Rice canopy temperature is affected by nitrogen fertilizer

Min Jiang ,Zhang Chen ,Yuan Li ,Xiaomin Huang,Lifen Huang,Zhongyang Huo#

1 Jiangsu Key Laboratory of Crop Genetics and Physiology/Jiangsu Key Laboratory of Crop Cultivation and Physiology/Agricultural College, Yangzhou University, Yangzhou 225009, China

2 Jiangsu Co-Innovation Center for Modern Production Technology of Grain Crops, Yangzhou University, Yangzhou 225009, China

Abstract Canopy temperature strongly influences crop yield formation and is closely related to plant physiological traits.However,the effects of nitrogen treatment on canopy temperature and rice growth have yet to be comprehensively examined. We conducted a two-year field experiment with three rice varieties (HD-5,NJ-9108,and YJ-805) and three nitrogen treatments (zero-N control (CK),200 kg ha–1 (MN),and 300 kg ha–1 (HN)). We measured canopy temperature using a drone equipped with a high-precision camera at the six stages of the growth period. Generally,canopy temperature was significantly higher for CK than for MN and HN during the tillering,jointing,booting,and heading stages. The temperature was not significantly different among the nitrogen treatments between the milky and waxy stages. The canopy temperature of different rice varieties was found to follow the order: HD-5>NJ-9108>YJ-805,but the difference was not significant. The canopy temperature of rice was mainly related to plant traits,such as shoot fresh weight (correlation coefficient r=–0.895),plant water content (–0.912),net photosynthesis(–0.84),stomatal conductance (–0.91),transpiration rate (–0.90),and leaf stomatal area (–0.83). A structural equation model (SEM) showed that nitrogen fertilizer was an important factor affecting the rice canopy temperature.Our study revealed: (1) A suite of plant traits was associated with the nitrogen effects on canopy temperature,(2) the heading stage was the best time to observe rice canopy temperature,and (3) at that stage,canopy temperature was negatively correlated with rice yield,panicle number,and grain number per panicle. This study suggests that canopy temperature can be a convenient and accurate indicator of rice growth and yield prediction.

Keywords: canopy temperature,rice,physiological and biochemical characteristics,yield

1.Introduction

Usually,temperature has been used to evaluate the growing environment of rice,but even at the same temperature,rice canopy temperature may be significantly different. The canopy temperature of rice consists of its thermal characteristics and physiological response to the environment,including two energy transfer processes:(1) the heat transfer between rice and the environment through conduction,convection,radiation,and latent heat evapotranspiration;(2) heat transfer within rice tissues. Environmental factors,such as water,heat,air,and fertilizer,affect the canopy temperature of rice(Winterhalteret al.2011;Yanet al.2012;Caiet al.2016;Monet al.2016). A good linear relationship has been observed between air and canopy temperatures under different growth conditions. Rice has a certain ability to fine-tune its canopy energy balance with changes in ambient temperature,and the changes in rice canopy temperature were consistent with air temperature.However,changes occurred slightly later and to a lesser extent than air temperature,which could maintain a relatively stable state (Liuet al.2018). Compared to climate,varieties and cultivation methods also impact rice canopy temperature. Common field management practices,such as fertilizer and water,can directly affect the growth of rice populations and thus affect the canopy temperature (Perdomoet al.2017;Wang Xet al.2020).For example,rice will reduce stomatal opening under drought conditions to save water,thereby reducing the CO2exchange. In this case,leaf cooling mainly relies on physical traits,while nitrogen fertilizer is the most active nutrient factor affecting rice growth. Hikosakaet al.(2014)found that optimizing the nitrogen distribution (increasing and reducing the nitrogen contents in the upper and lower leaves,respectively) and matching nitrogen distribution in the canopy with light distribution could significantly improve the photosynthetic capacity of the canopy. Many studies have also considered canopy temperature as an important indicator of stress physiology,such as drought and heat resistances (Ahmedet al.2020;Carvalhoet al.2020).

Therefore,studying the temperature difference of rice canopy and its mechanism is helpful for accurately understanding the microenvironment of rice. Previous studies on rice canopy temperature have mostly focused on one or several traits. However,in addition to transpiration,many factors,such as leaf area index,plant water content,dry matter accumulation,and photosynthetic characteristics,can also affect rice canopy temperature. Tanner first applied infrared thermometry to plant temperature measurement in 1963 (Tanneret al.1963),and this technique has the advantages of convenience,speed,non-destructiveness,and relatively small errors. It is widely used in the research on the canopy temperature of crops,such as gramineae and legumes (Kumaret al.2017),that were not caused by climate. However,due to the limitation of the efficiency and accuracy of canopy temperature measurement methods,they have been rarely used in production.With the development of drones and thermal imaging technology for agriculture,unmanned aerial vehicles(UAVs) equipped with high-precision infrared thermometry devices can achieve high-throughput,nondestructive,and large-area evaluation (Wanget al.2013;Deeryet al.2016),thus bridging the gap between rice canopy temperature and production applications,which provided an efficient measurement method and technical basis for this study.

This study aimed to explore the variation in rice canopy temperature during the entire growth period and analyze the relationship between canopy temperature,biological traits (e.g.,transpiration,photosynthesis,stomatal area,water content,dry matter accumulation,and leaf area index),and yield. In this study,the relationship model and physiological mechanism of “canopy temperaturebiological trait-yield” were finally constructed,which could explore the formation process and regulation mechanism of rice canopy temperature and provide a scientific basis for the application of canopy temperature as an important indicator for rice cultivation and variety selection(breeding).

2.Materials and methods

2.1.Plant materials and growth conditions

The experiment was conducted at the Yangzhou Wantou Experimental Farm at the Lixiahe Agricultural Research Institute of Jiangsu Province,China (32°39´N,119°42´E)during the rice growing season (May to October) of 2017 and was repeated in 2018. The farm is located in the center of the Lixia River in Jiangsu,which belongs to the humid climate zone of the northern subtropics. The average annual sunshine,temperature,and precipitation of the experimental site were 2,305.6 h,15°C,and 1,024.8 mm,respectively. The daily average temperature and precipitation during the experimental period across the two study years are shown in Fig.1. The soil type in the area was sandy loam with 25.5 g kg–1organic matter,1.21 g kg–1total N,25.7 mg kg–1Olsen-P,and 85.7 mg kg–1exchangeable K in the 0–15 cm layer of topsoil.

Threejaponicarice (O.sativaL.) varieties,Huaidao 5(HD-5),Nanjing 9108 (NJ-9108),and Yangjing 805 (YJ-805),being currently used for local production were grown in the paddy fields. The three varieties have a similar growth period,ranging from 150 to 153 d from sowing to physiological maturity. The seedlings were raised in the seedbeds on May 17,2017 and May 20,2018 and transplanted on June 13,2017 and June 15,2018,respectively,at a hill spacing of 0.12 m×0.30 m with two seedings per hill. Phosphorus and potassium were evenly spread over the topsoil at a rate of 30 and 40 kg ha–1as basal fertilizers before transplanting. The experiment followed local agronomically recommended management practices for rice production,and water,weeds,insects,and diseases were controlled as required to avoid yield loss. Plants were harvested on October 15,2017 and October 18,2018.

2.2.Treatments

The experiment was arranged in a completely randomized design and comprised three N rate treatments: zero-N control (CK),200 kg ha–1(MN),and 300 kg ha–1(HN). N was split into three application phases: 40% as basal fertilizer,10% at tillering,and 50% at panicle initiation.Each plot was 8 m long and 5 m wide,with an area of 40 m2. All treatments were replicated thrice. The plots were separated by ridges,and individual small plots were connected with irrigation channels,ensuring that each plot could be independently irrigated or drained.

2.3.Sampling and measurements

The test was conducted using an unmanned aerial vehicle (UAV) technology for an infrared temperature measuring device with high-precision and high flux to obtain nondestructive and large-scale rice canopy thermal infrared images;a ZENMUSE-XT thermal imaging camera (sensitivity<0.05°C) was installed directly to the“DJI M210” Inspire 1 interface of Dajiang UAV. Sunny,windless days were selected and measured at 12 noon.The flying height of the UAV was 15 m from the ground,which effectively prevented the airflow generated by the propeller from affecting the canopy temperature measurement. The determination periods were tillering,jointing,booting,heading,milky,and waxy stages. The starting time of the main growth period was determined by the growth relationship of each organ and leaf age or remaining leaf primordium number. When reading the canopy temperature data from infrared images collected by the UAV,the effect of bare land was avoided,especially during the tillering and jointing stages,when the effect of bare land was significantly obvious. The middle area was selected for each hill,and the canopy temperature of rice in 15 hills was measured for each treatment. The images collected in different periods are shown in Fig.2. A ZDR-20 data recorder was used for measuring the atmospheric temperature (Ta) at a distance of 1.5 m from the ground,simultaneously with the canopy temperature.

Have you not heard that my son feeds upon Christians9? But he said no, and that he would not go, for he was so miserable10 that it was all one to him whether he died or not; that he had lost everything, and especially a splendid palace like none other in the whole world, for it had laths of gold and tiles of diamond, and all the furniture was of silver and gold

Fig. 1 Daily temperature,sunshine hour,and precipitation during the rice growing seasons in the experimental fields in 2017 and 2018.

The five hills of the representative plants were selected based on the average number of tillers in each plot. They were water-removed for 30 min at 105°C and dried to a constant weight at 80°C. Dry matter accumulation,plant water content,root–shoot ratio,leaf area index,and plant height were measured.

The net photosynthetic rate (Pn),stomatal conductance(Gs),intercellular carbon dioxide concentration,and transpiration rate of the flag leaves were measured using an LI-6400 portable photosynthetic meter manufactured by LI-COR,USA. The test was conducted 9:00–11:00 a.m.on a sunny day using red and blue light sources when the temperature was 28–30°C,photosynthetic photon flux density was 1,400 μmol m–2s–1,and CO2concentration in the leaf chambers was 380 μmol mol–1.

The flag leaves were taken from rice at the heading stage to observe the stomata,and three rice plants with moderate growth were selected from different treatments.Flag leaves from the main stem of each plant were cut offand stored in the formalin–aceto–alcohol (FAA) fixation solution. The middle of the leaf surface was observed using a TM-100 table scanning electron microscope(ZEISS GeminiSEM 300,Carl Zeiss AG,Oberkochen,Germany),and stomatal length and width were measured.Stomatal density was observed in three visual fields of each leaf segment (small pieces of the middle segments of the leaves were cut).

A total of five observation points were taken from each plot during the maturity stage,and 10 hills from each point(50 hills in total) were taken to determine the number of panicles. Based on the average number of panicles,10 hills were selected in each plot,the number of grains per panicle and seed-setting rate were measured,and 1,000 solid grain samples (dry seeds) were weighed and repeated thrice (error<0.05 g).

2.4.Statistical analysis

Basic data processing and statistical analyses were performed using Microsoft Excel 2019 and SPSS 16.0.The figures were drawn using GraphPad Prism version 8.Structural equation modeling (SEM) was used to explore the potential mechanism and quantify the direct and indirect influences of N on rice canopy temperature and stomatal area. Rigorously,the linear models in SEM can only deal with continuous variables (Huanget al.2022),so we constructed an a priori model based on the literature review and our understanding of the relationship between these predictors. The model adequacy was determined byχ2tests (P>0.05),goodness-of-fit index (>0.9),and comparative fit index (>0.9). Additionally,we examined the factors (TrandGs) for our theoretical models one by one as an independent model rather than putting them together in only one structural equation model (Grace 2006). We performed SEM using Amos 24.

Fig. 2 Images indicating canopy temperatures of Nanjing 9108 under different nitrogen treatments in different stages captured by the drone. Each plot was 8 m long and 5 m wide with an area of 40 m2. CK,zero-N control;MN,200 kg ha–1;HN,300 kg ha–1.

3.Results

3.1.Variation in rice canopy temperature under different nitrogen treatments

In this experiment,the canopy temperature was measured in six main rice growth periods. In the two years,the canopy temperature of the three rice varieties was measured twelve times. The analysis of variance (Table 1)showed that the influence of air temperature and nitrogen treatment on the canopy temperature of rice reached a significant level,but no significant difference was observed in the canopy temperatures of different rice varieties. The interaction between nitrogen fertilizer and air temperature,variety,and air temperature reached a significant level,but the interaction between nitrogen fertilizer and variety was not found to be significant. The interaction of nitrogen fertilizer,variety,and air temperature was observed to have a significant effect on canopy temperature.

Table 1 Analysis of variance (F-values) of canopy temperature among air temperature,varieties,and nitrogen treatments

Canopy temperature variations in the rice tillering,jointing,booting,heading,milky,and waxy stages in 2017 and 2018 are shown in Fig.3. In the tillering,jointing,booting,and heading stages,canopy temperature variations under different nitrogen treatments for each year were consistent and followed the order of CK>MN>HN,with the difference reaching a significant level (P=0.05). In the milkyand waxy stages,no significant differences were observed in the rice canopy temperature under different nitrogen treatments,and the canopy temperature of different rice varieties followed the order of HD-5>NJ-9108>YJ-805,but the difference was not significant. The change in rice canopy temperature was synchronized with air temperature.At 12:00 p.m.,the rice canopy temperature was lower than the air temperature,with the difference for all the treatments being the largest at the jointing stage. The canopy temperature of YJ-805 under high nitrogen treatment was found to be 8°C lower than the air temperature.

3.2.Relationship between rice canopy temperature and biological traits

Fig. 3 Changes in rice canopy temperature for different nitrogen treatments in different stages. Ta,atmospheric temperature;CK,zero-N control;MN,200 kg ha–1;HN,300 kg ha–1. Bars mean SE (n=15).

The shoot fresh weight,plant water content,leaf area index,leaf SPAD (soil and plant analyzer development),and root-shoot ratio of rice under different nitrogen treatments were significantly different,with the correlation coefficients of rice canopy temperature and the above traits remaining consistent between different years. The period with the highest correlation was the heading stage(Table 2),during which the canopy temperature best reflected the growth characteristics of rice. Among the biological indexes of rice,shoot fresh weight and plant water content showed the highest correlation coefficients with canopy temperature (Fig.4),which were–0.902 and–0.913 in 2017,and–0.895 and–0.912 in 2018,respectively. The correlation coefficient between root–shoot ratio and canopy temperature was not found to be significant at each growth stage. The canopy temperature of rice mainly reflected the growth characteristics of above-ground plants,while the vegetative growth of rice slowed down or even stopped during the milky and waxy stages,with the correlation coefficients between canopy temperature and biological traits of rice being small.

Table 2 Correlation coefficient between canopy temperature and biological characteristics during different stages

3.3.Relationship between canopy temperature,yield,and its components in rice

Canopy temperatures during the booting and heading stages were negatively correlated with yield,panicle number,and grain number per panicle,positively correlated with thousand-grain weight,and showed some correlation with the setting percentage (Table 3).However,the correlation coefficients between canopy temperatures,yield,and its component factors during the milky and waxy stages were small.

Table 3 Correlation coefficients between canopy temperature,yield,and elements of yield during different stages

The heading stage was found to be the best stageto observe the rice canopy temperature,where the correlation coefficient between rice canopy temperature,yield,and its components was found to be the highest(Fig.5). The relationship between theoretical yield (Y)and canopy temperature (X) wasY=–1.471X+51.76,R2=0.8901 in 2017,andY=–0.9948X+28.33,R2=0.7756 in 2018,with both reaching significant negative correlations.A significant negative correlation was observed between canopy temperature and panicle number per hectare(R2=0.9138 and 0.7846 in 2017 and 2018,respectively)and grain number per panicle (R2=0.6011 and 0.4737 in 2017 and 2018,respectively);however,a significant positive correlation was observed between canopy temperature and thousand-grain weight (R2=0.6368 and 0.3911 in 2017 and 2018,respectively).

Fig. 4 Correlation between biological characters and rice canopy temperatures during the heading stage. CK,zero-N control;MN,200 kg ha–1;HN,300 kg ha–1 .

3.4.Relationship between canopy temperature,photosynthetic indexes,and stomatal characters of rice during the heading stage

With the increase in nitrogen fertilizer,the net photosynthetic rate,stomatal conductance,and transpiration rate of rice significantly increased during the heading stage,with the change in the intercellular carbon dioxide concentration being irregular. This trend was found to be consistent among the different varieties in different years (Table 4). Taking the data of 2018 as an example,Pnof HD-5 in the MN treatment was 26.71% higher than that in the CK treatment,and it was 37.86% higher in the HN treatment than in the MN treatment;Trof HD-5 in the MN treatment was 16.40% higher than that in the CK treatment,and it was 19.29% higher in the HN treatment than in the MN treatment;the change rule for all varieties was essentially the same.

Fig. 5 Correlation between yield and its constituent factors and rice canopy temperature during the booting and heading stages.CK,zero-N control;MN,200 kg ha–1;HN,300 kg ha–1. ＊＊,significant at P=0.01.

The stomatal length and width of the rice leaves did not significantly change under different nitrogen treatments,but their area and density significantly changed. The trends of the different varieties in different years were found to be consistent (Table 5;Fig.6). The stomatal length of rice did not significantly change under different nitrogen treatments and was approximately 15 μm. The stomatal width increased with increasing nitrogen application. The stomatal width of YJ-805 in the HN treatment was found to be the highest,which was 107.96% higher than the smallest width,which was for HR-5 in the CK treatment. The stomatal area of rice significantly increased with the application of N fertilizer.In 2018,the stomatal area of HD-5 was 62.13% higher in the HN treatment than in the CK treatment. The stomatal area of NJ-9108 was 42.28% higher in the HN treatment than in the CK treatment,and that of YJ-805 was 58.67%higher in the HN treatment than in the CK treatment.Stomatal density was found to be significantly higher in the MN treatment than in the HN and CK treatments.

Table 5 Effects of different nitrogen treatments on rice stomatal characters during the heading stage1)

The correlation coefficients ofPn,Gs,Tr,SA,and rice canopy temperature during the heading stage in 2018 were–0.84,–0.91,–0.90,and–0.83,respectively (Fig.7),showing a significant negative correlation,andPn,Gs,Tr,and SA indicators also showed high correlations with each other. The SEM analysis model showed that nitrogen fertilizer was the most important factor affectingthe rice canopy temperature (Fig.8),which directly impacted the stomatal area of the rice leaves,thus,affecting transpiration rate and stomatal conductance,and indirectly affecting canopy temperature.

4.Discussion

4.1.Effect of N on rice canopy temperature

Rice canopy temperature is not only determined by environmental factors but also reflected by its own characteristics (Yanget al.2019). N addition can stimulate rice growth and change the canopy microclimate of paddy fields. The results of this study confirmed the significant differences between the rice canopy temperatures under different nitrogen treatments. The rice canopy absorbs solar radiation and converts it into heat energy,which increases the canopy temperature.Therefore,the canopy temperature of rice is consistent with the air temperature,and the time of its maximum value occurs slightly later than that of the air temperature.On the other hand,the photosynthesis and transpiration of rice consume heat,cool the leaves,and reduce the canopy temperature (Caiet al.2020;Gaoet al.2022).

Fig. 6 Changes in rice stoma under different nitrogen treatments during the heading stage. CK,zero-N control;MN,200 kg ha–1;HN,300 kg ha–1.

Fig. 7 Correlation heat map between photosynthetic stomatal indexes and rice canopy temperature during the heading stage. Ct,Pn,Gs,Ci,Tr,SL,SW,SA,and SD represent the canopy temperature,net photosynthetic rate,stomatal conductance,intercellular CO2 concentration,transpiration rate,stomatal length,stomatal width,stomatal area,and stomatal density,respectively.

Fig. 8 Path diagrams of the structural equation modeling (SEM) of the relationship of N with stomatal area (SA),transpiration rate(Tr),stomatal conductance (Gs),and canopy temperature (CT).

The physiological and biochemical activities of rice were found to be significantly different under different nitrogen treatments,with the canopy temperatures also being significantly different. At the highest temperature time of the day,the ambient temperature and light intensity were high,and the physiological and biochemical effects of the rice population were strong. At this time,the rice canopy temperature for all treatments was lower than the air temperature,with the difference between the canopy and air temperatures being the highest,which was more conducive for observing and studying the rice canopy temperature.

4.2.Effect of rice growth on canopy temperature

Many studies have shown that temperature plays an important role in the growth and development of rice(Penget al.2004;Dinget al.2020;Xuet al.2021).In this study,the difference between canopy and air temperatures was found to be the largest in the shooting stage,with the correlation coefficient between canopy temperature,physiological and biochemical traits,and yield being the highest in the heading stage,which might be a convenient and accurate indicator to reflect rice growth and forecast yield. Rice canopy temperature is closely related to plant morphological structure,physiological metabolism,and microstructure,such as stoma. According to Pamplonaet al.(1995),when limiting factors,such as water,fertilizer,and light,were satisfied,the rice yield was increased by 1–1.5 t ha–1when the canopy temperature dropped by 1°C. This study found that rice with low canopy temperatures showed stronger tillering ability,a larger LAI index,and a smaller light transmittance in the field. Simultaneously,high plant water content and SPAD were also found to be conducive to the formation of lower canopy temperatures(Zhouet al.2021). However,the effect of nitrogen on rice growth is comprehensive,and when multiple leaf traits are involved,isolating the most important factor affecting the rice canopy temperature is difficult (Fukudaet al.2021).

4.3.Effect of rice physiological factors on canopy temperature

The canopy temperature of rice is closely related to its external traits,plant physiological and metabolic characteristics,and organ structure characteristics.Therefore,we further studied the photosynthetic capacity and stomatal characteristics of rice flag leaves during the heading stage and found a significant negative correlation between canopy temperature andPn,Gs,Tr,and SA indexes of the flag leaves. The leaf’s physiological–biochemical and physical traits affect leaf temperature by influencing heat transfer,storage,and received radiation (Levizou and Kyparissis 2016). The SEM results of this study also showed that nitrogen fertilizer mainly affected the stomatal area of the rice leaves,thus changing the leaf transpiration rate and stomatal conductance and affecting the rice canopy temperature. Hillet al.(2014) also found that larger vein density and stomatal conductance created a suitable leaf temperature,high photosynthesis,and transpiration,which in turn dominated the leaf cooling. Linet al.(2017)found that plants showed stronger cooling ability under the conditions of high leaf reflectance,low absorption rate,and high transpiration rate and also showed larger stomatal area in the leaf anatomical structure.These results combined to reveal that leaf movement also influences leaf temperature. Transpiration is an important index for reducing the surface temperature of rice,and it releases water and removes heat.Under high temperatures or water stress,rice reduces transpiration by opening and closing the stoma to maintain its relative water content (Calet al.2019). This study demonstrated that the higher the physiological and biochemical activities,the lower the rice canopy temperature. In particular,the correlation coefficient between canopy temperature and yield was the highest during the heading stage. Hence,the canopy temperature demonstrated an important indicating significance and can serve as an indicator for improving the screening method in rice breeding and cultivation management in the future.

5.Conclusion

This study emphasized the differences in rice canopy temperature and the relationship between rice canopy temperature and biological traits under the conditions of N addition. The repeated experimental data of two years showed that the effect of different nitrogen treatments on rice canopy temperature reached a very significant level,and the air temperature also had a significant effect on rice canopy temperature. The air temperature reached the highest at the jointing stage,whereas the rice canopy temperature was relatively low. The jointing stage demonstrated the largest temperature difference throughout the whole growth period. Rice changed from vegetative growth (growing roots,leaves,and tillers) to reproductive growth (growing jointing and stalks),and the canopy temperature was low at the jointing stage. At the heading stage,the correlation coefficient between canopy temperature and population index,photosynthetic stomatal index,and yield component of rice was the highest during the whole growth period. At this time,the rice canopy temperature was significantly negatively correlated with fresh population weight and plant water content,and it was negatively correlated withPn,Gs,Tr,and stomatal area of rice flag leaves. Nitrogen fertilizer affects canopy temperature by affecting the stomatal area,transpiration rate,and stomatal conductance of rice leaves,so canopy temperature is a comprehensive index that can reflect the growth status of the rice population and individual physiological and metabolic capacity. In addition,superior biological traits are associated with lower canopy temperature.

Acknowledgements

This work was supported by the National Key Research and Development Program of China (2022YFD1500404),the National Natural Science Foundation of China(31801310),the Natural Science Projects of Universities in Jiangsu Province,China (21KJA210001),and a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD),China.

Declaration of competing interest

The authors declare that they have no conflict of interest.