Mitigation of Elevated CO₂ Concentration on Warming-Induced Changes in Wheat Is Limited under Extreme Temperature during the Grain Filling Period

Abstract

Elevated CO₂ concentration (eCO₂) generally increases plant growth by improving photosynthesis, but it is unclear whether eCO₂ can alleviate the negative effects of elevated temperatures, especially in high-temperature years. Manipulative experiments with elevated [CO₂] and temperature were conducted in North China to understand the effect of elevated CO₂ concentration and temperature on wheat. The photosynthesis, A_n–PAR and A–C_i curve parameters, growth period, biomass, yield component, and yield of wheat were investigated under different [CO₂] (around 400 and 600 ppm) and temperatures (ambient temperature and ambient temperature +2 °C) for 3 years by using controlled chambers. Results showed that elevated temperature significantly shortened the growth period and decreased the yield and biomass of wheat. Elevated [CO₂] significantly increased the maximum net photosynthetic rate (A_nmax) but reduced the maximum carboxylation rate (V_cmax) and the maximum electron transport rate (J_max). The extremely high temperature during the grain filling period in 2019 exerted a serious negative impact on wheat production. Elevated [CO₂] stimulated photosynthesis, increased kernel number per spike, and extended the duration of the grain filling period, which consequently increased biomass and grain yield under elevated temperatures in normal years (2018 and 2020). Although the combination of CO₂ and temperature reduced photosynthesis and biomass, it also alleviated the negative impact of elevated temperatures on grain yield to some extent under extremely high temperature during the grain filling period in 2019. The mitigative effect of eCO₂ under extreme high temperature is limited, and planting early-maturing cultivars or increasing the genotypes of kernel number per spike help to escape the extreme high temperature of the critical growth period.

1. Introduction

Wheat (Triticum aestivum L.) is an important food crop worldwide. North China is the main producing area of winter wheat, and a change in its yield can significantly impact the food security of China and the world. The grain production of wheat is considerably affected by climate change. Since the industrial revolution, atmospheric [CO₂] has increased by more than 40% and reached 412 ppm in 2020 [1]. Elevated CO₂ concentration (eCO₂) increased the weight of the shoot, root and spike by enhancing photosynthesis, increasing leaf appearance rates, and accelerating tiller development after appearance [2]. Elevated [CO₂] increases plant growth and yield by improving photosynthesis [3,4,5,6]. Alonso et al. [7] reveal that the effects of eCO₂ on photosynthetic electron transport and RuBisCO kinetics may improve the photosynthetic response of wheat to global warming. A meta-analysis reports that wheat yield is estimated to increase by 25% on average under eCO₂ (average ambient concentration of 372 ppm and elevated 605 ppm), but wheat yield is not increased with eCO₂ exceeding the critical threshold (600 ppm) [8]. Elevated [CO₂] stimulates biomass (+35%) and grain yield (+30%) in wheat [9]. Elevated [CO₂] increased the yield of 7 cultivars by an average of 38% due to increases in both ear number and aboveground biomass in wheat [10].

The global mean temperature is expected to increase by approximately 2 °C by 2041 to 2060 under the very high greenhouse gas emissions scenario (SSP5–8.5) [11]. Elevated temperature facilitates crop ontogenetic development, shortens the growth stage, changes seed formation, and reduces yield, especially in tropical regions [12,13,14]. Temperatures exceeding the critical threshold impair photosynthetic activity, disrupt plasma membrane integrity, cause oxidative damage, and denature proteins, thus reducing the quantity and quality of crops [15,16,17,18,19]. Porter and Gawith [20] suggest a minimum temperature of 9.2 °C, an optimal temperature of 20.7 °C, and a maximum temperature of 35.4 °C for grain filling in wheat. An increase in mean temperature over the range of 12–26 °C during grain filling decreases grain weight at a rate of 4–8%/°C [21,22]. Kuchel et al. [23] and Bennett et al. [24] identified a reduction of up to 187 kg ha⁻¹ for each 1 °C increase in average temperature during anthesis and grain filling through multi-year field experiments on wheat in Australia. Asseng et al. [12] reveal that a 0.6 °C increase in the temperature decreases wheat yield by approximately 3.6%. Several studies argue that temperatures below (<10 °C) or above (>25 °C) the optimum (12–25 °C) alter the phenology, growth, and development of wheat varieties and finally decrease the yield [25,26,27]. Hasan [28] suggests that each 1 °C rise in the average mean air temperature from anthesis to maturity reduces grain yields by approximately 2.6–7.2% in wheat varieties compared with those under optimal growth conditions (<26 °C). High temperature also accelerate leaf senescence, which reduces plant photosynthetic capacity and wheat yield [29]. Temperatures above 30 °C during floret formation cause complete sterility, which reduces kernel weight [30]. Telfer et al. [31] also demonstrate that high temperature exert significant negative impacts on grain yield, with a reduction of 161 kg ha⁻¹ °C⁻¹ for each day with a maximum temperature above 30 °C during grain fill.

Few studies have investigated the combined effects of elevated [CO₂] and temperature on crop growth and yield. Some manipulative experiments have examined the responses of crop growth and yield to the combination of elevated [CO₂] and temperature by chamber, open-top chamber, or free-air enrichment CO₂ [14,32,33,34]. Elevated [CO₂] (560 ppm) and elevated temperature (average temperature increase by 1.7 °C) exert no impact on the yield of irrigated winter wheat in North China [14]. The yields of wheat and rice are decreased by 10–12% and 17–35%, respectively, under the effects of eCO₂ (500 ppm) and elevated temperature (average temperature increase by 1.5–2 °C) in South China [32]. Elevated [CO₂] (600 ppm) compensates for the negative impact of low-intensity heat waves (+5.0 °C) on soybean yield but not for that of high-intensity heat waves (+9.0 °C) [34]. These studies show that infrared radiator heaters can increase temperature, but they only increase canopy temperature. The increase in temperature of other non-canopy sites and soils is not the same as the increase in the canopy temperature. Thus, these results do not reflect the real effect of elevated temperature and elevated [CO₂] on crops.

Whether or not eCO₂ can compensate for the negative impact of elevated temperature on wheat needs to be further tested, especially in hot years. We hypothesized that (1) high temperature would have a negative effect on photosynthesis, the growth period, and yield in wheat, and (2) the benefits of elevated [CO₂] on wheat growth would be weak under high temperature. This study determined the effects of eCO₂ on photosynthesis, growth and development, and yield components under elevated temperature in different years.

2. Materials and Methods

2.1. Experimental Design

This experiment was carried out in a [CO₂]-controlled chamber located in the Shanxi Agricultural University (37.42° N, 112.55° E), Taigu County, Jinzhong City, Shanxi Province, China. The rectangular structure of the chamber was fixed in the field and fabricated using an aluminum frame. The control chambers had four independent chambers used to control [CO₂] and temperature. The length, width, and height of each chamber were 8.0, 4.0, and 3.0 m, respectively. The frame was lined with tempered glass (6.0 mm thick), which transmitted 80–90% of natural sunlight. There were four chambers: CK (ambient [CO₂]: [CO₂] was approximately 400 ppm; ambient temperature: temperature was approximately the ambient temperature outside the chamber); ET (ambient [CO₂]; elevated temperature: 2 °C above ambient temperature); EC (elevated [CO₂]: 200 ppm above ambient [CO₂] (approximately 600 ppm); ambient temperature); and ECT (elevated [CO₂]; elevated temperature) (Figure 1 and Figure 2). The air flow was about 28.0 m³ min⁻¹ inside the chamber through a fan device. The temperature and [CO₂] were controlled by an automatic control system (Automatic control system of CO₂ concentration and temperature, SY-QHS, Shengyan Electronic Science and Technology CO., LTD., Handan, China). An air conditioner (4000 W) was installed in each chamber to control the temperature, and [CO₂] sensors (Vaisala, Finland) and solenoid valves were used to control [CO₂] in each chamber. The temperature and [CO₂] were maintained from crop emergence to harvest by automatic [CO₂] and temperature control systems, and CO₂ gas was not released during the overwintering period. The relative humidity and VPD of all the chambers were measured throughout the experimental period, and their humidity fluctuated at the same level in the three growing seasons (Supplementary Materials, Figure S1, Table S1). The average temperature of the ET and ECT chambers was 1.8–2.1 °C higher than that of the CK and EC chambers during the wheat growing season in 2017–2018, 2018–2019, and 2019–2020 (Figure 2). The mean temperature of the growing season in 2019–2020 was higher than that of the two other growing seasons. The lowest temperature of the growing season in 2018–2019 was lower than those of the two other growing seasons, whereas the highest temperature was higher than those of the growing seasons in 2017–2018 and 2019–2020. The mean temperatures of the lowest temperatures were –11.7 °C, –14.5 °C, and –10.8 °C in 2017–2018, 2018–2019, and 2019–2020, respectively. The mean values of the highest temperatures were 39.6 °C, 42.1 °C, and 41.4 °C in 2018, 2019, and 2020, respectively.

Surface soil (0–20 cm) from nearby cropland was collected, sieved, homogenized, and packed in the pots. The soil was a clay loam with a pH (1:2.5 soil:water) of 8.2 and contained 2.45% organic carbon (C) and 1.4% total N. Fertilizers were applied at sowing at the rates of 105.32 N kg ha^–1, 65.49 P₂O₅ kg ha^–1, 74.00 K₂O kg ha^–1, and 105.32 N kg ha^–1 at the elongation phenophase. Irrigation equivalent to 15–20 mm of rainfall was applied every 5–10 days after seedling emergence to retain soil moisture and prevent drought in all chambers. A wheat cultivar (Liangxing 99, a high-yield wheat bred by the Liangxing Seed Research Institute, Dezhou, China) was selected for the experiment. In total, 100 seeds of wheat plants were sown on 25 October 2017, 13 October 2018, and 27 October 2019 in each pot (60 cm × 40 cm × 28 cm length × width × depth). Then, eight pots were randomly distributed to each air chamber.

2.2. Gas Exchange Measurements

Gas exchange was measured in the last fully expanded leaf of the primary tillers per pot by using the portable gas exchange system (LI–COR 6400, LI–COR, Lincoln, NE, USA) at 9:00–11:00 a.m. at the booting phenophase (to ensure the same period of development, ET, ECT in 9 April; CK, EC in 14 April; Z45, boots swollen) [35], grain filling phenophase (ET, ECT in 8 May; CK, EC in 18 May; Z75, medium milk) in 2018, and at the booting phenophase (ET, ECT in 13 April; CK, EC in 18 April ), filling phenophase (ET, ECT in May 10; CK, EC in 20 May) in 2019, and elongation phenophase (3 April, Z36, the 6th node detectable), booting phenophase (ET, ECT in 10 April; CK, EC in 16 April), and grain filling phenophase (ET, ECT in 14 May; CK, EC in 22 May) in 2020. The light intensity in the leaf chamber was set at 1600 μmol (photons) m^–2 s^–1 by using an in-built LED lamp (red/blue), and [CO₂] was 400 ppm in the ambient [CO₂] chambers and 600 ppm in the elevated [CO₂] chambers. The temperature of the leaf chamber was stabilized at the actual temperature of the chamber at 9:00 a.m. in the ambient temperature chambers and +2 °C in the elevated temperature chambers. The net photosynthetic rate (A_n) and stomatal conductance (g_s) were measured in accordance with the method described by Li et al. [36].

Meanwhile, the light response curves of photosynthesis and CO₂ were measured at the booting and grain filling phenophases in 2019 and at the elongation, booting, and grain filling phenophases in 2020. The last fully expanded leaves of the primary tillers were assessed at 9:00–14:00 by using the portable gas exchange system (LI–COR 6400, LI–COR, Lincoln, USA). The temperature of the leaf chamber was set at 25 °C and 27 °C in the ambient temperature and elevated temperature chambers, respectively. The [CO₂] levels in the leaf chamber were set at 400 and 600 ppm in the ambient [CO₂] and elevated [CO₂] chambers, respectively. An in-built LED lamp (red/blue) was used to supply different light intensities. Photosynthetically active radiation (PAR) gradients were set to 2000, 1800, 1600, 1400, 1200, 900, 700, 500, 300, 200, 100, 50, and 0 μmol m^–2 s^–1. The net photosynthetic rate (A_n) under different PAR gradients was measured. Meanwhile, the CO₂ response curve was measured. The light intensity was set at 1600 μmol m^–2 s^–1 using an in-built LED lamp (red/blue). The concentration of CO₂ decreased from 400 ppm to 0 ppm and then increased from 0 ppm to 1200 ppm. Ten concentration levels were set at 400, 300, 200, 100, 0, 400, 600, 800, 1000, and 1200 ppm, respectively. The measured net photosynthetic rate (A_n) was used to simulate the CO₂ response curve. The maximum net photosynthetic rate (A_nmax), maximum carboxylation rate (V_cmax), and maximum electron transport rate (J_max) were calculated by simulating the A_n–PAR and A–C_i curves in accordance with the modified rectangular hyperbolic model [37]. The expression of the modified rectangular hyperbola model of the photosynthetically active response was: A_n $=$ $\mathsf{\alpha }\frac{1-\mathsf{\beta }I}{1+\mathsf{\gamma }I}$I-R_d, where $\mathsf{\alpha }$ was the initial slope of irradiance-response curve of photosynthesis when irradiance approaches zero; β and γ were coefficients which were independent of I (m² s μmol^–1), β represents the photoinhibition item (dimensionless), γ represents the light saturation item (dimensionless), and γ = $\mathsf{\alpha }$/P_nmax (P_nmax represents the maximum net photosynthetic rate (μmol m^–2 s^–1)); I was photosynthetically active irradiance; R_d was dark respiration rate. The expression of the modified rectangular hyperbola model of the CO₂-response was: A_n $=$ $\mathrm{a}\frac{1-\mathrm{b}Ci}{1+\mathrm{cC}i}$C_i-R_p, where A_n, C_i, and R_p were net photosynthetic rate, intercellular CO₂ concentration, and potorespiration rate, respectively; $\mathrm{a}$ was initial carboxylation efficiency of the CO₂-response curve; b and c were coefficient (mol μmol^–1), c = a/A_imax (A_imax was the plant photosynthetic capacity calculated using a modified rectangular hyperbolic model of the intercellular CO₂ response), b was the adjusting factor; when “b = 0”, the formula could be changed to the rectangular hyperbola model.

2.3. Growth and Development in Wheat

The emergence Z10 (first leaf through coleoptile), heading Z55 (a half of the inflorescence emerged), flowering Z65 (anthesis half-way), and ripening phenophases Z92 (caryopsis hard can no longer be dented by thumb-nail) were recorded. The emergence phenophase was when the plant emerged from the soil. The heading phenophase was when more than 50% of the heads were out of the sheath. The flowering phenophase was when more than 50% of the heads were flowering, and most flowers started slightly above the middle portion of the head. The ripening phenophase Z92 was when the grains became hard and the green plant tissue became straw.

2.4. Harvesting

At full maturity Z92, wheat plants were harvested on 4 June 2018, 11 June 2019, and 7 June 2020. After air drying, thousand seed mass, number of kernels per spike, ear per square meter, above-ground biomass, and yield were determined for every pot.

2.5. Data Analysis

The data were expressed as means with standard deviations of treatments. A one-way ANOVA was used to compare the differences among CK, EC, ET, and ECT treatments at the same phenophase or same year. A two-way ANOVA was used to analyze the effects of elevated [CO₂], elevated temperature, and their interaction on A_nmax, V_cmax, and J_max (The value of A_nmax, V_cmax, and J_max under the elongation phenophase in 2019 were not measured). A three-way ANOVA was used to analyze the effects of elevated [CO₂], elevated temperature, year, and their interactions on yield components. A four-way ANOVA was used to analyze the effects of elevated [CO₂], elevated temperature, phenophases, year, and their interactions on gas exchange parameters (The value of A_n and g_s under the elongation phenophase in 2018 and 2019 were not measured). The level of p < 0.05 was considered statistically significant. Mean comparisons were performed using Duncan’s multiple comparison test (p < 0.05). All the above statistical analyses were performed using SPSS (IBM SPSS Statistics 23, IBM, Armonk, NY, USA).

3. Results

3.1. Gas Exchange Parameters

Elevated [CO₂] and phenophase on A_n were significant (p < 0.001) in the three years (

Table 1. Effects of elevated [CO₂], elevated temperature and the combination on gas exchange parameters of wheat.

Year	Treatment	An (μmol m−2 s−1)			gs (mol m−2 s−1)
		Elongation	Booting	Grain Filling	Elongation	Booting	Grain Filling
2018	CK		12.27 ± 0.98 Bb	14.28 ± 0.42 Ab		0.27 ± 0.03 Bb	0.56 ± 0.08 Ab
	ET		12.29 ± 0.93 Ab	9.69 ± 0.91 Bc		0.61 ± 0.09 Aa	0.38 ± 0.03 Bc
	EC		21.60 ± 0.74 Aa	16.84 ± 1.02 Ba		0.53 ± 0.04 Ba	0.84 ± 0.08 Aa
	ECT		24.03 ± 0.94 Aa	15.28 ± 0.71 Bab		0.20 ± 0.02 Bc	0.36 ± 0.02 Ac
2019	CK		25.87 ± 1.33 Aab	16.81 ± 0.29 Bb		0.48 ± 0.02 Ba	0.81 ± 0.03 Aa
	ET		23.91 ± 1.25 Ab	13.90 ± 0.33 Bc		0.37 ± 0.03 Bb	0.53 ± 0.02 Ab
	EC		27.91 ± 1.28 Aa	21.13 ± 0.38 Ba		0.24 ± 0.02 Bc	0.63 ± 0.04 Ab
	ECT		24.75 ± 0.23 Ab	16.10 ± 0.43 Bb		0.21 ± 0.00 Bc	0.33 ± 0.03 Ac
2020	CK	19.69 ± 0.51 Ac	18.90 ± 0.58 Ac	14.47 ± 1.23 Bc	0.20 ± 0.01 Ac	0.21 ± 0.01 Ab	0.14 ± 0.02 Bb
	ET	20.72 ± 0.51 Ac	22.08 ± 1.25 Ab	22.56 ± 0.97 Ab	0.28 ± 0.02 Ab	0.26 ± 0.03 Aa	0.28 ± 0.03 Aab
	EC	26.41 ± 0.44 Bb	29.67 ± 1.09 Aa	21.11 ± 1.11 Cb	0.52 ± 0.03 Aa	0.28 ± 0.01 Ba	0.25 ± 0.03 Bab
	ECT	30.04 ± 0.83 Aa	24.08 ± 0.79 Bb	30.21 ± 1.85 Aa	0.30 ± 0.02 Ab	0.10 ± 0.00 Bc	0.37 ± 0.09 Aa
Pvalue	PCO2	**			NS
	PT	NS			**
	PP	**			**
	PYear	**			**
	PCO2*T	NS			**
	PCO2*P	**			**
	PCO2*Year	**			**
	PT*P	**			**
	PT*Year	**			**
	PP*Year	**			**
	PCO2*T*P	**			**
	PCO2*T*Year	**			**
	PCO2*P*Year	**			**
	PT*P*Year	**			**
	PCO2*T*P*Year	**			**

Note: Different lowercase letters indicated significant differences between CK, EC, ET, and ECT treatments (p < 0.05) at same phenophase. Different capital letters indicated significant differences between different phenophases under same environmental conditions (p < 0.05). The data were presented as mean values and standard deviations (n = 4). A_n: Net photosynthetic rate; g_s: stomatal conductance. CK, control (ambient [CO₂], ambient temperature); ET, elevated temperature (ambient [CO₂], ambient temperature +2 °C); EC, elevated CO₂ concentration (ambient [CO₂] + 200 ppm, ambient temperature); ECT, the combination of elevated [CO₂] and temperature (ambient [CO₂] + 200 ppm, ambient temperature +2 °C). P_CO2, P_T, P_P, P_Year, P_CO2*T, P_CO2*P, P_CO2*Year, P_T*P, P_T*Year, P_P*Year, P_CO2*T*P, P_CO2*T*Year, P_CO2*P*Year, P_T*P*Year, P_CO2*T*P*Year represented the p values of the ANOVA results for the effects of CO₂ concentration, temperature, phenophase, year, and their interactions on gas exchange parameters, respectively. ** Significant at the 0.01 probability level; NS, nonsignificant.

). Elevated [CO₂] significantly increased A_n by 76.0% and 17.9% at the booting and grain filling phenophases in 2018, respectively (Table 1). Elevated [CO₂] also increased A_n by 34.1%, 57.0%, and 45.9% at the elongation, booting, and grain filling phenophases, respectively, in 2020. Meanwhile, it increased A_n by 25.7% at the grain filling phenophase in 2019. Elevated temperature reduced A_n by 7.6% and 17.3% at the booting and grain filling phenophases in 2019, respectively; it reduced the A_n by 32.1% at the grain filling phenophase in 2018, but did not reduce A_n in 2020. The mitigation effects of eCO₂ on A_n under elevated temperature at the grain filling phenophase in 2018 and 2020 (the increase in A_n by ECT (1.00 = 15.28–14.28), (15.74 = 30.21–14.47) were greater than those by ET (–4.59 = 9.69–14.28) plus EC (2.56 = 16.84–14.28) and by ET (8.09 = 22.56–14.47) plus EC (6.64 = 21.11–14.47), respectively). The mitigation effects at the grain filling phenophase in 2019 (the increase in A_n by ECT (–0.71) was less than that by ET (–2.91) plus EC (4.32)). The effect of year on A_n was significant (p < 0.001) (Table 1). Compared with elevated temperature, the combination of CO₂ and temperature increased A_n 57.7%, 15.8%, and 33.9% at the grain filling phenophase in 2018, 2019, and 2020, respectively. Significantly, the mitigative effects of the combination of CO₂ and temperature on A_n was greater in 2018 and 2020 than that of in 2019.

Elevated [CO₂] increased g_s by 50.0% at the grain filling phenophase in 2018, increased g_s by 33.3% and 78.6% at the booting and grain filling phenophases in 2020, respectively, but reduced g_s by 50.0% at the booting and 22.2% at the grain filling phenophases in 2019, respectively. Elevated temperature reduced g_s by 32.1% at the grain filling phenophase in 2018, by 22.9% at the booting phenophase, and by 34.6% at the grain filling phenophase in 2019, respectively, but did not reduce g_s in 2020. The interaction between CO₂ and temperature, temperature, phenophase, and year on g_s were significant (p < 0.001) in the three years (Table 1). In 2018 and 2020, g_s were increased under eCO₂, but decreased under the combined effects of ECT at the booting phenophase. In 2019, g_s were reduced under eCO₂ and the combined effects of ECT (Table 1). Compared with the control, g_s decreased in extremely hot year (2019) (high VPD) under ET, EC, and ECT.

3.2. A_n–PAR and A–C_i Curves Parameters

Elevated [CO₂] and the combined effects of ECT significantly increased A_nmax at the booting phenophase in 2019 and all three phenophases in 2020. Elevated temperatures exerted no significant effect on A_nmax in 2019 and 2020 (Figure 3a,b). Elevated [CO₂], elevated temperature, or the combination reduced V_cmax and J_max at the booting and grain filling phenophases in 2019 and the grain filling phenophase in 2020. The interaction between CO₂ and temperature on V_cmax and J_max was significant (V_cmax, p = 0.012 (2019), p = 0.044 (2020); J_max, p < 0.001 (2019), p = 0.013 (2020)) at the grain filling phenophase in 2019 and 2020. The decrease in V_cmax by ECT (–2.63) was significantly less than that by ET (–6.73) plus EC (–12.27) in 2019. The decrease in J_max by ECT (–1.76) was less than that by ET (–6.13) plus EC (–15.34) in 2019. The decrease in V_cmax by ECT (–32.63) was significantly less than that by ET (–18.79) plus EC (–43.11) in 2020. The decrease in J_max by ECT (–45.16) was significantly less than that by ET (–43.1) plus EC (–41.43) in 2020. Elevated [CO₂] mitigated the negative effects of ET on V_cmax and J_max, especially in 2020 (Figure 3c–f).

3.3. Growth and Development in Wheat

Elevated [CO₂], elevated temperature, and the combination shortened the growth period in wheat, such as heading, flowering, and ripening (

Table 2. Effects of elevated [CO₂], elevated temperature, and the combination on wheat phenophases for three growing years.

			Days after Sowing (Days)
Years	Treatment	Sowing Time	Emergence	Heading	Flowering	Ripeness	Flowering to Ripening (Days)	Days of Maximum Temperature (>35.5 °C)
2017–2018	CK	25 October 2017	9	178	182	222	40	10
	ET	25 October 2017	8	171	176	213	37	18
	EC	25 October 2017	9	173	177	218	41	12
	ECT	25 October 2017	8	169	173	211	38	18
2018–2019	CK	13 October 2018	9	190	196	232	36	17
	ET	13 October 2018	8	182	188	221	33	19
	EC	13 October 2018	9	187	193	227	34	17
	ECT	13 October 2018	8	185	191	224	33	19
2019–2020	CK	27 October 2019	9	174	180	215	35	11
	ET	27 October 2019	8	170	174	207	33	16
	EC	27 October 2019	9	174	179	215	36	14
	ECT	27 October 2019	8	168	172	207	35	17

Note: CK: ambient [CO₂], ambient temperature; ET: ambient [CO₂], ambient temperature +2 °C; EC: ambient [CO₂] + 200 ppm, ambient temperature; ECT: ambient [CO₂] + 200 ppm, ambient temperature +2 °C.

). The effect of elevated temperature on the growth period was greater than that of eCO₂. The time from flowering to ripening in 2019 was shorter than that in 2018 and 2020. Compared with ambient conditions (CK), ET shortened the time from flowering to maturity by 3 days in 2018 and 2019, and by 2 days in 2020. Compared with the CK, the EC increased the time from flowering to maturity by 1 day in 2018 and 2020 but shortened it by 2 days in 2019. The combination of ECT decreased the time from flowering to maturity by 2 days in 2018, decreased the time by 3 days in 2019, but did not change the time in 2020, compared with the CK. Compared with the ET, ECT increased the time from flowering to maturity by 1 day in 2018 and by 2 days in 2020, but did not change the time in 2019. The combination of CO₂ and temperature mitigated the negative effects of ET on filling days. The number of high-temperature (>35.5 °C) days from flowering to ripening in 2019 was more than that in 2018 and 2020 (Table 2).

3.4. Biomass, Yield, and Yield Components

Elevated temperature, elevated [CO₂], and their combination had no significant effect on leaf area index at the booting phenophase in 2019 and 2020 (Figure 4a,b). Elevated temperature significantly decreased leaf area index at the grain filling phenophase in 2019 (p = 0.038) and exerted no significant effect at the grain filling phenophase in 2020. Elevated [CO₂] and the combination also exerted no significant effect at the grain filling phenophase in 2019 and 2020.

Elevated [CO₂] reduced ear number by 3.4% while increasing kernel number per spike by 32.7% in 2018. Elevated [CO₂] increased ear number by 7.9% and 27.5% while reducing kernel number per spike by 11.4% and 6.7% in 2019 and 2020, respectively (

Table 3. Effects of elevated [CO₂], elevated temperature, and the combination on yield components wheat in 2018, 2019, and 2020.

Year	Treatment	Ear Number m−2	Kernel Number per Spike	Thousand Seed Mass (g)
2017–2018	CK	323.6 Aa	23.73 Ab	51.69 Aa
	ET	304.2 Ab	21.70 Ab	44.99 Ac
	EC	312.5 Aab	31.49 Aa	46.09 Abc
	ECT	304.2 Ab	29.57 Aa	47.11 Ab
2018–2019	CK	209.7 Cb	30.32 Aa	44.91 Ba
	ET	181.9 Cc	30.46 Aa	43.37 Ba
	EC	226.4 Ca	26.85 Ab	44.74 Ba
	ECT	206.9 Cb	28.53 Ab	44.26 Ba
2019–2020	CK	226.4 Bb	17.74 Ba	40.85 Ba
	ET	249.0 Bb	13.85 Ba	40.55 Ba
	EC	288.6 Bb	16.56 Ba	44.84 Ba
	ECT	359.4 Ba	18.30 Ba	38.73 Ba
Pvalue	PCO2	**	*	NS
	PT	NS	NS	*
	PYear	**	**	**
	PCO2*T	NS	NS	NS
	PCO2*Year	**	**	NS
	PT*Year	**	NS	NS
	PCO2*T*Year	NS	NS	NS

Note: Different lowercase letters indicated significant differences among CK, EC, ET, and ECT treatments (p < 0.05) at same year. Different capital letters indicated significant differences between years (p < 0.05). The data were presented as mean values and error bars indicate standard deviations (n = 3). CK: ambient [CO₂], ambient temperature; ET: ambient [CO₂], ambient temperature +2 °C; EC: ambient [CO₂] + 200 ppm, ambient temperature; ECT: ambient [CO₂] + 200 ppm, ambient temperature +2 °C. P_CO2, P_T, P_Year, P_CO2*T, P_CO2*Year, P_T*Year, P_CO2*T*Year represent the p values of the ANOVA results for the effects of CO₂ concentration, temperature, year, and their interactions on yield components, respectively. * Significant at the 0.05 probability level; ** Significant at the 0.01 probability level; NS, nonsignificant.

). The yield increased (14.3%) by eCO₂ due to the increase in kernel number per spike in 2018 and increased (24.6%) by eCO₂ due to the increase in ear number in 2020, but was not increased in 2019 (Table 3 and

Table 4. Effects of elevated [CO₂], elevated temperature and the combination on the yield, above-ground biomass and harvest index of wheat.

Year	Treatment	Yield (g × m−2)	Above-Ground Biomass (g × m−2)	Harvest Index
2017–2018	CK	396.11 Ac	697.51 Ac	0.57 Ab
	ET	295.15 Ad	535.88 Ad	0.55 Ab
	EC	452.85 Aa	770.81 Aa	0.59 Aa
	ECT	423.00 Ab	747.11 Ab	0.57 Ab
2018–2019	CK	284.74 Ba	700.76 Ab	0.41 Ba
	ET	239.32 Bc	563.11 Ac	0.42 Ba
	EC	271.46 Bab	767.76 Aa	0.35 Bc
	ECT	261.03 Bb	692.54 Ab	0.38 Bb
2019–2020	CK	372.97 Aab	794.08 Aab	0.47 Aa
	ET	306.75 Ab	627.35 Ab	0.49 Aa
	EC	464.74 Aa	971.61 Aa	0.48 Aa
	ECT	382.85 Aab	933.81 Aa	0.41 Ab
Pvalue	PCO2	**	**	*
	PT	**	**	NS
	PYear	**	**	**
	PCO2*T	*	**	NS
	PCO2*Year	*	*	*
	PT*Year	NS	NS	NS
	PCO2*T*Year	NS	NS	NS

Note: Harvest index = Yield/Above-ground biomass. Different lowercase letters indicated significant differences between CK, EC, ET, and ECT treatments at same year (p < 0.05). Different capital letters indicated significant differences between years (p < 0.05). The data were presented as mean values and error bars indicate standard deviations (n = 3). CK: ambient [CO₂], ambient temperature; ET: ambient [CO₂], ambient temperature +2 °C; EC: ambient [CO₂] + 200 ppm, ambient temperature; ECT: ambient [CO₂] + 200 ppm, ambient temperature +2 °C. P_CO2, P_T, P_Year, P_CO2*T, P_CO2*Year, P_T*Year, P_CO2*T*Year represented the p values of the ANOVA results for the effects of CO₂ concentration, temperature, year, and their interactions on the yield, above-ground biomass, and harvest index of wheat, respectively. * Significant at the 0.05 probability level; ** Significant at the 0.01 probability level; NS, nonsignificant.

). [CO₂] and the interaction between CO₂ and year on ear number (p < 0.001) and kernel number per spike (P_CO2 = 0.011, P_CO2*Year < 0.001) were significant across 3 years. [CO₂] exerted no effect on thousand-seed mass for 3 years. Temperature exerted no effect on ear number or kernel number per spike, but decreased thousand-seed mass across 3 years. The change in ear number/m², kernel number per spike, and thousand seed mass was significant (p = 0.001) across 3 years. Compared with 2018 and 2020, the mean ear number decreased in 2019 by 33.7% and 26.6%, but the mean kernel number per spike increased in 2019 by 9.1% and 74.8%, respectively. In 2019, the mean thousand seed mass decreased by 6.6% compared with 2018 and increased by 7.5% compared with 2020. The yield decreased in a high-temperature year (2019) due to the decrease in ear number, while eCO₂ increased ear number to compensate for the decline in grain yield under extremely high temperatures (Table 3 and Table 4).

Elevated temperature decreased the yield by approximately 25.5%, 16.0%, and 17.8% in 2018, 2019, and 2020, respectively. [CO₂], temperature, and the interaction between CO₂ and temperature on yield were significant (P_CO2 < 0.001, P_T = 0.002, P_CO2*T = 0.018) across 3 years. The yield increased by 26.9 with ECT combination, and it was significantly greater than that by ET (–100.9) plus EC (56.7) in 2018. The yield increased (9.9) with ECT combination or (91.8) with EC, while it decreased by ET (–66.2) in 2020. The decrease in yield by ECT (–23.7) was significantly less than that by ET (–45.4) plus EC (–13.3) in 2019. The combination of CO₂ and temperature increased yield by 43.3%, 9.1%, 24.8% in 2018, 2019, and 2020, respectively, compared with the ET. The mitigation effects of the combination of CO₂ and temperature on yield was greater in 2018 and 2020 than that of in 2019. The year and the interaction between CO₂ and year on yield were significant (P_Year < 0.001, P_CO2*Year = 0.025) across 3 years. Compared with 2018 and 2020, the yield decreased in 2019 by 32.6%, 30.8%. The extra temperature in 2019 severely decreased the yields accompanied by slight mitigation from eCO₂. Elevated [CO₂] increased above-ground biomass by 14.1% on average across 3 years, whereas elevated temperature decreased above-ground biomass by an average of 21.3%. The interaction between CO₂ and temperature on above-ground biomass was significant (p < 0.001) across 3 years. The above-ground biomass increased by 49.6, 139.7 under ECT, which was greater than that by ET (–161.6), (–166.7) plus EC (73.3), (177.5) in 2018 and 2020, respectively. The above-ground biomass decreased by –8.2 with ECT, which was less than that by ET (–137.7) plus EC (67) in 2019. Elevated [CO₂] reduced the negative effects of ET on above-ground biomass across 3 years, and it was greater in 2018 and 2020 than that of in 2019 (Table 4).

[CO₂], the year, and the interaction between CO₂ and year on the harvest index were significant (P_CO2 = 0.016, P_Year < 0.001, P_CO2*Year = 0.024) across 3 years. Temperature and the interaction between CO₂ and temperature on harvest index exerted no effect across 3 years. The harvest index was larger in 2018 and 2020 than in 2019 (Table 4).

4. Discussion

Elevated [CO₂] enhanced wheat growth and yield by improving photosynthesis [38,39,40,41,42]. Similar to these studies, eCO₂ enhanced wheat yield in normal-temperature years (2018 and 2020) but did not increase it in high-temperature year (2019). The growth stimulation by eCO₂ was a result of enhanced photosynthesis but also improved water use efficiency due to reduced stomatal conductance in normal-temperature years, similar to the study of Wang et al. [43]. However, reduced stomatal conductance under elevated [CO₂] decreased evapotranspiration by about 10% on average and produced an average 0.7 °C increase in leaf temperature across a wide array of species [44]. This increase in leaf temperature would not have a large effect on plants in normal-temperature years, whereas the increase under extreme temperature in 2019 would decrease capacity for transpiration and evaporative cooling, which might lead to higher canopy temperatures relative to air temperature. Higher canopy temperatures above the threshold adversely affected yield. This phenomenon has been confirmed in rice [45]. Ainsworth and Long [46] found that 18 C₃ crop yields from almost 250 observations were increased by about 18% under eCO₂ and by 10% under the combination of elevated [CO₂] and temperature (2 °C). ECT increased photosynthesis in 2018 and 2020 (except for the booting phenophase), thus enhancing the biomass and yield of wheat. However, ECT caused a decrease in photosynthesis because of the extreme temperature at the grain filling phenophase in 2019. The impacts of eCO₂ on photosynthesis in wheat response to elevated temperature depend on the background temperature. Alonso et al. [47] suggested that eCO₂ increased the V_cmax and enhanced the rubisco activity of wheat in response to increased temperature. In our study, elevated [CO₂] and elevated temperature, or the combination reduced V_cmax and J_max. Cai et al. [48] also indicated that V_cmax and J_max were significantly decreased with increasing growth temperature. This reduction in V_cmax and J_max was possibly because of plant photosynthetic acclimation to eCO₂ induced by sink capacity limitation [48,49,50] or ribulose-1,5-bisphosphate regeneration-limited (J_max) by an increase in carbon assimilates [51]. This phenomenon needs further confirmation in wheat. The decline in yield exposed to elevated temperature was recovered under the combination of ECT because of an increase in kernel number per spike in 2018 and 2020 and an increase in ear number in 2020. On the other hand, the recovery of yield exposed to extreme temperature year (2019) was limited under ECT due to a significant decrease in the kernel number per spike.

Extreme temperature or temperatures beyond the optimum temperature of physiological and developmental processes may have a strong negative effect on crops, leading to variable yields or even crop failure [16,17]. In the present study, the top temperature of one day (23 May) during the grain filling period was 43.3 °C under ET treatment in 2019, and the mean day temperature reached 32.3 °C (Figure 2). Meanwhile, the number of days at high temperature (>35.5 °C) from flowering to ripening in 2019 was greater than that in 2018 and 2020 (Table 2), leading to a significant decrease in A_n, leaf area index, growth period, ear number, and yield in 2019 (Figure 4, Table 1, Table 2, Table 3 and Table 4). Temperatures exceeding optimal growth temperatures reduced wheat yield even under eCO₂ in 2019. Similarly, when wheat was exposed to high temperatures above 32 °C after anthesis, grain yield decreased under eCO₂ [9,52]. Elevated temperature advanced wheat development and shortened the growth period, except for the grain filling period, thereby reducing the biomass and yield of wheat [13,53]. Other studies indicated that high temperature reduced grain weight by shortening the grain filling duration in wheat [14,54,55]. This confirmed our first hypothesis that high temperature had a negative effect on photosynthesis, the growth period, and yield in wheat. Elevated [CO₂] accelerated individual development by increasing leaf temperature and decreasing the stomatal conductance and transpiration rate of leaves, possibly shortening the growth period [53]. Some studies indicated that the positive impact of eCO₂ on growth and yield might be attributed to a longer vegetative phase due to the delays in the flowering time of rice [56], soybean [57], and pigeon pea [58]. However, our study suggested that eCO₂ accelerated individual development, shortened the growth period of vegetative phenophase, but increased the grain filling period from flowering to maturity in 2018 and 2020, thus increasing the photosynthetic metabolite accumulation and stimulating wheat yield in 2018 and 2020 (Table 1, Table 2 and Table 4). From double ridge to anthesis, Warrington et al. [59] revealed that wheat grown at 25 °C has only 40% of the kernel number in the main spike when compared with plants grown at 15 °C during this period. The kernel numbers per spike were significantly increased in 2018 and slightly increased in 2020. However, it was significantly lower under ECT due to the number of days at high temperature in 2019 was greater than that in 2018 and 2020. The combination of ET and EC in 2019 seriously reduced the kernel number per spike, which ultimately reduced the harvest index and limited the CO₂ fertilizer efficiency on wheat. This confirmed our second hypothesis that the mitigation of eCO₂ (600 ppm) on crop response to elevated temperature were weak in a warm season. Chavan et al. [9] also suggested that elevated [CO₂] alleviated the negative impact of heat stress on wheat physiology but not the yield due to grain abortion induced by heat.

Elevated [CO₂] could partly mitigated the negative impact of increased temperature on wheat in North China, but the yield was not increased under the combination of elevated [CO₂] and temperature [14]. Ruiz Vera et al. [60] suggested that the mitigation of eCO₂ on soybean response to elevated temperature was greater in a cool year than in a warm year. Our study also found that the combination of ECT on wheat varied greatly according to whether the growing season was warm or normal. During a normal season in 2018 and 2020, elevated temperature decreased yields, whereas eCO₂ mitigated this adverse effect. During a warm season in 2019, the extra temperature severely decreased the yields accompanied by slight mitigation from eCO₂. In South China, Cai et al. [32] argued that wheat yield increased by eCO₂ and decreased by elevated temperature; meanwhile, the yield was still somewhat decreased under the combination of elevated [CO₂] and temperature. However, in the relatively cool Heilongjiang province (the northernmost province of China), the combination of elevated [CO₂] and temperature increased the mean grain yield of soybean by 31% and of maize by approximately 25% across 5 years [61]. Taken together, the mitigation of eCO₂ (around 600 ppm) on crop response to elevated temperature may be greater in a cool season or relatively cold areas than that in a warm season or relatively warm areas. In general, eCO₂ might mitigate the adverse effect of elevated temperature on crop growth and yield when the mean temperature is below the optimal temperature level for crop growth. In addition, the fluctuations of the CO₂ concentration are normal in many CO₂ enrichment experiments. Allen et al. [62] and Bunce [63] suggested that fluctuations in the CO₂ concentration reduced crop photosynthesis and growth compared with constant eCO₂, and underestimated the benefits of eCO₂. In this study, although the oscillations and fluctuations of eCO₂ might decrease leaf photosynthetic rates in wheat compared to steady levels of eCO₂, eCO₂ increased A_n in wheat by 25.7–76.0% on average across 3 years compared to the control. Zhang et al. [64] suggested that the early-maturing cultivar, due to its shorter growth period, reduced the days of extreme heat during the growing season and escaped the extreme high temperature of the critical growth period to a certain extent. Extremely high temperature presents an even greater challenge to wheat production in the future climate, although eCO₂ could mitigate some negative impacts. Effective measures must be taken by planting early-maturing cultivar, or preferentially increasing the genotypes of kernel number per spike will display better yield responses under elevated [CO₂] and extremely high temperature.

5. Conclusions

Elevated temperature shortened the growth period and decreased wheat yield. Moreover, the extremely high temperature during grain filling exerted a serious negative impact on wheat production. This supported our first hypotheses. Elevated [CO₂] mitigated the negative impacts of high temperature on wheat photosynthesis, biomass, and grain yield in North China, and the mitigative effect was better in the normal-temperature year than in the high-temperature year. Elevated [CO₂] could compensate for the negative impact of elevated temperature on wheat but exerted a limited impact under the extremely high temperature in North China. This was consistent with our second hypothesis. Effective measures, such as planting early-maturing cultivars or preferentially increasing the genotypes of kernel number per spike of wheat, may be used to ensure the stability of wheat production under future climate change scenarios.