2015 metagenomic data from soils associated with the Chichaqua Bottoms Greenbelt Nutrient Network site

Site description: Chichaqua Bottoms Greenbelt (CBGB) is a complex of land managed by Polk County Conservation, approximately 30 km NNE of Des Moines, Iowa, USA (41.79 latitude, -93.39 longitude, 275 m elevation). The mean annual temperature is 9°C and mean annual precipitation is 855 mm (Hijmans et al. 2005). The soils are Farrar fine sandy loam (87.5 % sand, 7.4% silt, and 5.1% clay) and have a bulk density of 1.08 ± 0.03 g dry soil (Riggs et al. 2015, Natural Resources Conservation Service 2009). In 2015, Ackerman et al. (2019) estimated local ambient N-deposition (wet and dry) to be 0.97 g m-2.The plant community is a restored tallgrass prairie established in the mid-1990s in an area previously used for row crop agriculture. It is currently dominated by C4 grasses, with C3 grasses, legumes, and forbs as sub-dominants. Vegetation is managed with prescribed fire (typically every 2 or 3 years) and the plots of interest were burned in the spring (May) prior to data collection (August 2015).Fertilization: The research plots were established in 2009 as part of the Nutrient Network (Borer et al. 2014, nutnet.org) and annual fertilization application per protocols began in May 2010. The data presented here were collected in plots that received 0 (control) or 10 g N m-2 y-1 (timed-release urea [(NH2)2CO]), as part of the standard Nutrient Network scheme. Plots were randomly placed within each of four blocks in the same management unit.Specialized species: In 2015 we collected soil under two perennial, late successional prairie species: Ratibida pinnata ((Vent.) Barnhart) (RP), a sub-dominant forb or Schizachyrium scoparium (Michx.) Nash (SS), a C4 grass (Poaceae family). This species-specific data collection only occurred in blocks 2, 3, 5, 6.Soil collection: Soil samples for this study were taken from the two quadrats within the 25 m2 plot designated for site use. Here, we collected and bulked soils with a 2.5 cm soil probe (to 10 cm) from 10 random locations (denoted as NonTarget) in each plot on August 13, 2015. Although these were taken at random locations, sampling through the crown of a plant was typically avoided. At the same time, we also collected and bulked soils under the influence of several (3 to 4) SS or RP plants within these quadrats. This was done by either running the probe along the stem of the forb to get soils directly underneath it (RP) or directly through the heart of the grass crown (SS). The soil probe was wiped with ethanol and dried between sampling locations (NonTarget, RP, SS) and plots.We mixed each set of plot/plant soil cores, removing three subsamples (~10 g). The large sample was placed in a chilled cooler (temperature maintained by bottles of cold water) and subsamples were placed in a small cooler on dry ice. Upon return to the lab, we transferred all soils to a -80°C freezer and processed the large sample within 24 hours.Soil metagenomics (DNA): We sent the final subset of frozen soils to Argonne National Lab in December 18, 2015 (Job #151207-1) and was a MiSeq 151x151 Cycle Run. The soil was extracted using the PowerSoil DNA isolation kit from Mobio. Samples were then amplified with custom barcoded 16S gene sequences. A total of 13,691,304 paired-end Illumina sequences were obtained from Argonne National Lab (107,41154,076 sequences per sample). They were analyzed using the MG-RAST pipeline (Meyer et al. 2008) and base pair abundances for each phylum were obtained using ribosomal RNA similarities to entries in the REFseq protein database (See http://v4-web.metagenomics.anl.gov/index.html?stay=1) for details.