2.3 Field Sample Collection and Lab Analysis
At all sites, we collected water samples from May 05, 2018 to September 25, 2018 at approximately weekly intervals for a total of 21 sampling events. A total of 228 water samples were collected for aqueous geochemistry, filtered within 24 hours with a 0.45-μm PVDF filter (Millipore, HVLP04700) and frozen until analysis for major ions using a Dionex ICS-3000 ion chromatograph at the US Forest Service Rocky Mountain Research Station in Fort Collins, Colorado. A subset of 215 water samples were collected in sterile 60 mL falcon tubes for 16S rRNA amplicon analysis, kept cold until filtered within 12 hours onto a white polycarbonate GTTP 0.2-μm filter (Millipore, GTTP02500), flash frozen with liquid nitrogen and then stored in a -80°C freezer until analysis. To assess within site variability, for a subset of sites (Inflow, Outflow, Side-01, Pond-Con-01, and Main-Mid), we collected duplicate samples each week for 16S rRNA analyses. At all other sites, only individual samples were collected. Due to analysis constraints, Side-02 was sampled for 16S rRNA analysis at a bi-weekly frequency while other sites were sampled weekly.
We extracted DNA from each filter with a MoBio PowerSoil® DNA Isolation Kit using standard protocols. The 16S rRNA gene (V4 region) was amplified using 515F and 806R universal primers with the forward primer barcoded following the Earth Microbiome Project protocols (Caporaso et al., 2011). The forward primer 515F included the unique sample barcode following Parada et al. (2016) and both primers included degeneracies as described in Parada et al. (2016) and Apprill et al. (2015). For each sample, we ran a 50 μL PCR reaction using an Invitrogen PlatinumTM Hot Start PCR Master Mix with 10 μL of DNA. The PCR product was quantified and then combined into a single pool in equimolar concentrations and cleaned using a MinElute® PCR Purification kit. Cleaned, pooled DNA was sequenced with a MiSeq reagent v2 500 cycle kit on the Illumina MiSeq platform at the Colorado State University Next Generation Sequencing Core facility. Sequence reads were analyzed using MOTHUR (Schloss et al., 2009) and OTU counts defined at a 97% similarity of the sequence using the OptiClust algorithm. Generated OTUs were then aligned to a SILVA reference file (Quast et al., 2013). We then removed samples with limited sequences (<1000 reads), trimmed operational taxonomic units (OTUs) to remove samples not observed more than 3 times in 20% of the samples, and relativized OTU counts by the total OTUs in the sample. All 16S amplicon analyses were conducted using the phyloseq package in R (McMurdie & Holmes, 2013). Due to lack of duplication at all sites, we further merged duplicate samples into mean values to simplify subsequent analysis.