Influence of Nitrogen Applications during Grain-Filling Stage on Rice (Oryza sativa L.) Yield and Grain Quality under High Temperature

Abstract

High temperature frequently occurs during rice’s early grain-filling period in the south of China, negatively affecting rice yield and quality and posing a major threat to local rice production. This experiment researched the influence of 3.5 °C warming during the first 20 grain-filling days on rice yield and quality and emphatically investigated the effects of the low-broadcast nitrogen fertilizer application level (LBN), high-broadcast nitrogen fertilizer application level (HBN) and foliar nitrogen fertilizer application (FN) at heading on the rice organ temperature, leaf photosynthesis, chlorophyll fluorescence, yield and grain quality, pasting and thermal properties under high temperature in 2020 and 2022, with a widely planted japonica rice variety, “Wuyunjing31”, in order to explore the practical mitigation measures for reducing the adverse impact of high temperature on rice productivity. The results showed that high temperatures during grain filling increased the rice plant temperature, damaged the chlorophyll fluorescence system and decreased the net photosynthesis rate. This led to a decline in the seed-setting rate and grain weight, resulting in a 7.0% and 13.9% yield loss in 2020 and 2022, respectively. In addition, high temperature caused a decline in the head rice rate and an increase in chalk occurrence and pasting temperature, thereby deteriorating rice grain quality. Under high temperatures, HBN enhanced the rice yield by 3.6% and 13.0% in 2020 and 2022, respectively, while FN enhanced the rice yield by 11.5% in 2022. The increase in yield was linked to the increased seed-setting rate and 1000-grain weight. LBN did not significantly affect the rice yield under high temperatures. The positive effects of nitrogen fertilizer measures on rice yield were associated with their role in lowering plant temperature and protection against the damage to the chlorophyll fluorescence system. All three nitrogen application measures generally improved rice milling quality and appearance quality under high temperature, with HBN generally showing the greatest impact. Under high temperature, LBN and FN tended to make the texture of cooked rice softer due to the decreased consistency, retrogradation enthalpy and retrogradation percentage, and this was closely associated with the decline in amylose content. In summary, nitrogen supplementation at the heading could efficiently mitigate the adverse impact of high temperature during the early grain-filling period on rice yield and quality.

1. Introduction

Rice (Oryza sativa L.) is the staple food for over half of the global population, and approximately 90% of the world’s rice is produced in Asia [1]. China is the largest rice-producing and -consuming country in the world; its rice production plays an important role in global food security. In recent decades, improvements in breeding and cultivation techniques have successfully enhanced the rice yield and quality in China [2]. However, rice production faces severe challenges due to the frequent occurrence of unfavorable environmental conditions, currently and in the future [3,4].

The growth environment is another core factor affecting rice growth, in addition to genotype and cultivation measures. In Southern China, high temperatures are the most frequent abiotic stress incident during the rice reproductive period. Previous researches reported that high temperatures during booting have multiple negative effects on rice yield formation, including but not limited to a reduction in the quantity of rice spikelet differentiation [5], pollen development [6] and fertilization [7], thus reducing rice spikelets per panicle, seed-setting rate, grain weight and yield. Grain filling, the final phase of rice growth, has been found to be quite sensitive to both extreme high-temperature and moderate-temperature rises [8,9,10,11]. It has been stated that higher temperatures during this period would accelerate rice leaf and root senescence [12,13], weaken the leaf’s antioxidant system [14] and disrupt the balance between carbon and nitrogen metabolism [15]. As a result, the grain-filling duration is shortened [9,13], and there is a great loss in grain weight [8,9]. Rice plant temperature reflects its thermal characteristics and the physiological response to the growth environment [16,17,18], which could be a useful indicator of rice’s adaption ability for environmental changes. Leaf photosynthate products during the grain-filling period provide the most assimilate source for grain weight accumulation, while chlorophyll fluorescence characteristics significantly determine rice photosynthesis ability and duration, especially under abiotic stress [19,20].

In addition to yield, rice grain quality is also sensitive to unfavorable growth temperatures. It has been proved that high temperature during the grain-filling period increased the chalky rate [10,21] and decreased the head rice rate [22] and taste [23]. Amylose and protein content have been considered as reliable indicators for the judgement of cooked rice texture [24], which is also affected by high temperature during grain filling [9,24,25]. The RVA (Rapid Viscosity Analyzer) system has been widely used to evaluate rice cooking and eating properties [24] due to its high efficiency, repeatability and objectivity.

Accelerated senescence in vegetative organs is a major factor leading to rice yield reductions under higher temperatures during the grain-filling process [14,15,22]. Therefore, we assumed that measures that could postpone leaf senescence may relieve rice yield and quality loss under high temperature. Plant growth regulators have been testified to be beneficial in remitting the negative impact of high temperatures on rice growth and development, such as salicylic acid [26], jasmonates [27] and the compounds of several plant growth regulators [28], but the expensive price of plant growth regulators limited their popularization and application in rice production. Hence, efficient and cheap measures were urgently required for rice production to cope with the possible high-temperature scenario. Nitrogen is the core nutrient element for plant growth and development, and proper nitrogen fertilizer management could improve rice yield and quality under both normal growth conditions and abiotic stress [29,30,31,32,33]. Nitrogen fertilizer application can usually be divided into two patterns: broadcasting solid particle nitrogen fertilizer and spraying foliar nitrogen fertilizer. Whether these two nitrogen fertilizer application measures could mitigate the negative influence of high temperature on rice growth or not has been rarely reported, and whether there is a difference in their effects remains unclear. In accordance with the possible high-temperature scenario during the grain-filling period in the Yangtze River downstream, this experiment investigated the regulating effects of different nitrogen application measures on rice yield, plant organ temperature, leaf photosynthesis and grain quality, with the aim to explore the regulation impact of nitrogen application on rice productivity under high temperature and its possible mechanisms. The results of this research could enrich mitigation measures for rice to cope with high-temperature stress and assist in stable rice production under climate change.

2. Materials and Methods

2.1. Experimental Site, Plant Material and Soil Information

This experiment was carried out at Yangzhou University, Yangzhou City, Jiangsu Province, China (119.42° N, 32.39° E), in 2020 and 2022. Wuyunjing31, a widely planted japonica rice cultivar in the Yangtze River downstream, was used as the tested object in this research. In this study, rice was cultivated. The pot diameter and height were 30 cm and 26 cm, respectively, and each pot was filled with 14 kg soil, which was from the top 0~20 cm plough layer of the adjacent paddy field and was sieved by 10-mesh net before loading into pots. The soil was a clay loam of 24.1 g kg⁻¹ organic matter, 1.12 g kg⁻¹ total N, 12.36 mg kg⁻¹ available phosphorus, 113.42 mg kg⁻¹ rapidly available potassium and soil pH 6.58. Six healthy seedlings were transplanted into each pot, forming three holes with two seedlings in each hole. The distance from each planted hole to the inwall of the pot was 7.5 cm, so the seedlings could be equably distributed in a pot.

2.2. Experiment Design

In 2020, we designed four treatments: NT (normal temperature), HT (high temperature), HT + LBN (high temperature + low-broadcast nitrogen fertilizer application level at heading), HT + HBN (high temperature + high-broadcast nitrogen fertilizer application level at heading). Because the regulation effects of LBN were detected to be markedly inferior than for HBN in 2020, we retained the treatment of HT + HBN and used FN (foliar nitrogen fertilizer, BRT, Madrid, Spain) to substitute for LBN in 2022. Pots were placed under the outdoor ambient environment until heading. After heading, plants were transferred to artificial climate chambers for accepting two temperature treatments. The chambers are glass rooms with 5.3 m length and 4.0 m width, which could control temperature and relative humidity accurately.

There were 30 pots for each treatment in both experimental years. To guarantee that the development level of rice growth in pots was close to the medium–high-level growth in paddy field, the nitrogen application rates of NT and HT treatments were equal to 315 kg ha⁻¹ in paddy field, and the extra nitrogen application amount at heading was designed at medium level in consideration of avoiding the occurrence of “Unfavorably-delayed senescence”. P₂O₅ and K₂O application rates were also in accordance with the application level of high-yielding production in paddy field. The nitrogen, phosphatic and potassic fertilizers applied in the experiment were the same as those applied in field production, among which the nitrogen fertilizer was instant urea with 46% nitrogen content, the phosphatic fertilizer contained 12% P₂O₅ and the potassic fertilizer contained 60% K₂O. Basal fertilizer contained 1 g of pure nitrogen, 1 g of pure phosphorus and 1 g of pure potassium for each pot, and it was applied one day before transplanting. Panicle fertilizer contained 0.8 g of pure nitrogen, and 1 g of pure potassium was applied to each pot and applied at panicle initiation stage. The detailed fertilizer application amount and stage can be viewed in

Table 1. Fertilizer operation methods for treatments.

Year	Treatment	TNAR (g/Pot)	BNAR (g/Pot)	PNAR (g/Pot)	Nitrogen Application Rates at HS (g/Pot)	P2O5 Application Rates (g/Pot)	KCl Application Rates (g/Pot)
2020	NT	1.80	1.00	0.80	0	1.00	2.00
	HT	1.80	1.00	0.80	0	1.00	2.00
	HT + LBN	2.05	1.00	0.80	0.25	1.00	2.00
	HT + HBN	2.30	1.00	0.80	0.50	1.00	2.00
2022	NT	1.80	1.00	0.80	0	1.00	2.00
	HT	1.80	1.00	0.80	0	1.00	2.00
	HT + HBN	2.30	1.00	0.80	0.50	1.00	2.00
	HT + FN	1.82	1.00	0.80	0.023	1.00	2.00

Note: NT, normal temperature; HT, high temperature; HT + LBN, high temperature + low-broadcast nitrogen fertilizer application level at heading; HT + HBN, high temperature + high-broadcast nitrogen fertilizer application level at heading; HT + FN, high temperature + foliar nitrogen fertilizer application at heading. TNAR, total nitrogen application rates; BNAP, basal nitrogen application rate; PNAR, panicle nitrogen application rates; HS, heading stage.

. The total rate of fertilizer was calculated according to the content of effective ingredients of fertilizer and the number of pots before fertilization. Fertilizers were mixed into an appropriate amount of water and stirred until completely dissolved, and then the solution was fixed to a certain volume to ensure that the amount of fertilizer required for each pot was dissolved in the 200 mL solution. Before fertilization, we poured moderate amount water into pots to form a shallow water layer. Then, 200 mL of the prepared solution was measured and poured into each pot using a graduated cylinder without touching the rice plant. The normal temperature level was designed based on the average temperature in September across the past ten years, a month that covers rice early grain-filling period in Yangtze River downstream of China, and the temperature range between NT and HT was set as 3.5 °C (3.0 °C in daytime/4.0 °C at night), which was in consideration of the possible high-temperature scenario in this region and the asymmetry of global warming. The detailed temperature setting and fertilizer application information is displayed in Table 1 and

Table 2. Temperature setting of climate chamber during filling stage in two years.

Temperature Treatment	DAMTC	Temperature Setting of Different Time Periods (°C)								Nighttime (°C)	Daytime (°C)
		18:00~ 21:00	21:00~ 24:00	24:00~ 3:00	3:00~ 6:00	6:00~ 9:00	9:00~ 12:00	12:00~ 15:00	15:00~ 18:00
NT	1~5 d	28	25	23	21	28	31	34	31	24.25	31
	6~10 d	27	24	22	20	27	30	33	30	23.25	30
	11~15 d	25	22	20	18	25	28	31	28	21.25	28
	16~20 d	23	20	18	16	23	26	29	26	19.25	26
HT	1~5 d	32	29	27	25	31	34	37	34	28.25	34
	6~10 d	31	28	26	24	30	33	36	33	27.25	33
	11~15 d	29	26	24	22	28	31	34	31	25.25	31
	16~20 d	27	24	22	20	26	29	32	29	23.25	29

Note: NT, normal temperature; HT, high temperature. DAMTC, days after moving to chambers. 18:00–6:00 is considered as nighttime, 6:00–18:00 is considered as daytime.

. Relative humidity was kept at approximately 75% during rice growth in climate chambers. The temperature treatments lasted for 20 days, since higher temperature usually occurred during the early grain-filling period and presented more negative impact on rice growth than that occurring during late grain-filling period. After that, all pots were moved back to the outdoor environment. Figure 1 displays temperature and precipitation data for September and October of 2020 and 2022, indicating that the climatic conditions of two experimental years were generally normal.

The irrigation management was carried out as follows: keeping 0~3 cm water depth during seedling recovery period and tillering period, drying to soil surface cracking once tillers were enough, keeping 0~3 cm water depth from jointing to heading and implementing wet and dry alternation during grain-filling period. Weed was manually removed at tillering, jointing and heading stages. Chemical pesticides were used at 3 days after jointing, 6 days before rupturing stage and at the rupturing stage, with the aim to control local major pests and diseases, including borers, rice planthoppers, sheath blight diseases, rice blast and green smut.

2.3. Sampling and Measurement Methods

2.3.1. Rice Plant Temperature

An infrared thermometer (MT4 MAX, Fluke, Everett, WA, USA) was used to determine flag leaf and panicle temperatures at 15:30–16:00 in four growth stages: heading, 10 days after heading (10 DAH), 20 days after heading (20 DAH) and 30 days after heading (30 DAH). For each treatment, nine panicles and flag leaves from the similar single stems were selected for organ temperature measurement. Panicle temperature was calculated by averaging the temperatures of the dorsal and the ventral surfaces. The temperatures of both sides of a flag leaf were measured, and the average of the two sides was considered as the flag leaf temperature.

2.3.2. Rice Yield, Components and Grain-Filling Duration

There were three replications for yield measurement. At maturity stage, three pots were randomly selected to determine panicles per pot pikelets per panicle, seed-setting rate and 1000-grain weight for each replication of every treatment; 1000-grain weight was adjusted to the weight with 14.5% moisture. Rice yield was calculated using the following formula:

Rice yield (g/pot) = panicles per pot × spikelets per panicle × seed-setting rate × (1000-grain weight/1000)

(1)

Rice grain-filling duration was defined as the days from heading to maturity, heading was considered as when 50% of the panicles headed over 2/3 of their whole panicle and maturity stage was considered as when 95% of the spikelets in the whole plot had turned from green to yellow.

2.3.3. Flag Leaf Photosynthesis and Chlorophyll Fluorescence Parameters

At heading, 10 DAH, 20 DAH and 30 DAH, five flag leaves’ photosynthesis parameters and chlorophyll fluorescence parameters were determined for each treatment. Photosynthesis parameters were measured from 9:00 to 11:00 am using a Portable Photosynthesis System (Li-6400, LI-COR, Lincoln, NE, USA). The background parameters were set as follows: LeafFan at Fast, the air flow rate at 500 μmol s⁻¹, light intensity at 1000 μmol/(m²·s), CO₂ concentration inside leaf chamber at 400 ppm and temperature inside leaf chamber at 26 °C. The data were recorded after these parameter readings became relatively stable. A portable chlorophyll fluorometer (Mini-PAM, WALZ, Effelyrich, Germany) was applied to estimate flag leaf chlorophyll fluorescence parameters at nighttime, including Fv/Fo (potential photochemical activity of photosystem II), Fv/Fm (maximal photochemical efficiency), qP (photochemical quenching; photochemical quenching coefficient) and qN (non-photochemical quenching). The minimal fluorescence yield of the dark-adapted state (Fo) was measured with a low-intensity modulated beam (1.6 kHz, 0.5 μmol (photon) m⁻² s⁻¹). The maximal fluorescence yield of the dark-adapted state (Fm) was induced by saturated light (Schott lamp KL 1500 FL 103; 6000 μmol (photon)m⁻² s⁻¹, 2 s) followed by treatment with a lower intensity of light (300 μmol (photon) m² s⁻¹) and with light at 100 kHz supplied to improve the ratio of signal to noise and to stabilize the output signals.

2.3.4. Milling Quality, Chalk Characteristics, Amylose Content and Crude Protein Content

The brown rice rate (the weight percentage of brown rice to rough rice), milled rice rate (the weight percentage of milled rice to rough rice) and head rice rate (the weight percentage of head rice to rough rice) were determined to reflect the rice milling quality. Brown rice was separated from approximately 150 g rough rice using a rice huller (JLG-III, Zhongchuliang, Chengdu, China); then, milled rice was separated from brown rice using a milled rice machine (LTJM-2099, Xiba, Taizhou, China). Rice grain with at least 3/4 proportion of an intact grain body was considered as the head rice. Chalk characteristics and amylose content of milled rice were tested according to the China National Standard GB/T 17897-1999 [34]. Crude protein content of milled rice was measured using an automatic Kjeldahl nitrogen analyzer (Kjeltec 8400, FOSS, Copenhagen, Denmark).

2.3.5. Relative Crystallinity

Rice starch relative crystallinity was measured by an X-ray powder diffractometer (D8 Advance, Bruker, Karlsruhe, Germany) and calculated with the MDI Jade 6 software.

2.3.6. Pasting Properties

Milled rice was porphyrized and screened using a 100-mesh sieve; then, the rice flour pasting properties were tested using a Rapid Visco-Analyzer (RVA-Tech Master, Perten, Hägersten, Sweden). Following this, 3 g of rice flour was mixed with 25 mL of pure water and then put in an aluminum canister. The RVA program first heated at 50 °C for one minute, next heated to 95 °C in 3.75 min and then heated at 95 °C for 2.5 min. Then, it was cooled to 50 °C for 3.75 min and held for 1.4 min. Indicators of starch gelatinization included peak viscosity, hot viscosity, cool paste viscosity and pasting temperature, calculated by the computer. Breakdown = peak viscosity − hot viscosity, setback = final viscosity − peak viscosity, consistency = final paste viscosity − hot viscosity.

2.3.7. Thermal Properties

The thermal properties were measured with a differential scanning calorimetry (DSC) analyzer (Model 200 F3 Maia, Netzsch, Bavaria, Germany). Five milligrams of rice starch was mixed with pure water at two-times the weight of the starch and then sealed in an aluminum pan at 4 °C overnight. The DSC analyzer was first calibrated using a standard pan (empty pan) as reference and then heated from 20 °C to 100 °C at a rate of 10 °C /min. Onset temperature (To), peak temperature (Tp), conclusion temperature (Tc) and gelatinization enthalpy (ΔHg) were calculated by the TA Universal Analysis 2000 software. The samples were heated from 20 °C to 100 °C at a rate of 10 °C /min and then stored at 4 °C for 7 days to test the retrogradation properties. Retrogradation enthalpy (ΔHr) was calculated by the TA Universal Analysis 2000 software, and retrogradation percentage (%R) = ΔHr/ΔHg × 100%.

2.4. Statistical Analysis

Excel 2013 was used to calculate the average value. Statistical Package SPSS 25.0 was used for statistical analysis. One-way analysis of variance was conducted by Duncan’s new multiple-range test; significance was considered at p < 0.05. Pearson correlation analysis was undertaken to research the relation among parameters.

3. Results

3.1. Rice Yield, Components and Grain-Filling Duration

As displayed in

Table 3. Effects of nitrogen fertilizer at heading stage on yield and its components of rice under high temperature during grain-filling period.

Year	Treatment	Panicles per Pot	Spikelets per Panicle	Seed-Setting Rate (%)	1000-Grain Weight (g)	Yield (g/Pot)	Grain-Filling Duration (d)
2020	NT	20.2 a	119.4 b	90.27 a	31.54 a	68.68 a	51
	HT	20.6 a	122.3 a	86.56 b	29.30 b	63.87 c	47
	HT + LBN	20.0 a	121.0 ab	87.90 b	29.57 b	62.87 c	47
	HT + HBN	20.2 a	116.9 c	89.71 a	31.23 a	66.15 b	49
2022	NT	19.4 a	108.7 b	87.75 a	31.98 a	59.18 a	55
	HT	19.0 a	109.1 b	84.74 b	29.02 b	50.97 c	48
	HT + HBN	19.3 a	111.1 a	86.93 a	30.89 a	57.60 b	51
	HT + FN	19.0 a	109.1 b	86.46 a	31.73 a	56.85 b	49

Note: Values labelled by the different letters are significantly different (p < 0.05). NT, normal temperature; HT, high temperature; HT + LBN, high temperature + low-broadcast nitrogen fertilizer application level at heading; HT + HBN, high temperature + high-broadcast nitrogen fertilizer application level at heading; HT + FN, high temperature + foliar nitrogen fertilizer application at heading.

, high temperature during the early gran-filling period had a significant influence on the rice yield. The rice yield was significantly decreased by high temperature compared to rice grown under normal temperatures, with a decrease of 7.00% and 13.83%, respectively. In 2020, the order of rice yield was as follows: NT > HT + HBN > HT > HT + LBN; however, the yield difference between HT and HT + LBN did not reach a significant level. In 2022, rice yield was ordered as follows: NT > HT + HBN > HT + FN > HT; however, there was no significant yield difference between HT + HBN and HT + FN.

There was no significant difference in panicles per pot among the four treatments in each experimental year, and panicles per pot in 2020 and 2022 were similar. Rice grown in 2020 universally presented higher spikelets per panicle than that planted in 2022. Values of spikelets per panicle were generally similar among different treatments in the same year, although a significant difference was observed among some treatments. HT lowered the rice seed-setting rate in both experimental years relative to NT. HT + LBN, HT + HBN and HT + FN all enhanced the seed-setting rate when compared with HT, whereas the difference in the seed-setting rate between HT + LBN and HT + HBN did not reach a significant level. The 1000-grain wight was reduced by 7.10% and 9.25% under HT in comparison with NT in 2020 and 2022, respectively. HBN and FN measures showed increasing effects on 1000-grain weight when rice was exposed to high temperature, whereas rice grown under HT + LBN did not show significantly higher 1000-grain weight than HT.

The grain-filling durations were shortened by HT for 4 days and 7 days in 2020 and 2022, respectively, when compared with NT. Grain-filling durations were lengthened by HT + HBN for 2 days and 3 days in comparison with HT in 2020 and 2022, respectively. Rice grown under HT + FN had 1 more day of grain-filling duration than that grown under HT in 2020, while the grain-filling durations were the same under HT and HT + LBN in 2020.

3.2. Temperature of Rice Organs

In this experiment, the temperatures of the rice panicle and flag leaf were measured to reflect plant temperature (

Table 4. Effects of nitrogen fertilizer at heading stage on rice organ temperatures under high temperature during the grain-filling period.

Year	Treatment	Panicle (°C)				Flag Leaf (°C)
		HS	10 DAH	20 DAH	30 DAH	HS	10 DAH	20 DAH	30 DAH
2020	NT	27.88 a	30.01 c	27.24 c	25.16 b	27.32 a	29.97 c	26.76 c	24.98 b
	HT	27.92 a	32.95 a	29.63 a	25.90 a	27.40 a	32.62 a	29.30 a	25.93 a
	HT + LBN	27.87 a	32.45 a	29.63 a	25.65 a	27.45 a	32.45 a	29.53 a	25.58 a
	HT + HBN	27.82 a	31.80 b	28.21 b	25.23 b	27.20 a	31.60 b	28.41 b	25.15 b
2022	NT	27.40 a	29.42 d	26.92 d	21.78 b	27.20 a	29.30 d	26.69 d	21.72 b
	HT	27.56 a	32.35 a	29.80 b	22.42 a	27.12 a	32.27 a	29.73 b	22.34 a
	HT + HBN	27.38 a	31.93 b	28.83 c	21.90 b	27.57 a	31.67 b	28.76 c	21.41 b
	HT + FN	27.54 a	31.22 c	30.41 a	22.60 a	27.32 a	31.02 c	30.20 a	22.58 a

Note: different letters after the data in the same column indicate significant differences at the p < 0.05 level. NT, normal temperature; HT, high temperature; HT + LBN, high temperature + low-broadcast nitrogen fertilizer application level at heading; HT + HBN, high temperature + high-broadcast nitrogen fertilizer application level at heading; HT + FN, high temperature + foliar nitrogen fertilizer application at heading. HS, heading stage; 10 DAH, 10 days after heading; 20 DAH, 20 days after heading; 30 DAH, 30 days after heading.

). Flag leaf and panicle temperature gradually decreased during the grain-filling process. Flag leaf temperature was generally slightly lower than panicle temperature, but the differences were universally non-significant. There were no significant differences in both the flag leaf temperature and panicle temperature among treatments at the heading stage, since warming treatment did not start at this stage. At 10 DAH, 20 DAH and 30 DAH, panicle and flag leaf temperatures under HT were significantly higher than those under NT in both years, and the organ temperature differences between NT and HT tended to be smaller with the process of grain filling. No significant differences were observed in both the panicle and flag leaf temperatures between HT and HT + LB. In 2020, rice panicle and flag leaf temperatures under HT + HBN were significantly lower than those under HT and HT + LBN at each stage after heading, but they were still higher in comparison with NT. In 2022, we conducted HBN and FN to explore their effects on rice plant temperature under high temperatures, and both nitrogen application measures lowered rice panicle and leaf temperature under high temperatures at 10 DAH; the cooling influence induced by FN was even larger than that induced by HBN. At 20 DAH and 30 DAH, FN did not significantly change the panicle and flag leaf temperature for rice grown under high temperatures. Different from FN, HBN also significantly reduced the rice plant temperature under high temperatures at 20 DAH and 30 DAH.

3.3. Flag Leaf Photosynthesis and Chlorophyll Fluorescence Parameters

At the heading stage, there were no significant differences in flag leaf photosynthesis in both experimental years (Figure 2). However, at 10 DAH, 20 DAH and 30 DAH, rice grown under different temperature conditions and nitrogen application measures presented obvious differences in photosynthesis and chlorophyll fluorescence parameters. With the grain-filling process, Pn (net photosynthetic rate), Gs (stomatal conductivity), Ci (intercelluar CO₂ concentration) and Tr (transpiration rate) gradually declined under each treatment, while high temperature accelerated the Pn rate of decline. Rice grown under HT + LBN, HT + HBN (two years’ data) and HT + FN showed higher Pn than that grown under HT at 10 DAH. At 10 DAH, 20 DAH and 30 DAH, HT tended to decrease Pn in contrast to NT. LBN and HBN significantly enhanced Pn when rice was exposed to high temperature, while HT + LN showed little influence on Pn at 10 DAH and 20 DAH relative to HT and had an even lower Pn than that under HT at 30 DAH.

No significant difference was observed in chlorophyll fluorescence parameters among treatments at the heading stage (Figure 3). Fv/Fo, Fv/Fm and qP all showed the highest values at the heading stage. Compared with NT, HT significantly reduced Fv/Fo and Fv/Fm after heading in both years except for Fv/Fm at 10 DAH in 2020, and HT also decreased qP at 10 DAH and 20 DAH in two experimental years, but the qP difference between the two treatments did not reach a significant level in 2020. HT significantly enhanced qN after heading in both experimental years relative to NT. In 2020, Fv/Fo and Fv/Fm of HT + LBN and HT + HBN showed a decreasing tendency or no clear change at 10 DAH, respectively, compared with HT, but they both showed higher values than those under HT at 20 DAH and 30 DAH. LBN did not induce a significant change in qP under HT in 2020, whereas HBN significantly enhanced qP at 10 DAH in 2020. Under high temperatures, LBN and HBN both significantly increased qN at 10 DAH, and HBN significantly reduced qN at 20 DAH and 30 DAH, compared with that without nitrogen fertilizer application. In 2022, both HBN and FN significantly increased Fv/Fo and Fv/Fm after heading when rice was grown under high temperatures. qP was enhanced by HT + HBN and HT + FN at 10 DAH and 20 DAH relative to HT. qN was significantly increased by HT + FN at 10 DAH, 20 DAH and 30 DAH, compared with HT, while it was significantly increased and decreased by HT + HBN at 10 DAH and 20 DAH in comparison with HT, respectively.

3.4. Grain Quality

3.4.1. Rice Characteristics, Milling Quality, Amylose Content and Crude Protein Content

As displayed in

Table 5. Effects of nitrogen fertilizer at heading stage on rice milling quality, chalk characteristics, amylose content and crude protein content under high temperature during grain-filling period.

Year	Treatment	Brown Rice Rate (%)	Milled Rice Rate (%)	Head Rice Rate (%)	Chalky Rate (%)	Chalky Area (%)	Chalky Degree (%)	Amylose Content (%)	Crude Protein Content (%)	Relative Crystallinity (%)
2020	NT	83.33 ± 0.11 a	72.05 ± 0.37 a	68.51 ± 0.02 a	39.1 ± 1.4 d	15.6 ± 1.3 d	6.1 ± 0.8 d	18.13 ± 0.12 a	7.36 ± 0.15 c	20.54 ± 0.12 b
	HT	82.15 ± 0.06 b	66.13 ± 0.35 d	62.32 ± 0.15 d	63.9 ± 1.3 a	31.8 ± 1.4 a	20.3 ± 1.9 a	16.04 ± 0.11 b	8.19 ± 0.09 b	22.54 ± 0.27 a
	HT + LBN	82.33 ± 0.15 b	67.28 ± 0.09 c	63.25 ± 0.53 c	52.1 ± 1.3 b	26.2 ± 0.7 b	13.7 ± 1.2 b	15.55 ± 0.17 bc	9.71 ± 0.11 a	22.43 ± 0.19 a
	HT + HBN	83.24 ± 0.07 a	69.07 ± 0.35 b	66.43 ± 0.11 b	45.0 ± 1.7 c	23.1 ± 1.1 c	10.4 ± 0.6 c	15.29 ± 0.18 c	9.75 ± 0.07 a	22.94 ± 0.17 a
2022	NT	83.57 ± 0.23 a	72.04 ± 0.11 a	70.63 ± 0.76 a	33.3 ± 0.8 c	20.7 ± 0.9 c	6.9 ± 0.2 c	18.67 ± 0.20 a	7.13 ± 0.12 c	19.57 ± 0.07 c
	HT	82.15 ± 0.09 b	68.77 ± 0.61 b	65.46 ± 0.32 c	58.0 ± 1.0 a	33.9 ± 0.8 a	19.7 ± 0.2 a	16.77 ± 0.17 b	7.98 ± 0.14 b	20.05 ± 0.11 ab
	HT + HBN	83.74 ± 0.20 a	71.89 ± 0.26 a	68.84 ± 0.26 b	48.8 ± 2.2 c	23.8 ± 2.0 b	11.5 ± 0.6 b	15.75 ± 0.12 c	9.27 ± 0.08 a	20.80 ± 0.06 a
	HT + FN	82.03 ± 0.27 b	67.22 ± 0.69 b	64.11 ± 0.36 c	51.3 ± 1.2 b	24.4 ± 1.0 b	12.5 ± 0.2 b	15.42 ± 0.16 c	8.23 ± 0.06 b	19.76 ± 0.13 bc

Note: NT, normal temperature; HT, high temperature; HT + LBN, high temperature + low-broadcast nitrogen fertilizer application level at heading; HT + HBN, high temperature + high-broadcast nitrogen fertilizer application level at heading; HT + FN, high temperature + foliar nitrogen fertilizer application at heading. Different letters after the data in the same column indicate significant differences at the p < 0.05 level.

, rice grown under NT showed the highest brown rice rate, milled rice rate and head rice rate among all treatments in both years. Compared with NT, HT significantly reduced the rice brown rice rate, milled rice rate and head rice rate in both 2020 and 2022. LBN did not significantly alter the brown rice rate when rice was exposed to high temperature but markedly improved the milled rice rate and head rice rate. The brown rice rate, milled rice rate and head rice rate under HT + HBN were significantly higher than those under HT, HT + LBN and HT + FN. No significant difference was observed in the brown rice rate, milled rice rate and head rice rate between HT and HT + FN in 2022. Compared with NT, high temperature significantly increased the chalky rate, chalky area and chalky degree in both years. Different nitrogen application measures all showed decreasing effects on three rice grain chalk parameters when exposed to high temperature. The chalky rate, chalky area and chalky degree presented 18.5%, 17.6% and 32.5% lower values under HT + LBN than under HT in 2020 and showed 11.6%, 28.0% and 36.5% lower values under HT + FN than under HT in 2022. HBN also induced significant decreases for the chalky rate, chalky area and chalky degree when rice was exposed to high temperature in both 2020 and 2022, and the decreasing ranges were larger than those caused by LBN and FN. The above results indicate that HBN had larger decreasing effects on rice chalk occurrence and area than LBN and FN when rice was grown under high grain-filling temperatures.

Rice grown under NT always showed the highest amylose contents across all the treatments in both experimental years. Compared with NT, HT reduced rice amylose content by 11.5% and 10.2% in 2020 and 2022, respectively. In 2020, the rice amylose content presented a decreasing trend with the application of nitrogen fertilizer under high temperature, and a significant difference was detected between HT and HT + HBN treatments, whereas no significant difference was found between HT and HT + LBN treatments. In 2022, both HT + HBN and HT + FN induced significantly lower rice amylose contents than those under HT. In both years, HT significantly increased crude protein contents relative to NT, and rice protein contents were further markedly enhanced by each nitrogen application method under high temperature. Relative crystallinity was significantly increased by HT in contrast with NT in both years. The three nitrogen fertilizer application measures did not significantly change the relative crystallinity.

3.4.2. RVA Profiles

Table 6. Effects of nitrogen fertilizer at heading stage on rice RVA profiles under high temperature during grain-filling period.

Year	Treatment	Peak Viscosity (cP)	Hot Viscosity (cP)	Final Viscosity (cP)	Breakdown (cP)	Setback (cP)	Consistency (cP)	Pasting Temperature (°C)
2020	NT	2562.67 ± 25.44 b	1955.67 ± 16.76 a	2777.33 ± 17.89 b	607.00 ± 42.20 b	214.67 ± 19.60 b	821.67 ± 31.48 bc	76.02 ± 0.02 b
	HT	2632.67 ± 5.36 a	2018.67 ± 14.40 a	2905.67 ± 21.99 a	614.00 ± 19.09 b	273.00 ± 25.40 a	887.00 ± 26.21 ab	78.60 ± 0.69 a
	HT + LBN	2360.67 ± 12.44 c	1846.00 ± 26.50 b	2638.00 ± 19.66 c	514.67 ± 38.93 c	277.33 ± 32.10 a	792.00 ± 7.09 c	78.90 ± 0.50 a
	HT + HBN	2550.00 ± 7.55 b	1829.67 ± 29.33 b	2783.33 ± 2.40 b	720.33 ± 31.86 a	233.33 ± 5.90 b	953.67 ± 31.27 a	78.88 ± 0.56 a
2022	NT	2749.67 ± 18.81 d	2156.00 ± 18.56 c	3172.33 ± 32.33 b	593.67 ± 3.76 b	422.66 ± 43.28 a	1016.33 ± 45.49 a	73.87 ± 0.25 b
	HT	3085.67 ± 31.25 a	2280.67 ± 17.13 b	3242.67 ± 16.17 a	805.00 ± 41.86 a	157.00 ± 31.23 d	962.00 ± 11.93 a	77.03 ± 0.24 a
	HT + HBN	2825.33 ± 81.63 c	2034.00 ± 13.20 d	3018.00 ± 32.14 c	791.33 ± 74.80 a	192.67 ± 56.31 c	984.00 ± 20.84 a	77.25 ± 0.28 a
	HT + FN	2947.33 ± 24.03 b	2433.67 ± 85.89 a	3287.33 ± 32.11 a	513.66 ± 85.23 c	340.00 ± 25.06 b	853.66 ± 60.19 b	77.05 ± 0.53 a

Note: NT, normal temperature; HT, high temperature; HT + LBN, high temperature + low-broadcast nitrogen fertilizer application level at heading; HT + HBN, high temperature + high-broadcast nitrogen fertilizer application level at heading; HT + FN, high temperature + foliar nitrogen fertilizer application at heading. Different letters after the data in the same column indicate significant differences at the p < 0.05 level.

displayed the rice flour RVA profiles. Compared with NT, HT significantly increased the peak viscosity, hot viscosity, final viscosity, breakdown and pasting temperature, except for hot viscosity and breakdown in 2020. HT presented a different influence on setback in different years, with an increase in 2020 and a decrease in 2022. Consistency was not significantly altered by HT in comparison with NT in both years. Compared with HT, rice grown under HT + HBN did not show significant consistency changes, while LBN and FN significantly lowered consistency in 2020 and 2022, respectively. Pasting temperatures were significantly increased by 3.4% and 4.3% under HT in 2020 and 2022, respectively, compared with NT. No nitrogen application measures brought significant changes in pasting temperature under high temperatures. Rice grown under HT + LBN and HT + HBN both showed significantly lower peak viscosity, hot viscosity and final viscosity. In 2020, compared with HT, HT + HBN significantly increased and lowered rice breakdown and setback, respectively, but did not significantly change the consistency and pasting temperature. HT + LBN significantly lowered and increased rice breakdown and consistency, respectively, while it did not have significant influence on setback and pasting temperature. In 2022, both HT + HBN and HT + FN significantly enhanced setback, and HT + FN also significantly reduced breakdown and consistency.

3.4.3. Thermal Properties

Table 7. Effects of nitrogen fertilizer at heading stage on rice starch gelatinization and retrogradation parameters under high temperature during grain-filling period.

Year	Treatment	Gelatinization				Retrogradation
		ΔHg (J/g)	Tp (°C)	To (°C)	Tc (°C)	ΔHr (J/g)	R (%)
2020	NT	10.65 ± 0.15 b	67.68 ± 0.18 c	59.65 ± 0.28 b	75.57 ± 0.15 c	5.65 ± 0.01 d	53.04 ± 0.37 c
	HT	10.66 ± 0.25 b	69.85 ± 0.15 b	58.88 ± 0.03 c	78.93 ± 0.12 a	6.69 ± 0.04 a	62.75 ± 0.49 a
	HT + LBN	11.20 ± 0.17 a	70.63 ± 0.15 a	60.50 ± 0.48 a	78.25 ± 0.25 b	6.17 ± 0.04 c	55.13 ± 0.86 c
	HT + HBN	10.99 ± 0.05 ab	70.68 ± 0.28 a	58.90 ± 0.38 c	78.90 ± 0.21 a	6.47 ± 0.03 b	58.95 ± 0.08 b
2022	NT	11.79 ± 0.15 b	66.13 ± 0.18 d	58.47 ± 0.28 c	75.17 ± 0.15 c	3.77 ± 0.01 c	31.95 ± 0.37 d
	HT	11.66 ± 0.25 b	71.03 ± 0.15 b	59.43 ± 0.03 b	79.77 ± 0.12 a	5.38 ± 0.04 a	46.20 ± 0.49 a
	HT + HBN	12.94 ± 0.17 a	72.37 ± 0.15 a	60.37 ± 0.48 a	78.70 ± 0.25 b	5.48 ± 0.04 a	42.39 ± 0.86 c
	HT + FN	11.18 ± 0.05 c	70.47 ± 0.28 c	57.23 ± 0.38 d	78.80 ± 0.21 b	4.97 ± 0.03 b	44.48 ± 0.08 b

Note: NT, normal temperature; HT, high temperature; HT + LBN, high temperature + low-broadcast nitrogen fertilizer application level at heading; HT + HBN, high temperature + high-broadcast nitrogen fertilizer application level at heading; HT + FN, high temperature + foliar nitrogen fertilizer application at heading. Different letters after the data in the same column indicate significant differences at the p < 0.05 level.

showed the rice starch gelatinization and retrogradation parameters. For rice starch gelatinization parameters, HT significantly increased Tp, To and Tc when compared with NT in both years, but it did not have significant effects on ΔHg. In 2020, compared with rice grown under HT, rice grown under HT + LBN showed significantly higher ΔHg, Tp, To, and, meanwhile, showed significantly lower Tc. HT + HBN significantly increased Tp, while it did not significantly change the other three parameters in 2020, but it significantly increased the values of all the four rice starch gelatinization parameters in 2022. Compared with HT, HT + FN significantly reduced the values of all four rice starch gelatinization parameters. To investigate the cold rice texture, retrogradation enthalpy (ΔHr) and retrogradation percentage (R) were determined. Compared with NT, HT induced significant higher ΔHr and R. Under high temperatures, ΔHr and R were significantly decreased by all three nitrogen application measures, except for the fact that ΔHr was not significantly changed by HBN in 2022.

3.5. Correlation Analysis

3.5.1. Correlation Analysis between Pn of Flag Leaves and Rice Milling and Chalk Parameters

Table 8. Pearson correlation analysis between Pn of flag leaves and rice milling and chalk parameters.

	BRR	MRR	HRR	Chalky Rate	Chalky Area	Chalky Degree
Pn at 10 DAH	0.567 **	0.446 *	0.434 *	−0.691 **	−0.736 **	−0.706 **
Pn at 20 DAH	0.327	0.3	0.266	−0.475 *	−0.523 **	−0.478 *
Pn at 30 DAH	0.629 **	0.769 **	0.776 **	−0.544 **	−0.335	−0.380

Note: * represents significant at p < 0.05 level, ** represents significant level at p < 0.01 level, no * after number represents nonsignificant. Pn, net photosynthesis rate; DAH, days after heading. BRR, brown rice rate; MRR, milled rice rate; HRR, head rice rate.

showed the correlation between Pn of the flag leaves and rice milling and chalk parameters. The brown rice rate, milled rice rate and head rice rate all had a significant positive relationship with Pn at 10 DAH and 30 DAH, while no significant correlation was detected between Pn at 20 DAH and the three milling quality parameters. The chalk rate, chalky area and chalky degree were all significantly negatively correlated with Pn at 10 DAH and 20 DAH, while the chalky rate was also significantly negatively correlated with Pn at 30 DAH.

3.5.2. Correlation Analysis between Rice Amylose Content, Crude Protein Content, Relative Crystallinity and Pasting and Thermal Properties

Pearson correlation analysis was undertaken to investigate the relationship between amylose content, protein content, relative crystallinity and rice pasting and thermal properties (

Table 9. Pearson correlation analysis between rice amylose content, crude protein content, relative crystallinity and pasting and thermal properties.

	AC	CPC	RC	PKV	HTV	FLV	BD	SB	Cos	PT	ΔHg	Tp	To	Tc	ΔHr	R
AC	1	0.822 **	0.545 **	0.075	0.119	0.170	−0.058	0.234	0.159	−0.758 **	−0.069	−0.869 **	−0.007	−0.818 **	−0.826 **	−0.825 **
CPC	−0.822 **	1	0.693 **	−0.390	−0.505 *	−0.495 *	0.114	−0.266	−0.112	0.750 **	0.353	0.768 **	0.209	0.576 **	0.699 **	0.566 **
RC	−0.545 **	0.693 **	1	−0.686 **	−0.750 **	0.778 **	−0.019	−0.239	−0.272	0.761 **	0.089	0.461 *	0.160	0.430 *	0.456 *	0.424 *

Note: * represents significant at p < 0.05 level, ** represents significant level at p < 0.01 level, no * after number represents nonsignificant. AC, amylose content; CPC, crude protein content; RC, relative crystallinity; PKV, peak viscosity; HTV, hot viscosity; FLV, final viscosity; BD, breakdown; SB, setback; Cos, consistency; PT, pasting temperature; ΔHg, gelatinization enthalpy; Tp, peak temperature; To, onset temperature; Tc, conclusion temperature; ΔHr, retrogradation enthalpy; R, retrogradation percentage.

). Amylose content had a significantly negative relation with protein content and relative crystallinity, while protein content had a significantly positive relation with relative crystallinity. Amylose content was negatively correlated with pasting temperature, Tp, Tc, ΔH and R, while no significant correlation was detected between amylose content and other RVA parameters. RC was significantly negatively related with PKV, HTV and FLV, and it was positively related with pasting temperature, Tp, Tc, ΔHr, To and R.

4. Discussion

4.1. Regulation Influence of Nitrogen Application Measures on Rice Yield and Leaf Photosynthesis under High Grain-Filling Temperature

Nitrogen fertilizer application is a crucial management measure for rice cultivation. Scientific nitrogen fertilizer management, including proper application amount, timing and form, could improve crop yield, grain quality and lodging resistance [35,36,37]. Considering that the rice yield and grain quality loss were closely associated with the accelerated senescence of the rice vegetative organ under high temperatures, we assumed that supplementing nitrogen fertilizer in the late rice growth stage may alleviate the damage caused by high temperatures.

Yield loss was detected under high temperatures, and it was mainly due to the decline in seed-setting rate and 1000-grain weight. Panicles per pot were not significantly altered by both high temperature and nitrogen application measures under high temperature, since this yield component had already formed before heading. Though a significant difference was observed in spikelets per panicle between some treatments, the highest value of spikelets per panicle was just 4.6% and 2.2% higher than the lowest value in 2020 and 2022, respectively. It is foreseeable that spikelets per panicle varied little across treatments, considering that both temperature and nitrogen treatment started at heading in this study, missing the forming period of the spikelet number, which was from the jointing stage to the heading stage. Many studies have reported that the seed-setting rate is greatly reduced by high temperature [21,38,39], though we also detected a decline trend for the seed-setting rate under high temperature; the decreasing amplitudes were much smaller than those in other studies, which undertook warming experiments with ranges over 4 °C [27,31,38]. Evidently, significant grain weight loss was detected under high temperature, being consistent with the results reported by many previous studies [8,9,21]. Nitrogen fertilizer application measures, especially HBN, markedly enhanced grain weight under high temperatures.

Rice temperature has been considered as an indicator of crop physiological metabolism ability and even yield [16,17]. Jiang et al. (2023) found that the rice yield was negatively related to canopy temperature across different nitrogen levels, since they observed higher photosynthesis and transpiration rates due to larger stomatal area under abundant nitrogen application, which could effectively lower the rice temperature [18]. In this study, each nitrogen fertilizer application measure significantly lowered the rice flag leaf and panicle temperature, and this cooling influence may be attributed to higher stomatal conductivity.

Nitrogen is involved in photosynthetic enzymes, pigment content and the size, number and composition of chloroplasts, and leaf’s nitrogen content positively affects photosynthesis [40,41]. According to a report by Kim et al. (2021), high temperature could accelerate the leaf nitrogen loss during grain filling and, thereby, reduce leaf chlorophyll content, causing a rapid decline in the net photosynthesis rate during the middle and late grain-filling periods [13]. The rapid decline in the flag leaf net photosynthesis rate under high temperature was also detected in this study, whereas HBN tended to improve the flag leaf net photosynthesis rate, especially in 2020. High temperature presented multiple influences on the rice leaf photosynthesis system; it could negatively affect the process of electron transfer and photosynthetic phosphorylation, and it deteriorated the photosynthesis system. PSII (photosystem II) absorbs luminous energy and use it to spill water and transmit electrons to plastoquinone, which is a key process of photosynthesis. The PSII reaction center is located in the chloroplast thylakoid membrane, and it plays a decisive role during the photosynthesis process. Fv/Fo and Fv/Fm are two key parameters evaluating photochemical efficiency, and they decline once a plant encounters severe abiotic stress. In this study, the two photochemical efficiency indicators showed clear decreasing tendencies under high temperature relative to normal temperatures, suggesting that the rice leaf photosynthesis system was weakened by high temperature. This study found that Fv/Fo and Fv/Fm were improved by different nitrogen fertilizer application measures under high temperature, revealing the positive influence of nitrogen fertilizer on photochemical efficiency when rice was exposed to high temperature. Among the three nitrogen fertilizer application measures, HBN evidently had larger improvement effects on Fv/Fo and Fv/Fm, especially at 20 DAH and 30 DAH, providing sustained protection for PSII during the middle and late grain-filling stages.

The use of luminous energy by plants is divided in three ways: inverting luminous energy via photochemical reaction, non-photochemical heat and dissipating surplus luminous energy by chlorophyll fluorescence. Chlorophyll fluorescence quenching consists of photochemical quenching and non-photochemical quenching. Non-photochemical quenching is a self-protective mechanism for the photosynthesizer [42]. Previous studies have shown that effective heat dissipation in rice can prevent photoinhibition from occurring [43,44]. In both experimental years, high temperature significantly decreased qP at 10 DAH, and it also significantly lowered qP at 20 DAH in 2022. The unitive decreasing tendency of Fv/Fo, Fv/Fm and qP under high temperature indicated that PSII was impaired by the unfavorable temperature, and we detected that rice developed more heat dissipation to lower the damage to PSII caused by HT, in accordance with the increased qN. LBN did not improve qP; thus, it failed to increase more heat dissipation continuously. HBN showed a different effect on luminous energy utilization by the leaf in 2020 and 2022. In 2022, rice grown under HT + HBN presented significantly higher qP at 10 DAH and 20 DAH and showed higher qN at 10 DAH, showing a better translocation ability of excess luminous energy during the most active grain-filling period. There is no doubt that the protection of the photosynthesis system through nitrogen application, especially HBN, could mitigate the negative impact of high temperature on rice leaf photosynthesis and, thus, guarantee the assimilate supply required from grain filling.

4.2. Regulation Influence of Nitrogen Application Measures on Rice Grain Quality under High Grain-Filling Temperature

Rice appearance, as the first impression on consumers, significantly affects consumers’ purchasing desire [45]. Chalkiness is an opaque white part of the endosperm created by the loose packing of starch granules, which induces unfavorable grain appearance characteristics and even deteriorates rice milling quality and taste. Plenty of research has mentioned the increasing trend of the chalky rate and area under higher temperatures during the grain-filling period [10,21,22]. This study conducted a 3.5 °C warming experiment and also found that more chalky grains and larger chalky area occurred as the grain-filling temperature increased, further indicating the sensitivity of chalk characteristics to the temperature increase. Deteriorative grain chalk characteristics under high temperature were necessarily associated with the loosely packed starch granules, usually caused by reduced photosynthesis and an accelerated filling rate during the early grain-filling period [46,47]. Under the 3.5 °C warming of this study, the three tested nitrogen fertilizer application methods all cut down the rice chalky rate, chalky area and chalky degree, but they failed to adjust chalk characteristics to the level under normal temperature. HBN had the best influence on inhibiting the rice chalk rate and area among the three nitrogen application treatments. The remission mechanism of nitrogen fertilizer on chalkiness occurrence has been reported in recent studies, which found that the supplement of nitrogen fertilizer increased molecular chaperone (cpHSP70-2) accumulation, and protein body development could lower the space between the amyloplast in the rice endosperm [48,49], which may support the declining trend of chalk occurrence by nitrogen application measures under high temperature in this study. In addition, a negative relationship was found between the chalky rate and area and flag leaf net photosynthesis rate, especially Pn at 10 DAH (Table 8), the most active grain-filling timing. We believe that the enhanced photosynthesis caused by nitrogen application treatments is also responsible for the decreased grain chalkiness occurrence under high temperature, since a shortage of assimilate supply has also been recognized as a cause of higher chalkiness occurrence under high temperatures [50,51]. This research also detected that leaf photosynthesis was enhanced and grain-filling duration was prolonged by the nitrogen application measures, which may provide enough assimilates for grain filling and, meanwhile, alleviate the unfavorably quick filling rate, which could cause a loosened arrangement of starch granules in the rice endosperm. These changes could eventually lower the grain chalkiness occurrence. Milling quality contains brown rice rate, milled rice rate and head rice rate. The head rice rate directly embodies the rice economic yield in markets, which determines the unhusked rice purchasing price by rice milling factories. In this study, high temperature significantly lowered the values of the three parameters, inducing a deterioration in the rice milling quality, whereas the rice milling quality was distinctly enhanced by the two broadcast nitrogen application treatments; increasing the broadcast nitrogen fertilizer would amplify the beneficial impact on the rice milling quality. The above results found that the increase in the flag leaf net photosynthesis rate could lower the rice chalky rate and area, and this effect would also be beneficial to improve the head rice rate, because the chalky part of rice grains was more fragile during the milling process than the transparent part of rice grains [24]. The positive relation between the head rice rate and Pn (Table 8) could further support the milling quality improvement by high-broadcast fertilizer application under high temperature. Regrettably, FN did not improve the rice milling quality.

In this study, a law of alternating growth and decline between the amylose content and protein content was observed; this may be linked to their competition in some common substrates. Cooked rice with a high protein content tends to have a harder and rougher mouthfeel [52]. In this research, high temperature increased the grain protein content, while two broadcast nitrogen fertilizer methods further aggravated the protein increasing tendency, which was probably linked to the enhancement in the nitrogen metabolism-related enzymes by both high temperatures and more nitrogen fertilizer application [53,54]. Amylose was considered as a key factor in the estimation of rice cooking and eating quality [24,55]. Rice with a lower amylose content in a certain range is linked to a sticky and soft sensory experience and is usually preferred by consumers. Lower amylose content was detected under higher temperatures than normal temperatures in this research, consistent with other studies [9,10,21]. Evidently, a synergistic reduction in the rice amylose and crude protein content could not be achieved by nitrogen supplementation at heading, since nitrogen application would necessarily increase the grain protein content.

Rice pasting is a process in which rice starch undergoes irreversible swelling to lose its birefringence and crystallinity with the cooking temperature increase. In this research, the pasting temperature of rice grains always increased under elevated temperature, and higher Tp and Tc were also detected under HT in comparison with NT, suggesting that rice grown under a warming regime would require a higher cooking temperature and longer cooking time. In the rice endosperm, molecular amylose is easy to combine with lipoic acid to form a chelation structure, which would inhibit the pasting of starch granules [24]. Meanwhile, protein bodies exist around the amylopast, and they interact with the amylopast to form a network structure during rice pasting, thus restraining starch swelling, which would bring a harder mouthfeel and increase the difficulty for rice cooking [56]. The decrease in amylose content and rise in protein content jointly contributed to the increased pasting temperature, Tp for rice under a high growth temperature. There was little difference in the rice pasting temperature among the three nitrogen fertilizer application measures under high temperature; thereby, we summarized that rice cooking difficulty under higher temperature could not be improved by supplemental nitrogen fertilization at heading. The two broadcast nitrogen fertilizers (LBN and HBN) at heading both decreased the hot viscosity and final viscosity, which may be caused by their positive effects on protein accumulation as the results of correlation analysis (Table 9). In the DSC assessment system, ΔHr and R indicate the texture of cold rice, and the three nitrogen application measures had marked decreasing effects on the two parameters, inducing better grain quality in cold rice than that without nitrogen supplementation under high temperatures. In this research, consistency was not decreased by high temperature as the changing trend of amylose content, so we speculated that the increased protein content may inhibit the anticipative decreasing effects of lower amylose on consistency. LBN and FN significantly reduce consistency, and this tendency would induce a sticker mouthfeel of rice under the two treatments. However, HBN did not show a significant influence on consistency when rice was exposed to high temperature. Under high temperature, the response of consistency to HBN was not the same as its responses to LBN and FN, which may be explained by their different grain nitrogen contents. Evidently, HBN induced larger advantages in the competition of N metabolism with carbon metabolism when contesting their common substrate than LBN and FN. High temperature caused a significant increase in the rice pasting temperature, which is in agreement with previous studies [10,52,53].

Restricted to the limited climatic chambers, this experiment only tested the influence of two broadcast nitrogen application levels and one foliar nitrogen application level at heading on rice productivity, while foliar nitrogen application showed a positive regulation impact on yield and some grain quality parameters. More optimal nitrogen application schemes should be explored to cope with the adverse impact from high temperatures in future studies, and the regulation mechanism of nitrogen fertilizer is required to be further investigated. Additionally, the cost of nitrogen fertilizer application should also be weighted to satisfy the applicability in rice production.

5. Conclusions

High temperature during grain filling increased the rice plant temperature, damaged PSII, decreased net photosynthesis rate and caused a decline in the seed-setting rate and grain weight, resulting in yield loss. In addition, high temperature induced the decline in the head rice rate and the increase in chalk occurrence and pasting temperature, thereby deteriorating the rice quality. HBN and FN at heading both mitigated the negative influence of high temperatures on rice yield, and it was associated with lowering plant temperature and protection of the chlorophyll fluorescence system, whereas LBN did not present a positive impact on rice yield under high temperatures. The three nitrogen application measures at heading all improved the rice milling quality and reduced the chalk occurrence under high temperature, while HBN showed the best improvement effects. LBN and FN tended to make the texture of cooked rice softer by reducing the consistency, ΔHr and R under high temperature. Two broadcast nitrogen application methods did not have positive or even deteriorated rice cooking difficulty, while FN lowered all gelatinization parameters. Different from various plant growth regulators, particle and liquid nitrogen fertilizers are cheap and widely sold in rural agricultural materials markets, so that they could be easily adopted by farmers to deal with high-temperature stress. Recently, the nitrogen application method at heading has been enacted into core measures for rice to cope with high-temperature stress via an agronomic technique extension department of our province, which was based on our research results. Future studies should crystallize the detailed nitrogen application amounts for rice to deal with different high-temperature scenarios, and the deep remission physiological and molecular mechanism of nitrogen application must be explored in subsequent research.