This study was conducted at the research station of Shenyang Agricultural University (41°31’ N-123°24’ E), Liaoning province, China, from May to September 2018. The soil type at the site is typical Brown Earth (Chinese Soil Taxonomy). The site has a temperate semi-humid continental climate. The annual temperature ranged between 6.2 and 9.7 ℃, and the annual rainfall was between 584 and 692 mm. Air temperature and humidity in 2018 were shown in Fig. S1. The basic physical and chemical properties of the tested soil are shown in Table 1; they were determined by the methods specified by Bao (2001).
Air-dried and chopped maize straw (average length 3 cm; C:N = 75:1) from the preceding maize plants on the same research station was returned to field at 28,000 kg/ha in the autumn (year before the experiment took place) at the soil surface (cover), 0-20 cm (shallow) or 20-40 cm (deep), with the control treatment having no straw returned. The experiment was carried out in the field microplots (2.4 m × 1.1 m), with eight treatments (labelled and non-labelled sets of four treatments specified above) in three replicates. The labelled and non-labelled treatments were set apart by more than 10 m to avoid the interference. The straw was manually mixed with soil at 0-20 cm for the shallow treatment. In the deep-return treatment, 0-20 cm surface soil was removed, the straw was manually mixed with 20-40 cm soil, and then 0-20 cm surface soil was returned.
The amount of N, P and K fertilizers applied was based on the standard farming practice for growing maize in the area (N: 240 kg/ha, P: 33 kg/ha and K2O, 87 kg/ha; as urea, superphosphate and potassium sulfate, respectively). The K and P fertilizers were applied as basal fertilizer at sowing, and N fertilizer was applied in three splits (as basal fertilizer and at jointing and tasseling) in the 3:4:3 proportion.
Maize (hybrid Jingke 968) was sown by hand planters and was thinned at the seedling stage to stand density of 57,000 plants/ha. Plant distance within rows was 30 cm, and the distance between rows was 50 cm. Border plots were included on the sides of the experimental field. Weed growth was controlled manually during the experiment.
Photosynthetic C (13C) labelling method
In the maize early jointing stage (on 11th July), the 13C pulse labelling was done simultaneously on all four treatments within one replicate block. The pulse labelling method (shown in Fig. S2) followed the published description (An et al. 2015a; Zhang et al. 2020) with modifications. A sealed and transparent labelling chamber measured 2.2 m length, 0.5 m width and 3 m height. This portable labelling chamber covered nine plants in each treatment and consisted of a transparent vinyl sheet on a steel frame. In order to provide a seal around the edges of the chamber, excess vinyl covered the contours of soil surface (Kong and Six 2010) and was sealed with wet soil (McMahon et al. 2005). Before the start of labelling, the black plastic film mulch was used to cover the soil surface of the micro-plot to prevent the labelled CO2 from diffusing into the soil. The plastic black film was laid only during labelling and was removed immediately afterwards. To avoid any impact of plastic film cover, the non-labelled areas were also covered with black plastic film for the duration of the labelling period.
Labelling took place from 8:00 to 13:00 on a sunny day. An infrared gas analyzer was connected to the top of the labelling chamber to monitor the total CO2 concentration (Wu et al. 2009). NaOH was used to absorb CO2 in the chamber. After the CO2 concentration fell below 80 μL/L, the sodium hydroxide trap was removed and H2SO4 (50 mL, 1 mol/L) was added to the first beaker containing labelled Na213CO3 (99 atom% 13C, Sigma-Aldrich) to obtain 13CO2 concentration of approximately 400 μL/L. When the CO2 concentration in chamber fell below 80 μL/L again, H2SO4 (50 mL, 1 mol/L) was added to the second beaker containing labelled Na213CO3. This process was repeated five times, and each labelling chamber required 9.12 g Na213CO3. Finally, we added sulfuric acid to the No. 6 beaker filled with non-labelled sodium carbonate (1.81 g Na212CO3) to enhance the 13C assimilation efficiency and minimize the loss of 13CO2 (Butler et al. 2004). The entire labelling process ended, and the labelling chamber was removed, after the CO2 concentration dropped below 80 μL/L after the final adjustment.
Sample collection and processing
Destructive sampling of maize plants in each treatment was conducted three times. Maize plants and soil samples were taken on 13rd July (the early jointing stage; two days after labelling), 26th July (the late jointing stage; 15 days after labelling) and 27th September (the grain maturity stage; 80 days after labelling). In each straw treatment, three labelled and three non-labelled plants were randomly selected from the respective plots. Shoots were cut at the base, and then roots and soil cores were dug out as a monolith (50 cm long × 50 cm wide and 40 cm deep). The aboveground material included shoots (stems and leaves) and grains (at maturity). All the visible small roots in the soil sample were picked out. Shoots (stems and leaves) and roots were washed in deionized water, oven-dried at 70 °C for 3 days and weighed. Dried root and shoot samples were ground in a mill (RetschMM200, Dusseldorf, Germany) for determining organic C.
The soil samples (0-40 cm) represented the mixture of rhizosphere and non-rhizosphere soil. The residual straw was carefully picked out (about 90% of straw was decomposed at grain harvest). The soil samples were stored in plastic bags at 4 ℃ and processed within 5 days. A portion of each soil sample was used for determining DOC and MBC. The remaining portion of each soil sample was air-dried, ground and passed through 0.25 mm sieve for the determination of total soil organic C. An elemental analyzer – stable isotope ratio mass spectrometer (Elementar vario PYRO-isoPrime100, Manchester, UK) was used to determine total organic C content and δ13C value in soil and plant samples.
Determination of soil DOC and MBC contents and δ13C values
Microbial biomass C (MBC) was determined by the chloroform-fumigation extraction method (Vance et al. 1987). Fresh soil equivalent to 10 g oven-dried soil was fumigated for 24 h and then extracted with 0.5 mol L-1 K2SO4. The same amount of soil was also extracted without fumigation. The non-fumigated extract was used to determine dissolved organic C (DOC). The soil extracts were measured to determine the dissolved organic C content using a Total Organic Carbon Analyzer (Multi N/C UV HS, Analytik Jena AG, Jena, Eisfeld, Germany). The MBC was calculated as the difference in dissolved organic C content between fumigated and non-fumigated soil extracts, with the conversion coefficient kEC of 0.45 (Wu et al. 1990). All K2SO4 extracts were freeze-dried (EYELA Freeze Dryer FD-1, Tokyo, Japan) to analyze 13C abundance (253Plus, Thermo Fisher, California, USA).
(1) δ13C value and δ13C abundance (FC)：
δ13C (‰) = (RC-RPDB)/RPDB × 1000
FC (%) = ((δ13C + 1000) × RPDB)/((δ13C+1000) × RPDB+1000) × 100
where RC is the 13C/12C atomic ratio of the sample, and RPDB is 0.0112372 (Lu et al. 2002a).
(2) The amount of 13C (mg) fixed in photosynthesis partitioned to maize shoots, roots, grains and soil (without considering a loss due to respiration)
13Ci = Ci × (FlC - FnlC)/100 × 1000
where Ci is the C content (mg) of shoots, grains, roots or soil in the labelling treatment; FlC is the abundance (%) of 13C in shoots, grains, roots or soil in the labelling treatment; and FnlC is the abundance (%) of 13C in shoots, grains, roots or soil in the non-labelled treatment (Leake et al. 2006).
(3) Partitioning of 13C (%)
Partitioning of 13Ci = 13Ci/13Cfixed × 100
where 13Cfixed is the sum (mg) of 13C partitioned to shoots, grains, roots and soil in the labelling treatment, and 13Ci is the 13C content of individual plant parts or soil (Yu 2017).
(4) Soil microbial biomass C (CMBC, mg/kg), dissolved organic C (CDOC, mg/kg), and the content of 13C (13C-CMBC, µg/kg; 13C-CDOC, µg/kg)
MBC = (CfumC-CnfumC)/kEC
CDOC = CnfumC
13C-CMBC = ((FfumC,l - FfumC,nl) × CfumC-(FnfumC,l - FnfumC,nl) × CnfumC)/(kEC × 100)
13C-CDOC = ((FnfumC,l - FnfumC,nl) × CnfumC)/100
where CfumC and CnfumC are the DOC content (mg/kg) in the K2SO4 extracts from fumigated and non-fumigated soils, respectively, in the same treatment; FfumC,l and FnfumC,l are the 13C abundances (%) in DOC in the K2SO4 extracts from fumigated and non-fumigated soils, respectively, from the labelled treatment; FfumC,nl and FnfumC,nl are the 13C abundances (%) in DOC in the K2SO4 extracts from fumigated and non-fumigated soils, respectively, from the non-labelled treatment. kEC is the conversion coefficient, and its value is 0.45 (Wu et al. 1990).
Two-way ANOVA was done on shoot biomass, root biomass, organic C in shoots, roots and soil, amount of assimilated C, and C partitioning to maize roots and soil, using sampling dates and treatments as independent variables. One-way ANOVA was conducted on parameters relative to four different treatments on each sampling date, or on twelve treatments (3 sampling dates × 4 treatments), depending on significance of the interaction between treatments and sampling dates. Means were compared with the Tukey’s honestly significant differences test at the 5% level of probability. All statistical analyses were done using the SPSS statistical software version 20.0 (IBM Corp., Armonk, NY, USA).