Elevated CO2 induces down-regulation of photosynthesis and alleviates the effect of water deficit in Ceiba pentandra (Malvaceae)

ABSTRACT The simultaneous effect of elevated CO2 concentration and drought on trees is still under investigation in the Amazon. We evaluated the effect of CO2 levels (400 and 800 ppm) and water regimes (50% and 100% soil field capacity) on photosynthetic traits, chlorophyll fluorescence, and total biomass accumulation in Ceiba pentandra. In well-watered plants, light-saturated photosynthesis (PN-sat) increased in plants exposed to elevated CO2, but both PN-sat and stomatal conductance decreased in response to water deficit. The maximum carboxylation rate of Rubisco declined under elevated CO2, which indicates down-regulation of photosynthesis at elevated CO2. The Fv/Fm ratio was not affected by treatments. Notwithstanding, total plant biomass and leaf area were reduced by 34-37% under water deficit, but they were not affected by CO2 levels. The PN-sat values measured in well-irrigated plants at ambient CO2 were similar to those observed in plants subjected to elevated CO2 and water deficit (p = 0.26). We concluded that the effect of water deficit on PN-sat was mitigated by elevated CO2. These results suggest that the increase of atmospheric CO2 concentrations associated to climate changes can at least partly offset the negative effect of drought in this multiuse and widely distributed species.


INTRODUCTION
Climate models predict that temperature and atmospheric CO 2 concentration will continue to rise and at a global scale CO 2 concentrations can reach about 700 ppm by 2100.Because of the importance of the Amazon rainforest to global carbon and water cycle, eff orts have been made to improve the understanding of potential eff ects of climatic variables on photosynthesis and growth of Amazonian trees (Marenco et al., 2014;Silveira et al., 2023).
Plant responses to elevated CO 2 have been studied for years, and it has been found that in the short-term (minutes to hours) photosynthesis almost always increases under elevated CO 2 (Delucia et al., 1985;Rogers and Ellsworth, 2002;Krämer et al., 2022).However, in the long-term (days to weeks) down-regulation of photosynthesis to elevated CO 2 can occur (Rogers and Humphries, 2000;Wang and Wang, 2021;Krämer et al., 2022).Down-regulation of photosynthesis has been attributed to acclimation of the photosynthetic apparatus to elevated CO 2 (e.g.reduction in the amount of Rubisco), and it has been defi ned as a decrease in maximum carboxylation rate of Rubisco (Rogers and Ellsworth, 2002;Kitao et al., 2007).Although much research has been done to assess the eff ect of elevated CO 2 on photosynthesis, the simultaneous eff ect of elevated CO 2 and water defi cit is still under investigation in rainforest tropical trees.Indeed, the eff ect of elevated CO 2 on photosynthetic of trees has been evaluated only in a relatively small number of trees of the Amazon region (Oliveira and Marenco, 2019;Ferrer, 2021), which justifi es carrying out further studies on this subject.
The exposure of plants to high [CO 2 ] generally causes two physiological eff ects, a direct impact on photosynthesis and an indirect eff ect closely related to water economy.Subjecting a plant to high CO 2 concentration often leads to an increase in photosynthesis and plant biomass (Delucia et al., 1985;Curtis and Wang, 1998;Kitao et al., 2007;Way et al., 2015).The indirect eff ect of subjecting the plant to high CO 2 improves water use effi ciency, as stomatal conductance may decrease at elevated CO 2 (Medlyn et al., 2001;Gao et al., 2015;Oliveira and Marenco, 2019;Ainsworth and Long, 2021).The increase in water use effi ciency is important because associated with climatic variability the rainfall pattern can change.For instance, it seems that droughts may become more frequent in Northeastern Brazil and part of the Amazon region (Cai et al., 2020).
Droughts cause a reduction in soil water content and an increase in vapor pressure defi cit, which can lead to a decline in photosynthesis and tree growth, and eventually to an increase in mortality of trees under severe drought (Duff y et al., 2015;Marenco and Antezana-Vera, 2021;Camargo and Marenco, 2023).Plant response to water defi cit varies with plant species, severity and duration of the drought stress (Kaur and Asthir, 2017;Fernando and Marenco, 2023).Parameters such as stomatal conductance, photosynthesis and leaf expansion are often modifi ed even by mild water defi cit, while reduced growth is a common response to prolonged droughts (Kaur and Asthir, 2017;Flexas et al., 2012;Oliveira and Marenco, 2019).Mild water defi cit leads to a decline in photosynthesis due to partial stomatal closure, while under severe drought stomatal and non-stomatal limitation lead to a decline in carbon assimilation (Flexas et al., 2012;Wang and Wang, 2021).Moreover, under water defi cit a reduction in photosynthesis occurs in parallel with an increase in the amount of energy not used in photochemical reactions, and hence nonphotochemical quenching (energy dissipation as heat) is an essential mechanism to protect the leaf against high light stress (Ruban, 2016).Therefore, parameters of chlorophyll fl uorescence have been commonly used to assess the performance of a leaf under water defi cit (Rascher et al., 2004;Kitao et al., 2007;Yang et al., 2014).The aim of this study was to evaluate the eff ect of elevated CO 2 concentration and water defi cit on biomass accumulation, photosynthesis and water use effi ciency in Ceiba pentandra.

Plant material and experimental conditions
The experiment was carried out at the Instituto Nacional de Pesquisas da Amazônia -INPA (03° 05′30″ S, 59° 59′35″ W), Manaus, between March and August of 2019.We used a greenhouse and a growth chamber with electronic control of CO 2 concentration and temperature.Ceiba pentandra (L.) Gaertn.(Malvaceae) grows relatively fast in the juvenile stage (Ribeiro et al., 2023).Under fi eld conditions growth rates of 2.25 cm year -1 in diameter and 24 cm year -1 in height have been reported (Keefe et al., 2009;Román-Dañobeytia et al., 2015).We selected C. pentandra not only because of its growth rates and large distribution throughout the Amazon Basin, Central and South America, but mainly due to its socio-economic potential, as besides producing timber and fi bers, it also seems to have medicinal properties (Gómez-Maqueo and Gamboa-deBuen, 2022).
Seeds of C. pentandra were germinated in vermiculite and 15 days after emergence, the plants were transferred to pots (21 cm diameter and 17 cm deep), containing 4.5 kg of forest soil, which was fertilized with 5 g kg -1 , as previously described (Silveira et al., 2023).After transplanting the plants were kept for 90 days in a greenhouse at ambient temperature (~27 °C) and ambient CO 2 concentration under well-watered conditions.Then, the plants (53 cm height and 7.0 mm stem diameter) were randomly assigned into four groups: two CO 2 levels (400 ppm and 800 ppm) and two water regimes within each CO 2 level.The water regimes were: soil at 50% and 100% fi eld capacity, FC.Thus, the four treatments were: T 1 : 400/100, T 2 : 400/50, T 3 : 800/100, and T 4 : 800/50 (i.e.800 ppm CO 2 and soil at 50% FC).The plants grew under these conditions for 138 days (hereafter referred to as the experimental period).The plants at ambient CO 2 (400 ppm) were grown in a greenhouse (four metal-frame sides (4 m tall) supporting an oval roof -6 m tall in the middle).The roof was covered with a double layer of 150-µm polyethylene fi lm.In the greenhouse, mean photosynthetically active radiation (PAR) was 200 µmol m -2 s -1 (i.e.8.6 mol m -2 day -1 ), relative humidity was 70-80%, mean temperature of 27.5 °C (ranging from 26 °C at night to 29 °C at midday), and day/night ambient CO 2 concentration of 400/420ppm).Thus, we used the daytime CO 2 concentration measured in the greenhouse to set the growth chamber at twice this value.The plants to be subjected to 800 ppm CO 2 were transferred to a growth chamber (TPC-1, Winnipeg, Canada).The growth chamber had a working area of 1.72 m 2 , 1.52 m internal height, with automatic control of temperature (HMP60 Vaisala, Vantaa, Finland) and CO 2 (GMM222, Vaisala).

Growth chamber conditions
Following the light conditions of the greenhouse, the PAR intensity in the growth chamber was set to a constant value of 200 µmol m -2 s -1 with a photoperiod of 12 hours (6 am to 6 pm), while the CO 2 concentration was kept constant at 800 ppm (hereafter referred to as eCO 2 ).The day/night temperature was set at 27/25 o C (average of 26 °C), and relative humidity was 80% (daytime) and 90% (nighttime).The irradiance used in this experiment was about 26% of total PAR in the open during 2019 (~33.0 mol m -2 day -1 , computed from INMET´s data; INMET, 2023).

Soil water content and water regimes
Before subjecting the plants to the water regimes, we gravimetrically determined the amount of water (volume) the soil could hold at fi eld capacity -FC (100% FC).Half of this value was added to the soil to be kept at 50% FC.During the experiment, every other day, the pots were weighed (7 am to 8 am) to measure the amount of water consumed by the plant, and then it was replaced to keep the soil water content at its target value (50% or 100% FC); in this calculation, the daily fresh weight gain of the plant was taken into account.Evaporation from the soil surface was prevented by covering the pot with a plastic bag and tying it off at the base of the stem.To estimate the daily contribution of fresh plant growth (PFW) to the pot total mass (i.e.pot, soil, water and fresh plant mass) an allometric equation was used (Silveira et al., 2023): Eq.1 where D is the stem diameter in millimeter.
The plants were subjected to eCO 2 and water defi cit treatments for 138 days (March -August 2019), enough time for the plant to produce new leaves under the treatment conditions.At the end of the experimental period, we measured photosynthetic traits, fl uorescence parameters, SPAD values (a measure of the relative chlorophyll content), leaf nitrogen, leaf total non-structural carbohydrates (TNC), and water potential in leaves produced during the experimental period.

Gas-exchange measurements
Photosynthetic traits were measured using a portable infrared gas analyzer (Li-6400XT, Li-Cor, Lincoln, NE, USA).Gas-exchange measurements were carried out between 8 am and 2 pm on two fully expanded leaves per plant from the upper third.At light saturation (1000 μmol m -2 s -1 ) and [CO 2 ] of treatments (400 ppm CO 2 for ambient conditions and 800 ppm for elevated CO 2 ) we measured light-saturated photosynthesis (P N-sat ), stomatal conductance (g s-sat ), transpiration (E sat ), water use effi ciency (WUE, P N-sat / E sat ), intrinsic WUE (WUE i , P N-sat /g s-sat ), leaf to air vapor pressure diff erence (VPD L ), and dark respiration (R d , measured at a PAR value of zero, after a stabilization period of 3-5 min).While at light and CO 2 saturation (2000 ppm CO 2 ) we measured photosynthesis (P N-max ) and stomatal conductance (g s-max ).Gas-exchange data were measured after a stabilization period of about 10 min at [CO 2 ] of 400 ppm in the leaf chamber (about 240 ppm of internal CO 2 concentration -C i ) and PAR of 250-500 μmol m -2 s -1 .The PAR value at light saturation (1000 μmol m -2 s -1 ) was determined after constructing a light-response curve (500, 250, 100, 75, 50, 25, 12, 0, 500 1000 and 2000 μmol m -2 s -1 ).We also generated a P N /C i curve, by measuring P N at light saturation and varying CO 2 concentration in the leaf chamber [400, 250 100, 0*, 400, 1000, 1500 and 2000 ppm].The 0* corresponds to the [CO 2 ] obtained by turning the mixer off .We used the P N /C i data for computing the maximum carboxylation rate of Rubisco (V cmax , expressed at 25 °C) and the maximum electron transport rate (J max , at 25 °C), after Farquhar et al. (1980).

Fluorescence parameters
Chlorophyll fl uorescence was determined at environmental conditions (~ 400 ppm of CO 2 and 27 °C) with a modulated fl uorometer (PAM-2500, Walz GmbH, Eff eltrich, Germany).These measurements were made on the same leaves used to determine gasexchange.Early in the morning on a dark-adapted leaf (30 min dark-adaptation), the minimum (F 0 ) and maximum fl uorescence (F m ) were measured.The F m was obtained by applying a light pulse (6,000 μmol m -2 s -1 for 1.0 s) and the F v /F m ratio (maximal quantum yield of photosystem II) computed as described by Rascher et al. ( 2004): At midday and under actinic light (230 μmol m -2 s -1 ) we also measured the eff ective quantum yield of photosystem II, Φ PSII (Rascher et al., 2004): where F s and F m ´ denote the steady-state and maximal fl uorescence of the light-adapted state, respectively.While the non-photochemical quenching (NPQ) was computed as (Rascher et al., 2004): Eq. 4

Leaf nitrogen, SPAD, sugars, starch and water content
The amount of starch and sugars in a fresh leaf sample (~20 mg) were determined as described by Dubois et al. (1956).Then, total non-structural carbohydrate content was obtained by summing up the sugar and starch contents.Leaf nitrogen (N L ) was determined by the classical Kjeldahl method (Miyazawa et al., 2009), while the SPAD value was measured (eight readings per leaf) with a chlorophyll meter (SPAD-502, Minolta, Osaka, Japan) in the same leaves used for measuring gas-exchange.

Plant biomass, leaf area, LAM and leaf water potential
At the end of the experimental period, the total plant dry mass was recorded after oven drying at 72 °C until constant mass.The leaf area was measured (Li-3000, Li-Cor, Lincoln, NE, USA), and then the leaf area per mass (LAM, sometimes called specifi c leaf area) obtained.Leaf water potential (Ψ L ) was determined early in the morning (~6 am) and at midday (~12 noon) in one leaf per plant with a pressure chamber (1505 D, PMS Instrument Company, Albany, USA).In addition, the relative leaf water content (RWC) in two leaves per plant was also measured: Eq.5 where, LFM, LTM, and LDM represent the leaf fresh mass, leaf turgid mass and leaf dry mass, respectively, being the dry mass measured as previously described.The turgid mass was obtained after cutting the tip of the petiole under water, and then the leaf was covered with a plastic bag and placed in a dark room for 12 hours (overnight) for full hydration.

Experimental design and statistical analysis
The experimental design was a split-plot, with the CO 2 concentration (400 and 800 ppm) as the main plot and the water regime (50% and 100% FC) as the split-plot, with eight replications.Prior to statistical analyzes data were tested for normality (Shapiro-Wilks, α = 0.05), and log-transformed [log (x+1)] when necessary.Then the data were submitted to ANOVA and the means compared by the Fisher-LSD test.We used p = 0.05 to defi ne statistical signifi cance.Statistical analyzes were performed using Statistica version 7.0 (StatSoft, Inc.Tulsa OK, USA).

Gas-exchange and chlorophyll fl uorescence
Over water regimes, light-saturated photosynthesis (P N-sat ) increased by 38% at eCO 2 (mean of 6.18 μmol m -2 s -1 at ambient CO 2 versus 8.90 μmol m -2 s -1 at eCO 2 ), and over CO 2 levels it decreased by 26% under water defi cit (Fig. 1A).The eff ect of eCO 2 on g s-sat was non-signifi cant over water regimes (p = 0.246, Fig. 1B), but irrespective of the CO 2 level, g s-sat decreased by 50% under water defi cit.That is, under well-watered conditions g s-sat was as high at ambient CO 2 as at eCO 2 (0.099 and 0.106 mol m -2 s -1 , respectively; i.e., a g s-sat ratio (eCO 2 to ambient CO 2 ) of 1.07).Leaf transpiration (E sat ) decreased under water defi cit, following an increase in VPD L under limited water supply (Table 1).
Despite the decline of stomatal conductance (g s-sat ) in plants subjected to water defi cit, the P N-sat values measured in well-irrigated plants grown at ambient CO 2 were similar to those found in plants subjected to elevated CO 2 and water defi cit (7.12 and 7.64 µmol m -2 s -1 , p = 0.26), being the lowest P N-sat values observed in plants grown at ambient CO 2 and subjected to water defi cit (Fig. 1A).
Under light and CO 2 saturation, photosynthetic rates (P N-max ) were greatly reduced in plants subjected water defi cit, which ultimately led to a signifi cant eff ect of eCO 2 and water defi cit (p ≤ 0.015, Fig. 1C).It should be noted, however, that there was no diff erence between the P N-max values recorded in well irrigated plants and those recorded at ambient CO 2 , irrespective of water regime; for instance, there was no diff erence between plants grown at ambient CO 2 and limited water supply and those well-watered and exposed to eCO 2 (p = 0.35, Fig. 1C).The stomatal conductance measured at CO 2 saturation (g s-max ) was neutral to the eff ect of eCO 2 (p = 0.35), but it decreased under water defi cit (p < 0.001, Fig. 1D).Regarding the loss of carbon by respiration, both eCO 2 and water defi cit had no signifi cant eff ect on leaf respiration (Table 1).
Over water regimes, the V cmax values were lower in plants exposed to eCO 2 than in those grown at ambient CO 2 (22.3 versus 30.6 μmol m -2 s -1 , p < 0.01, Fig. 2A).Because the V cmax values recorded at ambient CO 2 were very close (~30.5 μmol m -2 s -1 , p = 0.98), on average the eff ect of water stress over CO 2 levels did not reach the level of signifi cance (p = 0.059, Fig. 2A).However, it is worth noting, that V cmax signifi cantly diminished at eCO 2 under water defi cit (p = 0.012, Fig. 2A).The J max -likewise P N-max -only decreased in plants subjected to eCO 2 and water defi cit (p < 0.001), and as a result, the eff ects of both eCO 2 and water defi cit on J max were signifi cant (Fig. 2B).Thus, it can be noted, that there was no diff erence between J max recorded at ambient CO 2 and the J max measured at eCO 2 in well-irrigated plants (Fig. 2B).
On average, water use effi ciency (WUE) was 40% higher in plants subjected to eCO 2 than in those grown at ambient CO 2 (9.10 mmol mol -1 versus 6.50 mmol mol -1 ).Although WUE tended to increase by 12% under water limitation, the eff ect of water defi cit on WUE was not signifi cant (Table 1).On average, WUE i (P N-sat /g s-sat ) was responsive to both eCO 2 and water regimes (Table 1), while the leaf water potential (Ψ L ) was lower under water defi cits than in wellirrigated plants, and lower at midday than early in the morning (Table 1).Likewise, the relative water content (RWC) of leaves varied from 85.1% in plants under water defi cit to 88.7% in well-irrigated plants (p ≤ 0.05, Table 1).
Regarding the eff ect of treatments on leaf photochemistry, both eCO 2 and water defi cit had a neutral eff ect on the F v /F m ratio (Table 1).On the other hand, the Φ PSII decreased under water defi cit (0.46 versus 0.53, p ≤ 0.05, Table 1), but the eff ect of eCO 2 was not signifi cant (p > 0.05).Whereas, NPQ decreased at eCO 2 and rose under water defi cit, particularly in plants grown at ambient CO 2 conditions (p ≤ 0.05, Table 1).

Nitrogen, TNC, and biomass accumulation
Sugar and starch content increased at eCO 2 , and hence the total non-structural carbohydrates (TNC) content also augmented (8.1 g m -2 at ambient CO 2 against 10.8 g m -2 at eCO 2 , Table 1).Water defi cit, however, had no eff ect on sugar or starch content.Likewise, the leaf area per mass (LAM) was only aff ected by eCO 2 .While, neither eCO 2 nor water defi cit had a signifi cant eff ect on leaf nitrogen (Table 1).The total plant dry matter (W T ) was reduced 34.5% under water defi cit (31.3 versus 20.5 g per plant), and similarly the total leaf area per plant declined 37.5% Figure 2 -Maximum carboxylation rate of Rubisco (V cmax , A) and maximum electron transport rate (J max , B) in Ceiba pentandra subjected to two CO 2 levels (400 and 800 ppm) and two water regimes -water (soil at 50% and 100% FC).Further information as shown in Fig. 1.Figura 2 -Taxa máxima de carboxilação da Rubisco (V cmax , A) e taxa máxima de transporte de elétrons (J max , B) em Ceiba pentandra submetida a dois níveis de CO 2 (400 e 800 ppm) e dois regimes de hídricos -water (solo a 50% e 100% FC ).Mais informações conforme mostrado na Fig. 1.
under limited water supply, but the eff ect of eCO 2 on W T and total leaf area was not signifi cant (Table 1).

DISCUSSION
We found that P N-sat increased at eCO 2 in well irrigated plants, but irrespective of the CO 2 level, it decreased under water defi cit.Moreover, the P N-sat values were as high in well-watered plants at ambient CO 2 as in those at eCO 2 under water stress.Stomatal conductance (g s-sat ) was unresponsive to eCO 2 , but it dropped under water defi cit.The V cmax values decreased at eCO 2 , but the eff ect of water defi cit was only signifi cant at eCO 2 , while the F v /F m ratio and the eff ective quantum yield (Φ PSII ) were unaff ected by the CO 2 treatments.Both plant dry matter and leaf area were unresponsive to eCO 2 , but plant biomass and leaf area diminished under water defi cit.
The increase in photosynthesis (P N-sat ) at eCO 2 is consistent with classic response of an increase in photosynthetic rates in response to exposure to eCO 2 (Rogers and Humphries, 2000;Gao et al., 2015;Oliveira and Marenco, 2019;Ainsworth and Long, 2021).Furthermore, subjecting the plants to eCO 2 mitigated the negative eff ect water defi cit on photosynthesis, as P N-sat values of well-irrigated plants at ambient CO 2 were comparable with those measured at eCO 2 under water defi cit, even when stomatal conductance dropped under limited water supply (Fig 1A), which is consistent with the results reported by Oliveira and Marenco (2019) and Wang and Wang (2021).This is important and suggests that the steady increase of atmospheric CO 2 concentrations may at least partially alleviate the negative eff ect of drought on carbon assimilation.
Stomatal conductance (g s-sat ) did not decrease at eCO 2 under well-watered conditions, which is contrary to what has been reported by others (Kitao et al., 2007;Wang and Wang, 2021;Ainsworth and Long, 2021).However, it has also been found that eCO 2 may have a neutral eff ect on stomatal conductance (Curtis and Wang, 1998).Not to mention that Santrucek and Sage (1996) reported that eCO 2 may attenuate the response of stomatal conductance (g s ) to CO 2 .The eCO 2 to ambient CO 2 g s ratio reported in this study (1.07) is consistent with results reported by Medlyn et al. (2001) for experiments with eCO 2 exposure time of less than one year, as they reported a g s ratio (eCO 2 /ambient CO 2 ) of 0.95 (95% confi dence interval of 0.83-1.09).The decline of g s-sat under water defi cit concurs with the classic response of a drop in stomatal conductance due to water limitation (Kaur and Asthir, 2017;Oliveira and Marenco, 2019;Fernando and Marenco, 2023).The reduction of P N-max at eCO 2 under water defi cit (Fig. 1C) may refl ect both a drop in stomatal conductance and a decrease in V cmax .Thus, the decline in P N-max may indicate the negative eff ect of stomatal and non-stomatal limitation.This is because a decrease in mesophyll conductance reduces carboxylation effi ciency.Perez-Martin et al. (2014) reported that in Olea europaea the decrease in photosynthesis under severe drought was related to a drop in mesophyll conductance, while Flexas et al. (2012) reported that Rubisco activity can be negatively aff ected by water defi cit, which can explain the decline in V cmax under limited water supply and eCO 2 .Altogether, the decline of J max and V cmax at eCO 2 under water defi cit indicates that non-stomatal factors can aff ect photosynthetic rates, as mesophyll conductance may decrease with increasing intercellular CO 2 concentration (Kitao et al., 2015).Although P N-sat increased at eCO 2 (under well-watered conditions), there was a decline in V cmax at eCO 2 (p < 0.001), which led us to concluded that the exposure to high CO 2 concentration caused downregulation of photosynthesis -the decline in V cmax (Rogers and Humphries, 2000;Rogers and Ellsworth, 2002;Kitao et al., 2007).Although, a down-regulation of photosynthesis is not an uncommon response to eCO 2 (Rogers and Humphries, 2000;Wang and Wang, 2021;Krämer et al., 2022), no acclimation of photosynthesis to eCO 2 has also been reported (Curtis and Wang, 1998).
Water use effi ciency (WUE) was improved at eCO 2 , which can be explained by the positive eff ect of eCO 2 on P N-sat (Way et al., 2015;Gao et al., 2015).WUE also tended to increase under water defi cit, but the eff ect of water scarcity did not reach the signifi cance level (p > 0.05).This occurred because the relative decline of P N-sat under water defi cit (25%, Fig. 1A) was lower than the drop in leaf transpiration (35%, Table 1).The increase in WUE i (P N-sat /g s-sat ) under water defi cit occurred because the relative drop in stomatal conductance was larger than the relative decrease in photosynthetic rates, which was also reported by Oliveira and Marenco (2019) and Wang and Wang (2021).An increase in WUE i under eCO 2 may contribute to improve plant tolerance to water defi cit.This is important because climate changes can lead to prolonged droughts in parts of the Amazon region (Cai et al., 2020).
The lack of an eff ect of treatments on the F v /F m ratio indicates that both eCO 2 and water defi cit did not alter leaf photochemistry, which was also observed in other species subjected to water defi cit, and indicates absence of pre-dawn photoinhibition (Rascher et al., 2004;Kitao et al., 2007;Oliveira and Marenco, 2019).A decline in Φ PSII under water defi cit (and hence an increase in NPQ) can be associated with a reduction in stomatal conductance and photosynthesis, as there is a direct relationship between stomatal conductance and Φ PSII (Jiang et al., 2006;Ahanger et al., 2020;Vanitha et al., 2022).In addition, water defi cit can eventually aff ect the stability of membranes in chloroplasts.In rice, severe water defi cit leads to an increase in superoxide radicals and membrane peroxidation, which can cause a reduction in Φ PSII and F v /F m values (Yang et al., 2014).However, in this experiment, it seems unlikely that membranes were damaged, as there was no change in the F v /F m ratio.Instead, the decline in Φ PSII under water defi cit can be attributed to a decline in stomatal conductance, as mentioned above.Compared with well-watered plants, the increase of NPQ under water defi cit shows that a larger amount of energy can be dissipated as heat when less energy is used in photochemical reactions (Ruban, 2016).
Starch has been reported to increase under CO 2enriched atmosphere (Kitao et al., 2007;Wang and Wang, 2021).The high TNC found at eCO 2 may have contributed to decrease LAM (i.e. to increase leaf mass per area).We found only a slight decrease in RWC, which is consistent with the small decline in leaf water potential (Ψ L ) under water defi cit.Taking well-irrigated plants as the baseline, at noon Ψ L declined only 23% under water defi cit (Table 1).This is in line with the results of Rascher et al. (2004), who observed that C. pentandra maintains relatively high leaf water potential during drought.A slight decline in Ψ L in plants subjected to water defi cit (Table 1) may indicate that C. pentandra has a tight control of leaf transpiration by reducing stomatal conductance under water defi cit.
Even though P N-sat increased at eCO 2 (p < 0.001), we found no eff ect of eCO 2 on W T (total plant biomass).This can occur because plant growth is a complex process that refl ects the balance between carbon gain by photosynthesis and total carbon loss, not only via plant respiration, but also due loss of carbon by root exudation, which can be higher under elevated CO 2 (Phillips et al., 2011).Also a high sensitivity of C. pentandra to ethylene may also contribute to the absence of a positive eff ect of eCO 2 on W T .This is because, it has been reported that an increase in CO 2 concentration may promote ethylene biosynthesis (Seneweera et al., 2003), and high ethylene concentrations may reduce growth (Dai et al., 2023).

CONCLUSIONS
We found that light-saturated photosynthesis increased at elevated CO 2 , but it decreased under water defi cit.Notwithstanding, plant biomass was neutral to the eff ect of elevated CO 2 .The elevated CO 2 improves water use effi ciency and alleviates the eff ect of water limitation.We reach this conclusion because well-irrigated plants at ambient CO 2 and plants under elevated CO 2 and water defi cit had similar P N-sat values.The V cmax decreased at elevated CO 2 , which suggests acclimatization of photosynthesis to elevated CO 2 .These results improve our understanding of the concomitant eff ect of elevated CO 2 and water defi cit on the ecophysiology of Ceiba pentandra, an important multipurpose tree.