Effects of high temperature and vapour pressure deficit on net ecosystem exchange and energy balance of an irrigated orange orchard in a semi-Arid climate (Southern Spain)

Abstract

The focus of the work reported here is the impact of severe heat stress conditions
on orchards’ carbon dioxide exchange rate (NEE, Net Ecosystem Exchange).
NEE was monitored by means of the eddy-covariance technique over an irrigated
orange-tree orchard during summer 2009 in Southern Spain. In that period, severe
heat spells occurred (maximum air temperature and vapour pressure deficit up to
38°C and 5 kPa, respectively). Under these conditions, orange trees maintained their
transpiration rates at levels similar to those observed for normal sunny days, while
canopy stomatal conductance and NEE were strongly reduced, thereby leading to a
marked decrease in water use efficiency. The experimental results are discussed in the
context of (i) stomatal and non-stomatal limitations to CO2 exchange and (ii) orchard
respiration loss. As the frequency of extreme events is expected to increase in the
Mediterranean Basin, our results suggest that water productivity of irrigated orchards
may be significantly affected by climate change.

Full text

149
Effects of High Temperature and Vapour Pressure Deficit on Net Ecosystem
Exchange and Energy Balance of an Irrigated Orange Orchard in a Semi-
Arid Climate (Southern Spain)
B. Marti
n
-Gorriz, G. Egea, P.A. Nortes, A. Baille,
M.M. González-Real and I. Ruiz-Salleres
Área de Ingeniería Agroforestal Universidad
Politécnica de Cartagena
Cartagena, 30203
Spain
A. Verhoef
Soil Research Centre
Department of Geography and
Environmental Science
The University of Reading
Reading RG6 6DW
U
K
Keywords: ecosystem carbon balance, bulk stomatal conductance, ecosystem water-use
efficiency, heat stress, eddy-covariance
Abstract
The focus of the work reported here is the impact of severe heat stress con-
ditions on orchards’ carbon dioxide exchange rate (NEE, Net Ecosystem Exchange).
NEE was monitored by means of the eddy-covariance technique over an irrigated
orange-tree orchard during summer 2009 in Southern Spain. In that period, severe
heat spells occurred (maximum air temperature and vapour pressure deficit up to
38°C and 5 kPa, respectively). Under these conditions, orange trees maintained their
transpiration rates at levels similar to those observed for normal sunny days, while
canopy stomatal conductance and NEE were strongly reduced, thereby leading to a
marked decrease in water use efficiency. The experimental results are discussed in the
context of (i) stomatal and non-stomatal limitations to CO
2
exchange and (ii) orchard
respiration loss. As the frequency of extreme events is expected to increase in the
Mediterranean Basin, our results suggest that water productivity of irrigated orchards
may be significantly affected by climate change.
INTRODUCTION
The last report of the Intergovernmental Panel on Climate Change (IPCC, 2007)
left little doubt on the unequivocal warming of the climate system, as is now evident from
observations of increases in global average air and ocean temperatures. The temperature
increase is widespread over the globe, and cold nights and frosts have become less
frequent over most land areas, while hot days and warm nights have become more
frequent. IPCC experts also predict that heat waves will become more prevalent over
most land areas. The latter trend is thought to be especially acute in the Mediterranean
countries, where irrigated agriculture is an important economic sector, especially where
fruit tree production is concerned. An important issue for Mediterranean horticulture is
therefore to assess to what extent the trend towards more frequent heat waves, associated
with an increase in daily average temperature, can affect orchard net CO
2
assimilation,
productivity and viability of irrigated orchards. Limitations on CO
2
assimilation under
warm and dry conditions when soil water is not a constraining factor, can be due to
several causes (Baldocchi, 1997): (i) stomatal limitation, when the plant reduces leaf
stomatal conductance in response to climatic factors, such as air and soil temperature, and
air vapour pressure deficit (VPD), and (ii) non-stomatal limitation, due to a decrease of
the photochemical efficiency of Photosystem II, or of the mesophyll conductance to CO
2
.
The focus of the work reported here is to assess the impact of severe heat stress
conditions on NEE (net ecosystem (CO
2
) exchange) over a well-irrigated orchard (orange
trees). The main objectives are:
(i) to examine the effects of environmental forcing variables (available energy,
temperature, and VPD) on the diurnal trends and magnitudes of orchard-scale carbon
fluxes (NEE) and latent heat flux (E) over a quasi-closed orange-tree orchard;
Proc. XXVIII
th
IHC – IS on Water Use in a Changing World
Eds.: J.E. Fernández and M.I. Ferreira
Acta Hort. 922, ISHS 2011

150
(ii) to interpret the results in terms of stomatal and non-stomatal limitation at the canopy
scale.
MATERIALS AND METHODS
Prevailing Climatic Conditions
Very severe heat stress conditions were observed in Southern Spain during the
summer of 2009, especially from the beginning of July to mid-August, with several heat
spells characterized by maximum daily air temperature (T
a
) and vapour pressure deficit
(VPD) reaching values of up to 38°C and 5 kPa, respectively. Over the last ten years, the
frequency of occurrence of such high values is about 4 days per year.
Site Characteristics
The commercial orchard (20 ha) is located near Cartagena (Murcia Region) in
Southern Spain (37°70’ N and 0°10’ W). The trees were 30-year old drip-irrigated orange
trees (Citrus sinensis var. Navelate) grafted onto Cleopatra rootstock and planted at a tree
spacing of 6×4 m, with a high cover fraction (90%) and high leaf area index (LAI ≈ 5).
The orchard was irrigated to satisfy 100% of the standard crop evapotranspiration, ET
c
,
throughout the whole year. ET
c
was calculated as the product of reference
evapotranspiration, ET
o
, and the crop coefficient for orange trees from the FAO Penman-
Monteith method (Allen et al., 1998).
Measurements and Instrumentation
The following microclimatic variables were monitored at 1.5 m above the tree
crowns: air temperature (T
a
) and relative humidity, solar radiation (R
G
, Kipp & Zonen
pyranometer CMP3, Delft, the Netherlands) and crown temperature (T
c
, Apogee infrared
thermometer (IRT), Logan, Utah, USA). Soil heat flux was measured by means of two
heat flux plates (REBS, model HFT-3.1, Seattle, WA, USA) buried 5 mm below the
surface, near a dripper (wet bulb) and in the middle of the row (dry soil), respectively.
The CO
2
, sensible and latent heat fluxes were measured at 1.5 m above the trees by eddy-
covariance system comprising a Campbell Scientific Inc. (Logan, UT, USA) CSAT-3
sonic anemometer measuring high-frequency (10 Hz) three-dimensional wind speed and
sonic virtual temperature, and a LICOR (Lincoln, NE, USA) LI-7500 open path infrared
gas analyzer measuring CO
2
and H
2
O mixing ratios in absolute mode at 10 Hz.
Calculations
The 10 Hz wind velocity, sonic temperature and gas concentrations were post-
processed to calculate mean values, turbulent statistics, and the vertical fluxes of sensible
heat (H), water vapour (E), and carbon dioxide (NEE). Density corrections were applied
point by point to the gas data before calculating the half-hourly covariances of H
2
O and
CO
2
with the vertical wind speed. We also calculated half-hour estimates of crown-to-air
temperature difference (T
c
– T
a
) and vapour pressure deficit (VPD
c-a
) and derived the bulk
canopy stomatal conductance to water vapour transfer, g
w
(m s
-1
), as
acp
w
VPDC
E
g




(1)
where

is the psychrometric constant (kPa K
-1
),

is the latent heat of vaporization of
water (J kg
-1
), E is evapotranspiration rate expressed per unit ground area (kg m
-2
s
-1
), C
p
is the mass heat capacity of air at constant pressure (J kg
-1
K
-1
), and

is the air mass
density (kg m
-3
).
Bulk canopy internal CO
2
concentration (C
i
, µmol mol
-1
) was determined as C
i
=C
a
– NEE/g
c
, with g
c
= g
w
/1.6 (where the factor 1.6 is the approximate ratio of binary
molecular diffusion coefficients for water vapour and CO
2
in the air), and water use
efficiency (WUE, g L
-1
) as WUE=NEE/E. Data were recorded continuously by means of

151
two data loggers (CR3000 and CR1000, both from Campbell Scientific, Logan, USA) and
averaged over 30 min intervals. Four sunny days with similar heat-stress conditions
(‘stress’ days, T
a,max
> 35°C, VPD
a,max
> 4.5 kPa) and four other sunny days with normal
conditions (‘no-stress’ days, T
a,max
< 30°C, VPD
a,max
< 2.5 kPa) were selected. An average
‘stress’ day and an average ‘no-stress’ day were derived by averaging over the four days
the values of the state and flux variables. In the following, results are presented and
analysed for these two average days.
RESULTS
The average stress and no-stress days presented identical time-courses of R
G
(Fig.
1A), but clearly distinct trends in T
a
(Fig. 1B) and VPD (Fig. 1B), which were
substantially higher in the average stress day during the central hours of the day. Both
average days did not differ in the crown-to-air temperature difference (T
c
-T
a
; Fig. 2A),
whereas the crown-to-air VPD presented a similar trend to that of VPD (Fig. 1), with the
stress day showing values that were about twice as high as the values observed for the no-
stress day, near midday.
Soil heat flux was negligible (data not shown), due to the high vegetation cover
fraction. As a consequence, the diurnal pattern of soil temperature was rather similar
among the two experimental average days (data not shown), which differed by approx-
imately 0.5°C only, at solar noon. We therefore assumed that the canopy available energy
(R
n
- S, S being the heat stored by the canopy and soil) could be partitioned into latent
(

E) and sensible (H) heat flux components (R
n
-S 

E + H) for this case study. No
distinct differences were found in energy partitioning between the two average days (Fig.
3A, B). A strong reduction of NEE was observed for the two average days, from early
morning (from 8:00h till 18:00h a.m, UT) (Fig. 3C), i.e. when photosynthetic photon flux
density (PPFD) was higher than 1000 mol m
-2
s
-1
. However, the observed down
regulation of NEE was much more severe during the heat-stress days.
Bulk stomatal conductance to water vapour (g
w
) decreased in both days during the
same time interval as NEE (Fig. 3D), the down-regulating process of g
w
being stronger
for the heat-stress day. WUE decreased between 6:00h and 9:00h a.m. (UT) for both days
(Fig. 4), but the reduction was sharper during the stress day. From 9:00h a.m. onwards,
WUE was relatively constant for both days, with mean values of 1.08 and 5.38 g L
-1
for
the stress and no-stress days, respectively. An explanation for the divergences observed in
WUE between both days can be found in the different sensitivities to VPD
c-a
of NEE and
g
w
. As shown in Figure 5, the normalized values of NEE (NEE*) and g
w
(g
w
*) correlated
negatively with VPD
c-a
, but the slope was markedly lower for NEE* (-0.25) than for g
w
*
(-0.16). Both NEE* and g
w
* decreased sharply at VPD
c-a
values above 1.5 kPa.
The diurnal pattern of bulk internal minus external CO
2
concentration (C
i
-C
a
) was
considerably shifted with heat stress (Fig. 6). Despite the higher stomatal regulation
observed during the heat-stress day (Fig. 3D), C
i
was substantially higher during this day
(Fig. 6). It is worthy to highlight that NEE, and hence C
i
, includes the contribution of soil
respiration, R
s
. However, for high LAI orchards (this study), we can assume that R
s
is
much lower than canopy respiration (R
canopy
), so C
i
may be considered as representative of
the bulk canopy internal concentration. Furthermore, the differences in C
i
between
average days are unlikely to be due to differences in R
s
because soil temperatures were
relatively similar among both days, as mentioned before.
DISCUSSION
Transpiration
Our results indicated that the orange tree canopy presented an upper limit (ca. 200
W m
-2
) for evapotranspiration (Fig. 3A), which exhibited a similar magnitude and time-
evolution under heat-stress and no-stress conditions. The concept of an upper limit for
whole-tree water transport was first suggested by Halevy (1956) who found that
transpiration rates of ‘Shamouti’ orange leaves were not greater on hot dry days than on

152
mild days. Hall et al. (1975) reported that leaf transpiration rate of orange trees stabilized
with increasing VPD. Sinclair and Allen (1982) observed that citrus trees tend to stabilize
maximum whole-tree transpiration rate by stomatal closure, which limits whole-tree water
transport and decreases leaf CO
2
assimilation (see below). The ‘Valencia’ orange data of
Machado et al. (2005) showed, however, a slight reduction in transpiration rate with VPD
increasing from 1.5 to 3.5 kPa.
CO
2
Assimilation
Amongst the plant physiological processes, photosynthesis is particularly sensitive
to inhibition by heat stress. Both photorespiration and dark respiration increase with
temperature, but increased respiratory activities are insufficient to account for the strong
reduction of the photosynthetic rate that occurs under heat stress (Berry and Björkman,
1980). Thus, photosynthesis is directly inhibited by heat stress because one or more of the
components of the photosynthetic apparatus is exceptionally labile (Salvucci and Crafts-
Brandner, 2004).
Our observations at canopy scale confirmed that the characteristic effects of high
temperature and VPD on leaf photosynthesis are also prevailing at the orchard scale. The
maximum value of NEE occurred early in the morning (8-9 a.m., UT) for the two days,
but decreased much more sharply under heat-stress (Fig. 3C). Orange tree photosynthesis
appeared to be especially sensitive to VPD (Fig. 5A). Canopy-to-air VPD values causing
decreases of NEE were found to be higher than 1.5-2 kPa (Fig. 5A), to be compared with
2.4 kPa for Washington Navel and Valencia oranges (Kriedemann, 1968). Not all Citrus
spp. and cultivars show the same sensitivity to VPD; Habermann et al. (2003) found that
mean leaf photosynthesis was 9.3 and 4.0 mol m
-2
s
-1
at VPD values of 1.2 and 2.5 kPa,
respectively.
Bulk Stomatal Conductance and Photosynthesis Limitations
The concomitant decrease of NEE and g
w
observed with heat stress in this study
(Figs. 3 and 5), suggest that stomatal limitations play an important role in the down-
regulation of orange tree photosynthesis under heat stress conditions. Jifon and Syvertsen
(2003) concluded that the direct effect of high temperature on citrus leaf photosynthesis
was more important than the secondary effects of leaf-to-air vapour pressure deficit
mediated via decreased stomatal conductance. Our results also support the involvement of
heat-mediated non-stomatal limitations on orange tree photosynthesis, as confirmed by
the higher C
i
values observed throughout the average stress day (Fig. 6). However, further
research is needed to discriminate the true contributions of stomatal and non-stomatal
limitations on CO
2
assimilation under heat conditions, as well as the mechanisms
involved (i.e. diffusional and/or biochemical limitations through decreased mesophyll
conductance and Rubisco kinetic impairment, respectively) in the non-stomatal regulation
(Grassi and Magnani, 2005).
Besides heat stress, photoinhibition may also be down-regulating CO
2
assimilation. Citrus sp. trees are characterised by low-light saturation of photosynthesis
(PPFD ≈ 600-700 mol m
-2
s
-1
, at leaf scale), and higher PPFD amounts are known to
induce reversible (dynamic) photoinhibition in citrus. PPFD above 1000 mol m
-2
s
-1
(R
G
≈ 435 W m
-2
, Fig. 1A) seemed to be the limit above which NEE was down- regulated
irrespective of the day (Fig. 3C). In this species, a linear increase of non-photochemical
quenching with an increase in relative light energy excess has been observed in irrigated
and non-irrigated trees (Ribeiro and Machado, 2007).
Concluding Remarks
Our results confirmed that the limitation to water transport in citrus species results
in stomatal closure under high VPD, to the point of stabilizing whole-tree water transport
with concomitant reductions in canopy CO
2
exchange. NEE was decreased more than
transpiration rate in such high temperature and VPD conditions. Although further research
is clearly needed, mechanisms for reduction in orange leaf CO
2
assimilation under heat

153
stress seem to involve both stomatal and non-stomatal limitations. Our data also suggest
that productivity of irrigated orchards might be significantly affected by climate change.
However, it should be noted that elevated CO
2
concentration might partly offset the
negative impacts of high VPD and high temperature on citrus canopy NEE (Allen and Vu,
2009). Therefore, the associated increase of CO
2
concentration might alleviate the
negative effects of high temperature and heat waves evidenced in our study.
ACKNOWLEDGMENTS
G. Egea and P. Nortes hold a postdoctoral fellowship from Fundación Ramón
Areces (Madrid, Spain) and the Spanish Ministry of Science and Innovation (Juan de la
Cierva Program), respectively. The study was partly financed by the Consejería de
Agricultura y Agua of the Murcia Region Government.
Literature Cited
Allen, R.G., Pereira, L.S., Raes, D. and Smith, M. 1998. Crop evapotranspiration.
Guidelines for computing crop water requirements. FAO Irrigation and Drainage
Paper nº 56. FAO. Roma (Italia). 300p.
Allen, L. and Vu, J. 2009. Carbon dioxide and high temperature effects on growth of
young orange trees in a humid, subtropical environment. Agric. Forest Meteor. 149:
820-830.
Baldocchi, D.D. 1997. Measuring and modelling carbon dioxide and water vapour
exchange over a temperate broad-leaved forest during the 1995 summer drought.
Plant, Cell Environ. 20:1108-1122.
Berry, J. and Björkman, O. 1980. Photosynthetic response and adaptation to temperature
in higher plants. Annu. Rev. Plant Physiol. 31:491-543.
Grassi, G. and Magnani, F. 2005. Stomatal, mesophyll conductance and biochemical
limitations to photosynthesis as affected by drought and leaf ontogeny in ash and oak
trees. Plant, Cell Environ. 28:834-849.
Habermann, G., Machado, E.C., Rodrigues, J.D. and Medina, C.L. 2003. Gas exchange
rates at different vapor pressure deficits and water relations of ‘Pera’ sweet orange
plants with citrus variegated chlorosis (CVC). Sci. Hort. 98:233-245.
Halevy, A. 1956. Orange leaf transpiration under orchard conditions. V. Influence of leaf
age and changing exposure to light on normal and dry summer days. Bull. Res.
Council Israel 5D, 165-175.
Hall, A.E., Camacho-B, S.E. and Kaufmann, M.R. 1975. Regulation of water loss by
citrus fruit. Physiol. Plant. 33:62-65.
IPCC. 2007. Summary for Policymakers. In: Climate Change 2007: The Physical Science
Basis. Contribution of Working Group I. Solomon S., Qin D., Manning M., Chen Z.,
Marquis M., Averyt K.B., Miller H.L. (eds.). Cambridge University Press
Jifon, J.L. and Syvertsen, J.P. 2003. Moderate shade can increase net gas exchange and
reduce photoinhibition in citrus leaves. Tree Physiol. 23:119-127.
Kriedemann, P.E., 1968. Some photosynthetic characteristics of citrus leaves. Aust. J.
Biol. Sci. 21:895-905.
Machado, E.C., Schmidt, P.T., Medina, C.L. and Ribeiro, R.V. 2005. Respostas da
fotossíntese de três espécies de citros a fatores ambientais. Pesq. Agropec. Bras. 40:
1161-1170.
Ribeiro, R.V. and Machado, E.C. 2007. Some aspects of citrus ecophysiology in
subtropical climates: re-visiting photosynthesis under natural conditions. Braz. J. Plant
Physiol. 19:393-411.
Salvucci, M.E. and Crafts-Brandner, S.J. 2004. Inhibition of photosynthesis by heat
stress: the activation state of Rubisco as a limiting factor in photosynthesis. Physiol.
Plant. 120:179-186.
Sinclair, T.R. and Allen, L.H. 1982. Carbon dioxide and water vapour exchange of leaves
on field-grown citrus trees. J Exp. Bot. 33:1166-1175.

154
Figures
e
Hour (UT)
00 04 08 12 16 20 00
VPD (kPa)
0
1
2
3
4
5
Hour (UT)
00 04 08 12 16 20 00
T
a
(ºC)
16
20
24
28
32
36
R
G
(W m
-2
)
0
250
500
750
1000
AB
Fig. 1. Diurnal time-course of (A) air temperature (T
a
, circles) and solar radiation (R
G
,
triangles) and (B) air vapour pressure deficit (VPD) for the average stress (filled
symbols) and no-stress (open symbols) days. The error bars denote  SE.
Hour (UT)
00 04 08 12 16 20 00
T
c
-T
a
(º C)
-2
-1
0
1
2
3
4
5
6
7
Hour (UT)
00 04 08 12 16 20 00
VPD
c-a
(kPa)
0
1
2
3
4
5
6
7
AB
Fig. 2. Diurnal time-course of (A) crown-to-air temperature difference (T
c
– T
a
) and (B)
crown-to-air vapour pressure deficit (VPD
c-a
) for the average stress (filled
symbols) and no-stress (open symbols) days. The error bars denote  SE.

155
Hour (UT)
00 04 08 12 16 20 00
NEE (gCO
2
m
-2
s
-1
)
-0.2
0.0
0.2
0.4
0.6
0.8
00 04 08 12 16 20 00

E (W m
-2
)
0
50
100
150
200
250
00 04 08 12 16 20 00
H (W m
-2
)
-50
0
50
100
150
200
250
300
350
Hour (UT)
00 04 08 12 16 20 00
g
w
(mm s
-1
)
0
1
2
3
4
5
6
7
AB
CD
Fig. 3. Diurnal time-course of (A) latent heat flux (E), (B) sensible heat flux (H), (C) net
ecosystem exchange (NEE) and (D) bulk stomatal conductance to water vapour
(g
w
) for the average stress (filled symbols) and no-stress (open symbols) days. The
error bars denote  SE.
Hour (UT)
04 08 12 16
WUE (g L
-1
)
0
4
8
12
16
20
Fig. 4. Diurnal pattern of water use efficiency (WUE) for the average stress (filled
symbols) and no-stress (open symbols) days. The error bars denote  SE.

156
VPD
c-a
(kPa)
0123456
NEE*
0.0
0.2
0.4
0.6
0.8
1.0
y=-0.25x+1.46
r
2
=0.92
VPD
c-a
(kPa)
0123456
g
w
*
0.0
0.2
0.4
0.6
0.8
1.0
y=-0.16x+1.16
r
2
=0.94
AB
Fig. 5. Relationship between crown-to-air vapour pressure deficit (VPD
c-a
) and
normalized (A) net ecosystem exchange (NEE*=NEE/NEE
max
) and (B) bulk
stomatal conductance to water vapour (g
w
*=g
w
/g
w,max
). Values for the diurnal
period 6-14 h and for the average stress (filled symbols) and no-stress (open
symbols) days are plotted. The lines represent the linear regressions to the data
(VPD
c-a
>1.5 kPa). g
w,max
is maximum bulk stomatal conductance taken as the
maximum value of g
w
determined experimentally.
Hour (UT)
06 08 10 12 14 16 18 20
C
i
-C
a
(

mol mol
-1
)
-160
-140
-120
-100
-80
-60
-40
-20
0
Fig. 6. Diurnal pattern of bulk internal minus external CO
2
concentration (C
i
-C
a
) for the
average stress (filled symbols) and no-stress (open symbols) days. The error bars
denote –SE of the mean.