Old-field soil microcosm microbial assembly

Parallel experiments were established at five sites across the eastern United States in spring 2012 to explore soil, competition, and climate impacts on old-field succession. At each site, we set up raised beds with randomly assigned set soil and herbaceous community treatments in a fully factorial design. There were 6 replicates of each treatment arranged in blocks. Beds had holes drilled in the bottom to allow for drainage but were placed on top of woven landscape cloth to isolate them from the native soil and limit weed growth between beds. Soil treatmentsPrior to addition of vegetation, each bed was filled with a randomly assigned soil treatment. These treatments consisted of a gradient of four soil textures (100% topsoil; 75:25 topsoil: sand; 50:50 topsoil: sand; and 25:75 topsoil: sand). Topsoil was sourced from local commercial sources, and so there were site differences in macro- and micro-nutrient availability. Nevertheless there were generally consistent trends in pH, soil moisture, soil organic matter, phosphorous, and cation availability between the treatments at all sites. Vegetation treatmentsThere were three herbaceous community treatments: bunchgrass dominated (including Andropogon virginicus and Schizachyrium scoparium), Solidago dominated (including S. altissima and S. nemoralis), or control plots, which were weeded bi-annually for the first two years of the experiment and then allowed to seed from the surrounding old-field. The vegetation treatments were chosen to emulate two stages in old-field succession and included plants present along the latitudinal gradient. Herbaceous treatments were randomly assigned and transplant occurred in spring 2012 with a subsequent addition of Solidago rhizomes in fall of 2013. In fall 2013, all pools received an addition of woody species seeds. Sample Collection Three and a half years after establishment of the pools, soil cores (2.5cm x 10cm) were collected from all grass and Solidago pools at all sites between September and December 2015. In addition, we collected soil cores from the control pools at the NC and SYR sites. Soil cores were stored on ice immediately after collection and preserved at -20°C. The soil corer was cleaned with ethanol between pools. Amplicon SequencingMicrobial community diversity and composition were assessed using targeted amplicon sequencing of the 16s rRNA gene for Bacteria and Archaea domains. Soil cores were homogenized and passed through a 2mm sieve. DNA was extracted from ~0.25g soil using the PowerSoil DNA extraction kit (Mo Bio Laboratories, Inc.) following manufacturer's instructions. The V4 region of the 16s rRNA gene was amplified, cleaned, and quantified using custom barcoded primers (515f/806r primer pair; Caporaso et al 2011 PNAS) following published protocols. All samples were sequenced in a single MiSeq lane with paired end 250bp reads and reagent kit v2 (Illumina, San Diego, CA)