Carbon isotope discrimination and the factors affecting it in a summer maize field under different tillage systems

Experimental site

The experiment was conducted in a continuous, long-term (14 years) conservation tillage experiment at the Experimental Station of Shandong Agricultural University (117°09′13.79″–117°09′12.02″E, 36°09′30.78–36°09′27.59″N); our experiment was performed over 2 years (2016–2017). The experimental area has a typical temperate continental climate with sufficient light and four distinct seasons. The annual average temperature is 13.6 °C. The annual average number of hours of sunlight is 2,462.3 h, and the annual average rainfall is 786.3 mm. This site has a climate that is typical for the NCP. The soil type tested was brown soil with a deep soil layer, and the groundwater level was below 5 m. Climate data for 2016–2017 is shown in Fig. 1. The soil texture is 40% sand, 44% silt, and 16% clay. The major initial properties within the 0–20 cm soil layer were as follows: pH of 7.09, 7.19 g·kg⁻¹ SOC, 1.3 g·kg⁻¹ total N, 0.79 mg·kg⁻¹ available P, and 41.32 mg·kg⁻¹ exchangeable K.

Experimental design

A split-plot design experiment was arranged with three replicates. The plot size was 15 m × 4 m for every replicate. The experiment was divided into the main plot factor: four tillage methods [no-tillage (NT), rotary tillage (RT), subsoiling (SS), and conventional tillage (CT)] and the subplot-factor: straw management [straw return (S) and straw removal (0)].

The winter wheat–summer maize was the typical double cropping rotation system. The winter wheat variety ‘Jimai22’ was sown in the middle of October each year and harvested in the middle of June the next year. The summer maize variety was ‘Zhengdan958’, which was sown in the middle and late June of each year and harvested in the first 10 days of October. During the summer maize growth period, basal fertilizer was applied at a rate of 120 kg N ha⁻¹, 120 kg P₂O₅ ha⁻¹, and 100 kg K₂O ha⁻¹ before sowing, and topdressing fertilizer were applied at a rate of 120 kg N ha⁻¹ at the large mouth stage.

The four tillage methods in this experiment were carried out only before winter wheat sowing, and summer maize was sowed with no-tillage iron stubble. After harvest, the wheat straw and maize straw from the two seasons were completely crushed (3–5 cm) and returned to the field. For the SS treatment, subsoiler (ZS-180) was used; the soil was plowed for the CT treatment and a rotavator (C250) was used for the RT. Winter wheat was grown using a machine (LXH-150). After the winter wheat was machine-harvested, the straw treatments were applied as described above, and summer maize was sown directly by machine (SHB-2). The field management strategies employed in this experiment were the same as those used in generally high-yielding fields. The specific operating procedures for the tillage treatments could view the supplementary document.

Test items and methods

In this study, through an ongoing field experiment involving long-term tillage and straw treatments, we measured the Δ¹³C of summer maize, soil water content (SWC), leaf area index (LAI), canopy microclimate, and CO₂ concentration, and systematically studied the characteristics of Δ¹³C among the different summer maize layers and the relationships between Δ¹³C and the factors that influence it under different tillage methods.

Collection and determination of gas samples

Two maize plants with good growth, basic consistency, and no pest or disease, were covered with a chamber in each plot during the jointing stage. The chamber was designed as a cabinet surrounded by a transparent plastic sheet without covering top to connect to the atmosphere (Fig. 2), the volume was 2.88 m³ (1.2 m in length, 0.8 m in width, and 3 m in height). The bottom of this chamber was not sealed, embedded into the ground and the joints are sealed with soil. The gas sample was collected by an L-shaped polyvinyl (PVC) pipe from the chamber.

There were divided into three layers [upper layer (U, 240 cm above the ground), middle layer (M, 120 cm above the ground), and lower layer (L, 30 cm above the ground)] to collect the gas samples by a micro-vacuum pump, each sample was packed in 0.5 L gas sampling bag. Then, we took them back to the laboratory, using a Shimadzu gas chromatograph and a stable isotope mass spectrometer to measure the CO₂ gas concentrations and δ¹³C. The measurements were completed within 2 weeks.

Statistical analysis

Paired T-test and One-way analysis of variance (one-way ANOVA) was performed using SPSS 20.0. Duncan’s multiple range test were used to perform multiple comparisons to calculated significant differences at the 5% level among mean value from various groups. Pearson method was used to determine the correlation between difference factors. The test data were processed using Microsoft Excel 2016. Origin 8.0 was used for drawing.

Changes in the Δ¹³C of summer maize

In both growing seasons, the Δ¹³C among the different summer maize layers was significantly influenced by both tillage methods and straw management (Table 1). Different straw management methods resulted in different Δ¹³C values, and straw return significantly increased Δ¹³C. Regardless of layer and growing seasons, the average Δ¹³C (5.21‰) under straw return was significantly higher than that under straw removal (5.09‰). The results showed that the fluctuations in Δ¹³C in summer maize were consistent across the four tillage methods. SS_S significantly increased the Δ¹³C. The Δ¹³C performance was ranked SS_S > CT_S > RT_S > NT_S, and there were significant differences in Δ¹³C under different tillage methods, which indicated that the tillage methods significantly affected the Δ¹³C of summer maize.

At different growth stages, the Δ¹³C among the different summer maize layers showed that the Δ¹³C at the pre-filling stage was higher than that at the post-filling stage. The lowest value of Δ¹³C was in M, and the order of Δ¹³C among the different summer maize layers was L > U > M.

Analysis of the factors affecting the Δ¹³C of summer maize

Soil water content

The SWC in the summer maize field decreased seasonally as the rainfall intensity decreased, and the trends for changes in SWC with time varied slightly between the 2 years of our study due to differences in the rainfall distributions between the 2 years (Figs. 1 and 3). The SWC at the pre-filling stage was higher than that at the post-filling stage. The SWC under the condition of straw return was higher than that under straw removal, which indicated that straw return enhanced soil water holding capacity, especially in the 0–10 cm soil layer. In the 0–10 cm soil layer, the 2 years’ average SWC of CT_S, SS_S, RT_S, and NT_S was increased by 4.82%, 5.68%, 5.30%, and 5.91%, respectively, under straw return, compared with the straw removal treatments.

Across the soil depth range of 0–60 cm, SWC decreased initially and then increased with increasing soil depth. In the 0–20 cm soil layer, the SWC was significantly lower under CTs than under the other treatments with straw return, and the SWC order was RT_S > NT_S > SS_S > CT_S. In the 20–40 cm soil layer, the SWC of SS_S was significantly higher than that of the other treatments, and the SWC of NT_S was the lowest among the different tillage treatments. In the 40–60 cm soil layer, the SWC performance ranking of the different tillage methods under straw return was NT_S > SS_S > CT_S > RT_S.

Leaf area index

The LAI of summer maize at the pre-filling stage was greater than that at the post-filling stage (Fig. 4). Overall, the LAI of the straw return treatment was greater than that of the straw removal treatment. At the pre-filling stage, LAI for SS_S, RT_S, and NT_S were higher than those for SS₀, RT₀, and NT₀, except for CT_S. Under straw return conditions, the order of the LAI of summer maize was NT_S > SS_S > CT_S > RT_S at the pre-filling stage, and it was SS_S > CT_S > RT_S > NT_S at the post-filling stage.

Under straw return conditions, the LAI of the NT_S treatment was significantly higher than that of other treatments at the pre-filling stage. However, at the post-filling stage, the LAI of NT_S treatment decreased and was the lowest among the tillage treatments. The average difference between the post-filling and pre-filling LAI for CT_S, SS_S, RT_S, and NT_S was 0.72, 0.73, 0.75, and 1.87, respectively. Hence, the LAI of NT_S decreased the most between pre-filling and post-filling, indicating that the leaf senescence rate under the NT_S treatment was higher than those under the other three tillage methods.

Air temperature in the summer maize canopy

The changes in air temperature among the different summer maize layers at the pre-filling and the post-filling stages for summer maize in 2016 and 2017 are shown in Fig. 5. The air temperature decreased with the increase in height, and the air temperature at the pre-filling stage was significantly higher than that at the post-filling stage (both straw removal and straw return treatments). The air temperature under straw return was significantly higher than that under straw removal, which indicated that straw return significantly affected air temperature.

At the pre-filling stage, the air temperature of CT_S was the highest in the lower layers of the summer maize canopy, whereas the highest temperature under NT_S was observed in the upper layer. At the post-filling stage, the air temperature of NT_S was highest among the three layers under straw removal in both two growing seasons, whereas SS_S was lowest.

Relative humidity in the summer maize canopy

Straw return decreased relative humidity, and the relative humidity was higher at the pre-filling stage than at the post-filling stage (Fig. 6). Relative humidity was higher in the lower layer of the summer maize canopy than it was in the middle and upper layers. The trends for the effects on relative humidity were similar among the different tillage methods and straw management conditions, with a high level of influence apparent in the lower layer.

At the pre-filling stage, the relative humidity under straw return was lower than that under straw removal. The results showed that the relative humidity of NT_S treatment was significantly lower than those of the CT_S, SS_S, and RT_S treatments in 2016 growing seasons under straw at the pre-filling stage. The average humidity in the CT_S, SS_S, RT_S, and ZT_S treatments was 49.96%, 48.16%, 48.28%, and 47.13%, respectively, which was 1.06%, 5.36%, 2.41%, and 1.67% lower, respectively, than that of average humidity under straw removal, indicating that straw return reduced soil water evaporative loss. Straw return and no-tillage treatments had high water retention, so the water dispersion under those treatments was lower during water shortages, and the relative humidity was low. In conclusion, the relative humidity of CT₀ was the highest, and NT_S was the lowest. Under the different tillage methods, straw return increased relative humidity in the summer maize canopy.

Analysis of Δ¹³C and its influencing factors

SWC, LAI, air temperature, relative humidity, and the CO₂ concentration can directly or indirectly affect the Δ¹³C of summer maize (Table 2). The results showed that the correlations of SWC, LAI, air temperature, relative humidity, and the CO₂ concentration with Δ¹³C were significantly different among the different layers of summer maize. Among them, SWC, LAI, and air temperature were significantly positively correlated with the Δ¹³C in the lower layer, and, for SWC and LAI, this correlation was highly significant. By contrast, the relative humidity and the CO₂ concentration had no significant correlation with Δ¹³C. In the middle layer, SWC, LAI, and air temperature significantly affected the Δ¹³C of summer maize, among which SWC and LAI with Δ¹³C had a very significant positive correlation. In the upper layer, the CO₂ concentration and air temperature had significant positive correlations with Δ¹³C, and SWC and LAI had an extremely significant positive correlation with Δ¹³C, whereas relative humidity had a very significant negative correlation with Δ¹³C. In addition, SWC and LAI were extremely significantly positively correlated with air temperature under the different summer maize layers, which indicated that the increases in SWC and LAI had a beneficial effect on the increase in air temperature, and they synergistically promoted or inhibited changes in air temperature, thereby affecting Δ¹³C; SWC and LAI had an extremely significant positive correlation, and the larger the SWC was, the larger the LAI was, indicating that water availability significantly affected the LAI of summer maize plants and promoted the growth of summer maize plants under good soil moisture conditions, thereby increasing the LAI. There was a significant positive correlation between the CO₂ concentration and LAI in the lower layer, and an extremely significant negative correlation between the CO₂ concentration and relative humidity.

According to the correlation coefficient, the rankings for the factors affecting the Δ¹³C of summer maize canopy in the lower, middle, and upper layers was LAI > SWC > air temperature > CO₂ concentration > relative humidity, LAI > SWC > air temperature > relative humidity > CO₂ concentration, and LAI > SWC > relative humidity > air temperature > CO₂ concentration, respectively. Under the different summer maize layers, various factors have different contributions to Δ¹³C. SWC, LAI, and air temperature were significantly or extremely significantly correlated with the Δ¹³C in L, M, and U, all of which were the main factors affecting Δ¹³C, and the CO₂ concentration played an important role in Δ¹³C in U. Interestingly, we found that there was a negative correlation between relative humidity and Δ¹³C at different canopy heights in summer maize, among which, in U, relative humidity and Δ¹³C had a significant negative correlation.

The relationship between the Δ¹³C of summer maize and yield

The results of the correlation analysis for summer maize Δ¹³C and yield under both straw removal and straw return treatments is shown in Fig. 7. It can be seen that the Δ¹³C of summer maize at the pre-filling stage showed a significant positive correlation with the summer maize yield. The Δ¹³C of summer maize in the middle of the canopy had a very significant positive correlation with the summer maize yield, with a significance level reaching 0.0001. However, there was no significant correlation between Δ¹³C and yield at the post-filling stage in summer maize.

The Δ¹³C at different layers in the summer maize canopy

Temperature, moisture, and the atmospheric CO₂ concentration all affect the plant leaf Δ¹³C value (Li et al., 2017). Under different tillage systems, Δ¹³C at the pre-filling stage was greater than that at the post-filling stage, and the plants fractionated heavy isotopes more vigorously at the pre-filling stage, that is, their ability to distinguish ¹³CO₂ was stronger. This was due to the senescence of summer maize plants at the post-filling stage and the accompanying waning of the photosynthetic assimilation capacities of the summer maize plants. The Δ¹³C values were significantly different among the different summer maize layers. The lowest value of Δ¹³C was found in the middle of the summer maize canopy, whereas the highest Δ¹³C was in the lower maize canopy. The Δ¹³C order among the different summer maize layers was L > U > M. There may be two main reasons: soil respiration emits large quantities of CO₂ to the atmosphere, resulting in a lower δ¹³C near the ground atmosphere. Liu et al. (2019c) suggested that the contribution of summer maize field soil emissions to photosynthesis is 20.37–29.03%. As the canopy height increases, the ratio of the CO₂ released from the soil to atmospheric CO₂ gradually declines. To a certain extent, increasing soil carbon emissions can increase the rate of maize photosynthesis. Plant photosynthetic carbon assimilation is discriminatory, and its reaction matrix is mainly ¹²CO₂. Similarly, if respiration releases more CO₂, predominantly containing ¹²C, the δ¹³C of soil CO₂ will be lower than that of the plant canopy; that is, the Δ¹³C will be greater in the lower layer than in the middle layer, as it was in our study. Secondly, due to photosynthesis, the canopy took the lead in assimilating ¹²CO₂, which gradually increased the δ¹³C value of the canopy CO₂. The upper canopy leaves had a large light-receiving area and were distant from the source of CO₂ generated by soil respiration (i.e., the ground). This is consistent with the research results of Silveira et al. (1989) that the forest canopy has obvious stratification phenomenon, and the closer to the forest surface, the more obvious the isotope depletion of plant leaves.

In the summer maize growing season, the Δ¹³C values of SS_S, CT_S, and RT_S were significantly higher than that of NT_S. The main reasons may be as follows: first, the degree of soil disturbance under SS_S, CT_S, and RT_S was higher than that under NT_S. During the summer maize growing season, the temperature was higher, the rainfall was frequent, so the soil temperature and soil moisture were also high, which provided the theoretical conditions for soil respiration. Soil respiration releases a large amount of CO₂. Second, tillage improves soil permeability, enhances the migration and diffusion of gases in the soil, promotes the decomposition of organic matter, and increases the release of CO₂ from soil respiration. Compared with the other tillage methods, NT_S effectively reduced soil disturbance and the chance of soil-air contact, reduced the decomposition of soil organic matter, and significantly reduced soil respiration, which was consistent with previous research results (Six et al., 2004; Yonemura et al., 2014). Different straw management conditions resulted in different Δ¹³C values. In the two periods, the Δ¹³C of the straw return treatment was significantly higher than that of the straw removal treatment. The main reasons may be as follows: firstly, the plant straw that was returned to the soil is a carbon source, and the straw was decomposed into various organic and inorganic substances, providing many nutrients for plant growth and releasing CO₂ (Negassa et al., 2015; Song et al., 2017); secondly, straw return improved the physical and chemical properties of soil, maintained soil structure and fertility (Jones et al., 2005), promoted the growth and extension of plant roots, and promoted the growth of the aboveground plant parts, increased the LAI, and improved photosynthesis. The increase in photosynthesis increased the capture of CO₂, and finally affected the Δ¹³C. The lower Δ¹³C for summer maize may reduce lower photosynthesis resulted lower yield. And Wei et al. (2019) found that the more ¹³C-photosynthate distribution to ears, and the less ¹³C-photosynthate distribution to stems could achieve a greater yield consequently. In future studies, we will investigate whether the decrease of Δ¹³C reduced photosynthesis for summer maize.

Influencing factors of Δ¹³C

Among the five factors (i.e., SWC, LAI, air temperature, relative humidity, and CO₂ concentration) that affect the Δ¹³C of the summer maize canopy, there was variation in the effects of these factors at different heights within the maize canopy, and the relative contribution of each factor to Δ¹³C differed. From the correlation analysis of these various factors to Δ¹³C, it can be seen that SWC, LAI, and air temperature were the important factors affecting Δ¹³C in the lower, middle, and upper layers. However, there was only a weak correlation between relative humidity and the Δ¹³C in U, which may be related to the influence of air flow and solar radiation in the upper level of the summer maize canopy. Studies have shown that temperature has a negative correlation with plant δ¹³C (Morecroft & Woodward, 1990). The results of this experiment demonstrated that temperature was significantly positively correlated with plant Δ¹³C. Plant growth was largely affected by water availability. There have been many studies on the relationship between water availability or precipitation and the composition of plant Δ¹³C. For example, Tambussi, Bort & Araus (2007) reported that Δ¹³C can be used to predict the WUE of C₃ plants. Badeck et al. (2005) found that different crops or crop organs have different isotopic compositions. The Δ¹³C values of different plant organs indicate that their WUE varies. In short, Δ¹³C can indirectly indicate crop yield and WUE (Khan et al., 2007; Chen, Chang & Anyia, 2011; Gresset et al., 2014). SWC and LAI have important effects on air temperature, and SWC, LAI, and air temperature are all very significantly related, indicating that there is some interaction between the three, and mutual adjustment affects the canopy Δ¹³C. When SWC decreases, crop physiological indicators, including air temperature and stomatal conductance, will first show some characteristics of change. Most of the water absorbed by crops is lost through leaf transpiration to maintain a balanced leaf temperature. If the soil moisture is insufficient, and the evapotranspiration intensity in the air is very large, it will cause the leaf transpiration to decrease and the leaf surface temperature to increase. Webber et al. (2016) showed in their study that the air temperature is related to the SWC, and the influence of SWC on crop air temperature in that study was more obvious.

The Δ¹³C of summer maize is comprehensively controlled by many factors such as SWC, LAI, air temperature, relative humidity, and other factors. Under different field conditions, the contribution of each factor to Δ¹³C is different, and the combined effects of multiple factors may have a complex interaction with Δ¹³C. Further studies are required to clearly define the quantitative relationship between factor interactions for a deeper study of the characteristics of Δ¹³C.