Introduction

Knowledge of the processes that control the rates of soil organic carbon (SOC) stabilization or decomposition in cultivated soils is key to understanding whether agricultural land use is a sink or source of carbon (C). Calculating the C fluxes in soils associated with maize cropping is important because of its global importance as a food (grain and forage) and biofuel crop; the total land area under maize production is estimated at 162 × 106 ha (FAO 2010). Plant roots are a major pathway for C input to soils (Jones et al. 2009) but the majority of photosynthesized C may be lost by rhizosphere respiration, i.e. the combined flux from autotrophic (root) respiration and heterotrophic (rhizomicrobial) respiration (Högberg and Read 2006). 13C- and 14C-isotope labeling studies have revealed the complex dynamics of photoassimilate routing during the growth cycles of cereal crops (see review by Kuzyakov and Domanski 2000). However, most studies used plants at early stages of maturity (less than 60 days after emergence) which may have caused under- or overestimation of C allocation to different soil pools because of changes in plant physiology during the growth cycle (Hütsch et al. 2002; Kuzyakov and Schneckenberger 2004; Amos and Walters 2006; Jones et al. 2009). For example, root respiration is decreased under mature plants (Kuzyakov 2006), when resources are diverted to reproduction and fruiting (Dungait et al. 2011), and the proportion of C translocated belowground and used for root growth, respiration and exudation decreases (Swinnen et al. 1994). In a survey of the literature, we found that above:belowground monitoring of pulse-labeled C ranged from 0.3 to 28 days (Table 1). Therefore, in order to properly understand the cycling of maize C, we hypothesised that longer term monitoring periods would provide a more reliable assessment of C dynamics in maize crops.

Table 1 Recovered pulse-C partitioning at elongation growth stage in maize. Incubation period is days after 13C/14C labeling to final sampling. Lit. ref. = literature reference. Literature references: *This study; aNguyen et al. 1999; bWhipps 1985; cLiljeroth et al. 1994; dWerth and Kuzyakov 2008; eHolland et al. 1996; fFan et al. 2008; gFischer et al. 2010; hWerth and Kuzyakov 2005; iPausch et al. 2012
Full size table

Partitioning rhizosphere respiration and SOC respiration is prerequisite to calculating gaseous losses derived from recently assimilated C or more stable SOC pools (Werth and Kuzyakov 2008). Estimates indicate that approximately 40 to 50 % (1.5 to 2.2 t C ha−1) of net photosynthate-C is allocated belowground for the total vegetative period in graminaceous crops: 19 % in the root biomass, 12 % lost as rhizosphere respiration and 5 % in soil residues (Kuzyakov and Domanski 2000; Jones et al. 2009). Total CO2 respiration rates of bare soil are often measured as a proxy for potential maximum hetereotrophic respiration (e.g. Doetterl et al. 2012), but the processes contributing to CO2 efflux from soil are greatly changed in the presence of growing roots (Kuzyakov 2006; Paterson et al. 2006). The contribution of root-derived CO2 to total soil CO2 efflux in experiments in grass species ranges from 36 to 70 % (Werth et al. 2006). The advent of experimental systems using 13C pulse labeling has shown the potential for investigations of 13C routing into the rhizosphere in intact, planted soils (e.g. Paterson et al. 2009; Hafner et al. 2012).

The highly sophisticated instrumentation required for continuous isotopic labeling and sampling are rare (e.g. Paterson et al. 2009), especially for periods equivalent to the life cycle of a large crop plant such as maize. Alternatively, 13C isotope pulse labeling at different growth stages of a plant provides discrete information about C dynamics associated with particular growth stages (Swinnen et al. 1994). Therefore, to test the hypothesis that longer monitoring times are required to quantify C dynamics in maize, we applied 13C-pulse labeling to pot grown maize plants at the beginning of the four major growth stages in the whole plant lifecycle, and followed plant:soil:atmosphere 13C dynamics by trapping 13CO2 from soil respiration and destructive analysis of plant biomass and soils over time.

Materials and methods

Soil preparation

Soil samples were collected from the plough layer (0–30 cm) of an arable plot at the China Agricultural University Quzhou Experimental Station in Hebei province, Northern China (36°52′N; 115°01′E). The plot had been under winter wheat-summer maize double cropping management since the 1990s. The station is located in a warm, semi-humid and continental temperate monsoon zone, with a mean temperature of 13.1 °C a−1 and a mean precipitation of 556 mm a−1. The soil is classified as a Cambisol and has a silty loam texture (sand 62 %, silt 29 % and clay 10 %), with a pH of 7.7 (soil:water ratio of 1:25), total organic carbon content of 7.6 g kg−1, total inorganic carbon content 7.8 g kg−1, and total nitrogen of 0.66 g kg−1. The δ13C values of the SOC and soil inorganic carbon (SIC) were −22.4 ± 0.2‰ and −3.4 ± 0.2‰, respectively. After collection the soil was bulked and air-dried then sieved (5 mm).

Maize growth conditions

Two maize seeds (Zea mays L., Cultivar Jiyuan 1) were sown directly into pots (35 × 20 cm i.d.) containing 9.5 kg air-dried soil pre-mixed with fertilizer at the rate of 0.55 g N, 0.19 g P, 0.31 g K kg−1 soil (equivalent to field application rates to the plough layer). An equal number of pots were prepared without seeds for the determination of soil respiration without plants. The maize was cultivated between May and September in a net house to provide similar light and temperature to field conditions. The soil water content of each container was controlled gravimetrically to simulate local growing conditions at each of the growth stages, and was adjusted daily to 60 % (emergence), 70–75 % (elongation), 75–80 % (heading) and 70–75 % (grainfilling) field capacity. Emergence occurred after 6 days. After emergence of the third leaf, the weaker of the seedlings was removed. The maize was grown for 100 days, and four different growth stages were recognized as days after emergence (DAE): (i) emergence (0–24 DAE), (ii) elongation (25–53 DAE), (iii) heading (44–66 DAE), and (iv) grainfilling (67–99 DAE).

13CO2 pulse labeling and sampling

The analysis of soil respiration in calcareous soils presents particular challenges for the source apportionment of 13CO2 (Setia et al. 2010; Tamir et al. 2011). The sources of CO2 respiration from soils with an alkali pH may be: (i) autotrophic respiration (i.e. maize in this experiment, δ13C = −14‰), (ii) heterotrophic respiration, (iii) CO2 from either autotrophic or heterotrophic respiration that has been retained in the soil solution or (iv) exchanged with soil carbonate, or (v) released directly from carbonate (i.e. in this experiment δ13C = −3.4 ‰). Therefore, we applied the method described by Fan et al. (2008) for tracking 13C-pulse label in an Aridisol. This approach accommodates the potential release of soil carbonate-derived CO2, by using a control treatment (unlabeled maize) and exploiting the Δ13C value (i.e. the difference between the δ13C values of the unlabeled versus 13C-labeled treatments) to determine the percentage contribution of the 13C-pulse label to rhizosphere respiration (which includes CO2 from root respiration and the decomposition of rhizodeposits).

The maize plants were 13C pulse labeled with 13CO2 at the beginning of each growth stage. On each occasion, three maize plants were randomly selected for 13C labeling, and an additional three maize plants were selected as unlabeled controls (the unlabeled controls were kept separately from the labeled plants). A chamber (1.05 m long × 2.15 m high) adapted from (Swinnen et al. 1994) was used for 13C labeling. The surface of the soil was covered with a PVC board and sealed with silicon, including around the maize stems. A leak check was carried out to confirm air tightness (Kuzyakov et al. 2002). A beaker containing Ba13CO3 (98 at.% 13C; 4.5, 8.5, 8.0 and 8.0 g each for the 4 labeling occasions, respectively) was placed into the chamber and an aliquot of 1 M HCl was injected into the beaker every 1.5 h for 7 h. CO2 concentration within the chamber was semi-quantitatively detected by GXH305 infrared detector (Beijing Analytical Equipment Co.) because of the differing wavelengths for maximum absorption of 13CO2 and 12CO2. If the rate of CO2 concentration decline slowed considerably, air (CO2 δ13C = −6.6‰) was pumped into the chamber until 13CO2 + 12CO2 concentrations increased to approximately 360 μL L−1. The 13C abundance of CO2 within the chamber was calculated within 55 at.% ~ 65 at. % for the 4 pulse labeling stages. An electric fan was used to homogenize the gases. The plants were removed from the chamber after labeling to prevent reassimilation of shoot-respired 13CO2. Sodium hydroxide (NaOH; 3.5 mol L−1) was placed in a small glass beaker under the PVC board to trap headspace CO2 (rhizosphere respiration) and changed every 3 days after the start of labeling, until the end of tracing period (27 days). Excess BaCl2 was added to the NaOH solutions containing trapped CO2 to precipitate BaCO3 which was washed with deionized water three times after back-titration with HCl (0.3 mol L−1) for determination of the total soil CO2 flux. The BaCO3 precipitates were combined into 4 homogenised samples for periods: 3 to 9, 12 to 18, 21 to 24, and 25 to 27 days after the start of each growth stage. The precipitates were dried at 60 °C and ground to <500 μm, before suspension in deionized water and acidification (1 mol L−1 H2SO4) to release CO2 into an air-tight syringe. The evolved CO2 was injected into an evacuated sample vial for 13C isotope analysis.

Shoot, root and soil sampling

Control and 13C labeled maize plants and soils were destructively sampled 27 days after labeling at each growth stage. Visible roots >2 mm were separated from the soil by careful hand-picking followed by repeated sieving (2 mm) to remove finer roots. All root material was soaked in 2.5 L deionized water, gently shaken for 20 min and then sieved (1 mm mesh). All the washing water was saved and added to the remaining pot soil to attempt to completely recover dissolved organic carbon and other rhizodeposits. The soil was spread on plastic film overnight to reduce water content. Then the soil was thoroughly mixed and repeatedly half sampled by gridding to achieve a fully representative 50 g sub-sample. Twenty grams of the 50 g sub-sample was placed on a white plate and any remaining visible fine roots were separated from soil by careful hand-picking. The 20 g soil sample was then soaked in 100 mL deionized water and shaken for 30 min and left to settle under gravity. The supernatant was collected in beaker and HCl (3 mol L−1) added until pH was less than 3 to removed inorganic carbon. For the soil slurry, 50 ml of HCl (3 mol L−1) was added and completely mixed. After 2 day, the slurry was washed to neutralise the pH, and mixed with the supernatant, and then dried at 105 °C. Shoots (stem and leaves) and roots were also oven-dried at 60 °C. All samples were ground (<500 μm) prior to analysis.

Analyses

Total organic carbon and nitrogen content of soils and plant materials were determined by elemental analyzer (Flash EA1112, ThermoFinnigan, Milan, Italy). The bulk δ13C values of soils, plant materials and gas samples were determined using isotope ratio mass spectrometry (DELTAplus XP, ThermoFinnigan, Bremen, Germany). The abundance of 13C was expressed as parts per mil (‰) relative to the international standard PDB (0‰) expressed as delta units (δ).

Calculations

The δ13C values of the biomass, CO2 and SOC was converted into mg 13C (R) using the method described by Simard et al. (1997) and used to calculate the fractional abundance (Fi) of 13C relative to 13C+12C from R. The 13C assimilated (mg) in plant biomass, soil and respired CO2 was calculated as the difference between the 13C contents of the labeled and unlabeled samples (Lu et al. 2002):

$$ {}^{13}\mathrm{C}i=\mathrm{C}i\times \left(\mathrm{F}i-\mathrm{F} ul\right)/100\times 1000 $$
(1)

Where i and ul indicate labeled and unlabeled samples, respectively, and Ci is the total C content of sample.

We defined the total amount of 13C allocated into the 4 measured pools (root biomass, shoot biomass, rhizosphere respiration and SOC) as 13C recovered. The 13C recovered did not include shoot respiration as the 13CO2 respired by the shoots was returned to the atmosphere and would not affect SOC as described in previous 14C and 13C pulse labeling studies (Fan et al. 2008; Werth and Kuzyakov 2008). The total 13C recovered was the sum of the 13C content in the measured pools: shoots and root biomass, SOC, and respired CO2 after labeling:

$$ \mathrm{Net}{\kern0.5em }^{13}\mathrm{C}\kern0.5em \mathrm{recovered}=\left({\kern0.5em }^{13}{\mathrm{C}}_{\mathrm{root}}+{\kern0.5em }^{13}{\mathrm{C}}_{\mathrm{shoot}}+{\kern0.5em }^{13}{\mathrm{C}}_{\mathrm{respiration}}+{\kern0.5em }^{13}{\mathrm{C}}_{\mathrm{soil}}\right) $$
(2)

The percentage distribution of 13C (13Ci%) between pools was calculated as:

$$ {}^{13}\mathrm{C} i\%={{}^{13}\mathrm{C}i/\mathrm{Net}\kern0.5em }^{13}\mathrm{C}\kern0.5em \mathrm{recovered} $$
(3)

where, 13Ci is the 13C content in the measured pools: shoots and root biomass, SOC, and respired CO2 after labeling.

The photosynthate-C distribution in the different pools of the maize-soil system at each growth stage was estimated using the method described by Gregory and Atwell (1991), Kuzyakov (2001), and Werth and Kuzyakov (2008):

$$ \mathrm{SC}i=\varDelta {\mathrm{C}}_{\mathrm{shoot}}\times {}^{13}\mathrm{C} i\%/{}^{13}{\mathrm{C}}_{\mathrm{shoot}}\% $$
(4)

where, SCi is the biomass-C in each pool, and ∆Cshoot is the biomass-C change during each stage of the maize.

Statistical analysis

The experiment was implemented with three replications. Data were subjected to one-way analysis of variance (ANOVA) in which the growth stage was the variable factor. Fisher’s LSD (least significance difference, P < 0.05) was used to test differences in variables among different growth stages of the maize. SAS 9.1 software (SAS Institute, Cary, NC, USA) was used for all statistical tests.

Results

Plant biomass

There was no significant difference between the mean weights of the aboveground (shoot) and belowground (root) biomass of the 13C pulse labeled plants and the unlabeled control plants (data not shown) when they were harvested 27 days after labeling during each of the growth stages (Table 2). This indicated that 13C pulse labeling did not affect plant growth. The shoot:root ratios were similar at emergence and elongation stages, and increased in the heading and grainfilling stages (Table 2).

Table 2 Biomass and recovered 13C budget in the maize-soil system at each growth stage. The figure in brackets is the standard error. Different letters in the same row denote a significant difference (LSD, P < 0.05) between growth stages
Full size table

The plant biomass δ13C values were around −14‰ for both shoots and roots in the control plants (Table 3) which is within the range of δ13C values for C4 plants (Boutton 1991). In the 13C labeled maize plants, the shoot δ13C values were more 13C enriched compared to the roots at emergence, heading, and grainfilling but not at elongation. This equated to more than half of the assimilated 13C allocated to the shoots at each growth stage, with significantly increased proportions at heading (69.8 %) and grainfilling stages (80.1 %) than at emergence (57.1 %) and elongation stages (53.5 %, Fig. 1). The allocation of 13C to the roots (not including root respiration and rhizodeposition) was greater at elongation (26.6 %) than at emergence (14.9 %) and heading stage (12.1 %), and lowest at grainfilling (3.3 %).

Table 3 Mean δ13C (‰) values (n = 3) of rhizosphere respiration at 9, 18, 24 and 27 days, and shoot, root and bulk SOC at 27 days for each maize growth stage: emergence, elongation, heading and grainfilling in (a) 13C pulse labeled treatments and (b) control (unlabeled). The figure in brackets is the standard error
Full size table
Fig. 1
figure 1

Percentage 13C (n = 3) shoot biomass, root biomass, rhizosphere respiration and bulk SOC 27 days after 13C labeling at emergence, elongation, heading and grainfilling growth stages in maize

Full size image

Bulk soil organic carbon (SOC)

The total organic carbon content of the original field soil was 0.8 %, and this did not change with incubation. The δ13C values of the bulk SOC in the control soils at all growth stages were not significantly different from the value before planting (−22‰; Table 3). The δ13C values of the bulk SOC in the 13C labeled soils at all growth stages were significantly 13C enriched by ~2‰ at emergence, heading and grainfilling, and 6‰ at elongation (Table 3). The 13C remained in the soil was 2.3 %, 3.8 %, 1.5 % and 1.2 % for emergence, elongation, heading and grainfilling stage respectively (Fig. 1).

Rhizosphere respiration

In the control treatments, rhizosphere respiration was 13C enriched (δ13C value range = −13.5 to −15.3‰) compared to bulk soil values (δ13C = −22.4‰), and similar at all growth stages and sample times to the shoot and root values (−14.1‰), indicating the use of the natural abundance 13C labeled maize-C for respiration by the root and microbial biomass (Table 3). In the 13C pulse labeled treatments, rhizosphere respiration had the greatest δ13C values up to 9 days after 13C pulse labeling at all growth stages with the greatest values at emergence and elongation, compared with heading and grainfilling stages (Table 3). The δ13C values of the respired CO2 at all growth stages reduced rapidly at 18 days, and more slowly thereafter.

The 13C recovered as 13CO2 (rhizosphere respiration) declined with time in all growth stages after pulse-labeling, and there were significant differences between 13C recovered between growth stages (Fig. 2). The contribution of 13C to rhizosphere respiration at 9 days was significantly increased during emergence, and significantly decreased during grainfilling compared with elongation and heading. The opposite was observed at 18 days when rhizosphere respiration of 13C was significantly decreased during emergence, and significantly increased during grainfilling, compared with elongation and heading. After 27 days, the amount of 13C lost as rhizosphere respiration (Fig. 1) was only significantly different at the emergence stage, accounting for 25.8 % of the total recovered 13C, or 14.9 mg 13C plant-1 (Table 2), compared with that at other three stages (elongation: 16.0 %, heading: 16.6 %, and grainfilling: 15.5 %).

Fig. 2
figure 2

Recovery of 13C label in rhizosphere respiration at 9, 18, 24 and 27 days after the start of labeling for 4 maize growth stages. Error bars indicate standard error (n = 3)

Full size image

Above: below ground allocation of photosynthate after 27 days

The amount of photosynthate allocated below and aboveground after 27 days changed with each growth stage (Fig. 1). The percentage 13C of the total recovered 13C in the shoot biomass was similar at emergence and elongation and increased during heading and grainfilling stages. This equated to 33.0, 66.4, 81.5 and 86.5 mg 13C plant−1 at emergence, elongation, heading and grainfilling, respectively. Allocation belowground was greatest at elongation (57.6 mg 13C plant−1; Table 2), when 26.6 % of the total recovered 13C was determined in the root biomass, 16.1 % in rhizosphere respiration and 3.8 % in soil, compared with the smallest quantity of 21.6 mg plant−1 to belowground during grainfilling.

Discussion

Determining 13C emission from soil with inorganic carbon

The δ13C value of the SOC calculated in the acid washed soil used in the experiment was −22 ‰. Published values for C3 soils are more 13C-depleted, e.g. ~ −27 ‰, which is similar to the δ13C value of C3 vegetation (Boutton 1991). Therefore, the comparative 13C-enrichment of the soil was assumed to reflect the contribution of maize cultivation (δ13C = −13‰ ~ −14‰) to the previous predominantly C3 SOC pool. The rhizosphere respiration in the control treatments had δ13C values in the range −13.5 to −15.3‰ compared to bulk soil values and was similar at all growth stages and sample times to the shoot and root values (−14.1‰). This suggested that maize-C was the principal source of the respired CO2 from the soil because its δ13C value was very similar to the value of the maize. However, the soil was slightly alkaline (pH 7.7), and consideration needs to be given to the influence of soil carbonate on the quantity and source of CO2 in soils with a pH >6.7 (Bertrand et al. 2007; Tamir et al. 2011). Setia et al. (2010) found that in soils with low carbonate and inorganic N contents, CO2-C emission was not affected by CaCO3 addition. Similar to Setia et al. (2010), the soil in this experiment had a low content of inorganic C and low inorganic N content (NH4-N <5 mg kg−1, NO3-N <40 mg kg−1, reported in Liu et al. 2003), indicating carbonate-derived CO2 was unlikely to contribute substantially to total soil respiration. Moreover, the potential influence of carbonate to the respired 13CO2 was resolved by subtracting the δ13C value of the CO2 from the unlabeled maize treatment in our study. However, the specific contribution of inorganic C to CO2 was not determined and admittedly may decrease the reliability of the results herein. The application of 14C-pulse labeling (e.g. Werth and Kuzyakov 2008) rather than 13C would help to resolve this concern, but it is not possible to use this approach under field conditions.

Incorporation of 13CO2 into maize biomass

The majority of photosynthate was incorporated into the aboveground (shoot) biomass at each growth stage of maize in this experiment, and this increased with maturity of the plant to a maximum of >80 % in the grainfilling stage. Correspondingly, shoot biomass was generally more 13C enriched than the roots in the 13C pulse labeled treatments, except during elongation. This suggests that there was active routing of photosynthate belowground through the early establishment (i.e. elongation) of the maize during a period of active root growth, but there is lower allocation to the roots as plant becomes mature. This is similar to the plant:soil C dynamics observed previously during the cultivation of rice (Lu et al. 2002), wheat and barley (Keith et al. 1986; Gregory and Atwell 1991; Swinnen et al. 1994, 1995) and maize (Yang et al. 2006), but not for pasture grasses (Kuzyakov et al. 1999; Kuzyakov 2001; Warembourg and Estelrich 2001).

Incorporation of 13CO2 into soil pools

A minor proportion of the total 13C was recovered from the SOC pool at the four labeling stages, with the largest (~4 %) at elongation. This was lower than the proportions reported in other studies, especially compared to those with shorter labeling periods (Table 1). This small but significant increase in δ13C amount provided evidence of substantial C4-C (i.e. 13C labeled) from photosynthate translocation belowground. Rhizosphere respiration (CO2) in the planted control treatments was 13C enriched compared with the bulk SOC at all stages, suggesting that the majority of the maize C4-C in the soil was labile and available for use as a respiratory substrate within the rhizosphere. Derrien et al. (2004) determined that 40 % of root-derived C from wheat was soluble neutral sugars which are highly available for microbial respiration. However, a component will also have been derived from the decomposing maize roots and aerial crop residues within the SOC (Dungait et al. 2009). To determine the contribution of this pool requires differentiation of C4-SOC from novel maize rhizoexudates and root turnover, which is possible though the application of 14C pulse labeling (e.g. Werth and Kuzyakov 2008). The proportion of maize photosynthate allocated to rhizosphere respiration changed with growth stage, but was greatest (25.8 %) during emergence, and similar (15.5 to 16.6 %) during elongation, heading and grainfilling growth stages. The latter is comparable to the value (12 %) reported by Jones et al. (2009). However, the content of 13C in rhizosphere respiration fluctuated less between growth stages, with a mean value of 18 ± 2 mg 13C plant−1 at each growth stage (Table 2), suggesting a sustained supply of plant-C was directed to root exudation.

Longer term and entire growing season monitoring of 13CO2 in plant: soil systems

The δ13C values of respired CO2 determined after 13C pulse labeling were most elevated after the first sampling period of 9 days, and declined rapidly thereafter. However, even after 27 days, there was still evidence of the 13C label in all C compartments, i.e. shoot biomass, root biomass, rhizosphere respiration and SOC at all growth stages. The majority of the 13C allocated belowground at all growth stages was lost as CO2 via rhizosphere respiration (autotrophic root respiration or heterotrophic mineralisation of rhizodeposits) rather than retained in SOC after 27 days. At least 98 % of the total respired 13C collected over the whole monitoring period (27 days) was recovered after 24 days and for last 3 days period (25nd to 27nd days after labeling), only about 1.1 %–1.4 % was recovered (Fig. 2). This monitoring period described specifically for the elongation growth stage in maize is the longest in duration reported in published studies (Table 1). Similar sustained sampling periods have been reported that rely on the hypothesis that distribution of assimilated C is assumed to be complete if losses of labeled CO2 by respiration can no longer be detected (Kuzyakov 2001). Swinnen et al. (1994) reported that 3 weeks is appropriate to estimate C assimilate distribution at different development stages using 14C pulse labeling under field condition for spring wheat. Neergaard and Gorissen (2004) confirmed that 2 to 3 weeks is adequate to quantify 14C allocation for white clover and ryegrass in a pot experiment. Particularly, comparing the recovered 13C/14C allocation proportion within shoots, roots, rhizosphere respiration and SOC pools in our study and other literatures indicated the significance of longer labeling periods. The mean of the published values for recovered 13C/14C in the aboveground pool was 67.7±9.2 % (Table 1), significantly higher than this study. This was the opposite for root and rhizosphere respiration pools which were 13.9±7.8 % and 11.2±6.1 %, respectively, significantly lower than 26.7 % and 16.0 % in our study. The proportion for the SOC pool was similar (3.9±2.3 %) with our data (3.8 %). The metadata analysis reported by Amos and Walters (2006) found that net rhizodeposited C values for maize cultivated in growth chambers was 22.4 %, and significantly higher than that calculated NRC (ratio of net rhizodeposited C, i.e., soil microbial biomass and as soil residue, to total net root-derived belowground C, i.e., standing root biomass C+rhizodeposited C) for the emergence, elongation, heading and grainfilling stages in this study (i.e. 0.13, 0.13, 0.11 and 0.26, respectively). However, of the published values, only 4 were sampled for longer than 99 days. This concurs with our findings that sampling for the entire growth season of maize is required to accurately quantify the photosynthate distribution within crop-soil systems.

Conclusions

The longer timescale of the monitoring in this experiment provided a much more complete assessment of photoassimilate-C dynamics in the maize plant:soil system than previously described. We estimated the C budget for the entire life cycle of the maize plants in the experimental system based on the 27 day monitoring period of each growth stage. The total photosynthate-C inputs were aboveground 57.5 g pot−1 (63 %), root 16.2 g pot−1 (17.8 %), rhizosphere respiration 2.4 g pot−1 (2.7 %) and soil organic carbon 15.1 g pot−1 (16.6 %). Using this calculation, we estimated the input of photosynthesized C to the maize-soil system under field conditions for the intensive winter wheat-summer maize farming systems of the North China Plain, where the typical aboveground net biomass of maize is 18 t ha−1 (7.56 t C ha−1). This translates to 4.6 t C ha−1 translocated belowground annually, which is partitioned into 2.1 t C ha−1 lost as CO2 by rhizosphere respiration, 2.2 t C ha−1 allocated to root biomass and 0.3 t C ha−1 incorporated into SOC. However, the longevity of the latter maize-C pool in soil post-27 days remains to be determined, including the potential positive priming effect of maize rhizoexudation on extant SOC (Pausch et al. 2012).