired weight. Imposed drought levels were maintained
*
during 30 days.
k
Sali et al, 6
Seedlings’ height was measured daily, and predawn leaf water potential (Ypd) was
*
estimated with a pressure chamber on 5 and 12 seedlings per treatment respectively in
1984 and 1985. Height growth and water stress intensity relationships were assessed
through mean growth and mean Ypd on each treatment.
2.3.Plantation experiments.
The plantation was installed in the Bandia forest resort, about 70 km southeast
from Dakar. This region is submitted to a sahelo-sudanian climate, where annual rainfall
exhibits very important variations; for instance, during a six year period (1981-1986),
mean annual rainfall was limited to 3958 mm, with a minimal value of 246,6 mm during
1983 ( see table 2). In fact, rainfalls are decreasing in this area, from mean levels of about
600 mm during the fifties, two the low levels of about 400 mm in the eighties.
Annual evapotranspiration has been estimated to about 2000 mm, according to
Riou (1975).
Two plots have been defined,corresponding to two different reference soils of
Bandia: the fiist one (plot 1 ) is a haplaquent, and plot 2; a typic eutrochrept (according to
american classification, S-111, 1988).
In each plot, an experimental area of 28*28 m. was defined, with 49 trees spaced
4m*4m. Four months old seedlings were planted at the end of July 1981, after a
mechanical deforestation and ploughing. 42 trees survived the whole experiment on plot 1
and 39 on plot 2.
Soi1 humidity was measured from December 1981 to December 1986 with a
Campbell 501 B neutron probe. Five 3 m. long duralumin tubes were installed on each
plot. During dry season, water content was measured once monthly, and during the rain
period (from June to October), once weekly. Humidity measurements were made on 3 m.
profiles, but water stocks were calculated over 1.75 m only, as we use them exclusively
c.
c
.
.
;.
.:
r
. :
_.
. ,
.
.
Sa11 et al, 7
to assess the relationship between mean sojl water content and tree water status.
Furthermore, direct observations revealed that no roots grew deeper (Sall, 19SS).
Meanwhile, tree heights and circumferences were monitored monthly.
__.
During 1985 and 1986, predawn leaf water potential (Ypd) was monitored in
partiel with soi1 water content, and the daily minimum potential (ym) was measured at 1
~ - - -
PM UT. Measurements were made with a pressure chamber on 5 trees on plot 1 and 11
on plot 2.
:
Nine characteristic days were chosen during rain season and dry period, and water
potential evolution was followed hourly during these days.
The critical Ybp, as defined by Aussenac and Grnnier (1978) has been
’
established graphically, by plotting together Ybp and maximal’ daily variations of Y.(X),
and assuming that this threshold value of Ybp is reached when aY is lower than 0.3
*
MPa. This leaf water status level may involve a complete stomatal closure (Aussenac
and Granier, 1978).
*
Sall et al, 8
3.ResuIts.
3.1.Effects of drought on height gromth of Young seedlings.
Figure 2 sho\\i;s results of both experiments (1984 and 1985); a very close
relationship appears between mean cumulated elongation and mean Ypd for the different
treatments, in spite of some slight differences due to microclimate from one year to the
other. Elongation was in both cases drastically limited for Ypd values below -1.9 MPa.; a
predawn water potential threshold completly impeding elongation seemed to appear
between -1.9 and -2.1 MPa.
3.2.Water potential evoIution in the fieId.
Seasonal evolutions of Ypd and Ym have been followed during 1985 and 1986. A
very close relationship appears on each plot between predawn leaf water potential and
water stock in the profile (fig. 4). This closerelationship allowed an extrapolation of
predawn water potentials over the previous growth periods of 1983 and 1984, for each
plot. Differences appearing between plots on figure 4 may be attributed to variations in
soi1 texture and soi1 water characteristics between both plots.
Seasonal fluctuations of Ypd displayed very different pattems from one year to the
other, depending on the importance and distribution of rainfall (fig 3). During 1983, Ypb
reached
-2.35 MPa during dry season, and did not fully recover during min period (
maximal value of -1.6 MPa). Evolution was similar during 1984 with a minimal value of
-2.9 MPa, but a better recovery during tht rains (-1.4). Year 1985 was far more
favorable, as dry season minimum reached -3.0 MPa, but recovery values attained 0.3
MPa, due to important rainfalls. Following dry season was marked by a limited stress (-
1.95 MPa.); and rains induced a good recovery up to -0.6 MPa.
c
I
‘.
-
Sali et al, 9
Hourly variations of Ym are shown on figure 5. During rain season (July 23,
c
August 21 and September 27 1985) Yw showed important variations with time, reaching
.-
an amplitude of about 1.7 MPa. During dry season, in opposition (June, 5 1985), Yw
varied only slightly, staying between -2.9 and -3.4 MPa.
As shown on figure 6, critical Ypd may be located at a level of about -3.0 MPa, for
which daily variation (aY> remains below -0.3 MPa.
3.3.Effects of drought on growth in the plantation.
Mean yearly productivity, followed during 6 consecutive years, remained very
low, reaching only 1.1 m3 ha-l yr-1 on plot 1, and slightly higher values of 4.6 m3 ha-l
yr-1 on plot 2.
Figure 7 shows the parallel evolution of mean height and circumference growth and
of Ypd for both plots. From these data, we have dctermined the value of Ypd which was
accompanied by a stop of elongation and circumference growth. Table 3 indicates the
values obtained during successive years. They vary only slightly, with a mean of about -
c
2.0 MPa. for elongation and -1.9 MPa. for the diametér. Inversely, after the first rains,,
growth resumes immediately after the moment where Ypd reached the same values.
From these data, way may determine duration of periods favorable to growth,
marked by different threshold values (-1.0, -1.5 and -2.0 MPa., table 3). These periods
appear very short during 84 and 85, and far from optimum.
It is of great interest to observe that the Ypd thresholds for growth observed on
nees are very narrow to those estimated on Young seedlings .
.
.
.
.‘.
c Sa11 et al, 10
:.
4.Discussion and conclusion.
E. camaldulensis appeared in our experiments, to be submitted to very drastic
.
important water stresses in the sahelo-sudanian zone, which is characterized by low-. ’
annual rain falls, grouped over a very short period (mid June-mid September). In fact, the
predawn leaf water potential (Ypd) remained at levels of about -0.3 MPa. during the rain
season, but reached values as low as -3.2 MPa. during the dry period
The presented results show that a critical predawn water potential may appear at
levels of about -3.O.MPa. Although no published data are available about a critical Ybp in
Eucalyptus camaldulensis , some results may support our iden.bi Runwald and
Karshon (1982) observed on the same species that, as soon as Ypd reached -310 MPa,
Ym never fell below 3.4 MPa. In a other work, Quraishi and Kramer (1980)
observed a complete stomatal closure at water potentials below -2.8 MPa. Our results
therefore are in accordance with these observations. Critical Ypd levels have been
estimated to values between -3.0 to -3.5 for a set of species like Cedrru arlantica, Qrrercus
ilex, Quercus prrbescens (Aussenac and Valette, 1952).
This notion of critical Ypd may be questioned, as seasonal variations in
osmoregulation may influence the relations between water potential, turgor and therefore
stomatal conductance. Myers and Neales (l?SG), for instance, demonstrated the
importance of osmotlc adjustment after preconditioning by drought, on seedlings of
various Eucalyptus species (E. behriana, E. microcarpa, E. polyanthemos). But on the
other hand, Cuyon (1987) showed a good agreement between estimated critical Ypd,’
and Yw at loss of turgor, determined during pressure volume experiments.
Our measures allowed a relatively precise determination of a Ypd threshold value,
below which no more growth occured. This threshold is located near -2.0 MPa., as well
for shoot elongation as for diameter growth. Quraishi’and Kramer (19SO) observed
growth cessation in the same species at Yw below -2.4 MPa. We may therefore assume
. .
i Sa11 et al, 11
0
that growth is at least drastically limited below -2.0 MPa.. In this respect, E.
camaldulensis appears less adapted than other Eucalyptus, like E. polyantlzemos or E.
sideroxylon (Quraishi and Kramer, 1980). In related species as E. melliodora and
E. microcarpa, Claytgn-Greene (1983) observed loss of turgor at ‘i! values of about -
2,68 MPa., and -2.99 MPa., which seem very narrow to our critical Ypd values. Very
low Ypd values are also reported for E. globulus in Portugal (-4.0 MPa.,Pereira et al,
1986), for E. incrassata in Victoria, Australia (-3.5 MPa,, Wellington, 1983), and
for E. obliqua and E. fasciculosa in Western Australia (-4.0 MPa., Sinclair, 1980).
But growth and productivity are ver-y seldom reported together with water status, and no
final conclusion may be adopted.
Comparisons with some European trees show that growm cessation occurs at Ypd
values of about -0.8 MPa on Juglans regia (Dreyer, 1984), -l.lMPa. on Fraxinus
”
excelsior and Quercus robur (Aussenac and Levy, 1983), -1.5 MPa. on Pseudotsuga
c’
menziesii (Grieu, 1986), and -2.0 MPa on Cedrus atfantica. (Aussenac and
Finkelstein, 1983)
Therefore, even if E. camaldulensis may grow till quite low Ypd, it is not as
competitive as it could have been necessary for a realistic production under sahelo-
sudanian conditions. Would other provenances have displayed better productivities?
Moreshet (1981) showed that under the exueme conditions of Neguev desert (228 mm
annuial min), a provenance from southern Australia (347 mm) displayed better survival
and growth, than one from tropical Australia (958 mm). The provenance used here seems
intermediate between both of them, as in Derby annual rainfall reaches 610 mm.
These limited growth abilities of Eucalyptus camaldulensis at Ypd below -2.0
MPa, combined with the short period during which water availability is adequate for tree
3
life, explains the very low productions measured on the plots: after 6 years of plantation,
* Sali et al, I2
the mean productivity never exceeded 4.6 m-3 ha-l yr-1. Use of the same seed on other
r-
sites with same water supply, confirmed that production never exceeded this value.
At the same time, in Senegal, studies conducted with the same species, under
. .
sudanian climate (annÜa1 rainfall between 600 and 750 mm), showed that the production
could reach 7 to 8 m3 ha-l yr-1, and even 17 to 18 m3 ha-l yr-1 under guinean climate
(1200 mm rainfall) (Sall, 19SS).
These data confirm clearly that, for a significant wood production in the sahelian
zone, it is of no interest to use E. camaldulensis, Question arise to know which species,
from the local flora, or imported, may have a significant wood production potential with
less than 500 mm rainfall. Trials have to be undertaken in coming years with different
-
.
Acacia species(A. albida, A. sene@), which present the important advantage of being of
. .
great utility in aftican agroforestry systems.
Acknowledgements.
The authors thank particularly the Centre Technique Forestier Tropical, an
Institute of the Centre International de Recherche en Agronomie Tropicale
pour le Développement (CIRAD) for financial and technical help in the realization of
this research programme, and in particular M. Bnilly, who has initiated this work in
Senegal. They thank also the Institut SMgalais de Recherches Agronomiques
(ISRA) for facilities offered to P. Sa11 during his work.
.
.
.
. Sa11 et al, 13
REFERENCES.
c
Awe J.O., K.R. Shephered K.R., and Florence R.G., 1976. Roo t
development in provenances ‘of Eucalyptus camaldulensis Dehn. Austr. For., 39,201-
209.
Aussenac G and Granier A., 1978. Quelques resultats de cinétiques journalières du
potentiel de sève chez les arbres forestiers. Ann. Sci. for.,35,19-32.
Aussenac G. and Valette J. C.,1982. Comportements hydriques estivaux de
Cedrus atlantica, Quercus ilex et Quercus pubescens, et de divers pins dans le Mont
Ventoux. Ann. Sci. for.,39, 41-42.
Aussenac G. and Finkelstein D., 1983. Influence de la secheresse sur la
croissance et la photosynthèse du cèdre. Ann. Sci. for., 40,67-77.
Aussenac G. and Lévy G., 1953. Influence du déss&chement du sol sur le
comportement hydrique et la croissance du chêne pédonculé (Quercus pedunculata Ehrl.)
et du frêne (Fraxinus excelsior L.) cultivés en vases de vegétation. Ann. Sci. for.,40,251-
264.
Cissokho C.,1983. La place de 1’Eucalyptus dans la ‘politique forestière du Sénégal.
Ministère de la Protection de la Nature, Dakar.
Clayton-Greene K.A.,1983. The tissue water relationship of Callitris columellaris,
Eucalyptus melliodora and Eucalyptus microcarpa investigated using the pressure volume
technique. Oecologia,57,368-373.
Dreyer E., 1984. Comportement d’une plante perenne soumise Zt des contraintes
hydriques: réponses physiologiques de jeunes noyers soumis & des pt’riodes de
secheresse. Thése UniversitC de Clermont Ferrand, 160 pp.
Grieu P., 1956. Ecophysiologie du Douglas: contribution a 1’Ctude de l’influence des
d&ïcits hydriques sur les &Changes gazeux, la croissance et l’accumulation des metabolites
organiques. Thèse Université de Nancy, 169 pp.
Guyon J.P;,1987. Analyse des courbes pression volume de rameaux de trois espkes
forestières. Oecol. Applic.,8,363-370.
:
.
t.
..,‘.
._.
-
Sa11 eral, 14
Kreeb K., 1965. Untersuchungen über die Hitze und Trockenresistenz an
c
Australischen Immergriinen im Keimlingsstadium. Ber. dt. Bot. Ges.,78,90-98.
Moreshet S. 1981. Physiological activity in a serni arid environment of Eucalyptus
camaldulensis Dehn..from two provenances. Aust. J. Bot., 29,97-l 10. -
Myers B.A. and Neales T.F., 1986. Osmotic adjustment, induced by drought, in
seedlings of three Eucalyptus species. Aust. J. Plant Physiol.,13,597-603.
Pereira J.S,,Tenhunen J.D., Lange O.L. Beyschlag W., Meyer A. and
David M.M., 19% Seasonal and diumal patterns in leaf gas exchange of Eucalyptus
glubulus trees growing in Portugal. Can. J. Forest Res., 16, 177-184.
Poupon H., 1968. Premières études des caractéristiques hydriques des feuilles
d’Eucalyptus camaldulensis en Tunisie. Ann. INA Tunisie, 1, l-22.
Quraishi M.A., and Kramer J.T., 1970. Water stress in three species of .
.
Eucalyptus. For. Sci., 16,74-78.
Riedacker A., 1972. Physiologie des souches d’Eucalyptus camaldulensis traitées en
taillis au Maroc. Le problème de leur vieillissement. C.R. VIIIeme Congrès Forestier
Mondial, Buenos Aires.
Riou C., 1975. La determination pratique de l’evaporation. ORSTOM, 236~.
Runmald C. and Karschon A., 1982. Leaf xylem water potentials and water
saturation defïcits as related to seed origin of Eucalypnu camaldulensis. Aust. For. Res.,
12,175181.
Sa11 P.N., 198s. Etude écophysiologique de Eucalyptus camaldulensis Dehn. en zone
sahelo-soudanienne. Thèse Université de Nancy, 202 pp.
Scholander P.F., Hammel H.T., Bradstreet ED. and Hemmingsen E.A.,
1965. Sap pressure in vascular plants. Science, 148,339-346.
Sinclair R., 1980. Water potential and stomatal conductance of three Eucalyptus
species in the Mount Lofty Ranges, South Australia: responses to summer drought. Aust.
J. Bot., 18, 261-273.
Wellington A.B., 1954. Leaf water potentials, fire, and the regeneration of mallee
c
Eucalyptus in semi-arid, South-eastem Australia. Oecologia, 64, 360-362.
Sa11 et al, 15
a
TABLES.
Table 1. Characterization of drought treatments applied to Eucalyptus
camaldulensis seedlings cultivated in pots during 1984 and 1985.
YSOil
Nlean Ypd
(MPa.)
(MPa.)
-0.06
-0.7kO.17
-0.16
-0.9f0.13
-0.35
-1.lf0.18
-0.89
-1.87rt0.22
1985
10
-0.35
-0.92+0.05
10 .
;;
-0.45
-1.10+0.05
-0.56
-1.28+0 03
ii
:o
-0.79
-1.60+0’05
10
5
-0.89
-1.91+0:11
REW: relative extractable water content, expressed as fraction of total extractable
water (840 g per pot)
Ysoil: calculated mean soi1 water potential in the pots; these values are determined
according to potential-water content relation plotted on figure 1.
Ypd: mean predawn leaf water potential,measured with a pressure chamber during
the treatments, indicated values display confidence limits at 5%.
.
TabIe 2. Annual rainfall evolution at Bandia, during the experiment (mm) .
years
1981
1982
1983
1984
1985
lYS6
Mean
rainfall
403.0
445.5
246.6
327.9
474.2
478.9
395.8
:
.
.-
(
: 8
.
SaO et al, 16
Table 3. Predawn leaf water potential (Ypd, MPa.) levels for which growth
cessation and resuming occurred, during successive years, at Bandia.
Year
Height growtl
Diametix- growth
cessation
resuming
cessation
_ .
resuming
1983
plot 1
-1.7
-1.7
plot 2
-2.35
-2.15
-2.15
-2.15
1984
plot 1
-1.8
-1.75
-1.7
-1.75
plot 2
-2.35
-2.35
-2.0
-2.35
1985
plot 1
: -1.85
-1.85
-1.85
-1.85
plot 2
-1.85
-1.85
1986
plot 1
plot 2
-1.85
-1.85
.
mean
-2.01fO.3
-1.97kO.23 -1.1)7&0.17
-2.0-10.28
Table 4. Duration of periods (months) marked by predawn Ieaf water potentials
(Ypd) above threshold values of -1.0, -1.5 and -2.0 MPa. in the plantation at Bandia.
<,
I
Plot 1
Plot 2
I
Ypd
-1.0
-1.5
-2.0
-1.0
-1.5
-2.0
years
*-
1983
1984
0
01
ii
5 5
1985
1986
2.75
:
55
3::
5
12’
.
. .
Sa11 et al, 17
.z.,
n
FIGURES.
Figure 1. Relationship between soi1 water potential (MPa), and water content for
the soi1 used to grow potted Eucalyptus seedlings. Drawn lines represent the limits used
to define extractable soil water content.
Figure 2. Relationship between mean predawn leaf water potential (Ypd) and
mean elongation for potted Eucalyptus seedlings submitted to drought treatments as
described in table 1. Mean values are estimated for each treatment; bars represent
confidence intervals at 5%. Note that results of year 1985 (open symbols) and 1984
(closed symbols) display only a slight drift.
.
Figure 3. Relationship between measured predawn leaf water potential (Ypd) of
Eucalyptus grown in Bandia, and water content of soils, expiessed as mm in a 1.85 m
profile. Open symbols: plot 1, closed symbols: plot 2.; curve fitted by eye. On each plot,
the relation appears very well defined ail over measurement years 1985 and 1986. Plots
differ because of there different soi1 qualities: These relations were used to extrapolate
Ypd over two years previous to measurements.
’
Figure 4. Evolution over 4 years of monthly precipitations (bars), and measured
or estimated predawn leaf water potential (Upd) of Eucalyptus camaldulensis grown in
Bandia; 4a, plot 1; 4b plot 2. Notice the length of peri0d.s with very loti Ypd.
Figure 5. Daily variations of leaf water potential (Y) on Eucalyptus camaldulensis
grown in Bandia, during different days along rain and dry season.1985. Selected trees
were located on plot2.
Figure 6. Relation between leaf predawn water potential (Ypd) of Eucalyptus
camaldulensis grown in Bandia, and daily amplitude of variations (ay>. A critical Ypd
may be defined as the value below which JY remains below approximately 0.3 MPa.
:._
.
I.
. . .<
::.
,’
‘.
I
‘-_
.,.._
1’
Figure 7. Evolution of mean height (h), mean circumference (c) of Eucalyptus
.
camaZduZensis grown in Bandia on plot 1, during 1985 and 1986, and their mean predawn
leaf water potential (Ypd).Growth cessation and recovery occured at Ypd values of about
-1.9 MPa -
._<
_
.
- --
.
-
.._
. .
MPa
1.6
0.6
0 .1
0.06
0.032
0
5
10
15
20
Soi1 water content f%)
I
0 .
1.
I
1
I
0
I 0 . u1
I d . 0
I d . m
I N . 0
I rd3 . ul
I w . 0
a
.
l
I
1
1
t3
I
I
I
,
I
E; 0) cn
l
c-
v)
._
--.-
‘-..’
l,-:‘_ -. .
7
*
_.
.
_
,
_
-.
..-
,...
-
.,
---.
.
<’
.
.
_. Y.
c.
6 )::8 '10 12 li 16 .18 t
6 8 10 12 14 16 18 :/
a
1
1
1
1
I
1
r5.0 6.1985
. ..
- - - -. - -
.
.
-
.
_.
:.:
‘_
‘_ . -
m
.
0
-0.5 , -85
27.093985
- 1.0
I<
0
-0.5
17.12.1985
18.02.1985
- 1.0
-2.0
,
-3.0
.
.”
w
(MPa)
*
.
.
. .._.. -
:
:
1:: I. -7’ *-
.-.
-.
*
‘-
.:
:-
:
,. :
:
.
_
:-
‘.iom2
:.: -0.4 .‘- - 0 . 6
-4.8
-1.0
- 1 . 2
0
l
-0.5
/
l
l
__
4.0
- 1 . 5
r
.«
i.0
- 2 . 5
- 3 . 0
v PD
‘,
:JMPa)
‘:
-5c. -
t*c..
.
_ ” . _
..!’
.
.
.
.
. . .
.
.,
/
.
d
.
‘*I