Plant photosynthetic overcompensation under nocturnal warming: lack of evidence in subtropical evergreen trees

Abstract

Background and Aims

Increased plant photosynthesis under nocturnal warming is a negative feedback mechanism to overcompensate for night-time carbon loss to mitigate climate warming. This photosynthetic overcompensation effect has been observed in dry deciduous ecosystems but whether it exists in subtropical wet forest trees is unclear.

Methods

Two subtropical evergreen tree species (Schima superba and Castanopsis sclerophylla) were grown in a greenhouse and exposed to ambient and elevated night-time temperature. The occurrence of the photosynthetic overcompensation effect was determined by measuring daytime and night-time leaf gas exchange and non-structural carbohydrate (NSC) concentration.

Key Results

A reduction in leaf photosynthesis for both species and an absence of persistent photosynthetic overcompensation were observed. The photosynthetic overcompensation effect was transient in S. superba due to respiratory acclimation and stomatal limitation. For S. superba, nocturnal warming resulted in insufficient changes in night-time respiration and NSC concentration to stimulate overcompensation and inhibited leaf stomatal conductance by increasing the leaf-to-air vapour pressure deficit.

Conclusions

The results indicate that leaf stomatal conductance is important for the photosynthetic overcompensation effect in different tree species. The photosynthetic overcompensation effect under nocturnal warming may be a transient occurrence rather than a persistent mechanism in subtropical forest ecosystems.

INTRODUCTION

The global mean surface temperature was ~0.99 °C higher in 2001–2020 compared with 1850–1900 (IPCC, 2021). Significant attention has been given to the effects of nocturnal warming on the terrestrial ecosystem and its feedback on climate change (Alward, 1999; Wan et al., 2009; Peng et al., 2013; Anderegg et al., 2015) because temperatures rise faster at night than during the day (Xia et al., 2014; Thorne et al., 2016). Previous studies have demonstrated that nocturnal warming increases the respiratory rate of leaves at night (Atkin and Tjoelker, 2003; Atkin et al., 2005; Jing et al., 2016). According to the sink–source hypothesis of photosynthesis (Paul and Foyer, 2001; Paul et al., 2001; McCormick et al., 2006), the overdepletion of substrates during the previous night results in a compensatory enhancement of photosynthesis the following day (Turnbull et al., 2002, 2004; Wan et al., 2009). Therefore, nocturnal warming may enhance daytime leaf photosynthesis to overcompensate for night-time carbon loss, resulting in increased net carbon uptake on the diurnal scale (Griffin et al., 2002; Turnbull et al., 2002, 2004; Wan et al., 2009). This phenomenon, the photosynthetic overcompensation effect (POE), occurs under nocturnal warming in temperate ecosystems (Wan et al., 2009; Xia et al., 2014). However, whether POE occurs in other regions, especially in the wet tropical and subtropical ecosystems, is unclear. Thus, a mechanistic understanding of POE under nocturnal warming conditions is essential for establishing and predicting vegetation productivity in subtropical forests during future climate change events.

The enhanced leaf respiratory rate regulates the occurrence of POE on warm nights (Xia et al., 2014). However, night-time carbon loss does not always increase under nocturnal warming because of the thermal acclimation of plant respiration (Atkin and Tjoelker, 2003; Smith and Dukes, 2017). The thermal acclimation may occur quickly, even within a few days when plants are exposed to a higher temperature (Slot et al., 2014; Slot and Kitajima, 2015). Also, the acclimation of leaf respiration is reversible during changes in ambient temperature (Bostad et al., 2003). As a product of daytime photosynthesis and the primary substrate of night-time respiration (Göttlicher et al., 2006; Gibon et al., 2009), the content of non-structural carbohydrates (NSC) is an important trait that regulates photosynthesis and links the day–night physiological process (Turnbull et al., 2002, 2004; Wan et al., 2009; Mohammed et al., 2013; Jing et al., 2016). The two main NSC components, soluble sugar and starch, account for NSC fluctuations over time (Hartmann and Trumbore, 2016; Du et al., 2020). Soluble sugar is synthesized during daytime photosynthesis (Chapin et al., 1990), which supports immediate plant metabolism, and it is translocated to various organs at night (Geiger and Servaites, 1994; Geiger et al., 2000). Moreover, starch is produced from sugar that accumulates when the photosynthetic carbon gain exceeds the carbon demand for metabolic activities during the daytime (Chapin et al., 1990; Hartmann et al., 2016). It is then converted to soluble sugar and consumed as the primary carbon source during the night (Gibon et al., 2009; Sulpice et al., 2009). Climate warming stimulates more starch consumption compared with soluble sugar in plants, as revealed in a recent meta-analysis (Du et al., 2020). Thus, starch content is a better substrate depletion indicator under nocturnal warming conditions compared with soluble sugar. Besides participating in metabolism, translocation at night contributes to carbohydrate consumption. Nocturnal warming inhibits the export of carbohydrates from the leaves to other tissues in the plants (Tombesi et al., 2018), resulting in the accumulation of carbohydrates in the leaves. Thus, the occurrence of POE under nocturnal warming depends on the extent of carbohydrate depletion throughout the night.

In addition to carbon metabolism, the use of water is essential for linking daytime and night-time processes. Stomatal conductance is critical to regulating leaf carbon dioxide (CO₂) and water (H₂O) exchange and is sensitive to microclimate changes (Urban et al., 2017; Merilo et al., 2018). The stomata are traditionally assumed to be closed at night because of the absence of photosynthetic carbon gain. However, many studies have shown that stomata are not fully closed during the night, although this increases water loss (Caird et al., 2007; Resco do Dios et al., 2019). Night-time leaf conductance is sensitive to air humidity and is negatively associated with night-time vapour pressure deficit (VPD; Caird et al., 2007; Mott and Peak, 2010; Ogle et al., 2012). The inhibited stomatal conductance may limit respiratory rates under elevated VPD because of the linear relationship between stomatal conductance and night-time respiration (Yu et al., 2019). A tight correlation also exists between daytime and night-time transpiration or stomatal conductance (Christman et al., 2008; Schoppach et al., 2014; Coupel-Ledru et al., 2016). This suggests that leaf photosynthesis may be affected by the day–night linkage of stomatal conductance under nocturnal warming conditions. A previous study indicates that daytime stomatal conductance is increased and reduced under nocturnal warming in the short (hours to days) and long (months to years) terms, respectively (Jing et al., 2016). Therefore, plant stomata closure may downregulate the probability of POE occurrence under nocturnal warming conditions.

A nocturnal warming experiment was conducted in this study to determine whether POE occurs in the seedlings of two subtropical tree species, i.e. Schima superba (Theaceae) and Castanopsis sclerophylla (Fagaceae). The two tree species are common in subtropical evergreen broad-leaved forests (Ni and Song, 1997; Wang et al., 2019; Lu et al., 2021; Cui et al., 2022). Three hypotheses (Fig. 1) were generated based on the literature: (1) POE will not occur without stimulated night-time respiration (R_night); (2) although R_night is stimulated, POE will not occur when leaf photosynthesis is not enhanced because of insufficient substrate depletion or stomatal limitation; and (3) POE occurs when R_night and leaf photosynthesis are simultaneously enhanced. Thus, whether POE under nocturnal warming occurs persistently in the seedlings of two widely distributed evergreen trees from subtropical forests was determined.

Fig. 1.

Conceptual diagram of warming effects on night-time respiration (R_night, 1800–0600 h) and photosynthesis (A_sat, 0600–1800 h).

MATERIALS AND METHODS

Plant materials

Three-year-old S. superba and C. sclerophylla seedlings were obtained from a forestry centre in Zhejiang, China (29°52ʹN, 121°34ʹE), in March 2018. The seedlings were transplanted into 19-L cylindrical pots containing homogenized soil and grown in a naturally lit greenhouse at the East China Normal University, Shanghai, China (31°14ʹN, 121°24ʹE). Only one seedling was placed in each pot. The plants were watered for 5 d to prevent water deficiency. After 2 months, 36 healthy seedlings of similar size for each species were selected for the experiment. At the beginning of the experiment, the plant height and basal stem diameter (mean ± s.e.) were 70.0 ± 1.4 and 8.2 ± 0.2 cm (S. superba) and 70.3 ± 1.6 and 6.9 ± 0.2 cm (C. sclerophylla), respectively.

Experimental design

A complete random block design was used to divide the greenhouse (18 m × 4 m) into 12 blocks of equal spacing, which were assigned to control or nocturnal warming treatments (1800–0600 h the next day, local time). Each treatment for each species was done in triplicate, and each replicate contained six seedlings. Infrared heaters (Langpu Co. Ltd, Guangzhou, China) were suspended 2.2 m above the ground. Heaters without lamps were used in control blocks to avoid shading effects. Soil temperature at a depth of 5 cm was measured and automatically recorded every hour with a data logger (DS1923, Bobang Co. Ltd, Shanghai, China). The average of 24 measurements per day was calculated as the daily mean. Nocturnal warming elevated the average soil temperature by 1.0 °C (P < 0.05, Supplementary Data Fig. S1). The air temperature in the glasshouse was monitored with a greenhouse controller (TD400C, Siemens, Germany). The average air temperature during the experiment was 32.58 and 27.72 °C during the day and night, respectively (Supplementary Data Fig. S2). The experiment was conducted from May 2018 to January 2019.

Plant growth metrics

Plant height and basal stem diameter were measured on 17 May 2018 (before the beginning of the experiment), 23 September 2018 (day 117 of the experiment) and 13 January 2019 (day 221 of the experiment) to determine the growth of the two species at the initial, medium-term and final phases of the experiment, respectively. Moreover, four or five fully expanded upper leaves in each well-grown seedling were sampled after sunset on 5 July 2018 (day 38), 25 August 2018 (day 87) and 12 January 2019 (day 220) and before sunrise on 6 July 2018 (day 39), 26 August 2018 (day 88) and 13 January 2019 (day 221). The leaf area was immediately measured with a leaf area meter (LI-3100, Lincoln, NE, USA). The leaves were then dried at 105 and 70 °C for 1 and 48 h, respectively, to a constant mass. After measuring the dry mass, the oven-dried leaves were ground and passed through a 100-mesh screen for NSC (i.e. soluble sugar and starch) analysis.

Gas exchange and chlorophyll fluorescence measurements

One fully expanded leaf at the upper part of each seedling was selected for gas exchange measurements. According to the diurnal pattern of leaf gas exchange (Supplementary Data Figs S3 and S4), night-time measurements (e.g. night-time respiration, R_night; night-time leaf conductance, g_night) were taken weekly between 2100 and 2400 h, whereas daytime measurements (e.g. net photosynthesis under saturating light, A_sat; daytime stomatal conductance, g_day) were taken weekly between 0900 and 1200 h on the following day. The temperature of each leaf (T_leaf) was measured with a portable photosynthesis system (LI-6800, Lincoln, NE, USA), which was compared with T_leaf measured using an infrared gun (H10, Hikmicro Co. Ltd, Hangzhou, China). The difference was insignificant (Supplementary Data Fig. S5). The leaf-to-air vapour pressure deficit (VPD_leaf) was recorded during each gas exchange measurement. A calibrated T_leaf was used to recalculate VPD_leaf. The calibrated VPD_leaf did not alter the original findings (Supplementary Data Fig. S6). The environmental conditions for the gas exchange measurements in the LI-6800 were air relative humidity (kept at 60 %), reference CO₂ (maintained at 400 p.p.m.), and flow rate (500 µmol s⁻¹) through the leaf chamber. These environmental parameters were similar for daytime and night-time measurements. Light intensity was maintained at 1000 µmol m⁻² s⁻¹ when measuring daytime gas exchange. The dynamics of the environmental conditions and CO₂ and H₂O exchange rates were simultaneously monitored during the measurements. Data were recorded after environmental conditions had been stabilized. To reduce the bias in the measurements, the standard for judging the stability of the gas exchange data was the slope of respiration or photosynthesis versus time (<0.02 µmol m⁻² s⁻¹ min⁻¹) and the slope of stomatal conductance versus time (<0.0025 mol m⁻² s⁻¹ min⁻¹).

Five or six seedlings were selected from each species to measure the short-term temperature response of R_night with LI-6800. Temperature response curves were measured from 22 to 38 °C with an increment of 2 °C by adjusting leaf temperature. The leaf was allowed to equilibrate for 20 min following a temperature change for each measurement. Data were recorded with similar judging standards as regular gas exchange measurements. Q₁₀ represents the temperature sensitivity of respiration, indicating the proportional change in respiration per 10 °C rise in temperature. The observations of temperature and R_night were used to fit the following equation (Slot et al., 2014):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{l}\mathrm{o}{\mathrm{g}}_{10}\left(R_{\mathrm{n}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}\right)=a+b{T}_{\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{f}}}\end{array}}$$

where R_night is the night-time respiratory rate, T_leaf is the leaf temperature to measure R_night, and a and b are coefficients describing the temperature response of R_night. The Q₁₀ value was calculated as:

$${\displaystyle \begin{array}{l}{\displaystyle Q_{10}={10}^{10b}}\end{array}}$$

Chlorophyll fluorescence can reflect the functionality of the photosynthetic system (Chen et al., 2010). For example, the maximum quantum efficiency of photosystem II (F_v/F_m) indicates whether the plants are healthy. The F_v/F_m was measured with an LI-6800 before night-time gas exchange measurements on dark-adapted leaves (2000–2100 h).

NSC analysis

NSC was estimated from 100 mg of ground sample. Soluble sugar was extracted with water, and the water-insoluble residual was used to extract starch using perchloric acid. The soluble sugar and starch concentrations were quantified by the anthrone method using a UV-5500 spectrophotometer (Metash, Shanghai, China) at a wavelength of 620 nm. After sunset, the carbohydrate concentration reflects the accumulation of daytime carbohydrates and the available substrate for night-time respiration. Before sunrise, the carbohydrate concentration exhibits a potential sink regulation on photosynthesis on the following day. Thus, the concentration of NSC and its components were analysed before sunrise. The daily consumptions of NSC and its components were calculated as:

Carbohydrate consumption = NSC_sunset − NSC_sunrise

where $\mathrm{N}\mathrm{S}{\mathrm{C}}_{\mathrm{s}\mathrm{u}\mathrm{n}\mathrm{s}\mathrm{e}\mathrm{t}}$and $\mathrm{N}\mathrm{S}{\mathrm{C}}_{\mathrm{s}\mathrm{u}\mathrm{n}\mathrm{r}\mathrm{i}\mathrm{s}\mathrm{e}}$ are the concentration of NSC and its components sampled after sunset and before sunrise, respectively.

In addition to participating in night-time respiration, NSC is translocated from the source (mainly leaves) to sink organs at night (Geiger and Servaites, 1994; Geiger et al., 2000). Consumption of NSC used in carbon metabolism was calculated based on night-time respiration (Tombesi et al., 2018). Moreover, 1 mol of glucose was assumed to produce 6 mol of CO₂. The respiratory NSC consumption was calculated as follows: The night-time respiratory rate of each seedling (umol m^–2 s^–1) was multiplied by time (12 h), by the average specific leaf area (SLA) of each seedling [m² g^–1 dry matter (DM)] and the molar weight of glucose (180.16 g mol ⁻¹), then dividing 6. . After unit conversion, the respiratory NSC consumption was expressed as mg g^–1 DM. In addition, NSC consumption used in translocation was estimated as the difference between the overall and respiratory NSC consumption.

Phenological analysis

The A_sat time-series was used to fit the logistic function and extract the phenological metrics using the phenopix package in R 3.6 (Filippa et al., 2016). A widely used logistic model was selected to capture the seasonal dynamics (Zhang et al., 2003). Based on local extremes in the first derivative, the ending date of the growing season was extracted. The start of the growing season was not calculated because the experiment was conducted in May when plants had already begun to grow.

Statistical analysis

In this study, intrinsic water use efficiency (WUE_i) for each measurement was calculated as:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{W}\mathrm{U}{\mathrm{E}}_i=\frac{A_{\mathrm{s}\mathrm{a}\mathrm{t}}}{g_{\mathrm{d}\mathrm{a}\mathrm{y}}}}\end{array}}$$

where A_sat is net photosynthesis at a saturated light intensity and g_day is daytime stomatal conductance.

Time-series measurements (i.e. R_night, A_sat, g_night, g_day, WUE_i, F_v/F_m, T_leaf, VPD_leaf, consumption of NSC and its components, the concentration of NSC and its components, plant height, basal stem diameter, leaf area and SLA) were analysed using a repeated-measures mixed-effect model using the nlme package in R 3.6 (R Development Core Team, 2016). Time and warming were considered fixed factors for the analysis. The subject (seedling here) was a random factor. A one-way analysis of variance was conducted to test the warming effects on A_sat, R_night, T_leaf, VPD_leaf, the consumption of NSC and its components, the concentration of NSC and its components, plant height, basal stem diameter, leaf area and SLA for each measurement in R 3.6.

To generate a mechanistic understanding of how g_night is affected by nocturnal warming, a modified Lohammar’s function was used to describe stomatal responses to VPD (Oren et al., 1999):

$${\displaystyle \begin{array}{l}{\displaystyle g_s=-m\times \ln D+b}\end{array}}$$

where g_s is stomal conductance, D is VPD, m is stomatal sensitivity and b is the reference conductance.

The relative contributions of several factors (R_night, ΔNSC, g_night, and g_day) to A_sat changes were quantified. The relative contributions in multiple linear regression models were estimated with a relative importance analysis method using the relaimpo package in R 3.6 (R Development Core Team, 2016) and based on variance decomposition (Grömping, 2007). The metric known as the Lindeman–Merenda–Gold method in the relaimpo package was used to assess the relative importance of these factors in multiple linear regression models (Grömping, 2007).

RESULTS

Effects of nocturnal warming on temperature and VPD

Warming significantly increased night-time T_leaf on most of the measurement dates (P < 0.05) for the two species. Overall, nocturnal warming significantly increased T_leaf by 2.50 °C (P < 0.001) in S. superba (P < 0.001) and by 2.24 °C (P < 0.001) in C. sclerophylla (P < 0.001). Daytime T_leaf was unchanged in the two species during the experimental period (Supplementary Data Fig. S7). Nocturnal warming significantly elevated VPD_leaf at night on most measurement dates (P < 0.05). Overall, night-time VPD_leaf increased by 0.28 kPa in both species (both P < 0.001), whereas daytime VPD_leaf was unaffected by nocturnal warming (Supplementary Data Fig. S8).

Effects of nocturnal warming on plant growth

No difference in plant height or basal stem diameter was noted between temperature treatments (P > 0.05) at the beginning of the experiment. The warming treatment did not alter the plant height or basal stem diameter of either species at the medium (day 117) or final (day 221) experimental phase (P > 0.05, Supplementary Data Fig. S9). The leaf area was the highest at the second measurement (25 August). Nocturnal warming did not affect the leaf area or SLA in either species during the experiment (P > 0.05, Supplementary Data Fig. S10). The end of the growing season was estimated as day 150 of the experiment for both species. However, nocturnal warming did not affect the ending date of the growing season (Supplementary Data Fig. S11).

Leaf gas exchange and controlling factors

For S. superba, nocturnal warming stimulated R_night on days 1, 20, 109 and 125 and A_sat on day 1, whereas it inhibited A_sat on days 144, 165, 193, 207 and 235 (Fig. 2). Overall, R_night and A_sat significantly increased and decreased by 20.3 and 12.6 %, respectively (Fig. 2; Table 1). For C. sclerophylla, nocturnal warming stimulated R_night on day 193 but inhibited A_sat on days 109, 138, 151, 165, 179 and 235. Overall, R_night was unchanged and A_sat was reduced by 10.1 % (Fig. 2; Table 1). A significant difference in the warming effect on R_night was observed between the two species (P < 0.05; Fig. 2).

Fig. 2.

Dynamics of night-time respiration (R_night) and photosynthesis (A_sat) in two species. Time series of R_night and A_sat in S. superba (A, B) and C. sclerophylla (C, D). Values are mean ± s.e. (n = 12–18). The solid curves are Gaussian distributions fitted to the warming effect on R_night (E) and A_sat (F) in the two species (P < 0.01). The warming effect was calculated as [(warming − control)/control]. The asterisks in A–D denote significant difference of treatment effects: *P < 0.05; **P < 0.01; ***P < 0.001.

Table 1.

Results (P-values) of repeated measures linear mixed-effects models for the effects of nocturnal warming (NW), time (T) and their interactions on time-series measurements [i.e. photosynthesis (A_sat), night-time respiration (R_night), night-time leaf conductance (g_night), daytime stomatal conductance (g_day), night-time VPD_leaf, and night-time leaf temperature (T_leaf)]

Treatment	A sat	R night	g night	g day	VPDleaf	T leaf
S. superba
NW	<0.050	<0.001	<0.001	<0.050	<0.001	<0.001
T	<0.001	<0.001	<0.001	<0.010	<0.001	<0.001
NW × T	<0.001	0.181	0.122	<0.050	0.165	<0.001
C. sclerophylla
NW	<0.001	0.162	<0.010	<0.050	<0.001	<0.001
T	<0.010	<0.001	<0.001	<0.050	<0.001	<0.001
NW × T	0.082	0.777	<0.001	0.955	0.070	<0.001

A regression analysis revealed a non-linear relationship between the warming effects on R_night and warming duration in S. superba (Fig. 3). The minimum warming effect occurred when the duration was between 100 and 150 d. The warming effect on A_sat decreased linearly with experimental duration (Fig. 3A). Taking these results together, R_night and A_sat were stimulated within the first 50 d of the experiment. Subsequently, the warming effect on R_night began to decrease, and the effect on A_sat turned out to be negative. Around day 125 of the experiment, R_night was stimulated again, but A_sat was reduced persistently by nocturnal warming. For C. sclerophylla, the warming effects on R_night and A_sat were not affected by the experimental duration (Fig. 3B).

Fig. 3.

Relationship between warming effect on carbon exchange rates and experimental duration in S. superba (A) and C. sclerophylla (B). The purple and green symbols indicate night-time respiration and net photosynthesis, respectively. The warming effect was calculated as [(warming − control)/control].

Nocturnal warming significantly reduced the g_night and g_day of S. superba by 31.6 and 25.3 %, respectively (Table 1; Fig. 4A). For C. sclerophylla, nocturnal warming significantly reduced g_night and g_day by 13.7 and 16.0 %, respectively (Table 1; Fig. 4B). Moreover, g_night exhibited a significant negative relationship with VPD_leaf under nocturnal warming conditions in both species (Fig. 5). A positive linear relationship between warming effects on g_night (Δg_night) and g_day (Δg_day) was observed in both species (Supplementary Data Fig. S12). However, no clear relationship was noted between the changes in g_night (Δg_night) and night-time respiration (ΔR_night; Supplementary Data Fig. S13). A_sat and g_day both decreased, whereas WUE_i was unaffected by nocturnal warming (Supplementary Data Fig. S14). The WUE_i in S. superba was significantly higher compared with that in C. sclerophylla (P < 0.001). The relative importance analysis suggested that g_day contributed >50 % of A_sat for the two species under ambient temperature. In addition, g_night was the most important of all factors in explaining A_sat under nocturnal warming (Fig. 6).

Fig. 4.

Temporal dynamics of the warming effect on night-time leaf conductance (g_night) and daytime stomatal conductance (g_day) in S. superba (A) and C. sclerophylla (B). The warming effect was calculated as [(warming − control)/control].

Fig. 5.

Relationships between night-time leaf conductance (g_night) and night-time VPD_leaf in S. superba (A) and C. sclerophylla (B).

Fig. 6.

Relative importance of different factors to light-saturated photosynthesis (A_sat) based on multiple regression in S. superba (A) and C. sclerophylla (B). Factors include night-time respiration (R_night), NSC consumption (ΔNSC), night-time leaf conductance (g_night) and daytime stomatal conductance (g_day).

Consumption of leaf NSC

The concentration and consumption of NSC and its components were analysed (Supplementary Data Figs S15 and 16; Supplementary Data Table S1). Starch and NSC concentrations for S. superba were significantly decreased by nocturnal warming on day 39 (P < 0.05). However, lower NSC concentration at elevated temperature for C. sclerophylla was observed compared with ambient temperature on day 221 (P < 0.05). Consumption of NSC was significantly affected by nocturnal warming on day 221. For example, soluble sugar and NSC consumption were significantly increased by nocturnal warming in S. superba (P < 0.05), and starch consumption was significantly increased by warming in C. sclerophylla (P < 0.01).

Consumption of NSC at night was separated into consumption from respiration and translocation. NSC was primarily used in night-time respiration when leaf temperature was >28 °C in the experiment of the current study. NSC was translocated when leaf temperature was downregulated. The increased NSC translocation was observed when leaf temperature was increased to 25 °C by nocturnal warming (Fig. 7).

Fig. 7.

Overall NSC consumption and respiratory NSC consumption in the two species. The colour of symbols indicates leaf temperature (indicated in T_leaf scale bar). Plants were grown under ambient temperature or nocturnal warming (symbols indicated in the key). Each symbol represents the mean ± s.e. of n = 3–4.

DISCUSSION

Species-specific occurrence of photosynthetic overcompensation under nocturnal warming

A species-specific occurrence of POE under nocturnal warming was demonstrated in this study. For the two species, a transient POE was observed in S. superba, but not in C. sclerophylla. Such a species-specific occurrence of POE resulted from the different response of R_night to nocturnal warming in the two species (Fig. 1). For example, the temperature sensitivity of R_night was higher in S. superba (Q₁₀ = 2.17) compared with C. sclerophylla (Q₁₀ = 1.72; Supplementary Data Fig. S17). This finding suggests that the POE under nocturnal warming is more likely to occur in thermosensitive plants. For example, Deschampsia antarctica rather than Colobanthus quitensis upregulates photosynthesis under nocturnal warming because the former plant is more thermosensitive (Sanhueza et al., 2019). On a global scale, the temperature sensitivity of plant respiration increases from tropical to arctic regions (Heskel et al., 2016). Thus, one implication of this finding is the probability that the occurrence of POE may be lower in subtropical compared with temperate and boreal regions (Turnbull et al., 2004; Wan et al., 2009; Sanhueza et al., 2019).

Role of NSC in POE occurrence under nocturnal warming

The occurrence of POE under nocturnal warming conditions may be reflected by the warming response of the NSC status before sunrise and NSC consumption at night (Turnbull et al., 2004). The concentration of NSC and its response to nocturnal warming was measured on days 39, 88 and 221 following treatment (Supplementary Data Figs S15 and S16). On day 39, nocturnal warming reduced leaf NSC concentration in S. superba before sunrise (Supplementary Data Fig. S15) but did not affect A_sat (Fig. 2B). For C. sclerophylla, nocturnal warming reduced NSC concentration before sunrise, but not A_sat on the following morning on day 221. These results suggest that the reduction of NSC is not the primary cause of POE under nocturnal warming conditions in this study. However, this result is inconsistent with a previous experiment in a temperate steppe, which reported unchanged NSC concentrations before sunrise with enhanced A_sat under nocturnal warming (Wan et al., 2009).

Nocturnal warming was found to increase the night-time consumption of NSC and its components on a cold night (Supplementary Data Fig. S16). The consumption of NSC results from the enhanced R_night and translocation in the leaves to other plant tissues (Bunce, 2007). Further analysis revealed that NSC consumption resulting from translocation may be significant at low ambient temperatures (Fig. 7). The temperature dependence of translocation, affecting night-time NSC consumption, has also been reported in other plant species, e.g. Vitis vinifera (Tombesi et al., 2018). On a global scale, climate warming can stimulate greater plant NSC consumption in boreal and temperate compared with subtropical biomes (Martínez-Vilalta et al., 2016; Du et al., 2020; Fermaniuk et al., 2021). Thus, NSC consumption may play a more important role in driving POE under nocturnal warming in boreal and temperate ecosystems (Turnbull et al., 2004; Wan et al., 2009; Sanhueza et al., 2019) compared with subtropical forests.

Stomatal limitation on POE occurrence under nocturnal warming

A reduction in plant photosynthesis was observed in both species under nocturnal warming (Fig. 2). The primary cause was the lower g_night and g_day on warmer nights (Fig. 4). The important role of stomatal limitation is supported by the positive linear relationship between warming effects on g_night (Δg_night) and g_day (Δg_day) in the two species (Supplementary Data Fig. S12). Water loss at night is typically 5–30 % of daytime loss (Caird et al., 2007; Howard et al., 2009). Plants close night-time stomata and reduce g_night in response to a higher VPD_leaf (Caird et al., 2007). Plants use water with a similar strategy at night as in the daytime (Christman et al., 2008; Schoppach et al., 2014), so the increased water loss at night downregulates predawn leaf water potential (Sadok and Jagadish, 2020; Lu et al., 2022). In the experiment of the current study, C. sclerophylla is a rather anisohydric species characterized by profligate water use, whereas S. superba, with a more isohydric behaviour, regulates stomata more tightly in response to water stress (Lu et al., 2022). Thus, WUE_i in S. superba was higher compared with that in C. sclerophylla (P < 0.001; Supplementary Data Fig. S14).

Nocturnal warming increased R_night but decreased g_night in S. superba in the present study (Figs 2–4), suggesting that the change in R_night is unrelated to g_night (Supplementary Data Fig. S13). The g_night of plants under the warming treatment was 0.02–0.04 mol m^− 2 s^− 1, which may not be low enough to trigger the inhibitive effect on respiration (Resco de Dios et al., 2019). However, the stomatal responses to increasing evaporative demands in the air and leaves are an essential contributor to plant productivity and growth (Ocheltree et al., 2014; Sanginés de Cárcer et al., 2018). The regulation of photosynthesis by gnight occurs through multiple processes, e.g. nutrient supply, hydraulic redistribution, water use and carbon gain (Caird et al., 2007). Although g_night was inhibited because of high VPD_leaf in the current study (Fig. 5), g_night may be unchanged or increased with higher VPD (Barbour et al., 2005; Dawson et al., 2007; Rogiers and Clarke, 2013). The different stomatal responses may be related to plant functional type (Hoshika et al., 2018), circadian rhythm (Chowdhury et al., 2022), genetic regulation (Resco de Dios et al., 2016) or phylogenetic and biogeographic controls (Yu et al., 2019). The findings of the current study warrant further studies to explore the mechanisms of stomatal regulation of POE occurrence in tree species.

Other traits influencing photosynthesis under nocturnal warming conditions

Several plant traits have been reported to affect leaf photosynthesis under nocturnal warming aside from the day–night linkage between carbon and water processes. Leaf fluorescence parameters, e.g. maximal photochemical efficiency of PSII (i.e. F_v/F_m), mediate the photosynthetic response to temperature changes (Prasad et al., 2008; Djanaguiraman et al., 2013). The damage to PSII typically occurs in herbs or from extreme heat stress (Satoh et al., 1998; Djanaguiraman et al., 2013; Jing et al., 2016). However, F_v/F_m was unchanged in the two species in the present study (Supplementary Data Fig. S18). High night-time T_leaf exerts a positive influence on SLA, leaf area ratio and the number of leaves (Cheesman and Winter, 2013; Jing et al., 2016). Capturing light and increasing photosynthesis is beneficial for plants (Jing et al., 2016; Tang et al., 2016). However, SLA was unchanged in the current study and had no impact on photosynthesis (Supplementary Data Fig. S10).

Conclusions

The current study determined whether POE under nocturnal warming occurs in the seedlings of subtropical evergreen trees. POE in S. superba, which was transient because of respiratory acclimation and stomatal limitation, was detected. The stomatal limitation under nocturnal warming induces a persistent reduction in photosynthesis. This finding highlights the important role of water processes in regulating day–night carbon cycling under non-uniform diurnal climate warming. However, the attention given to night-time ecology is limited (Xia et al., 2014; Gaston, 2019). The lack of a persistent POE occurrence in the experiment of the current study supports a large-scale determination of accelerated respiratory carbon loss in tropical regions under nocturnal warming (Anderegg et al., 2015). Thus, more studies to explore the differential response and connections of plant ecophysiological processes on the diurnal scale during climate warming are recommended.

SUPPLEMENTARY DATA

Supplementary data are available online at https://academic.oup.com/aob and consist of the following. Figure S1: daily soil temperature at 5 cm depth in control and warming groups. Figure S2: temporal dynamics of air temperature during daytime and night in the glasshouse. Figure S3: diurnal patterns of leaf gas exchange rates measured at 3-h intervals in S. superba and C. sclerophylla. Figure S4: temporal dynamics of stomatal conductance in S. superba and C. sclerophylla. Figure S5: comparison of leaf temperature measured with an LI-6800 and an infrared gun. Figure S6: relationships between night-time stomatal conductance and night-time VPD_leaf measured with an LI-6800 or recalculated with calibrated T_leaf. Figure S7: night-time and daytime T_leaf for S. superba and C. sclerophylla. Figure S8: night-time and daytime VPD_leaf for S. superba and C. sclerophylla. Figure S9: plant height and stem diameter of two species in the initial, medium-term and final experimental phases. Figure S10: leaf area and specific leaf area on days 39, 88 and 221. Figure S11: average timing of the end of the growing season in the two species. Figure S12: relationships between change in night-time stomatal conductance and change in daytime stomatal conductance in S. superba and C. sclerophylla. Figure S13: relationships between change in night-time stomatal conductance and change in night-time respiration in S. superba and C. sclerophylla. Figure S14: intrinsic water use efficiency in S. superba and C. sclerophylla. Figure S15: mean concentration of NSC and its components before sunrise in S. superba and C. sclerophylla on three measurement dates. Figure S16: mean consumption of NSC and its components in S. superba and C. sclerophylla on three measurement dates. Figure S17: temperature response of night-time respiration in S. superba and C. sclerophylla. Figure S18: relationships between change in night-time stomatal conductance and change in night-time respiration in S. superba and C. sclerophylla. Table S1: results of repeated-measures linear mixed-effects models for the effects of nocturnal warming, time, and their interactions on consumptions of NSC and its components.

FUNDING

This research was financially supported by the National Natural Science Foundation of China (31722009 and 31800400), the China Postdoctoral Science Foundation (2021M700045) and the Natural Science Foundation of Shanghai (18ZR1412100).

CONFLICT OF INTEREST

The authors declare no competing financial interests.

AUTHOR CONTRIBUTIONS

J.X. designed the study. Y.D., R.L. and H.S. contributed to experimental measurements. Y.D. conducted the analysis and wrote the manuscript. Y.D., E.C., L.Y. and J.X. contributed to the revisions.