The methods summarized below are part of the following publication, currently in review and available as a pre-print: Tucker, S. J. et al. Sharp transitions in phytoplankton communities across estuarine to open ocean waters of the tropical Pacific. (2024) doi:10.1101/2024.05.23.595464.

The methods employed in this study were collaboratively developed with Heʻeia Fishpond stewards and the Heʻeia National Estuarine Research Reserve (NERR; Winter et al. 2020). Sampling campaigns were conducted with permission from Paepae o Heʻeia, the stewards of Heʻeia Fishpond, and the private landowner, Kamehameha Schools.

At all stations, seawater samples for biogeochemical analyses and nucleic acids were collected, as were in situ measurements of seawater temperature, pH, and salinity with a YSI 6600 or ProDSS multi-parameter sonde (YSI Incorporated, Yellow Springs, OH, USA). Approximately one liter of seawater was prefiltered with 85-μm Nitex mesh and subsequently filtered through a 25-mm diameter, 0.1-μm pore-sized polyethersulfone (PES) filter membrane (Supor-100, Pall Gelman Inc., Ann Arbor, MI, USA) to collect microbial cells for DNA isolation. The filters were subsequently submerged in DNA lysis buffer (Suzuki et al. 2001; Yeo et al. 2013) and stored in −80°C until further processing.

Seawater subsamples for fluorometric chlorophyll a concentrations (125 mL) and photosynthetic pigments via high-performance liquid chromatography (HPLC; 2 L) were collected on 25-mm diameter GF/F glass microfiber filters (Whatman, GE Healthcare Life Sciences, Chicago, IL, USA) and stored in aluminum foil at −80°C until extraction. The collection of phytoplankton pigments on the GF/F glass microfiber filters allow for comparisons with the Hawaii Ocean Time-series data. However, because the filters have a pore size of 0.7µm, we acknowledge that most small cyanobacteria were likely missed. Chlorophyll a was extracted with 100% acetone and measured with a Turner 10-AU fluorometer (Turner Designs, Sunnyvale, CA, USA) following standard techniques (Welschmeyer 1994). Photosynthetic pigments measured via high performance liquid chromatography were extracted in 100% acetone and analyzed on a Waters 2690 separations module equipped with a C18 column and full spectrum photodiode array detector, following (Mantoura and Llewellyn 1983) and modified according to (Bidigare et al. 1989).

For cellular enumeration, seawater was preserved in 2 mL aliquots in a final concentration of 0.95% (v:v) paraformaldehyde (Electron Microscopy Services, Hatfield, PA, USA) at −80 °C until analyzed via flow cytometry. Cellular enumeration of cyanobacterial picophytoplankton (Synechococcus and Prochlorococcus), eukaryotic picophytoplankton, and non-cyanobacterial (presumably heterotrophic) bacteria and archaea (hereafter referred to as heterotrophic bacteria) was performed on a Beckman Coulter CytoFLEX S, following the method of (Monger and Landry 1993). Inorganic nutrients were measured using a Seal Analytical AA3 HR Nutrient Autoanalyzer (detection limits: NO2− + NO3− , 0.009 µM; SiO4, 0.09 µM; PO43 –, 0.009 µM; NH4, 0.03 µM).

DNA extraction and 16S rRNA gene sequencing followed previously published methods (Tucker et al. 2021). Briefly, amplicon libraries were made from polymerase chain reactions of the 16S rRNA gene using barcoded 515F and 926R universal primers (Parada et al. 2016) and paired-end sequenced with MiSeq v2 2x250 technology (Illumina, San Diego, CA, USA). Genomic DNA from a subset of 32 of the 368 total samples collected between 2017-2021 were used for metagenomic sequencing. This included samples from four sampling events between 2017 and 2019 at 6-10 stations. Libraries were constructed from approximately 100 ng of genomic DNA using the Kappa HyperPrep Kit (Roche, Pleasanton, CA, USA) with mechanical shearing (Covaris, Woburn, MA, USA) and paired-end sequenced on a single lane of the NovaSeq 6000 SP 150 (Illumina, San Diego, CA, USA).