These processes and their controls on soil structure and nutrient availability reflect the possible indirect effects of soil microorganisms on plant growth, health and fruit development . Vineyard management practices and production systems that alter the soil environment, and thus may contribute to shaping the microbial community, include: cover crop use, tillage, compost application, and conventional, organic, or bio-dynamic systems. Here, we focus on establishing a baseline understanding of the relationships between management practices and changes soil microorganisms within winegrowing regions. This represents the baseline from which we can subsequently delineate the ecological roles of specific taxa to elicit desired outcomes in wine grape production. In other words, altering management practices to change soil properties, which in turn shift key individual or consortia of soil microorganisms, could tune interactions among wine grapes, the soil environment, and associated microorganisms to influence wine grape production. The soil microbial roles discussed above are intrinsically coupled to both soil quality and soil health. Soil quality refers to the fitness of soil for a particular purpose , and thus, requires a specific definition for each purpose. In viticulture, hydroponic nft channel soil quality is defined as “the soil’s ability to support the production of a crop while minimizing negative effects on the environment” . Soil health is subtly distinct from soil quality.

Soil health is defined as “the continued capacity of soil to function as a vital living system, within ecosystem and land-use boundaries, to sustain biological productivity, maintain the quality of air and water environments, and promote plant, animal, and human health” . Microorganisms are intimately linked with the cycling and stability of soil organic matter, among other functions related to soil health, and are sensitive to changes in soil attributes and management . Therefore, measurements of soil-borne microbial communities, such as biomass, structure and functions, have been recommended as good indicators of soil quality . In order to implement assessments of soil microbial community structure for soil health monitoring, additional research is needed to understand the link between soil microbial community structure and soil functions, as they relate to soil health. Studies have begun to show empirically that soil microbial community structure and function are linked , and soil biodiversity is assumed to improve ecosystem resilience by offering functional redundancy . Soil biodiversity is recognized for its importance to agricultural sustainability in an economic, social, and ecological context . By describing the effect of agricultural management practices on the soil microbial community structure, we aim to form the foundation from which linkages among soil quality, agroecosystem function, and soil biodiversity can be built to better define soil health for wine grape production. Recent work has shown that climate, region, soil type, and wine grape variety can play strong roles in structuring microbial communities in vineyard soil, the vine phyllosphere, must and wine, and that soil microbial activities and wine metabolome are correlated with microbial community structure .

However, no single study examines the vineyard microbiome from soil to wine nor do they examine effects of vineyard management practice on soil microbial communities. Numerous studies have assessed the effects of land use and agricultural management practices on soil quality, soil properties, and soil microbial communities . Land-use effects on soil microbial communities are thought to be mediated mostly through alteration of soil properties. Soil properties correlated with soil microbial community structure include soil texture, pH, water content, carbon and nitrogen content, and C:N ratio . Plants alter many soil properties as well as soil aggregation and soil nutrient status, through root exudation and fine root turnover. In turn, this affects the soil microbial environment, resulting in shifts in the soil microbial community . Tillage disturbance also alters the distribution of soil organic matter and soil structure, thereby causing shifts in aggregate size, composition, and stability, and changing soil nutrient availability . Compost amendments add labile carbon and nitrogen, nutrients, and active microbial communities to soil . Consequently, these changes mediate shifts in microbial communities and microbial processes . These practices are embedded within conventional, organic, and bio-dynamic agricultural management systems, which differ primarily in their methods of fertilization and control of disease, insects, and weeds. Though effects of pesticides and fertilizers on soil microbial communities are well studied with clear effects , studies based on a comparison of conventional, organic, or bio-dynamic systems, have not been consistent in showing the same effects on soil microbial communities . Vineyard management in Napa Valley, California, includes this array of management practices and production systems across a range of soil types, allowing us to examine how vineyard floor management practices influence soil bacterial community structure in the context of environmental and edaphic factors. We measured differences in the soil-borne bacterial and archaeal community composition and diversity by sequencing the V4 small subunit ribosomal RNA gene .

We hypothesized that variations in soil bacterial communities, at the landscape scale, result from different agricultural management practices, as mediated through changes in soil properties. The scope is delineated in this manner to extend the observations of Burns et al. , who recently examined the roles of wine growing region, or appellation, climate and topography on soil bacterial communities across this same suite of sites.Soil samples were collected from 57 sites in 19 wine grape vineyards, with three sites per vineyard, throughout Napa Valley, California, and treated as a completely randomized design. See Burns et al. for a complete description of the experimental design, approach and details on specific practices at each vineyard. Details of management practices were gathered through interviews with vineyard managers. Soil samples were collected MarcheJune 2011, at a depth of 0e5 cm, from the centers of the vineyard alleyways. Plant residues and shoots, if present, were removed priorto soil collection. At each site, three soil samples were collected approximately 2-m apart and mixed into a composite sample. Samples were kept on ice until representative sub-samples were divided for laboratory analyses. For microbial community assessment, 50 g of soil from each composite sample was stored in sealed plastic bags at ! 80 ” C.Raw Illumina fastq files were demultiplexed and quality filtered using QIIME v1.6.0 and analyzed using QIIME v1.7.0 , as described in Burns et al. . Reads with a Phred quality of <20 were discarded. Operational taxonomic units were assigned using QIIME’s UCLUST-based open reference OTU-picking workflow, with a threshold of 97% pairwise identity. Sequence prefiltering and open reference-based OTU picking were performed using the Greengenes 16S rRNA gene database . OTUs were classified taxonomically using a QIIME-based wrapper of the Ribosomal Database Project classifier and the Greengenes 16S rRNA gene reference database , using a 0.80 confidence threshold for taxonomic assignment. 16S rRNA gene sequences were aligned using PyNAST against a template alignment of the Greengenes core set filtered at 97% similarity, and a phylogenic tree was generated from the filtered alignment using FastTree . Each subreplicate was collapsed into its composite sample . Any OTU representing less than 0.001% of the total filtered sequences was removed to avoid inclusion of erroneous reads that would otherwise lead to inflated estimates of diversity , as were samples with less than 28,008 sequences following all quality-filtering steps. Richness was estimated by the number of observed phylotypes and by the Chao1 richness estimate . The b-diversity , using the weighted UniFrac distance between samples, was calculated in QIIME. To enable visualization of sample relationships, the resulting weighted UniFrac distance matrix was used to perform non-metric multidimensional scaling in the R vegan package using four dimensions as determined based on the elbow of the scree plot in PC-ORD . NMDS is considered the most robust unconstrained ordination method . The impact of vineyard management practices on soil properties, a-diversity and richness, and b-diversity, nft growing system was determined by examining differences in spread along NMDS axes, using the Kruskal-Wallis rank sum test or Wilcoxon rank sum test for the special case of two groupings . Differences in b-diversity, based on the weighted UniFrac distance matrix, among soil properties and the vineyard management sample groups were also tested using non-parametric multivariate analysis of variance with 999 permutations. Relationships also were investigated within one sub-appellation Rutherford, as it contained the greatest number of samples. The effect of appellation was examined directly by Burns et al. . To determine which relative taxa abundances differed between vineyard management practices at various levels of taxonomy, oneway analysis of variance was performed in QIIME. Canonical discriminant analysis was performed using the candisc and heplots R packages to graphically reveal differences between sample groups of vineyard management practices and to identify high-level taxa associated with each practice .

To help elucidate the relative importance of soil attributes, management, and location in structuring the communities, variation partitioning using canonical correspondence analysis , a constrained unimodal approach, was performed using CANOCO 5 for Windows at a coarse taxonomic level, consisting of phyla, except for Proteobacteria, which were divided into a-, b-, g-, and d-classes. This coarse level was selected based on the conclusions of Philippot et al. , which suggested high levels of bacterial taxonomy are ecologically coherent and each different level of taxonomy may offer a different piece of information on the underlying mechanisms driving establishment of bacterial populations. An additional analysis using a fine taxonomic resolution, consisting of genera or the finest level of classification available for each group, also was conducted. However, the results revealed overall patterns similar to the results of the coarse taxonomic resolution; therefore, details of the finer-level taxonomic analysis are omitted from this report .The structure of soil-borne microbial communities is influenced by soil properties typically affected by crop management practices . Management practices or other factors were identified that were more relevant to structuring the microbial community in one subset compared to another subset of that same type. This suggests that the microbial community was more responsive to a given management practice or factor due to inherent characteristics associated with that subset . This also suggests that there is a relative hierarchy of effect of these management practices or factors on microbial community structure, where one factor may have a stronger effect within one subset versus another . Because of the strong relationship of sub-appellation with soil microbial communities, and the great variation and uneven representation among sub-appellations , it is also useful to look at the influence of the other factors, such as management and soil type, within a single sub-appellation instead of across all sub-appellations. For example, general effects of tillage presence/absence and specific effects that occurred based on its recency of application in the field emerged within the Rutherford American Viticultural Area , as the best represented sub-appellation . These findings suggest that an agricultural practice could have different effects with respect to AVA, which may not be too surprising given that soil attributes also differ with respect to AVA in this region . If we assumed that tillage, for example, affects all AVAs the same by only examining its overall effect rather than within individual AVAs, we would overlook important information regarding the structuring of the soil bacterial communities.It is known that plants differentially affect soil structure, and hence, the soil microbial environment. Plants differ in contributions to labile soil C and soil organic matter, and hence, the soil microbial resources through differential root exudation and fine root turnover, as suggested by distinctions in total C and N and the <53 mm fractions among cover crops in this study . Plant-soil-microbe interactions are particularly pronounced in the rhizosphere, where soil microbial community compositions are often plant-specific and distinct from the bulk soil . For example, Firmicutes had high relative abundance in association with mustard cover crops, similar to that observed in Brassica juncea . These plant-soil-microbe interactions enable cover crops to distinctly affect the soil, soil microbial communities, and microbially-mediated soil processes . Here, we observed an effect of cover crop mix and the general presence of cover crop on soil physicochemical properties . In turn, soil bacterial community structure, taxa abundances and soil C and N pools differed by cover crop mix, suggesting that there is an interaction among the cover crop type, soil resource pools, and the microbial community . For example, relative abundance of Actinobacteria can be associated with enriched C and N pools , as observed here in the grass cover crop soils with the highest soil C and N content. However, the opposite was observed in vineyard soils in Spain enriched in soil C pools after 13 years of compost application . Effects of cover crops on soil microorganisms and microbially-mediated processes have been observed in other vineyard studies and in annual cropping systems .