Soil organic carbon and nitrogen storage under a wheat (Triticum aestivum L.)—maize (Zea mays L.) cropping system in northern China was modified by nitrogen application rates

Experimental site and climatic conditions

This long-term study was established in July 2012 at Zhangpan Town, Jian’an District, Xuchang City, Henan Province (33°59′N, 113°58′E), China, and field management practices were treated in the same way. Observations for this study were made during the 2017–2018 growing season. The area has a temperate monsoon climate, with annual mean precipitation and temperature of 712 mm and 14.6 °C, respectively. About 60% of the precipitation falls between the months of July and September. The predominant cropping system is double-cropping rotation of winter wheat and summer maize.

The soil type is mortar black soil, with 13.04 and 10.43 g kg⁻¹ organic matter, 0.72 and 0.85 g kg⁻¹ total N, 0.79 and 0.84 g kg⁻¹ total phosphorus, 6.58 and 6.99 g kg⁻¹ total potassium, and bulk density were 1.39 and 1.47 g cm⁻³ for wheat and maize in the 0–20 cm soil profile, respectively. Agricultural production levels and the natural conditions of the experimental site are representative of high-yielding winter wheat production in this region.

Experimental design and field management

The high-yielding wheat and maize cultivars of Aikang 58 and Denghai 678 respectively were used in this study, representing the most commonly planted wheat and maize cultivar in Huanghuai area, wheat and maize were sown in sequence in the same plot. The field experiment consisted of three replicates for each N treatment applied in a randomized block design with four N application rates of 0, 180, 240, and 300 kg ha⁻¹ (N0, N180, N240, and N300, respectively). Each plot was 10 × 11 m, with a seeding rate of 187.5 and 30 kg ha⁻¹ for wheat and maize, respectively. At the time of sowing, chemical fertilizer comprising 60% urea and 40% sulfur-coated urea was applied by hand before plowing at a rate of 66 kg P ha⁻¹ and 72 kg K ha⁻¹. Prior to sowing, plowing was carried out to remove pests and weeds, and during the growing period, weeds were hand-hoed several times.

Soil samples and preparation

After the wheat and maize were harvested, the root system was excavated with shovel to a size of 20 cm (length) × 20 cm (width) × 40 cm (depth) then washed with deionized water, and the above-ground parts were sampled with scissor or axe from the root. Soil samples were collected from the soil layers (every 20 cm up to a depth of 60 cm) in all 12 subplots on May 28 and September 28, 2018 subsequently. Three soil cores (inner diameter: 5.0 cm) were randomly sampled from each subplot then mixed to form one fresh sample for each depth increment. Visible plant material was removed then soil samples were divided into three parts and transported to the laboratory, where they were air dried before passing through a two mm screen. Roots and other debris were then removed before further analysis. The bulk density (BD) of the soil was measured in different layers using a soil bulk sampler with a stainless steel cutting ring (inner diameter: 5.0 cm, height: 5.0 cm) at points adjacent to the soil sampling plots. First, the volume of each soil core was measured then the dry mass was determined after oven drying at 105 °C.

Analysis of soil physical and chemical properties

The dichromate oxidation method was used to measure the SOC and the Kjeldahl method (Kjeltec 2300 Analyzer Unit; Foss, Hogana, Sweden) was used to analyze the TN (Bremner, 1996; Nelson & Sommers, 1982). Soil pH was measured using a pH meter after shaking for 30 min with a soil: water suspension of 1: 2.5 w/v. Each measurement was performed in triplicate.

Analysis of soil C and N stocks and soil C and N sequestration

Since no coarse fraction (>2 mm) was found in the soil samples, the following equation was used to calculate the soil C stock (C_s, Mg ha⁻¹): (1)${\displaystyle {\mathrm{C}}_{\mathrm{s}}=\left(\mathrm{SOC}\times \mathrm{BD}\times \mathrm{D}\right)/10}$ where BD is the soil bulk density (g cm⁻³), SOC is the soil organic carbon concentration (g kg⁻¹) and D is the depth of the sampled soil layer (cm). The following equation was used to calculate the soil N stock (N_s, Mg ha⁻¹): (2)${\displaystyle {\mathrm{N}}_{\mathrm{s}}=\left(\mathrm{TN}\times \mathrm{BD}\times \mathrm{D}\right)/10}$ where BD and D are as above and TN is the total N concentration (g kg⁻¹) of the soil.

The C_s and N_s values were then used to determine C and N sequestration (ΔC_s and ΔN_s, Mg ha⁻¹) after maize planting as follows: (3)${\displaystyle \Delta {\mathrm{C}}_{\mathrm{s}}={\mathrm{C}}_{\mathrm{m}}-{\mathrm{C}}_{\mathrm{w}}}$ (4)${\displaystyle \Delta {\mathrm{N}}_{\mathrm{s}}={\mathrm{N}}_{\mathrm{m}}-{\mathrm{N}}_{\mathrm{w}}}$ where C_m is the soil C stock after maize planting (Mg ha⁻¹), C_w is the C stock after wheat planting (Mg ha⁻¹), N_m is the soil N stock after maize planting (Mg ha⁻¹) and N_w is the N stock at maize planting (Mg ha⁻¹).

Statistical analyses

Data were analyzed using SPSS software version 17.0 (IBM Corp., Armonk, NY, USA). Duncan’s multiple range test was used for variance analysis and to compare mean values (n = 3) at a significance level of 0.05.

Crop yield

For wheat field, the grain yields under N180, N240 and N300 were significantly higher than that under N0 by 184.67, 191.11 and 173.24%, respectively; the yields under N180 and N240 were significantly higher than that under N300 (P < 0.05). For maize field, the grain yields under N180, N240 and N300 were significantly higher than that under N0 by 79.79, 85.23 and 86.85%, respectively (P < 0.05); there were no significant differences among N180, N240 and N300 (Fig. 1).

The SOC and TN content

In wheat growing season, SOC and TN were highest in 0–20 cm soil layer, with no significant differences among N levels, the TN was significantly lower under N240 than that under other N levels in 20–40 cm layer, but was significantly higher in 40–60 cm layer (P < 0.05) (Fig. 2). In maize growing season, SOC and TN were also highest in 0–20 cm, the SOC under N180, N240 and N300 were significant higher than that under N0 in 0–20 cm layer, the values under N180 and N240 were significant higher than N300 in 20-40 and 40–60 cm (P < 0.05). In the entire profile, SOC values were higher after wheat planting than maize planting, while TN were higher under N240 and N300 after maize planting than that after wheat planting.

The SOC and N storage

In wheat growing season, SOC and TN storage were highest in the 0–20 cm soil layer, the SOC and TN storage were significant lower than that under other N treatments in 20–40 cm under N240, and the TN storage were significant higher in 40–60 cm under N180 and N240 (P < 0.05) (Figs. 3A and 3B). The SOC and TN storage was highest in the entire profile under N180. In maize growing season, the SOC storage was significant higher in 0–20 cm layer under N300, but that under N0, N180 and N240 were significant higher than that under N300 in 20–40 cm (P < 0.05) (Fig. 3), the SOC storage was highest in the entire profile under N180, with lowest values under N300. In contrast, TN storage was higher under N240 and N300. The SOC storage in the entire profile was higher after wheat planting than maize planting under all N levels, while TN storage was higher after maize planting, especially under N240 and N300 treatment.

Relationships between SOC and TN content, and SOC and TN storage after wheat and maize planting

The SOC and TN content plus SOC and TN storage were significantly positively correlated, respectively (P < 0.01, Fig. 4). Correlations between soil SOC and TN, and soil C and N storage were also stronger after wheat planting than maize planting. Moreover, the correlations between SOC and TN storage were greater than those between SOC and TN content in wheat and maize growing seasons.

The biomass, SOC content and storage of above and belowground organs

The biomass of wheat and maize were significant increased after adding N, but the SOC content and storage have no significant difference among N treatments for wheat. For maize,the SOC storage of above and belowground were significant higher under N180 and N240 than that under N0 treatment, and the SOC content was highest under N240 than that under other N treatments (Table 1).

Differences in SOC and TN sequestration after maize planting

The SOC and TN sequestration were different in soil depths and N levels (Fig. 5). The SOC sequestration in the 0–20 cm soil layer decreased following maize harvest , while in the subsoil layers (>20 cm), SOC and TN sequestration increased under almost all N levels. The SOC sequestration was highest in the 20–40 cm layer under N240 (4.81 Mg ha⁻¹), followed by the 40–60 cm under N180 treatment. In the 20-40 and 0–60 cm soil layer, TN sequestration was highest at N240 level. The higher TN sequestration were observed under N240 and N300 compared to N0 and N180 across 0-60 soil profile.

Correlations between SOC, TN storage and soil properties

The SOC (or TN) storage and SOC (or TN) content were significantly positively correlated at each soil depth in both wheat and maize; in wheat, SOC storage and C/N were significantly positively correlated across the entire soil profile. No significant correlations were observed between SOC storage and BD in maize or between N storage and SOC content in both wheat and maize (Table 2).

Effect of N fertilizer on crop yield

The findings of this study revealed that annual N application increased crop yield compared with control treatment (N0) during the study period. Mean wheat yield was in the order of N240 > N180 > N300 > N0, with significantly higher yield under N180 and N240, similar to a previous report (Wang et al., 2018). In contrast, maize yield was highest under N300 treatment, and was in the order of N300 > N240 > N180 > N0, suggesting that maize has a higher demand for N fertilizer. It was previously revealed that fertilizer treatment significantly increased crop production compared to no fertilizer application (Chen et al., 2015), although the yields under different fertilizer treatments were not significant in both wheat and maize, thus differing from our results. This difference was attributed to the larger amount of crop stubble returned to the soil in the fertilized plots as a result of the increase in crop yield. In other words, a larger grain yield, especially in terms of straw and stalk biomass, inevitably leads to an increase in residue return to the soil compared to no fertilizer or synthetic fertilizer treatment (Huang et al., 2010; Xie et al., 2017), thereby increasing C sequestration.

Content and storage of SOC and N

Soil organic C affects soil functions and processes, thereby affecting nutrient availability for plants (Troeh & Thompson, 2005). Fertilization effects on SOC show regional differences, and in our study, SOC and TN content decreased gradually with increasing soil depth under all N treatments, with highest contents in the 0–20 cm layer. This is thought to be because the rate of decay of soil C is faster in deeper soil layers, consistent with previous findings in northeast China and the Loess Plateau (Wang et al., 2011a; Wang, Wang & Zu, 2014). In northern China, soil organic C and N conditions in both upper and deeper soil layers are significantly affected by N levels, and in this study, TN values were higher under N240 and N300 treatment during the maize compared to wheat season. This is probably why maize yield increased with increasing N fertilizer application. Moreover, SOC values were higher during the wheat compared to maize season in the entire profile. This is thought to be because increased rainfall during the maize season have resulted in a relatively higher soil moisture content, resulting in a more favorable soil environment for rapid propagation of soil microorganisms, and subsequent acceleration of SOC mineralization and decomposition (Silver & Miya, 2001; Wang et al., 2020; Zhang et al., 2010). In contrast, during the winter wheat growth period, rainfall was lower, and activity of microorganisms in the soil was likely to be lower, thereby slowing the rate of SOC decomposition. Moreover, wheat stubble returned to the maize field has a higher ratio of C: N than maize stubble, which is also likely to have had an effect on SOC decomposition (Zhao et al., 2016).

Cropping systems have a significant effect on SOC stocks, influencing the balance between C input from litter and C losses via decomposition (Huang, Sun & Zhang, 2012). In general, soil depth is stable, suggesting that soil C stocks are determined by SOC and soil BD (Deng et al., 2013). In this study, soil SOC and TN storage was highest in the 0–20 cm layer under all N levels in both wheat and maize. Nutrient and organic matter accumulation in the topsoil is the result of complex interactions between biotic processes regulated by plants and soil biota, and abiotic processes driven by environmental processes (Hooper, Bignell & Brown, 2000). Land use and the depth of sampling are therefore important in terms of SOC and soil C measurements (Strahm et al., 2009; VandenBygaart et al., 2010). In this study, SOC and TN storage were high under N180 for wheat and SOC storage showed no further increases with increasing N application. Soil nutrients are gradually depleted during continuous cropping under no fertilizer application, thereby affecting SOC dynamics and stability (Zhang et al., 2010). Thus, overall, balanced chemical fertilizer use combined with straw application is the most beneficial to SOC stocks.

Correlations between SOC and TN content, SOC and TN storage after wheat and maize planting

After wheat and maize planting, soil SOC and TN, and soil C and N storage were significantly positive correlated (P < 0.01). Deng et al. (2018) also observed a significant positive correlation between soil C and N on the Loess Plateau in vegetation succession, while a significant positive effect of N enrichment on SOC was also revealed in globally syntheses (Liu & Greaver, 2010; Lu et al., 2011). The correlations of soil C and N storage were different between wheat and maize planting, probably because SOC stocks are affected by a number of soil chemical and physical properties (Wang et al., 2013), and they differed following long-term N treatments in this study.

The biomass, SOC content and storage of aboveground and belowground organs

Different crop types result in different litter decomposition processes, thereby affecting the release of C and N into the soil (Zhang et al., 2013). Aboveground vegetation plays an essential role in regulating the biogeochemistry of different ecosystems, fixing C and nutrients and preventing nutrient losses following disturbance. In this study, the above and belowground biomass of both wheat and maize were significantly higher following N application compared to N0 treatment. It is well known that N fertilization has a positive effect on plant growth, with aboveground net primary productivity and belowground biomass both increasing in response to increasing N application (Peng, Guo & Yang, 2017; Tang et al., 2017).

In this study, the SOC content and storage of above and belowground wheat organs did not significantly differ between N treatments, similar to a previous report showing that the level of N application had no effect on SOC, with N addition increasing aboveground plant productivity but not root input into the soil C pool (King et al., 2005). Meanwhile, in maize, no significant differences in aboveground SOC was observed between N treatments, although belowground SOC content and storage was significantly higher under N180 and N240. A significant increase in C stocks with N application was previously observed in both above and belowground biomass, suggesting that N treatment stimulates biomass accumulation, thereby increasing C and nutrient retention in the soil (Lu et al., 2011). Overall, these findings suggest that results differ between crop types.

Differences between SOC and TN sequestration under different N levels after maize planting

The N levels and soil layers had a significant effect on SOC and N sequestration. The primary productivity of litter and roots have a significant effect on soil carbon sequestration, resulting primarily in an accumulation of SOC and N storage in the topsoil layer (Cheng et al., 2006; Dou et al., 2018). In this study, soil SOC sequestration in the 0–20 cm soil layer decreased under all N treatments, while both SOC and TN sequestration almost tended to increase in the 20–60 cm layer after maize planting for all N treatments. This is thought to be because larger maize roots are distributed at a depth of 20–30 cm (Dou et al., 2016), while soil C and N in the topsoil layer are lost rapidly, possibly due to soil respiration and incomplete decomposition soil organic matter in residues (Cheng et al., 2006). The N dynamics are another key factor affecting long-term terrestrial C sequestration (Luo, Field & Jackson, 2006). In this study, SOC and TN sequestration were at their highest in the 20–40 cm layer at N240 under all the treatments. The SOC sequestration almost tended to decrease in the 0–60 cm soil profile under N treatments; however, N sequestration were all increased, and was significantly higher under N240 and N300 treatment. These findings suggest that the consumption of SOC is greater than production, with combined application of N fertilizer and straw working together to increase N sequestration during the maize growing season. In addition, soil particle size, and the composition and dynamics of microbial communities as well as the increase in biomass with increasing N levels also have an effect on the biogeochemical cycles of soil carbon and N (Cleveland et al., 2004; Li et al., 2005), thereby increasing rates of carbon sequestration.

Pearson’s correlation coefficients between SOC, TN storage and soil properties

In this study, SOC and TN storage and contents of SOC and TN were significantly positively correlated in each soil layer after both wheat and maize planting, suggesting that SOC and TN storage are the result of soil C and N accumulation. Meanwhile, there were respective negative correlations between SOC and TN storage and BD, except in the 20–40 cm layer after wheat planting and the 0–20 cm layer after maize planting (Table 2). Wang et al. (2011b) revealed a negative relationship between SOC and TN with soil bulk density, while Singh et al. (2012) and Singh, Singh & Singh (2012) reported a significant increase in SOC and TN but a significant decrease in BD during the restoration of degraded land. In this study, BD differed between soil layers (Fig. 6), suggesting that the differences between studies were due to differences in soil characteristics and/or the crop type.