Fluorescence Responses to Water Stress in Grapevine Leaves:

(1)

Fluorescence Responses to Water Stress in Grapevine Leaves:

A New Remote Sensing System

Jaume Flexas, Jean-Marie Briantais,*

^†

Zoran Cerovic,

^†

Hipo´lito Medrano,*

and Ismael Moya

^†

A

new fluorimeter built at Orsay allowed us to mea- gressed, thus serving as a good indicator of plant water status. (3) A simple fluorescence parameter, Fs, accu- sure at a distance of up to 6 m both the steady-state and

rately reflected the plant physiological state. Over the the maximum chlorophyll fluorescence. This instrument

range of light intensities used in this study, this parame- has been applied continuously during 17 days of water

ter changed in parallel with irradiance in well-watered stress development to follow the chlorophyll fluorescence

plants. With increasing water stress, Fs changed in oppo- parameters of a potted grapevine. Gas-exchange rates for

site direction to irradiance changes. The response of Fs H2O and CO2and chlorophyll fluorescence parameters of

to rapid changes in irradiance was fast (within seconds).

the same leaf were recorded concurrently. It was shown

The potential of this parameter for remote sensing of wa- that: (1) Under well-watered conditions, before noon, a

ter stress is discussed. Elsevier Science Inc., 2000 correlation was found between net photosynthetic rate

and the rate of electron transport calculated from fluores- cence measurements. After several hours of high light ex-

INTRODUCTION posure, CO2 assimilation (A) started to decrease more

than the rate of electron transport (ETR). Under drought

The interest of chlorophyll fluorescence as a useful signal conditions, the above-mentioned correspondence was lost:

reflecting plant photochemistry has been widely re- when A almost vanished due to high stomatal closure, the

viewed (Bolha`r-Nordenkampf et al., 1989; Krause and ETR was still about 50% of the control value. It is sug-

Weis, 1991; Schreiber et al., 1994; Lichtenthaler, 1996).

gested that under these conditions, the ratio of photores-

This is a nondestructive and nonintrusive signal, easy to piration to CO2assimilation increased. (2) Light response use for many purposes in laboratory and fieldwork. For of the quantum yield of ETR became increasingly differ- these reasons efforts have been made to relate chloro- ent between morning and afternoon as water stress pro- phyll fluorescence parameters, mainly the electron transport rate from PS II (ETR), with actual rates of CO2 assimilation (Edwards and Baker, 1993; Genty et al., 1989;

* Instituto Mediterra´neo de Estudios Avanzados—Universitat de

Harbinson et al., 1989; O¨quist and Chow, 1992; Schin-

les Illes Balears (UIB-CSIC), Departament de Biologia Ambiental,

dler and Lichtenthaler, 1996; Valentini et al., 1995; Weis

Carretera Valldemossa Km. 7,5, 07071 Palma de Mallorca, Baleares,

Spain and Berry, 1987). The results have shown good agree-

† Laboratoire pour l’Utilisation du Rayonnement Electromag-

ment between CO2 assimilation and ETR in C4 plants,

ne´tique (LURE), Centre Universitaire Paris-Sud, B.P. 34, 91898 Orsay

but not as good agreement in C3plants, due to the con-

Cedex, France

tribution of other processes to electron use.

Address correspondence to J. Flexas, Instituto Mediterra´neo de

Estudios Avanzados—Universitat de les Illes Balears (UIB-CSIC), De- Photorespiration and the Mehler reaction are the

partament de Biologia Ambiental, Carretera Valldemossa Km. 7,5,

main processes related to the imbalance between CO2

07071 Palma de Mallorca, Baleares, Spain. E-mail dbajfs4@ps.uib.es

Received 20 September 1999; revised 28 January 2000. assimilation and ETR. The first consists of the oxygen-

REMOTE SENS. ENVIRON. 73:283–297 (2000)

Elsevier Science Inc., 2000 0034-4257/00/$–see front matter

655 Avenue of the Americas, New York, NY 10010 PII S0034-4257(00)00104-8

(2)

ation of ribulose-1,5-bisphosphate by Rubisco, which, ac- pulse amplitude modulation fluorimeter (FIPAM) and gas-exchange rates with a CO2/H2O porometer (LI-6400, cording to the enzyme properties, is likely to increase

when CO2 availability in the chloroplast is reduced, as Li-Cor Inc., Lincoln, NE, USA) were performed continuously (night and day) during the 17 days of a drought occurs under water stress due to stomatal closure. The

photorespiratory pathway itself consumes only about half cycle. Some experiments with artificial light were performed to complete the study.

of the NADPH synthesized by the chloroplastic electron transport chain in respect to the consumption by CO2as- similation. However, this cycle, which evolves one mole-

MATERIAL AND METHODS cule of CO2 per each two molecules of O2 reduced, is

always coupled to CO2 assimilation through the recycling Plant Material

of the evolved CO2. Both processes together represent a One-year-old plants of Vitis vinifera (L.) cultivar Ca- combined cycle, the C2-C3 cycle that according to a bernet Sauvignon were grown in a greenhouse at Orsay steady-state biochemical model recently presented (Ta- (vicinity of Paris, France), under natural light and tem- keba and Kozaki, 1998) is able to maintain about 75% of perature conditions in small pots (0.5 L) with horticul- maximum ETR in a situation in which no net CO2assim- tural substrate. Pots were covered with aluminum foil to ilation is observed (that is, only by internally recycling avoid soil water evaporation, and periodically irrigated to the CO2 evolved by the photorespiratory pathway). The maintain them at field capacity until the onset of mea- Mehler reaction consists of a direct reduction of O2 by surements, which were performed during the summer of the electron transport chain at the ferredoxin level 1996. The last fully expanded leaf of the main shoot was (Asada, 1999). Both processes increase under water used for measurements.

stress as a consequence of reduced CO2 availability in the chloroplast, which increases the ratio O2/CO2, and

both have been suggested many times as important elec- Environmental Conditions tron consumers under water stress (Cornic and Briantais,

Measurements were performed under natural greenhouse 1991; Flexas et al., 1999a; Flexas et al., 1999b; Flexas et

conditions during the summer. Air and leaf temperature al., 1999c; Osmond et al., 1997; Wingler et al., 1999),

(inside and outside the Li-6400 chamber) were continu- and their importance in water-stressed grapevines has

ously recorded using thermocouples coupled both to the been recently demonstrated (Flexas et al., 1999c).

FIPAM (measurements every 30 s) and to the Li-6400 Despite these difficulties with C3plants, chlorophyll

(measurements every 5 minutes). Photosynthetic Photon fluorescence has been shown to be an interesting tool for

Flux Density (PPFD) was also recorded on the leaf sur- plant stress detection (Cecchi et al., 1994; Cerovic et al.,

face with a quantum meter coupled to the Li-6400.

1996; Gu¨nther et al., 1994; Moya et al., 1992; Moya et

Environmental heterogeneity was present during the al., 1995). The potential of fluorosensing water stress has

experiment, with sunny and cloudy days, as well as sunny been reported recently for several plants including

and cloudy intervals within the same day. The iron glass- grapevines using parameters other than ETR, such as the

house structure also caused a temporally unavoidable chlorophyll fluorescence mean lifetime (Cerovic et al.,

light interception that shaded the leaf for short periods 1996; Schmuck et al., 1992) or nonphotochemical quench-

during the diurnal time courses, which caused disconti- ing of chlorophyll fluorescence (Flexas et al., 1998;

nuities in the profile of light interception by the leaf.

Flexas et al., 1999a; Flexas et al., 1999b; Schultz, 1997).

These sudden discontinuities served to aid observation of Moreover, it has been noticed that water stress induces

the rapid response of photosynthesis to changes in the marked effects on the daily pattern of steady-state chlo-

light environment. About 1,200 lmol photon m^⫺² s^⫺¹ rophyll fluorescence (Fs) (Cerovic et al., 1996; Flexas et

PPFD were recorded at midday sunny peaks during al., 1999a; Flexas et al., 1999b; Rosema et al., 1998).

sunny days. Air temperature inside the greenhouse var- The aims of the present work were:

ied between 15⬚C and 20⬚C during the night and dawn, 1. To test the capacity of the new fluorimeter devel- to peaks of 30⬚C to 35⬚C at midday (data not shown).

oped at the LURE (Orsay, France) by I. Moya for measuring at a distance of 0.5 m to 6 m both

the steady state and the maximum chlorophyll flu- Plant Water Status

orescence in vivo. Water stress was induced by withholding watering. Daily 2. To test the utility of several chlorophyll fluores- water loss was followed by successive pot weighing dur- cence parameters for plant water stress detection, ing the experiment. Leaf discs were taken periodically paying special attention to the time resolution and from leaves similar to those used for photosynthetic mea- spatial correlation of Fs changes to light over a surements. The samples were taken always in the early range of values of the plant water status. morning to avoid differences in water content due to water loss during the day. The leaf water deficit (LWD) Simultaneous measurements of chlorophyll fluores-

cence with the newly constructed frequency-induced was estimated from disc fresh weight and the weight of

(3)

the same discs after 24 h in distilled water at 4⬚C (full by specially designed electronics locked to the frequency of excitation pulses, which make the pulsed response in- turgor), as follows: LWD⫽(turgid weight⫺fresh weight)/

turgid weight. sensitive to continuous illumination even under condi-

tions that saturate fluorescence. Two signals are obtained in parallel: a fluorescence signal (Fs) and a continuous signal (Rcont), which is proportional to the ambient light Chlorophyll Fluorescence Measurements

reflected by the leaf that passes through the detection A new fluorimeter was designed and built at LURE (Or-

filter. It has been observed that Rcont is proportional to say, France) by I. Moya, with the aim of continuously

the PPFD intensity. Under our experimental conditions, recording fluorescence parameters (Fs and Fm) during

changes in the incident light due to solar position in- several days, at a distance from the leaf sufficient to

duced only minor decorrelation between Rcont and avoid any interference with the natural illumination of

PPFD. Therefore, after calibration, Rcont can be used the leaf. This distance was 0.6 m in our particular ex-

to follow changes in PPFD. The fact that Rcont and Fs periment.

originate from exactly the same leaf area enables a pre- The FIPAM fluorimeter is based on fluorescence ex-

cise correlation between these two signals.

citation by a laser diode (635 nm, 10 mW, SDL Inc.).

The instrument is controlled by a computer with a The beam is modulated at different frequencies with

specially designed program, which allows continuous constant amplitude and duration (2 ls) and focused on

measurement over several days. Every second the Fs and the leaf by a microscope objective, from distances adjust-

Rcont signals are measured together with the air, leaf, able in the range of 0.5 m to 6 m, depending on the

detector, and laser temperatures. Corrections were ap- laser source. The resulting spot has a rectangular shape

plied to make the experiments insensitive to temperature of 0.5 mm by 4 mm at a distance of 1 m. The new con-

changes of the instrument. The mean of 30 measure- cept of saturating the fluorescence yield by increasing

ments is calculated every 30 s. The zero of the fluores- the frequency of modulation makes a bridge between the

cence signal is measured for each cycle by triggering the PAM technique, widely applied among plant physiolo-

measurement in the absence of the excitation pulse. This gists (Schreiber, 1983) and other LIDAR systems capa-

value is automatically subtracted from the fluorescence ble of detecting chlorophyll fluorescence at distances

signal. Thus, the Fs value is free of any electronic drift.

greater than 10 m, but restricted to measuring the sta-

This is of particular importance since the experiment tionary fluorescence level (Fs). We have already used the

lasted for several days.

FIPAM fluorimeter at distances of about 6 m using a

In this experiment the frequency of Fm measure- 100-mW laser diode (Philips CQL 822/D, Eindhoven,

ments was initially set to one measurement each 10 min- The Netherlands) instead of the 10-mW one used in the

utes. The same procedure was used to measure Fv/Fm present work. The market availability of high-power laser

and ⌬F/Fm⬘. We refer to Fv/Fm when measurements diodes and other solid-state, high frequency modulation

are taken by night (i.e., all photochemical quenching re- lasers, which can be used for chlorophyll excitation, is

laxed) and to ⌬F/Fm⬘ (Genty et al., 1989) when mea- growing very fast. There is no doubt that remote sensing

surements were taken in the presence of actinic light measurements with the FIPAM method over distances

(i.e., after dawn). Since⌬F/Fm⬘ represents the quantum higher than 10 m will be feasible in the near future.

yield of PSII photochemistry, the electron transport rate With our system, the basal fluorescence value when

from PSII was calculated by multiplying⌬F/Fm⬘by inci- all photosystems are closed—that is, at complete dark-

dent PPFD. The result is expressed in relative units be- ness (Fo), as well as at the steady-state chlorophyll fluo-

cause it considers neither the leaf absorbance nor the rescence emission under a given irradiance (Fs)—are

factor of PSI-PSII excitation distribution. Most workers measured with a frequency of only 1 Hz, which corre-

accept this parameter as a good estimate of the linear sponds to an average intensity of 0.05lmol photons m^⫺²

electron transport from PSII (Bilger et al., 1995; Cornic s^⫺¹. At this frequency no actinic effect is observed even

and Briantais, 1991; Flexas et al., 1999a; Flexas et al., in complete darkness. Maximum fluorescence when all

1999b; Flexas et al., 1999c; Genty et al. , 1989; Krall and centers are closed, both in darkness (Fm) or under a

Edwards, 1992), although it has been recently suggested given irradiance (Fm⬘), are induced by increasing the

that this does not hold under CO2-limited photosynthesis frequency to 100 kHz. Under these conditions, the aver-

and high irradiance (Rosema et al., 1998).

age intensity ranges between 2,000 and 10,000lmol m^⫺² s^⫺¹, depending on focusing. We ensure that actual maximum chlorophyll fluorescence has been reached by re-

Gas-Exchange Parameters cording the complete induction kinetics, with a time res-

olution of 10 ms (not shown). The leaf fluorescence is The photosynthetic performance (both fluorescence and gas exchange) of a single leaf was followed during the 17 collected by a 15-cm Frenel lens and focused on a PIN

photodiode (Hamamatsu S3590) after passing through a days of the experiment to avoid any effect due to plant or leaf variability. The Li-6400 chamber was placed in a high-pass filter (Schott RG665). The signal is processed

(4)

experiment was to confirm the results obtained pre- viously in the absence of Fm quenching induced during the night by excessive frequency of saturating pulses.

The frequency of saturation pulses was decreased to one pulse each 20 minutes and laser focus was slightly changed. Again, plants were grown under greenhouse light and temperature conditions and irrigated periodically to maintain the soil at field capacity, and water stress was induced by withholding watering. Leaf water deficit was estimated as described above.

For measurements, plants were dark-adapted for 2 hours in a dark room. The temperature was maintained constant at 25⬚C throughout the experiment. The artificial diurnal cycle on a single leaf was provided by a 250-W slide projector whose light intensity was varied Figure 1. Decreases in pot weight during the studied pe- using a rotating dimmer coupled to a stepping motor.

riod, showing pot water loss by the plant (experiment with

The motor was controlled by a program run on a Hew- natural light). Numbers indicate the day of the month,

lett Packard 9816 computer. The light beam was filtered which correspond to 3 days represented in the following

figures. through a 2-cm layer of a copper sulphate solution (1 M)

to minimize spectral changes when varying light intensity. A different plant was used for each diurnal cycle, way that the natural position of the leaf was not modi-

during which gas-exchange and chlorophyll fluorescence fied. CO2concentration of air flowing through the system

measurements were performed as described above.

was maintained constant at about 360 lmol mol^⫺¹. The

“Autolog” setting of the instrument was used to record

the parameters every 5 minutes. The ability of the FI- RESULTS PAM to measure at a distance enabled fluorescence

Leaf Water Status measurements on the same area of the leaf through the

Figure 1 shows the decrease in soil water content (esti- transparent window of the Li-6400 chamber. No differ-

ences were observed between leaf temperature inside mated as pot weight loss) during the water stress devel- and outside the chamber. opment. Water loss was due only to plant transpiration, as the pots were covered with aluminum foil to prevent Experiments with Artificial Light evaporation from the soil surface. The progressive reduction of the slope of weight decrease revealed that leaves One year later (summer 1997), the same plants were

transferred to larger pots (3 L). The aim of this second adjusted their transpiration rate gradually in response to

Table 1. Changes in Leaf Parameters During Drought Development^a

LWD (%) A/g ETR/A

Day Sunlight (⫾1.5%) (⫾15%) (⫾15%)

5/08/96 S 5.1 not m. not m.

8/08/96 S 5.8 103.3 (130) not m.

9/08/96 S 7.6 101.5 (147.1) 13.5 (33.4)

10/08/96 C not m. 116.7 (136.4) 12 (13.1)

11/08/96 S not m. 123.8 (99.6) 15 (20.8)

12/08/96 C not m. 116.6 (184) 16.4 (20.4)

13/08/96 S not m. 155.8 (143.5) 18.7 (26.1)

14/08/96 S not m. 249.0 (193.3) 22.1 (37.9)

15/08/96 S not m. 293.5 (437.5) 38.7 (104.8)

16/08/96 S 10.0 366.7 (1125.5) not m.

17/08/96 S 10.0 437.5 (4500.0) 154.8 (82)

25/06/97 S 5.9 75.0 (66.5) 6.8 (13.3)

5/07/97 S 10.7 1095.9 (n.s.) 221.7 (n.s.)

aLeaf water deficit (LWD, mean for three replicates), water use efficiency, and the ratio of electron transport to CO2assimilation at 200 lmol photon m^⫺²s^⫺¹ (A/g in lmol CO2 lmol H2O^⫺¹; ETR/A inlmol electronslmol CO^⫺2¹, assuming a leaf absorptance of 0.84 and equal distribution of energy between the two photosystems). Values in brackets represent afternoon data. Sunlight (S), sunny days, up to 800lmol photon m^⫺²s^⫺¹or more; cloudy days (C), less than 500lmol photon m^⫺²s^⫺¹.

not m.⫽not measured; n.s.⫽nonsignificant because of the low and scattered values of both A and g (see Fib. 8b).

(5)

Figure 2. Diurnal time course of chlorophyll fluorescence and gas-exchange under irrigation conditions on a sunny day (9 Au- gust 1996). (A) Chlorophyll fluorescence. Dots represent values of Fm and Fm⬘. Continuous line represents values of Fo and Fs. The spikes of Fo during the night are due to incomplete reopening of closed centres during the 30 seconds after a saturating pulse. (B) Variable fluorescence, Fv/Fm and⌬F/Fm⬘ (dots). The dotted line is the PPFD measured with the FIPAM.

(C) Relative electron transport rate (ETR) estimated from chlorophyll fluorescence measurements (continuous thin line), rate of CO2assimilation (A) measured by gas exchange (continuous thick line) and PPFD measured with the internal quantum meter of the gas-exchange analyzer chamber (dotted line). (D) The relationship between⌬F/Fm⬘ and PPFD, replotted from Fig. 2B. Solid triangles are morning data and empty circles are afternoon data.

soil water availability. Recorded diurnal time courses of a substantial decrease in Fv/Fm (Fig. 2B). The origin of leaf transpiration and stomatal conductance confirmed these phenomena, present in most of the recorded cycles this adjustment (data not shown). As can be seen (Table (see Figs. 3 and 4), seems to be the repetition of saturat- 1), the studied range of LWD (from 5 to 10%) was far ing pulses in the same leaf area during the whole night.

from the 30% known to cause strong reductions in pho- During the morning, the Fs pattern followed quite well tosynthetic capacity (Cornic, 1994). The 1997 LWD that of PPFD, but this relationship was not completely values were within the range of the 1996 experiment maintained during the afternoon. The relationship be-

(Table 1). tween ⌬F/Fm⬘ and irradiance during the day (Fig. 2D)

shows that for any given irradiance, values corresponding to the morning were similar to those of the afternoon.

Effects of Water Stress on the Diurnal Time Only points corresponding to the dawn showed a differ- Course of Chlorophyll Fluorescence and Gas ent pattern. The diurnal time course of electron trans- Exchange (Experiment with Natural Light) port rate (Fig. 2C) followed the diurnal pattern of irradiance. The rate of CO2 assimilation also followed the Figures 2A and 2B show the diurnal pattern of chloro-

same pattern during the morning (Fig. 2C). However, phyll fluorescence parameters under irrigated conditions

from midday on, a progressive decrease in CO2 assimila- during a sunny day. Through the night, a marked de-

crease of Fm and a slight increase in Fo (Fig. 2A) caused tion was recorded and was not accompanied by concomi-

(6)

Figure 3. Diurnal time course of chlorophyll fluorescence and gas exchange under irrigation conditions on a cloudy day (10 August 1996). (A) Chlorophyll fluorescence. Dots represent values of Fm and Fm⬘. Continuous line represents values of Fo and Fs. The spikes of Fo during the night are due to incomplete reopening of closed centres during the 30 seconds after a saturating pulse. (B) Variable fluorescence, Fv/Fm and⌬F/Fm⬘ (dots). The dotted line is the PPFD measured with the FI- PAM. (C) Relative electron transport rate (ETR) estimated from chlorophyll fluorescence measurements (continuous thin line), rate of CO2assimilation (A) measured by gas exchange (continuous thick line) and PPFD measured with the internal quantum meter of the gas-exchange analyzer chamber (dotted line). (D) The relationship between⌬F/Fm⬘ and PPFD, replotted from Fig. 3B. Solid triangles are morning data and empty circles are afternoon data.

tant decreases in electron transport rate, which caused orescence parameters to light intensity (dawn increase of Fs, diurnal time course of ⌬F/Fm⬘; Figs. 4A and 4B).

an increase of the ratio ETR/A during the afternoon (see

also Table 1). An interesting aspect was observed in the relationship

between ⌬F/Fm⬘ and light intensity by comparing Figs.

The chlorophyll fluorescence diurnal pattern of irri-

gated plants during a cloudy day (Figs. 3A and 3B) 2D and 4D. Two different patterns, corresponding to morning and afternoon data, were clearly distinguished showed the same trends and relationships described for

a sunny day, but with changes not so marked along the for the stressed plant, whereas only one pattern was present for the control plant. During the morning the day. Interestingly, on this day, which had maximum irra-

diances lower than 400lmol photons m^⫺²s^⫺¹, the diur- quantum yield of PSII was similar to that of irrigated plants for any given irradiance, but clearly lower in the nal pattern of CO2 assimilation followed quite well that

of electron transport rate during the whole day, with no afternoon. Such afternoon quenching of⌬F/Fm⬘did not reverse after several hours of darkness, so the maximum imbalance detected in the afternoon (Fig. 3C). The slope

of the relationship between ⌬F/Fm⬘ and irradiance was Fv/Fm recorded during the following night was 0.52 (not shown). Diurnal patterns of CO2 assimilation and elec- similar to that of the previous day (Fig. 3D).

In comparison with irrigated plants, drought stress tron transport rate were clearly different under water stress conditions (compare Figs. 3C and 4C). While ETR induced a more pronounced response of chlorophyll flu-

(7)

Figure 4. Diurnal time course of chlorophyll fluorescence and gas exchange under drought conditions on a sunny day (17 Au- gust 1996). (A) Chlorophyll fluorescence. Dots represent values of Fm and Fm⬘. Continuous line represents values of Fo and Fs. The spikes of Fo during the night are due to incomplete reopening of closed centres during the 30 seconds after a saturating pulse. (B) Variable fluorescence, Fv/Fm and⌬F/Fm⬘ (dots). Dotted line is the PPFD measured with the FIPAM. (C) Relative electron transport rate (ETR) estimated from chlorophyll fluorescence measurements (continuous thin line), rate of CO2assimilation (A) measured by gas exchange (continuous thick line) and PPFD measured with the internal quantum meter of the gas-exchange analyzer chamber (dotted line). (D) The relationship between⌬F/Fm⬘ and PPFD, replotted from (B).

Solid triangles are morning data and empty circles are afternoon data.

followed the diurnal pattern of irradiance, CO2 assimila- a negative correlation was found even at low light intensities (Fig. 5C).

tion was almost absent during most of the day, due to complete stomatal closure only 3 hours after dawn (plot

not shown). Effects of Progressive Soil Drying along the Days

In water-stressed plants the diurnal time course of on Stomatal Conductance, CO2Assimilation, Fs showed an opposite pattern to that of well-watered and Electron Transport Rate

conditions; that is, it showed an inverse correlation with Progressive soil drying was accompanied by different de- PPFD. Figure 5 shows details (periods of 4 hours) of grees of reduction in A, g, and ETR (Figs. 6A and 6B).

drought-associated change in the Fs response to PPFD Average values of these parameters were obtained from for three different days after withholding water. It is measurements taken at 200lmol photon m^⫺²s^⫺¹for the clear that under irrigated conditions, there was a positive full period of water stress development. This PPFD was correlation between Fs and irradiance, in spite of the chosen because it is present both on sunny and cloudy large variations in irradiance (Fig. 5A). However, only 5 days. Stomatal closure was an early response to soil dry- days later, under a mild water stress, this pattern had ing, accompanied by a concomitant decrease of CO2 as- changed. The positive correlation was maintained at low similation. However, CO2 assimilation showed a slightly light intensities (below 250 lmol photons m^⫺² s^⫺¹), but lower decrease, and therefore the intrinsic water use effi- at high light intensities there was an inverse correlation ciency (A/g) progressively increased with water stress (Fig.

6 and Table 1). Such increases were high when taken between the two parameters (Fig. 5B). Three days later,

(8)

Experiment with Artificial Light

The “diurnal” cycles performed with artificial light were made to verify the above results under controlled conditions, avoiding frequent changes in incident light intensity over the leaf, as well as large variations of temperature. In addition, we could check that the Fv/Fm quenching en- dured by oversaturating pulses during the night did not affect the dependence of other fluorescence parameters on water stress. Figure 7A shows the diurnal time course of Fs and Fm for a well-irrigated plant. Both Fo and Fm values remained constant during the night, and thus Fv/

Fm was stable (Fig. 7B). It is interesting to note that similar to the experiment performed with natural light, Fs increased suddenly at dawn, when light intensity was less than 10lmol photon m^⫺² s^⫺¹, but to a much lower extent than in the previous experiment. Also, this effect was reversed in a shorter time. The quenching of Fm observed at dawn was much lower than in the previous experiment, and its duration was much shorter. The diurnal time course of Fs followed that of PPFD (Fig. 7A).

Again, under water stress conditions, this pattern was in- verted (not shown).

Figure 8A shows the diurnal time course of CO2 assimilation and electron transport rate for irrigated plants.

A good coincidence was observed between both parameters during the morning. At higher light intensities (midday), electron transport continued to rise until the midday light peak, while CO2assimilation reached its maximum value at about 500lmol photon m^⫺²s^⫺¹. The decorrelation is larger during the afternoon, as noticed in the experiment with natural light.

When the plant reached a water deficit similar to Figure 5. The relation between Fs (continuous line) and

PPFD (dotted line). Detail of 4-hour periods. (A) Well- that of the experiment with natural light, CO2 assimila- watered plant, corresponding to 9 August 1996. (B) Mild tion was almost completely absent (Fig. 8B), in accor- water stress situation, 14 August 1996. (C) Severe water

dance with an almost complete stomatal closure (not stress situation, 17 August 1996.

shown). Electron transport rate, however, was still maintained at about 50% to 60% of control values (Fig. 8B).

The plot of ⌬F/Fm⬘ against light intensity showed from morning data (from values around 100 under well- that the afternoon data coincide with those of the morn- watered conditions, to more than 400 under drought). ing in the well-watered plant (Fig. 8C). For the stressed For the afternoon data, taking into account the standard plants, however,⌬F/Fm⬘was lower in the afternoon than error, such increases were similar up to the fourteenth in the morning for the same light intensity, as in the ex- day, and were larger thereafter. This is due to the well-

periment with natural light. Moreover, at high light in- known effect of water stress on midday stomatal closure

tensities, ⌬F/Fm⬘ values were lower than for irrigated (Chaves, 1991). Electron transport rate oscillated within a

plant (Fig. 8D).

narrow range of values during the period. Only the values corresponding to the afternoon of the last days of the

drought cycle decreased clearly. A progressive increase in DISCUSSION the ratio ETR/A was observed with soil drying (Fig. 6

Water Stress Effects on Leaf Photosynthesis and Table 1). Moreover, afternoon values were higher

It is shown that under well-watered conditions, ETR, A, than morning values even in well-watered plants, proba-

and g followed the diurnal time course of PPFD during bly due to increased photorespiration after midday sto-

the whole morning. During the afternoon, however, matal closure (Flexas et al., 1999a). Only on day 17 did

there was a consistent decrease in A that matched a de- a decrease of the ratio occur in the afternoon. This is

crease in g. It is well established that even for irrigated probably a consequence of photoinhibition under such

drastic conditions, as already discussed. plants some degree of water stress is achieved at midday,

(9)

Figure 6. Averaged values⫾standard error of electron transport rate (ETR, solid circles), CO2assimilation (A, empty squares) and stomatal conductance (g, solid squares) at 200lmol photon m^⫺² s^⫺¹during the days of water stress development. This PPFD was chosen to include comparable data of both sunny and cloudy days: (A) represents morning data; (B) represents afternoon data.

as a consequence of excess atmosphere water demand Vicia faba and Hordeum vulgare (Lal et al., 1996), as well as for grapevines (Flexas et al., 1998; Flexas et al., (Chaves, 1991). Down-regulation of A by photosynthate

accumulation has also been claimed to take place (Az- 1999a) and has been associated with relative increases in photorespiration and/or Mehler reaction rates, which co´n-Bieto, 1983), although it has been reported by Chau-

mont et al. (1994) and Downton et al. (1987) that such might help to maintain PSII stability under conditions of drought and excess light (Kozaki and Takeba, 1996; Park an accumulation does not occur in grapevines. Under our

experimental conditions, an evaporative demand ex- et al., 1996; Takeba and Kozaki, 1998). The results given here show that when net CO2assimilation is close to zero ceeding the water flux into the leaf seems to be the

cause of the decrease in g and A. This is in agreement under severe water stress, ETR is still about 75% of control values (see Fig. 6A). This is exactly the percentage with the observed decreases in g, concomitant to an in-

crease of A/g during the afternoon, and also such an af- of maintained ETR expected at the compensation point, that is, when the only CO2 assimilation corresponds to ternoon depression did not appear in cloudy days, when

leaf-to-air vapor pressure deficit had been lower during recycling of internally produced CO2, with no net exchange between the leaf and the atmosphere (Takeba the morning.

Even when decreases in A were measured, ETR re- and Kozaki, 1998). These results, together with the proven importance of both photorespiration and the mained unaltered. Such an imbalance between electron

transport rate and CO2 assimilation as a response to wa- Mehler reaction as electron consumers in chilled and water-stressed grapevines (Flexas et al., 1999c), makes us ter stress has already been reported for the C3 plants

(10)

Figure 7. Diurnal time courses of chlorophyll fluorescence and gas exchange under irrigation conditions. Experiment with artificial light. (A) Chlorophyll fluorescence. Dots represent values of Fm and Fm⬘. Continuous line represents values of Fo and Fs. (B) Variable fluorescence, Fv/

Fm and ⌬F/Fm⬘ (dots). Dotted line is the PPFD measured with the FIPAM.

assume that the ETR calculation is quite accurate even antais (1991), electron transport to O2 should be rela- under water stress, in contrast to that suggested by Ro- tively increased during the desiccation of the leaf. This sema et al. (1998). An important implication of this is alternative sink for electrons should be large enough to that the imbalance between A and ETR is due to real, maintain high rates of electron transport during most of physiological events, and not to an invalidation of the the day. In the afternoon, a slight decrease of ETR was Genty model (Genty et al., 1989) for PSII photochemis- observed. By contrast to what happened under irrigated try under water stress. Thus, it will not be possible to esti- conditions, such an increase in electron transport to O2

mate actual CO2 assimilation from ETR measurements. was not able to protect leaves from photoinhibition dur- Under water stress, almost all measured parameters ing drought stress, as witnessed by only partial recovery showed marked decreases throughout the day. Especially of the afternoon quenching during the night. This is con- remarkable was the decline in g, which reached values sistent with recent reports of Brestic et al. (1995). How- near zero only a few hours after dawn. Also, a large de- ever, photoinhibitory effects did not appear until the af- cline in A was found, but the ratio A/g increased dramat- ternoon, whereas photosynthesis was almost totally ically. In addition ETR/A strongly increased as a conse- inhibited since early morning, indicating that electron quence of the low reduction in ETR. These results transport to oxygen could help mitigate the damage of confirm that water stress does not cause important inhi- photosystem II at least during a large part of the day bition of the photochemical mechanism (Cornic et al. (Kozaki and Takeba, 1996; Park et al., 1996).

In general, the experiment with artificial light 1989; Genty et al., 1987). According to Cornic and Bri-

(11)

Figure 8. Experiment with artificial light. (A) Irrigated plant and (B) water-stressed plant: Relative electron transport rate (ETR) estimated from chlorophyll fluorescence measurements (continuous thin line), rate of CO2assimilation (A) measured by gas-exchange (continuous thick line), and PPFD measured with the internal quantum meter of the gas-exchange analyzer chamber (dotted line). (C) Irrigated plant and (D) water-stressed plant: the relationship between ⌬F/Fm⬘ and PPFD. Solid triangles are morning data and empty circles are afternoon data.

became of nonnegligible importance. The clearest con- yielded very similar results, confirming that the observed

sists in a decrease of Fv/Fm at night (Figs. 2–4). This effects were due to water stress and not to the heteroge-

effect was observed in the first experiment, but not in neity of experimental conditions. These results confirm

the one with artificial light, after changing two laser con- that water stress-induced decreases in CO2 assimilation

ditions (Fig. 7) (i.e., the focalization and the frequency were mainly due to stomatal closure and not to de-

of saturating pulses). This suggests that it was caused by creased photochemical efficiency of PS II (Cornic et al.,

excessive intensity and frequency of repetition of the sat- 1989; Cornic and Briantais, 1991; Cornic, 1994; Lal et

urating light pulses. It is important to note that in the al., 1996). However, some down-regulation of PS II ac-

set of experiments presented in this work, this quenching tivity occurred as a consequence of the decreased CO2

availability (Foyer et al., 1990), since the rates of elec- is reversible under normal daylight conditions, as the tron transport under severe water stress were slightly same Fv/Fm value is observed after 24 h (see Figs. 2 lower than those of the irrigated one. and 3, which correspond to two consecutive days). This quenching results from two different effects: an increase in Fo and a decrease in Fm.

Night and Dawn Quenching of Chlorophyll The increase of Fo through the night seems to be a Fluorescence due simply to a cumulative noncomplete relaxation between saturating pulses. Indeed, 10 minutes should not The possibility offered by the FIPAM to determine chlo-

be enough for complete relaxation in dark-adapted sam- rophyll fluorescence without interfering with the light

ples, due to slow reoxidation of plastoquinone (Bukhov climate of the leaf lead us to apply saturating pulses day

et al., 1996), especially if there is an accumulation ofQB- and night during several days, over the course of the first

nonreducing centers under water stress, as recently sug- experiment (“natural illumination”). As a result, problems

that may not appear under short duration measurements gested (Lu and Zhang, 1998; Lu et al., 1998). It is likely

(12)

that the height of the “comb” effect of night reduction sity, and the diurnal pattern of steady-state chlorophyll of Fo could be an indicator of the relative abundance of fluorescence (Fs). We assume that these two approaches QB-nonreducing centers. Also, there was an increase of can be useful tools for water stress assessment, although Fs at dawn because of a lack of activation of photosyn- it remains to be tested if other stresses would lead to thetic enzymes after several hours darkness (Figs. 2–4). similar results.

A decrease of Fm through the night was also ob- Under irrigation conditions the relationship between served, as well as soon after dawn. As a result the ⌬F/ ⌬F/Fm⬘and light intensity showed low scatter, and both Fm⬘ curve exhibited a sort of “hole” during ca. 1 hour morning and afternoon points fitted the same relation- to 2 hours at the beginning of the day (Figs. 2–4). This ship. When water stress was present, the points corre- phenomenon could be tentatively related to a State1⫺ sponding to morning measurements fitted a curve clearly State2 transition (that is, a disconnection of a part of different from that for afternoon data. For the same light PSII antennas that are transported and coupled to PSI intensity, the values of ⌬F/Fm⬘ measured in the after- reaction centers), as suggested under similar conditions noon were lower than those in the morning, indicating to those encountered here (Bukhov et al., 1996). down-regulation of PS II efficiency after a large period In addition to those effects, the Fo level tended to under excess light. When water stress became more pro- increase through the days of continuous recording on the nounced, this difference increased. All these characteris- same part of the leaf, together with a progressive de- tics contribute to qualify the relationships between ⌬F/

crease of the maximum Fv/Fm achieved during the night Fm⬘ and light intensity as a robust tool for water stress

(not shown). detection.

We have recently studied these phenomena using The diurnal response of Fs to light intensity could several plant species, and we have confirmed that they also be a sensitive indicator of water deficit. The inverse were entirely due to excessive frequency of saturating correlation between Fs and light intensity is a character- pulses (Apostol et al., 1999). In any case, it is important istic signal of water stress, which can be related to a to stress that although these effects lowered the Fv/Fm

strong increase of the nonphotochemical quenching to a value of only 0.6 after 15 days of continuous re-

(Cerovic et al., 1996; Flexas et al., 1998). Although this cording over the same leaf, they did not change the main

behavior of Fs under water stress has been reported ear- photosynthetic responses to water stress, as demon-

lier (Cerovic et al., 1996; Flexas et al., 1999a; Flexas et strated by the similarity of results between the first ex-

al., 1999b; Rosema et al., 1998), here we show (thanks periment and the second one, as well as previous results

to the ability of FIPAM to measure Fs and PPFD from (Flexas et al., 1998; Flexas et al., 1999a).

the same leaf area and every second) that the response of Fs to sudden changes in PPFD takes place in seconds, so the light response of Fs should be an accurate and The Importance of Fluorescence Parameters for

simple signal to detect water stress that can be used even Water Stress Assessment

in cloudy days or with heterogeneous structures such as The present results show that it is not possible to esti-

those of the glasshouse used here.

mate the rate of CO2 assimilation from chlorophyll fluo-

To illustrate the correlation between Fs and PPFD rescence measurements in grapevines, at least in the ab-

even during short periods of light variation (several min- sence of complementary approaches. A similar conclusion

utes), we have depicted such a correlation with data from has been recently pointed out by Rosema et al. (1998).

Figs. 5a and 5c (Fig. 9). We have chosen data from mo- Even in irrigated plants there was an impairment be-

notonous light transitions since the changes at high tween electron transport and CO2assimilation, consistent

PPFD are too rapid and cause high hysteresis. The dif- with previous reports (Flexas et al., 1998; Flexas et al.,

ferent response of well-watered and water-stressed plant 1999a; Lal et al., 1996), which is likely due to an increase

is quite clear. It is shown that under water stress, there in alternative ways for electron consumption, such as

is a saturation of the minimum value of Fs above 400 photorespiration during the afternoon, and not to an in-

lmol m^⫺² s^⫺¹. This may coincide with the saturation of correct determination of PSII ETR, as suggested by Ro-

nonphotochemical quenching. Under irrigation, with sema et al. (1998). In drought plants there was a general-

these particular plants and conditions, the relationship ized lack of relationship between these two parameters.

between Fs and PPFD is poor above 600lmol m^⫺² s^⫺¹, In spite of these results, chlorophyll fluorescence as-

due to the slow development of a high nonphotochemi- sessment can be a very useful tool for stress detection,

cal quenching (not shown in Fig. 9). However, in field- especially with instruments that allow a continuous re-

grown plants, this relationship is clear at much higher cording under natural light conditions, such as the FI-

PPFD values (Flexas et al., 1999a). This technique is es- PAM tested here. Some fluorescence parameters clearly

pecially easy to use with the FIPAM fluorimeter, which reflect plant water status, and we will focus on two of

them: the relationship between⌬F/Fm⬘and light inten- allows the measuring of Fs in a same leaf continuously

(13)

tion and Science (MEC) and Beca de Investigacio´n of UIB for JF. JF wishes to thanks Miguel Mansilla, from the Education Department of Govern Balear, for administrative help during his PSS. We are indebted to Prof. G. Cornic and his research group for the use of their gas-exchange analyser and to Dr. Na- thalie Ollat (INRA-Bourdeaux) for providing the grapevine plants. Language corrections by Dr. E. Descals are gratefully acknowledged.

REFERENCES

Apostol, S. L., Briantais, J.-M., and Moya, I. (1999), Photoinhi- bition of pea leaves by accumulation of over-saturating light pulses in darkness. InProceedings of the XIth International Figure 9. Typical response of Fs to PPFD during a light

Congress on Photosynthesis (G. Garab, Ed.), Kluwer Aca- to shade variation of several minutes duration. Data are

demic Publishers, Dordrecht pp. 2127–2130.

shown for a well-watered plant (triangles, corresponding

Asada, K. (1999), The water-water cycle in chloroplasts: Scav- to data from Fig. 5a) and a water-stressed plant (circles,

corresponding to data from Fig. 5c). enging of active oxygen and dissipation of excess photons.

Annu. Rev. Plant Physiol. Plant Mol. Biol.50:601–639.

Azco´n-Bieto, J. (1983), Inhibition of photosynthesis by carbohy- drates in wheat leaves.Plant Physiol.73:681–686.

Bilger, W., Schreiber, U., and Bock, M. (1995), Determination and at a distance during long periods without interfering

of the quantum efficiency of photosystem II and of non- with leaf physiology.

photochemical quenching of chlorophyll fluorescence in the field.Oecologia102:425–432.

Bolha`r-Nordenkampf, H. R., Long, S. P., Baker, N. R., O¨quist, G., Schreiber, U., and Lechner, E. G. (1989), Chlorophyll CONCLUSIONS

fluorescence as a probe of the photosynthetic competence The present work shows that the new technique dis- of leaves in the field: A review of current instrumentation.

cussed here is a very useful tool for remote sensing of Funct. Ecol.3:497–514.

vegetation stress. Brestic, M., Cornic, G., Fryer, M. J., and Baker, N. R. (1995), The present results confirm and extend previous Does photorespiration protect the photosynthetic apparatus in French bean leaves from photoinhibition during drought work by Cerovic et al. (1996) and Flexas et al. (1999a).

stress?Planta 196:450–457.

The interest of Fs response to light as a putative indica-

Bukhov, N. G., Wiese, C., Neimanis, S., and Heber, U. (1996), tor of water stress is shown again. This idea has also

Control of photosystem II in spinach leaves by continuous been proposed by Rosema et al. (1998).

light and by light pulses given in the dark. Photosynth.

In addition, this work presents novel aspects in re-

Res.50:181–191.

spect to our previous work: Cecchi, G., Mazzinghi, P., Pantani, L., Valentini, R., Tirelli, D., and De Angelis, P. (1994), Remote sensing of chlorophylla 1. The possibility of measuring Fm at distance has

fluorescence of vegetation canopies: 1. Near and far field been shown for the first time thanks to the new

measurement techniques.Remote Sens. Environ.47:18–28.

concept of saturating fluorescence by changing

Cerovic, Z. G., Goulas, Y., Gorbunov, M., Briantais, J.-M., Ca- the frequency of the excitation source.

menen, L., and Moya, I. (1996), Fluorosensing of water 2. Fs and PPFD are obtained from the same leaf

stress in plants: Diurnal changes of the mean lifetime and piece with this new instrument. This has allowed yield of chlorophyll fluorescence, measured simultaneously us to resolve rapid variations (within seconds) of and at distance with a c¸-LIDAR and a modified PAM-fluo- Fs in response to sudden changes of incident rimeter, in maize, sugar beet and Kalanchoe¨.Remote Sens.

light. Such variations are shown to be a simple _Environ.58:311–321.

and rapid way to detect the dominant type of Chaumont, M., Morot-Gaudry, J.-F., and Foyer, C. (1994), quenching regulating leaf photochemistry, which Seasonal and diurnal changes in photosynthesis and carbon partitioning inVitis viniferaleaves in vines with and without is a sensitive indicator of plant stress.

fruit.J. Exp. Bot. 45(278):1235–1243.

Chaves, M. M. (1991), Effects of water deficits on carbon assimilation.J. Exp. Bot.42(234):1–16.

This work has benefited from DG XII—European Community

Cornic, G. (1994), Drought stress and high light effects on leaf TMR-LSF Programme, and is part of LURE Projects BF037-

photosynthesis. InPhotoinhibition of Photosynthesis: From 96, BF028-97, and BF026-98 and CICYT Projects AGF94-0687

Molecular Mechanisms to the Field(N. R. Baker, J. R. Bow- and AGF97-1180 of the Plan Nacional (Spain). Financial sup-

yer, Eds.), Bios Scientific Publishers, Oxford, pp. 297–313.

port from CNRS (GDR 1536) is gratefully acknowledged, as

well as Beca de Colaboracio´n of the Spanish Ministry of Educa- Cornic, G., and Briantais, J. M. (1991), Partitioning of electron

(14)

flow between CO2and O2reduction in a C3leaf (Phaseolus CO2 fixation and carboxylation capacity. Photosynth. Res.

49:57–69.

vulgaris L.) at different CO2 concentrations and during

drought stress.Planta 183:178–184. Lichtenthaler, H. K. (1996), Vegetation stress: An introduction to the stress concept in plants. J. Plant Physiol. 148:4–14.

Cornic, G., Le Gouallec, J.-L., Briantais, J. M., and Hodges,

M. (1989), Effect of dehydration and high light on photo- Lu, C., and Zhang, J. (1998), Effects of water stress on photosynthesis, chlorophyll fluorescence and photoinhibition in synthesis of two C3plants (Phaseolus vulgarisL. andElatos-

tema repens(Lour.) Hall f.). Planta177:84–90. wheat plants. Aust. J. Plant Physiol.25:883–892.

Lu, C., Zhang, J., and Vonshak, A. (1998), Inhibition of the Downton, W. J. S., Grant, W. J. R., and Loveys, B. R. (1987),

Diurnal changes in the photosynthesis of field-grown grape quantum yield of PS II electron transport inSpirulina plat- ensis by osmotic stress may be explained mainly by an in- vines.New Phytol. 105:71–80.

Edwards, G. E., and Baker, N. R. (1993), Can CO2assimilation crease in the proportion of the QB-non-reducing PS II reaction centres.Aust. J. Plant Physiol.25:689–694.

in maize leaves be predicted accurately from chlorophyll

fluorescence analyses?Photosynthesis Res.37:89–102. Moya, I., Goulas, Y., Morales, F., Camenen, L., Guyot, G., and Schmuck, G. (1995), Remote sensing of time-resolved chlo- Flexas, J., Escalona, J. M., and Medrano, H. (1998), Down-reg-

ulation of photosynthesis by drought stress under field rophyll fluorescence and back-scattering of the laser excitation by the vegetation.EARSeL Adv. Remote Sens. 3:188–

conditions in grapevine leaves. Aust. J. Plant Physiol. 25:

893–900. 197.

Moya, I., Guyot, G., and Goulas, Y. (1992), Remotely sensed Flexas, J., Escalona, J. M., and Medrano, H. (1999a), Water

stress induces different levels of photosynthesis and electron blue and red fluorescence emission for monitoring vegetation.ISPRS J. Photogram. Remote Sens. 47:205–231.

transport rate regulations in grapevines. Plant Cell Envi-

ron.22:39–48. O¨quist, G., and Chow, W. S. (1992), On the relationship be-

tween the quantum yield of Photosystem II electron trans- Flexas, J., Escalona, J. M., Cifre, J., and Medrano, H. (1999b),

Water stress effects on variations of steady-state chlorophyll port, as determined by chlorophyll fluorescence and the quantum yield of CO2-dependent O2evolution.Photosynth.

fluorescence (Fs) in response to light. InProceedings of the

XIth International Congress on Photosynthesis (G. Garab, Res.33:51–62.

Osmond, C. B., Maxwell, K., Bjo¨rkman, O., Badger, M., and Ed.), Kluwer Academic Publishers, Dordrecht, pp. 2589–

2592. Leegood, R. (1997), Too many photons: Photorespiration,

photoinhibition and photooxidation. Trends Plant Sci.

Flexas, J., Badger, M., Chow, W. S., Medrano, H., and Os-

mond, C. B. (1999c), Analysis of the relative increase in 4:119–121.

Park, Y.-I., Chow, W. S., Osmond, C. B., and Anderson, J. M.

photosynthetic O2uptake when photosynthesis in grapevine

leaves is inhibited following low night temperatures and/or (1996), Electron transport to oxygen mitigates against the photoinactivation of Photosystem II in vivo. Photosynth.

water stress.Plant Physiol.121:1–10.

Foyer, C., Furbank, R., Harbinson, J., and Horton, P. (1990), Res. 50:23–32.

Rosema, A., Snel, J. F. H., Zahn, H., Buurmeijer, W. F., and The mechanisms contributing to photosynthetic control of

electron transport by carbon assimilation in leaves. Pho- Van Hove, L. W. A. (1998), The relation between laser- induced chlorophyll fluorescence and photosynthesis. Re- tosynth. Res.25:83–100.

Genty, B., Briantais, J. M., and Baker, N. R. (1989), The rela- mote Sens. Environ. 65:143–154.

Schindler, C., and Lichtenthaler, H. K. (1996), Photosynthetic tionship between the quantum yield of photosynthetic elec-

tron transport and quenching of chlorophyll fluorescence. CO2 assimilation, chlorophyll fluorescence and zeaxanthin accumulation in field grown maple trees in the course of a Biochim. Biophys. Acta990:87–92.

Genty, B., Briantais, J.-M., and Vieira da Silva, J. B. (1987), sunny and a cloudy day. J. Plant Physiol. 148:399–412.

Schmuck, G., Moya, I., Pedrini, A., van der Linde, D., Lichten- Effects of drought on primary photosynthetic processes of

cotton leaves.Plant Physiol.83:360–364. thaler, H. K., Stober, F., Schindler, C., and Goulas, Y.

(1992), Chlorophyll fluorescence lifetime determination of Gu¨nther, K. P., Dahn, H.-G., and Ludeker, W. (1994), Remote

sensing vegetation status by laser-induced fluorescence.Re- waterstressed C3- and C4-plants. Rad. Environ. Bioph. 31:

141–151.

mote Sens. Environ.47:10–17.

Harbinson, J., Genty, B., and Baker, N. R. (1989), Relationship Schreiber, U. (1983), Chlorophyll fluorescence yield changes as a tool in plant physiology. I. The measuring system. Pho- between the quantum efficiencies of photosystems I and II

in pea leaves.Plant Physiol.90:1029–1034. tosynth. Res.4:361–373.

Schreiber, U., Bilger, W., and Neubauer, C. (1994), Chloro- Kozaki, A., and Takeba, G. (1996), Photorespiration protects C3

plants from photooxidation.Nature 384:557–560. phyll fluorescence as a nonintrusive indicator for rapid assessment of in vivo photosynthesis. InEcophysiology of Pho- Krall, J. P., and Edwards, G. E. (1992) Relationship between

photosystem II activity and CO2fixation in leaves.Physiol. tosynthesis(E-D. Schulze, M. M. Caldwell, Eds.), Springer- Verlag, Berlin, pp. 49–70.

Plantarum86:180–187.

Krause, G. H., and Weis, E. (1991) Chlorophyll fluorescence Schultz, H. R. (1997), Water relations and photosynthetic responses of two grapevine cultivars of different geographical and photosynthesis: The Basics. Annu. Rev. Plant Physiol.

Plant Mol. Biol.42:313–349. origin during water stress. Acta Hortic.427:251–266.

Takeba, G., and Kozaki, A. (1998), Photorespiration is an es- Lal, A., Ku, M. S. B., and Edwards, G. E. (1996), Analysis of

inhibition of photosynthesis due to water stress in the C3 sential mechanism for the protection of C3plants from photooxidation. In Stress Responses of Photosynthetic Organ- speciesHordeum vulgareandVicia faba: Electron transport,

(15)

isms (K. Satoh, N. Murata, Eds.), Elsevier Science B.V., tosystem II in relation to “energy”-dependent quenching of The Netherlands, pp. 15–36. chlorophyll fluorescence. Biochim. Biophys. Acta 894:198–

Valentini, R., Epron, D., De Angelis, P., Matteucci, G., and 208.

Dreyer, E. (1995), In situ estimation of net CO2 assimila- Wingler, A., Quick, W. P., Bungard, R. A., Bailey, K. J., Lea, tion, photosynthetic electron flow and photorespiration in P. J., and Leegood, R. C. (1999), The role of photorespira- Turkey oak (Quercus cerrisL.) leaves: Diurnal cycles under tion during drought stress: An analysis utilising barley mu- different levels of water supply. Plant Cell Environ. 18: tants with reduced activities of photorespiratory enzymes.

631–640. Plant Cell Environ.22:361–373.

Weis, E., and Berry, J. A. (1987), Quantum efficiency of pho-

Fluorescence Responses to Water Stress in Grapevine Leaves: