Impact of soil inoculation on crop residue breakdown and carbon and nitrogen cycling in organically and conventionally managed agricultural soils

Organic agriculture relies on organic fertilizers and amendments to provide nutrients to plants and will therefore depend on decomposer communities to release nutrients from these organic inputs. However, after conversion of conventional to organic agriculture it may take up to decades before decomposer communities become adapted to the new resource inputs. The aim of the present study is to investigate if the functional capacity of soil communities for decomposing recalcitrant crop residue types can be enhanced by inoculating soil communities from organically into conventionally managed soils. We used a microcosm incubation experiment to test how soil inoculation, agricultural management history, and crop residue type affect carbon and nitrogen cycling with crop residue addition. We collected soil samples from 5 pairs of conventional and nearby organic fields and set up a reciprocal inoculation experiment under controlled lab conditions. We inoculated soil from each conventional field with soil from the paired organic field and vice versa. To each soil mix, five types of crop residues were added: a cover crop mixture, carrot leaves (Daucus carota), alfalfa (Medicago sativa), hay (Lolium perenne), and straw (Triticum aestivum). There was one control treatment without any addition. Soils were incubated for 34 days and we measured mass loss of the crop residues from litter bags, cumulative soil respiration, cumulative potential plant available nutrients, permanganate oxidizable carbon (POXC), and substrate-induced respiration (SIR). Initial soil abiotic conditions (soil organic matter content, pH, C:N ratio, plant available nutrients), soil microbial biomass and soil bacterial and fungal community composition were also determined. We did not find clear effects of inoculation on mass loss and cumulative respiration. Instead, effects of crop residue type on all parameters were substantial. Crop residues with higher C:N ratios generally had lower mass loss and cumulative respiration, and resulted in lower nitrogen availability but higher POXC contents. Organic management enhanced cumulative respiration. There was little overlap in bacterial and fungal ASVs between the organic and conventional soils within each pair, resulting in a potential increase in diversity as a result of soil inoculation. We conclude that decomposition of crop residues declined with their recalcitrance, and that soils from organically managed fields did not increase the capacity of the soil community to decompose recalcitrant residues. Further studies are needed to determine whether compositional differences between soils from organic and conventional fields are a response to farming practices or whether management also has functional implications for soil fertility.

Methods
We sampled ten fields in the east of the Netherlands. Five organic and five conventionally managed arable fields that were a subselection from the Chronosequence 2017 (sampled 1.5 years earlier).

These soil samples are characterized.

Then; we used these soils for an mesocosm incubation experiment; see material and methods section below.

We set up a 34-day incubation experiment of soils from fields with contrasting arable management (organic vs conventional) and inoculated each soil with a 10% inoculum from the contrasting management regime. We then added plant residues ranging in recalcitrance (based on lignin content and C:N ratio) and measured litter mass loss, C respiration at different time points, and potential cumulative plant available nitrogen using resin bags. Before and directly after the experiment we determined permanganate-oxidizable carbon (POXC) as an indicator for carbon stabilization and substrate induced respiration (SIR) as a proxy for microbial biomass (Hurisso et al., 2016; Neely et al., 1991). We also characterized initial soil microbial communities by phospholipid fatty acids analysis (PLFA) and 16S and ITS metabarcoding.

SOIL SAMPLING

Soils were taken from sandy fields in the east of the Netherlands near Oploo and Deventer. Soils were classified as Anthrosols with a very low elutriable fraction and an A-horizon of at least 30cm. Gross coordinates can be found in table S1. We sampled 10 fields; 5 organic and 5 conventional which were paired. Soil cores (0-30 cm) were taken at 3 locations in each field with a minimum distance of 25 m between the cores and 10 m between the cores and the field border. Per field, cores were combined to a composite sample per field. Sampling depth was 0-30 cm. Soil was transported in a cooler box, stored at 4 °C, and sieved the day after collection (4 mm).

EXPERIMENTAL SET-UP

We used 500 mL air-tight pots with gas-sampling septa which were filled with 230 gram dry-weight equivalent soil. We had five replicated pairs of field fresh soil from the conventional field and the organic field. For both we created inoculation treatments by mixing soils within a pair from the contrasting management in ratio 1:9. The inoculation of conventional (10%) into organic soil was actually a control for inoculation, because we expected only an inoculation effect in conventional soils that were inoculated with organic soils.

We added five types of crop residues to the soils, with a single residue per pot. In addition, for each inoculation treatment per field we had control pots where no crop residues were added. 5 replicates x 2 management types x with/without inoculation x 5 substrates plus 1 control resulted in total in 120 pots. Crop residues that we used were a cover crop mixture, carrot leaves (Daucus carota), alfalfa (Medicago sativa), hay (Lolium perenne), straw (Triticum aestivum). We collected them from the shop, but the cover crop mixture was self-grown at the field. This was the TERRALIFE® - SOLARIGOL-mixture from DSV Zaden (DSV Zaden, Venzelderheide). The amount of residues added to each pot was standardized based on C content, with an addition of 1.04 gram C / 230 gram soil. To determine the C content of the crop residues we collected xxx subsamples for each residue, which were ground and measured for total C and N content using a CN elemental analyser (Interscience flashEA© 1112 series), after grinding using TissueLyser II (Qiagen, Hilden Germany). In total we added 2.78 g carrot leaves (Daucus carota), 2.73 g cover crop mixture, 2.58 alfalfa (Medicago sativa), 2.67 g hay (Lolium perenne) and 2.47 g wheat straw (Triticum aestivum). As the amount of added residue was standardized by carbon, amounts of added nitrogen varied between the crop residues and ranged from 0.013 for wheat straw to 0.116 gram for the cover mixture per pot (Table S2).

Before the addition of the crop residues to the pots, all residues were cut into small pieces (5-10 mm). For all residues, 25% was put into a litter bag and the rest was mixed though the soil with a sterilized spoon. Litter bags were buried in the soils. Pots were places in an incubator at 18 °C and 65% water content in the dark for 34 days. We left the lids of the pot slightly open to prevent moisture loss, but allow gas diffusion. Pots were randomized.

On days 1, 3, 4, 7, 10, 14, 23 and 32 we collected head space gas samples, to measure soil respiration. Before collecting the samples, we closed the lids for four hours. Then we took a 12 mL headspace gas sample using a syringe and 0.6mm needle. Samples were kept in pre-evacuated 6 ml exetainers. In each sample, the CO2 concentration was measured using a Trace GC ultra (Thermo Scientific, Waltham US). Cumulative potential plant available nitrogen was measured using resin bags that were buried in the pots and incubated during the whole experiment. Resin bags were constructed of nylon stockings, and contained 10 g of a 1:1 mixture of cation and anion exchange resins, preloaded with H+ and Cl-, respectively.