Methods

\label{methods}

Studied sites and soil collection (JT?)

\label{studied-sites-and-soil-collection-jt}
The two studied sites are located in the central-West area of Belgium at Gaurain-Ramecroix (coordinates?) and Wiers (region of Tournai). The region is characterized by an oceanic temperate climate with an average annual rainfall of 800 mm and an average annual temperature of 10°C (IRM, Begium). According to the WRB-FAO, the soil type of the both sites is classified as Endogleyic Retisol. The two sites are characterized by contrasting pedological conditions explained by different soil textures: Fine Sandy Loam (so-called “Sandy site”) and Silt Loam (so-called “Silty site”).. Within each site, we selected three pairs of field, each pairs being composed of two farming systems: organic versus conventional. . The soil samples were collected in three replicates, each replicates being a composite of 6 sub-samples collected from the surface horizon (0–20 cm) in July 2015 (before harvest).

Farming system description (Fanny?)

\label{farming-system-description-fanny}
The organic fields (at the both site) are managed by only one farmer, while the conventional fields are managed by different farmers.
At the “Sandy” site, fields under organic farming were intensively and conventionally managed until 1995 including a deep plowing (30 cm) and the use of chemical inputs (fertilizers, fungicides, herbicides and insecticides). Since 1992, the organic farmer decided to gradually replace intensive plowing by minimum tillage (3 to 5 cm) and finally direct seeding with zero tillage in 2002. In 2011, the use of chemical input was stopped and in 2015 the farm obtained the official status of “organic farming”. At the “Silty” site, the history of farming until 2010 is unknown, but since this period the fields are managed in the same way as for the “Sandy” site: direct seeding and no chemical inputs. The organic farming is continuously covered by a mulch made up of a mix of leguminous and ? A compléter!!!!

Soil chemical properties

\label{soil-chemical-properties}
The soil samples were dried at 40°C and sieved to <2mm. The water content was measured by weighing the sample before and after drying it at 105°C according to ISO 11465 (1993). Total C and total N contents were measured by dry combustion (ISO 10694, 1995) under the combined action of elevated temperature (more than 900°C) and oxygen flow. CO2 produced is separated by gas chromatography. Given the absence of carbonates in the soil samples, the total C measured corresponds to total organic C. Exchangeable cations (Ca, Mg, K and Na) and P extraction was carried out with the modified Metson method (Pansu & Gautheyrou, 2006) and their concentrations were determined by atomic absorption spectroscopy and P concentration was determined by spectrophotometry. Cation-exchange capacity (CEC) was measured using the cobaltihexamine chloride method according to ISO 23470 (2007). Determination of pH(water) and pH(KCl) were carried out by using a soil/solution (water or KCl 1M) ratio of 1:5 (ISO 10390, 2005). Particle size distribution was determined by sedimentation using the pipette method.

Soil biological properties

\label{soil-biological-properties}

High-throughput sequencing of bacterial and fungal ribosomal markers

\label{high-throughput-sequencing-of-bacterial-and-fungal-ribosomal-markers}
DNA was isolated from the soil samples (8 g wet weight) with the PowerMax® soil DNA isolation kit (MO BIO Laboratories, Solana Beach, CA) according to the manufacturer’s recommendations.
Amplicon generation (bacterial V3-F4 and fungal ITS2 of the ribosomal RNA operon) was performed as described by Frey et al. (2016). The amplicons were sent to the Génome Québec Innovation Center at McGill University (Montréal, Canada) for barcoding and paired-end sequencing on the Illumina MiSeq V3 platform (Illumina Inc., San Diego, CA, USA). The sequence data were denoised according to Frey et al. (2016), including paired-end read assembly (REF), elimination of sequencing errors (REF) and chimeras (Edgar, 2013), as well as target verification and extraction (Hartmann et al., 2010; Nilsson et al., 2010). Denoised sequences were clustered into operational taxonomic units (OTUs) at 97% of identity and queried against Greengenes (DeSantis et al., 2006; McDonal et al., 2011) and UNITE (Abarenkov et. al, 2010) using the naive Bayesian classifier (Wang et al. 2007) and a minimum bootstrap support of 60%.

Statistics

\label{statistics}
The experimental design was as follow: The sites (sandy and silty ) and farming systems (organic and conventional ) were considered as fixed factors, and the spatial component (plot) was considered as random factor nested within the site factor (plots 1 to 3 at sandy site and 4 to 6 at silty site).
Estimates of alpha-diversity, i.e. richness Sobs and Smith-Wilson evenness E (Smith and Wilson, 1996), were based on evenly rarefied OTU abundance matrix and computed in mothur. The significance of the factors effect on alpha-diversity was examined using univariate Permanova based on Euclidean distance. Differences in beta-diversity were detected using the Bray-Curtis ecological distance calculated on the basis of normalized and square root transformed OTU abundance and after rare OTUs (singletons) were removed. The major variance components of bacterial and fungal beta-diversity was examined using principal coordinate analyses (PCO). The hypothesis were tested using the Permanova routine (Anderson, 2001) implemented in the software Primer6+ (Clarke and Gorley, 2006) with 99999 permutations. The presence of a priori groups with reference to the hypothesis was visualized with constrained analysis (CAP). The significance of the test was determined based on the Monte-Carlo empirical P value. The positive or negative response of taxa to farming system for each site was identified using the function adonis with 9999 permutations on centered and scaled OTUs relative abundance. Rare OTUs (<0.01% and those that occurred in less than 3 samples) were removed. Adjustments for multiple testing were performed using the Benjamini-Hochberg correction (Benjamini and Hochberg, 1995).