Location: Jardines de la Reina reef system south of the island of Cuba; General location: 20.8333° N, 78.9167° W

Sampling and analytical procedures:

Reef seawater microbial biogeochemistry and extracellular metabolite compositions (see "Related Datasets" section) were surveyed at nine, shallow (6 – 14 m in depth) forereef sites during a cruise to Jardines de la Reina (JR), Cuba in November of 2017. Seawater was also collected from two surface ‘off reef’ sites (800 – 1600 m depth).

At each reef, CTD (YSI Exo Sonde, Xylem Inc., Yellow Springs, OH, USA) casts were completed to measure the physicochemical properties of the water column. Seawater samples were collected from surface and reef depth to enumerate the number of microbial cells (1 mL samples) and assess macronutrient concentrations (30 mL for inorganic macronutrients and 40 mL for organic carbon and total nitrogen samples) using a submersible groundwater pump. Seawater samples (4L) were collected from reef depth and off reef samples for chlorophyll and phaeophytin analysis.

Samples collected for total organic carbon and nitrogen analyses were acidified with 75 µL of concentrated phosphoric acid, capped, and stored at room temperature. Surface and reef depth seawater samples collected for analyses of inorganic macronutrient concentrations (30 mL) were immediately frozen. Seawater, collected for enumeration of Prochlorococcus, Synechococcus, picoeukaryotic cells, and unpigmented cells (heterotrophic bacteria and archaea) was fixed with paraformaldehyde (1% final volume), incubated at 4 °C in the dark for 30 minutes, frozen at -50 °C on the research vessel, and then stored at -80 °C prior to analysis.

Non-purgeable total organic carbon (TOC, unfiltered), dissolved organic carbon (DOC, 0.2 µm filtered), total nitrogen (TN, unfiltered), and total dissolved nitrogen (TDN, 0.2 µm filtered) concentrations were analyzed using a Shimadzu TOC-VCSH total organic carbon analyzer (Hansell & Carlson, 2001) with a TNM-1 module. Inorganic macronutrient (phosphate, nitrite + nitrate, nitrite, ammonium, silicate) concentrations were measured with a continuous segmented flow system (as used in Apprill & Rappé, 2011). Nitrite was subtracted from the nitrite + nitrate concentrations to obtain the nitrate concentrations. Concentrations of total organic nitrogen were obtained by subtracting the sum of the inorganic nitrogen species (nitrite + nitrate and ammonium) from the total nitrogen concentrations per sample. If the measured concentrations fell beneath the detection limits of the instrument (ammonium = 0.02 M, phosphate = 0.01 M, nitrite + nitrate = 0.07 M, nitrite = 0.01 M), these measurements were removed from the analysis.

A few erroneous data values related to bottle/sample misidentification were included in the publication Weber et al. (2020). The differences between values are minor. The original values in Weber et al. (2020) are listed below and the corrected values are included within the dataset published here.

Reef name: JR4B_surf
The value for PO4 was 0.18
The value for SI was 1.6
The value for NO2 was 0.04
The value for NH4 was 0.03
The value for TON was 2.83

Reef name: OR2_surf
The value for TN was 0.18
The value for TON was -0.012

Acetone (90%) was used to extract Chlorophyll a and phaeophytin and the optical density (OD) values were measured on a calibrated spectrophotometer using standard optics (Lambda 18, Perkin Elmer, Waltham, MA, USA). The concentration ratios of chlorophyll a to phaeophytin were calculated and incorporated into the analyses. To obtain cell counts, flow cytometry was conducted using a collinear analysis method and a UV wavelength of 488 nm on an Altra flow cytometer at the University of Hawaii. Each sample was divided so that pigmented, fluorescent cells and unpigmented cells could be run separately. Analyzed volume was 100 ul. Unpigmented cells were stained with Hoechst stain at a final concentration of 1 µg mL-1. Abundances of each cell type were estimated by binning populations using FlowJo (v. 6.4.7) software.

Surface and reef depth seawater samples were also collected to quantify the concentrations of known metabolites and survey trends in untargeted metabolite feature composition using liquid chromatography mass spectrometry. Detailed methods are included on the MetaboLights project page (Project MTBLS1820) and within the open access paper.