Vapor-pressure-deficit-controlled temperature response of photosynthesis in tropical trees

Abstract

Rising temperatures can affect stomatal and nonstomatal control over photosynthesis, through stomatal closure in response to increasing vapor pressure deficit (VPD), and biochemical limitations, respectively. To explore the independent effects of temperature and VPD, we conducted leaf-level temperature-response measurements while controlling VPD on three tropical tree species. Photosynthesis and stomatal conductance consistently decreased with increasing VPD, whereas photosynthesis typically responded weakly to changes in temperature when a stable VPD was maintained during measurements, resulting in wide parabolic temperature-response curves. We have shown that the negative effect of temperature on photosynthesis in tropical forests across ecologically important temperature ranges does not stem from direct warming effects on biochemical processes but from the indirect effect of warming, through changes in VPD. Understanding the acclimation potential of tropical trees to elevated VPD will be critical to anticipate the consequences of global warming for tropical forests.

Highlights

• Vapor pressure deficit limits tropical tree photosynthesis

• Photosynthesis and stomatal conductance decrease as vapor pressure deficit rises

• Warming impacts photosynthesis via vapor pressure deficit in tropical trees

Introduction

Tropical forests are responsible for one-third of the global terrestrial primary production (Beer et al. 2010) but are currently experiencing reduced growth rates and increased mortality due to atmospheric and climate change (Sullivan et al. 2020, Gora and Esquivel-Muelbert 2021). The gross primary productivity of tropical forests results from a delicate balance of large fluxes of CO₂ exchanged between the tropical biosphere and the atmosphere, where both fluxes are strongly affected by environmental factors such as sunlight, temperature, and humidity. Rising temperatures and their effects on tropical forest growth and gross primary productivity have interested tropical plant scientists for decades (e.g., Clark et al. 2003, 2010, 2013; Corlett 2011, Pau et al. 2018, Sullivan et al. 2020). In recent years, vapor pressure deficit (VPD) has emerged as a potentially decisive environmental variable that may significantly affect the physiology of tropical forests (Rowland et al. 2015, Slot and Winter 2017b, Smith et al. 2020). A better understanding of the mechanisms underlying photosynthetic responses to increasing VPD is critical for improving our ability to predict the future of tropical forests.

The saturating vapor pressure of the atmosphere increases exponentially with temperature, so without an increase in atmospheric water vapor content, VPD, the difference between the saturation pressure and actual vapor pressure, increases with increasing temperature. Hence as global temperatures rise, the VPD increases (Barkhordarian et al. 2019). High VPD induces stomatal closure, minimizing water loss but reducing CO₂ uptake and leaf cooling. High VPD associated with climate change has been discussed as a major contributor to recently observed drought-induced plant mortality in several studies (Breshears et al. 2013, Eamus et al. 2013, Stovall et al. 2019, Grossiord et al. 2020, Hammond et al. 2022). Furthermore, by limiting cell turgor pressure, high VPD has been recently implicated in reducing stem growth in tropical forest trees (Peters et al. 2023).

Separating the direct effects of elevated temperatures on photosynthetic CO₂ uptake from indirect effects through changes in VPD is challenging due to the strong correlation between both factors in natural environments (Grossiord et al. 2020, Mills et al. 2024). Comparison of Gross Ecosystem Productivity between low VPD conditions at the Biosphere 2 experimental facility in Arizona with forests in Brazil and Mexico in which VPD increased with temperature suggested that stomatal responses to VPD drove the decrease in stand-level photosynthetic carbon uptake (Smith et al. 2020). Likewise, statistical separation of VPD and temperature effects on photosynthesis in tropical forest canopies showed that stomatal sensitivity to rising VPD lowered the estimated temperature optimum by ~4°C (Slot et al. 2024). These studies suggest that across leaves and plants, those that experience higher temperatures have lower stomatal conductance and lower rates of photosynthesis. What remains unknown, however, is how consistent this pattern is across and within species when individual leaves undergo gradual warming.

To explore the independent short-term effects of VPD and temperature on photosynthetic CO₂ uptake of tropical trees, we experimentally controlled VPD across ranges of temperatures during leaf-level photosynthesis measurements on three tropical tree species in central Panama, using the enhanced humidity control feature of the LI-6800 photosynthesis system (LI-COR, Lincoln, NE, USA). We hypothesized that VPD consistently drives the apparent temperature response of photosynthesis in tropical forest trees.

Materials and methods

Plant material

Three tree species were examined in Panama City, Panama (8.9824°N, 79.5199°W), characterized by a seasonally dry climate with mean annual temperature (MAT) of 25.9°C, mean annual precipitation (MAP) of ~1,900 mm, and a distinct 4-month dry season (Paton 2020): Persea americana Mill., Plumeria rubra L. (white and pink varieties), and Terminalia catappa L. In the following, species will be referred to only by their genus name. The species were selected based on their proximity to one another in the same soil and microclimate, and the accessibility of sun-exposed branches of reproductively mature trees. We prioritized the number of temperature and VPD conditions measured across multiple species, over replication at the species level. Measurements were taken in situ on free-growing (i.e., not potted) trees from March to May 2022.

VPD-controlled temperature response measurements

All measurements were taken with an LI-6800 portable photosynthesis system (Li-Cor, Lincoln, NE, USA) with a CO₂ mixer controlling the incoming (reference) CO₂ concentration at approximately ambient concentration (410 μmol mol^–1). For each species, we first measured 1–3 light-response curves to determine a saturating and non-photoinhibitory level of photosynthetically active radiation (PAR). Based on these light-response curves, we set the PAR level for all three species to 1,500 μmol(quantum) m^–2 s^–1, provided by 90% red and 10% blue LED lights in the leaf gas-exchange chamber.

To determine light-saturated net photosynthesis (P_Nmax) over a range of different temperatures while maintaining a constant VPD, we used the humidifying column of the LI-6800 with water-saturated Stuttgarter Masse, which enabled the maintenance of a low VPD at high temperature. There are, however, limits to the temperature range over which a given VPD can be maintained, due to the risk of condensation at low temperatures, and an insufficient capacity of the humidifying column to maintain low VPD at high temperatures. To extend the humidity control, the airflow rate was reduced from 600 μmol s^–1 to a minimum of 200 μmol s^–1 at high temperatures, when necessary. See Fig. 1S (supplement) for the target VPD levels and associated leaf temperature (T_leaf) ranges, and the data selected for further analysis. At each target VPD level, P_Nmax was measured at a series of temperatures sequentially on the same leaf. At each temperature, we waited for stabilization of stomatal conductance (g_s) and P_Nmax before recording values, typically after 5–15 min, but occasionally exceeding 20 min.

We determined the effect of temperature on the maximum rate of RuBP carboxylation (V_cmax) and maximum rate of RuBP regeneration (J_max) by measuring CO₂ response (P–C_i) curves at ambient (~31°C) and high (~38°C) temperatures. Photosynthesis was first measured at 410 μmol(CO₂) mol^–1, after which values were recorded at 13 additional CO₂ concentrations, first below ambient CO₂ concentrations, after which the curve was completed through step-wise increases above ambient levels to a maximum of 1,800 μmol mol^–1. For each species (variety), two curves were measured at each temperature. V_cmax and J_max were estimated from the relationship between P_Nmax and substomatal CO₂ concentration (C_i) using the fitaci function from the Plantecophys package in R (Duursma 2015). The chloroplast CO₂ concentration is lower than C_i because of mesophyll resistance to CO₂ transport, and as a result, V_cmax derived from P–C_i curves is underestimated (Niinemets et al. 2009). Reliable mesophyll conductance measurements are, however, challenging at best (Leverett and Kromdijk 2024), and doing so in the field was not feasible in this project.

Data analyses

Even with extended humidity control and reduced flow rates, it was not always possible to maintain VPD within a narrow range of the target as temperature increased. We restricted the analyses to cases where VPD was maintained within 0.5 kPa of the target (see Fig. 1S for the VPD–T_leaf relations, and the selected data).

For each target VPD, a temperature-response curve was fitted using Eq. 1 (Gunderson et al. 2010) to estimate T_opt (optimum temperature) and P_opt (photosynthetic rate at the optimum temperature) for each replicate leaf, as:

$$P_{\text{N}}=P_{\text{opt}}-\text{b}\times {\left(T_{\text{leaf}}-T_{\text{opt}}\right)}^2$$ (1)

where b is a constant that is proportional to the width of the curve. The curves were fitted with nonlinear least squares regression analyses, using the ‘nls_multstart’ function in the ‘nls.multstart’ package (Padfield and Matheson 2018) in R (R Development Core Team 2021). To test for stomatal limitation of photosynthesis, we plotted photosynthesis against the C_i/C_a ratio at each target VPD.

We further calculated the stomatal slope parameter g1 by Medlyn et al. (2011) as the slope of g_s vs. $1.6\times P_{\text{N}}/\left(C_{\text{a}}\times \sqrt{\text{VPD}}\right)$, where C_a is the CO₂ concentration of the atmosphere surrounding the leaf. The intercept, g0, was also recorded. g1 represents the carbon cost of water supply and is inversely related to water-use efficiency. As such, VPD sensitivity might scale with g1, and g1 might predict species differences in VPD effects on photosynthesis.

The relative limitations placed on CO₂ uptake by the stomatal diffusion (L_s) and the biochemical capacity (L_b) were estimated from the P–C_i response curves according to the approach of Grassi and Magnani (2005):

$$L_{\text{s}}=\frac{\frac{g_{\text{tc}}}{g_{\text{sc}}}\cdot \partial P_{\text{N}}\text{/}\partial C_{\text{i}}}{g_{\text{tc}}+\partial P_{\text{N}}\text{/}\partial C_{\text{i}}}$$ (2)

$$L_{\text{b}}=\frac{g_{\text{tc}}}{g_{\text{tc}}+\partial P_{\text{N}}\text{/}\partial C_{\text{i}}}$$ (3)

where g_tc is the total conductance to CO₂, g_sc is the stomatal conductance to CO₂, and ∂P_N/∂C_i is the slope of the P–Ci curves calculated over a range of 50–100 ppm CO₂ as suggested by Tomás et al. (2013).

All data were analyzed using R version 4.1.2 (R Development Core Team 2021).

Results

Leaf level measurements with VPD control

When VPD was controlled experimentally in the cuvette of the LI-6800 (i.e., when different levels of VPD were maintained over certain temperature ranges), leaf-level temperature response curves of P_N were typically shallow and P_N only approached zero at very high temperatures when VPD was also high, e.g., at 38°C in Plumeria (white) and 40°C in Terminalia, both at 4.0 kPa VPD (Fig. 1). On average, photosynthesis could be maintained at ≥80% of the maximum rate across a 14°C range from 25.2 to 39.2°C; for measurements <3.0 kPa VPD the range was even wider, going from 23.4 to 39.6°C.

Fig. 1. Leaf-level temperature responses of photosynthesis at different controlled levels of vapor pressure deficit (VPD) for three species, including two varieties of Plumeria rubra. Temperature-response curves are fitted at the leaf level using Eq. 1.

P_opt decreased significantly with increasing VPD for all species, by 2.0, 2.3, 1.3, and 0.7 μmol m^–2 s^–1 kPa^–1 for Persea, Plumeria (pink), Plumeria (white), and Terminalia, respectively (when only including curves for which P_opt could be reliably estimated with P<0.05) (Fig. 2). While the rate of photosynthesis decreased with increasing VPD, there was a tendency for T_opt to increase, by on average 1.8°C per kPa VPD (Fig. 2S, supplement). The increase in T_opt was significant with P<0.05 in Persea.

Fig. 2. Photosynthetic CO₂ uptake at optimum temperature (P_opt) plotted against the vapor pressure deficit (VPD) at which the parameters were determined, for three species, including two varieties of Plumeria rubra, for which leaf-level temperature response curves at different VPD were measured. Solid and dotted lines indicate linear regressions with P<0.05 and P>0.05, respectively.

Photosynthesis correlated positively with the C_i/C_a ratio at almost all VPD levels in Plumeria (white) and Terminalia, without a systematic change in the slope (Fig. 3S, supplement), indicating that decreases in photosynthesis at any given VPD level were associated with increasing stomatal limitations. The strong correlation between P_N and C_i/C_a across all VPD levels with the lowest C_i/C_a ratios was observed at the highest VPD in both species. In contrast, in Persea, there was a small positive relationship at most VPD levels consistent with stomatal limitations, but while photosynthesis decreased with increasing VPD, there was no parallel decrease in C_i/C_a. Finally, in Plumeria (pink), no consistent patterns were found, possibly due to a smaller sample size for this species. The stomatal slope parameter g1, proportional to the marginal water cost of carbon gain, varied considerably among species, ranging from 2.1 kPa^0.5 in Persea to 7.4 kPa^0.5 in Plumeria (white) (Table 1).

Table 1. Biochemical parameters measured at local ambient (31°C) and elevated (38°C) temperatures, and water-use parameters of the species for which photosynthesis was measured with VPD control. Maximum rates of RuBP carboxylation (V_cmax) and RuBP regeneration (J_max) ± SEM (n = 2), the mean temperature during the measurements, their ratio, the relative limitation to P_N by stomatal diffusion (L_s) and biochemical capacity (L_b), and the stomatal slope parameter, g1.

Species	Tleaf [°C]	Vcmax [μmol m–2 s–1]	Jmax [μmol m–2 s–1]	Jmax/Vcmax	Ls [%]	Lb [%]	g1 [kPa0.5]
Persea americana	31.0	69 ± 3	73 ± 6	1.07	53.6	45.7	2.1
	38.0	162 ± 70	45 ± 10	0.38	57.6	41.8
Plumeria rubra	31.0	105 ± 9	118 ± 1	1.13	29.7	68.7	4.9
(pink variety)	38.0	192 ± 8	157 ± 27	0.82	38.0	60.8
Plumeria rubra	31.0	112 ± 4	125 ± 7	1.12	31.2	67.3	7.4
(white variety)	38.0	189 ± 18	148 ± 6	0.78	38.8	60.0
Terminalia catappa	31.0	80 ± 3	106 ± 1	1.32	31.9	66.9	4.2
	38.0	153 ± 7	126 ± 8	0.83	44.0	55.0

Temperature and VPD responses were not consistent among individuals of the same species (Fig. 3), despite growing close to one another, in the same soil, hydraulic, and light conditions. For example, whereas Plumeria (white) tree 1 was able to maintain high P_N and g_s with increasing temperature when VPD was controlled, in tree 2, P_N and g_s decreased with increasing temperature despite maintaining a stable VPD. The VPD response of P_N and g_s was almost completely independent of temperature in tree 1. Only at the very highest temperatures, a moderate reduction in the elevation of the curve (Fig. 3C,G) was observed. In tree 2, the slopes of the VPD responses were independent of temperature, but the elevation of the curves decreased with increasing temperature across the entire temperature range (Fig. 3D,H). These patterns represented the extremes across all trees measured (Fig. 3; Fig. 4S, supplement).

Fig. 3. Temperature and vapor pressure deficit (VPD) responses of photosynthesis (A–D) and stomatal conductance (g_s) (E–H) of two different Plumeria rubra (white) trees.

Biochemical parameters

Between 31 and 38°C, V_cmax increased in all species by an average of 94%, from an average of 91 to 174 μmol m^–2 s^–1, respectively (Table 1). J_max of Persea decreased by 39% between 31 and 38°C, while J_max of the remaining species increased by, on average, 24%, from an average of 116 μmol m^–2 s^–1 at 31°C to 144 μmol m^–2 s^–1 at 38°C. The J_max/V_cmax ratio at 38°C was lower than at 31°C in all species, by on average 40%.

Relative limitations to photosynthesis by CO₂ diffusion

The relative limitation to photosynthesis by stomatal diffusion (L_s) increased in all species by an average of 24% (range: 7.5–37.9%) between 31 and 38°C while the limitation by biochemical capacity (L_b) decreased by an average of 12% (8.5–17.8%; Table 1). At both temperatures, the L_b values were consistently higher than the L_s values for each species except Persea, where the reverse was the case.

Discussion

We have presented experimental support for our hypothesis that the closure of stomata in response to rising vapor pressure deficit (VPD) is the primary driver of the short-term temperature response of photosynthesis of tropical trees. When controlling VPD, photosynthesis exhibited relatively weak responses to temperature, especially below ~35°C, although exact patterns varied among species and individual trees. Across ecologically-relevant temperature ranges the photosynthetic response of sun-exposed tropical forest trees to temperature is thus predominantly governed by the response of stomatal conductance to VPD. Here we showed that temperature manipulation at the leaf level yields this same pattern as observed from pooling measurements across canopy leaves (Slot et al. 2024).

Temperature responses and cumulative heat effects

The extended humidity control of the LI-6800 enabled us to measure photosynthesis across a wide enough temperature range to estimate T_opt and P_opt at each target VPD. The decrease in P_opt with increasing VPD was consistent with our hypothesis, but the increasing T_opt was contrary to expectations (e.g., Kumarathunge et al. 2020, Slot et al. 2024). However, firstly, when studying P_N–temperature relationships under controlled VPD conditions, temperature-response curves were relatively flat in the temperature range around T_opt, so T_opt becomes a somewhat notional concept, with P frequently at ≥80% of P_opt across the entire measurement range. Secondly, the T_opt increase was observed when restricting the analysis to T_opt values that could be estimated with confidence; T_opt values outside the measured temperature range were not included, because their estimates were poorly constrained. With increasing VPD, the measured temperature ranges shifted up. For example, in Persea, photosynthesis was measured between 28 and 34°C when VPD was 1.0 kPa, but between 35 and 44°C when VPD was 4.5 kPa (see Fig. 1S), and the T_opt values within those ranges also increased with increasing VPD (see Fig. 2S). Thus, despite the extended humidity control of the LI-6800, the VPD control approach has its limits, as the risk of condensation and lack of humidity set limits to the minimum and maximum temperatures achievable at a given target VPD, respectively. Even with experimental control of VPD, temperature, and VPD thus still covaried across the full dataset, i.e., our dataset did not include measurements at high temperature and low VPD or low temperature and high VPD (Fig. 1S).

Despite VPD control, increasing temperature frequently decreased photosynthesis, resulting in negative temperature-response slopes at most VPD levels. The decrease in photosynthesis at a given VPD tended to be accompanied by decreasing C_i/C_a, suggesting that photosynthesis was limited by stomatal conductance. The effect of temperature on stomatal conductance independent of VPD has been less explored (Grossiord et al. 2020, Mills et al. 2024), and the mechanism is still a subject of debate (Buckley 2019). Some authors report increasing stomatal conductance with increasing temperature at a constant VPD (Fredeen and Sage 1999, Mott and Peak 2010, Urban et al. 2017, Mills et al. 2024). Sadok et al. (2021) suggest that such an increase in stomatal conductance due to an increase in temperature might be associated with increased hydraulic conductivity as the viscosity of water decreases, and because the transmembrane water movement enabled by aquaporins increases with increasing temperature. In contrast, and in line with some of the temperature responses in the current study, Eamus et al. (2008) reported that at a constant VPD of 2.1 kPa, stomatal conductance of Eucalyptus haemastoma leaves declined as temperature increased from 18 to 38°C. They argue that increased cuticular transpiration at high temperatures causes a reduction in water supply to guard cells, thereby decreasing guard cell turgor and stomatal aperture.

Cuticular conductance varies widely among species (Schuster et al. 2017, Duursma et al. 2019), and so does its temperature sensitivity (Riederer 2006, Bueno et al. 2019, Slot et al. 2021). However, the stomatal response to temperature also varies within species (Fig. 3). Intraspecific variation in cuticle conductance has been linked to growth conditions (Bueno et al. 2020) and the presence of leaf endophytes (Arnold and Engelbrecht 2007), neither of which are expected to vary enough among neighboring trees to cause significant differences in cuticular water loss in the current study. Besides potential genetic differences, which can be substantial within species (Alonso-Blanco et al. 2009), it thus remains unclear what explains the stomatal temperature response at fixed VPD and its variation across and within species.

Differences in temperature responses of photosynthesis and stomatal conductance may also be affected by differences in mesophyll conductance and its temperature response (e.g., von Caemmerer and Evans 2015). There is some evidence for intraspecific variation in mesophyll conductance, but only across a species distribution range (Peguero-Pina et al. 2017), and not between neighboring individuals. To evaluate the relative importance of inter and intraspecific variation in controls over the temperature response of photosynthesis, greater species-level replication is required.

The temperature-response curves of photosynthesis shown in Fig. 1 might be influenced by cumulative heat effects. Under natural conditions, leaf temperatures are highly dynamic (e.g., Vogel 2009, Fauset et al. 2018) and even forest canopy leaves in the tropics are unlikely to experience sustained high temperatures. For example, five-minute averages of leaf temperature monitored at a semi-deciduous forest in Panama rarely exceed 35°C (Rey-Sánchez et al. 2016). In contrast, in our experiments, individual leaves were sequentially exposed to a series of increasing high temperatures, including four or more temperatures at ≥35°C, and measurements were not taken until g_s and P had stabilized at each target temperature. Increased heat exposure duration could increase cuticular water loss, and/or lead to changes in the expression of aquaporins and heat shock proteins (e.g., Araújo et al. 2019). This cumulative heat exposure might have contributed to VPD-independent changes in stomatal conductance and photosynthesis during leaf-level temperature responses. Given the dynamic nature of leaf temperatures within forest canopies, parameters estimated from exposing leaves to a series of increasing high temperatures might need validation to confirm the behavior of canopy leaves.

Persea americana was the only species for which J_max was lower at 38°C than at 31°C, suggesting that reduced electron transport rate capacity may have limited photosynthesis at high temperatures, consistent with a recent model by Scafaro et al. (2023), in which Rubisco activation state and the electron transport capacity were identified as the key drivers of the decrease in photosynthesis above T_opt. In this species, growing alongside the other species in the same soil and microclimate, measurements at high VPD levels were not associated with greater stomatal limitation and a decrease in C_i/C_a. The decrease in photosynthesis with VPD being independent of C_i/C_a might reflect the higher temperature ranges across which high VPD values could be maintained (Fig. 1S), and the associated decrease of electron transport capacity in this species.

Long-term effects of rising temperature and VPD

Our study investigated the effect of VPD on the short-term temperature response of photosynthesis. In response to ongoing climate change, over time, forest trees may exhibit acclimation or adaptation responses to rising temperature and VPD that differ from the short-term response (Berry and Björkman 1980, Hikosaka et al. 2006, Kumarathunge et al. 2019, Crous et al. 2022). The differential response between the long- and short-term has been shown to vary between tropical (see Slot and Winter 2017c) and temperate forests (see Marchin et al. 2016 and Schönbeck et al. 2022). While experiments have been conducted on the acclimation of temperate forest tree species to warming with VPD manipulation (Marchin et al. 2016, Dusenge et al. 2021, Schönbeck et al. 2022), similar experiments are rare in the tropics (Middleby et al. 2024). Growing tropical forest species under a range of temperature and VPD conditions is clearly warranted.

Concluding remarks

Photosynthetic carbon uptake by tropical forests is a critical regulator of the earth's climate, especially in the context of anthropogenic climate change (Malhi et al. 2008). The stabilizing influence of tropical forests is threatened by rising temperatures and associated increases in VPD (Tan et al. 2017, Smith et al. 2020). Model predictions could be fine-tuned with improved mechanistic understanding of the independent roles of temperature and VPD in affecting the photosynthetic CO₂ uptake of tropical trees. It is well established that photosynthesis decreases above a temperature optimum that corresponds roughly to local mean temperatures (Slot and Winter 2017a, Tan et al. 2017, Huang et al. 2019, Kumarathunge et al. 2019), but the processes responsible for this decrease are not yet entirely clear (Slot and Winter 2016, 2017b; Scafaro et al. 2023). We showed here that independent of temperature, VPD adversely affects the photosynthesis of tropical trees. Conversely, photosynthesis exhibited relatively weak responses to temperature when controlling VPD, especially at moderate temperatures.

High VPD can result in water-deficit stress in plants, and lead to decreased vegetation productivity through reduced stomatal conductance and thus photosynthesis (Yuan et al. 2019, Gharun et al. 2020, Grossiord et al. 2020, Schönbeck et al. 2022). It is currently unknown to what extent the photosynthesis of tropical vegetation can acclimate to changes in VPD, so more experimental data on the long-term effects of high VPD and temperature on tropical tree photosynthesis are needed.

Abbreviations

C _a

CO₂ concentration of the atmosphere surrounding the leaf

C _i

intercellular CO₂ concentration

g _s

stomatal conductance

g _sc

stomatal conductance to CO₂

g _tc

total conductance to CO₂

J _max

maximum rate of RuBP regeneration

L _b

relative limitation to P_N by biochemical capacity

L _s

relative limitation to P_N by stomatal diffusion

P _N

net photosynthesis

P _Nmax

light-saturated net photosynthetic rate

P _opt

photosynthetic rate at the optimum temperature

T _leaf

leaf temperature

T _opt

optimum temperature

V _cmax

maximum rate of RuBP carboxylation

VPD

vapor pressure deficit

Conflict of interest

The authors declare that they have no conflict of interest.