Surficial and Downcore Sedimentological and Foraminiferal Microfossil Data from St. Marks National Wildlife Refuge, Florida

Estuarine surface grab samples for sedimentological analyses (denoted with G, STM044G-STM053G) and foraminiferal census data (denoted with F) were collected using a petite PONAR grab sampler, deployed according to manufacturer specifications. The sediment recovered in the grab sampler was inspected for an undisturbed (for example, free of slumping, washout, scouring, cracking, and/or other disturbance features) sediment-water interface to ensure the bulk sample collected was representative of the surface and material was not lost. If the sediment was disturbed, the sediment was discarded, and a new grab sample was collected and assessed. If the sediment surface was intact, the overlying water and the mesh screens were slowly removed in accordance with the product manual, and the uppermost one centimeter of sediment was subsampled with a spoon or scoopula for sediment characterization and foraminiferal census, as described in Osbourne and DeLaune (2013), and for consistency with related products from the northern Gulf of Mexico (Ellis and Smith, 2021; Haller and others, 2019) and as is standard and recommended for recent surficial sediment and microfossil analyses (Schönfeld and others, 2012). Water quality properties at each estuarine site were measured with a YSI Pro-DSS and values were recorded. At nine sites in the St. Marks National Wildlife Refuge marsh, sediment surface samples (top 0-1 cm of sediment, denoted with a S, samples STM001S-STM014MS) were collected by hand using a spoon and/or scoopula and transferred and stored in a labeled and sealed baggie on ice for sedimentological analyses (for example, organic matter content, grain-size) for surface sediment characterization. At each surface sample marsh site, a surface sediment sample (0-1 cm) was collected separately for foraminiferal microfossil analyses (denoted with F), consistent with related products from the northern Gulf of Mexico (Ellis and Smith, 2021; Haller and others, 2019), and as is standard and recommended for recent surficial sediment and microfossil analyses (Schönfeld and others, 2012) to characterize the modern foraminiferal distribution. All estuarine grab and marsh surface foraminiferal samples, collected in duplicate at each site, were preserved in ethanol and stained with rose Bengal for the determination of live (stained) and dead (unstained) foraminifera (Schönfeld and others, 2012). At four of the marsh sites, five marsh push cores were collected with 10.16-cm diameter polycarbonate barrels. Upon retrieval, similar to methods described in Osbourne and DeLaune (2013; with the exception of not adding water for extraction when sediments were already saturated) and calculation of compaction due to coring, the cores were visually inspected for disturbances (for example, slumping, washout, scouring, cracking, bubbling, and/or discontinuities) to ensure the core was intact and representative of the site. If the core appeared disturbed, it was discarded, and a new core was collected and inspected. Core lengths ranged between 36 and 55 cm. The cores were transported upright, in order to avoid slumping and preserve the natural sediment orientation, to SPCMSC for sectioning. At each marsh site, Russian peat augers were collected in agreement with the methods described in Osbourne and DeLaune (2013) and manufacturer recommendations. Visual characteristics of the peat augers were described (for example, general color; visual organic matter texture and type such as roots, bivalves, and level of decomposition; and sediment texture such as sandy silt or clayey silt) and thickness of the upper organic-bearing unit (peat) was recorded in field forms, in centimeters. Once described and photographed horizontally with a scale bar and label, peat augers were discarded in the field. The field forms associated with those peat augers include handwritten notes visible on the scanned field forms for each site and are available upon request. Select single 50-cm peat auger segments were collected in the field and immediately stored in halved acrylic tubes and sealed for microbial analyses. Sample identifiers consist of the USGS project ID (19CCT05), a site-specific identifier (for example, STM014), and appended with an alphabetic identifier to differentiate the sediment collection method (S for marsh surface sample, M for marsh push core, G for estuarine grab sample, R for Russian peat auger, R50 for 50-cm section of peat auger for microbial analyses, F for foraminiferal sample). Estuarine site coordinates and water depth were recorded with a Garmin boat GPS and depth sounder; marsh site coordinates were recorded using a handheld GPS in addition to an RTK antenna in the WGS84 datum. Site location information includes sample type, date collected, location relative to the shoreline, latitude, longitude, elevation, core lengths, and YSI measurements which are reported in an Excel spreadsheet. Comma-separated values data files containing the tabular data in plain text are included in the download files.

Upon return from the field, all marsh push cores were X-rayed vertically using a stand, to avoid sediment slumping, onto an iCRco 11 x 14-inch cassette using an Ecotron EPX-F2800 X-ray unit at a distance of 79 cm for a 1:1.015 ratio. The cassette was inserted into and scanned using an iCR3600+ cassette scanner and processed using iCRco QPC XSCAN32 version 2.10. Images were then exported as .tiff files and edited in accordance with a standard operating procedure developed by staff at the SPCMSC in Adobe Photoshop and Illustrator. Image alterations and edits include using grayscale color inversion, inserting a reference scale bar, and image cropping. X-ray images were exported as Joint Photographic Experts Group (JPEG) images and used as visual aids prior to extraction and subsampling to assess peat thickness in comparison with peat augers, presence of macrofossils which may impede sectioning, and to ensure cores are intact for the length of the core. Following image processing, a second individual assessed the images for visual anomalies, workflow consistency, precision and accuracy, and quality.

Following x-raying, all cores were vertically extruded and sectioned into 1-cm intervals using a serrated knife, pre-measured polycarbonate ring, and extruder (in accordance with methods described in Osbourne and DeLaune, 2013 and in 1-cm intervals as is standard for sediment and radiochemical analyses; Nittrouer and others, 1979) at the USGS SPCMSC sediment core laboratory. The outer circumference of each sample interval was removed to avoid use of sediment that was in contact with the polycarbonate barrel, which could result in contamination by sediment from other depths due to movement within the barrel both during collection and extruding. Each sediment interval was bagged in a zipped baggie, homogenized, and refrigerated (Osbourne and DeLaune, 2013).

In the laboratory, marsh core samples were homogenized in the sample bag to ensure a representative subsample from the 1-cm interval, and the subsample for sediment parameters was extracted and processed for basic sediment characteristics: dry bulk density and porosity. Water content, and dry bulk density were calculated by determining water mass lost during drying using the same methods as outlined in Osterman and Smith (2012). To calculate, known volumes of each wet subsample, usually 30-60 milliliters (mL), were packed into a graduated syringe with 0.5 cubic centimeter (cm^3) resolution. The wet sediment sample was then extruded into a pre-weighed aluminum tray/weigh boat, and the weight was recorded. The wet sediment sample and tray were placed in a drying oven for 48 hours at 60 °Celsius to remove water content. A 1 mL split from the original wet sample was also taken for diatom analysis and was dried using the same procedure. Water content (θ) was determined as the mass of water (lost when dried) relative to the initial wet sediment mass. Dry bulk density (g/cm^3), also referred to as sediment bulk density, was determined by the ratio of dry sediment to the known volume of wet sediment packed into the syringe, resulting in a mass/volume ratio. Water content, porosity and dry bulk density are reported in the Excel spreadsheet. A comma-separated values data file containing the tabular data in plain text is included in the download file.

Organic matter (OM) content was determined with a mass loss technique referred to as loss on ignition (LOI). The dry sediment subsample from the previous process step, measuring dry bulk density, was homogenized with a porcelain mortar and pestle. Approximately 2 to 6 grams (g) of the dry sediment was placed into a pre-weighed porcelain crucible. The mass of the dried sediment was recorded with a precision of 0.01 g on an analytical balance. The sample was then placed inside a laboratory muffle furnace with stabilizing temperature control. The furnace was heated to 110 °C for a minimum of 6 hours to remove hygroscopic water adsorbed onto the sediment particles. The furnace temperature was then lowered to 60 °C, at which point the sediments could be reweighed safely (modified from Dean, 1974 who heated the furnace to 100 °C for 1 hour). The dried sediment was returned to the muffle furnace and heated to 550 °C over a period of 30 minutes and kept at 550 °C (Galle and Runnels, 1960) for 6 hours (optimal exposure times for complete combustion of organic carbon are reported ranging between 1–12 hours; Dean, 1974; Wang and others, 2011; Heiri and others, 2001; Santisteban and others, 2004). Following the 6-hour burn time for removal of organic carbon, the furnace temperature was lowered to 60 °C, at which point the sediments could be reweighed safely while preventing the absorption of moisture, which can affect the measurement. The mass lost during the 6-hour baking period relative to the 110 °C-dried mass is used as a metric of OM content (Dean, 1974); OM content for core GB541M may be overestimated due to the increased furnace temperature. Approximately 15.5% percent of the field samples were run in triplicate and 1.4% percent of the field samples were run in replicate LOI to assess precision. Data are reported as a ratio of mass (g) of organic matter to mass (g) of dry sediment (post-110 °C drying). Replicate analyses of loss on ignition are reported for quality assurance in the Excel spreadsheet. A comma-separated values data file containing the tabular data in plain text is included in the download file.

Down-core particle size analysis was performed on 43% of the 1-cm depth intervals for the five marsh push cores based on dry bulk density variations, and on all 18 surficial sediment samples. Prior to analyses, sediment samples were digested with 8 mL of 30 percent hydrogen peroxide (H2O2) overnight to remove excess organics. The H2O2 was then evaporated slowly on a hot plate, and the sediment was washed and centrifuged twice with deionized water. Grain-size analyses on the sediment cores were performed using a Coulter LS 13 320 (https://www.beckmancoulter.com/) particle-size analyzer (PSA), which uses laser diffraction to measure the size distribution of sediments ranging in size from 0.4 microns to 2 millimeters (mm; clay to very coarse-grained sand). To prevent shell fragments from damaging the Coulter instrument, particles greater than 1 mm in diameter were separated from all samples prior to analysis using a number 18 (1000 microns, or 1 mm) U.S. standard sieve, which meets the American Society for Testing and Materials (ASTM) E11 standard specifications for determining particle size using woven-wire test sieves. Two subsamples from each depth interval were processed through the instrument a minimum of three runs each. The sediment slurry made from the digested sample and deionized water was sonicated with a wand sonicator for 1 minute before being introduced into the Coulter PSA to breakdown aggregated particles. The Coulter PSA measured the particle-size distribution of each sample by passing sediment suspended in solution between two narrow panes of glass in front of a laser. Light is scattered by the particles into characteristic refraction patterns measured by an array of photodetectors as intensity per unit area and recorded as relative volume for 92 size-related channels (bins). The bin boundaries are determined by the manufacturer, and GRADISTAT groups the data from the bins into sediment-size classifications.

The raw grain-size data were processed with the free software program, GRADISTAT version 8, (Blott and Pye, 2001), which calculated the mean, median, sorting, skewness, and kurtosis of each sample geometrically in metric units and logarithmically in phi units (Φ) (Krumbien, 1934) using a modified Folk and Ward (1957) scale. GRADISTAT also calculated the fraction of sediment from each sample by size category (for example, clay, coarse silt, fine sand) based on Friedman and Saunders (1978), a modified Wentworth (1922) size scale. A macro function in Microsoft Excel, developed by the USGS SPCMSC, was applied to the data to calculate the average and standard deviation for each sample set (6-8 runs per sample), and highlight runs that varied from the set average by more than ± 1.5 standard deviations. Excessive deviations from the mean are likely the result of equipment error or extraneous organic material in the sample and are not considered representative of the sample. The highlighted runs were removed from the results, and the sample average was recalculated using the remaining runs. The averaged results for all samples, including the number of averaged runs and the standard deviation of the averaged results were summarized in an of Excel workbook with each core on its own tab. A comma-separated values data file containing the tabular data in plain text is included in the download file.

Dried, ground sediment from the 1-cm depth intervals of the 5 push cores were analyzed for the detection of radionuclides by standard gamma-ray spectrometry (Cutshall and Larsen, 1986) at the USGS SPCMSC radioisotope lab. Intervals from the uppermost 30 cm were analyzed from each core with alternating intervals analyzed from lower depths in the cores. The sediments were sealed in airtight polypropylene containers with plumbers' tape around the threads for the planar detectors or polystyrene test tubes for the well detector. Sediments placed in the test tubes were sealed with a layer of epoxy. To assess the airtightness of the polypropylene jars used on the planar detectors, core STM002M was run a second time with electrical tape around the outside of the jar where the cap screws on; this core is identified as STM002M-taped. The sample weights and counting container geometries were matched to pre-determined calibration standards. The sealed samples were stored for a minimum of 3 weeks prior to analysis to allow radium-226 (Ra-226) to come into secular equilibrium with its progeny isotopes lead-214 (Pb-214) and bismuth-214 (Bi-214). The sealed samples were then counted for 24-200 hours on a 16 x 40-millimeter well or 50-millimeter diameter planar-style, low energy, high-purity germanium, gamma-ray spectrometer. The suite of naturally occurring and anthropogenic radioisotopes measured along with their corresponding photopeak energies in kiloelectron volts (keV) are lead-210 (Pb-210, 46.5 keV), thorium-234 (Th-234, 63.3 keV), Pb-214 (295.7 and 352.5 keV; proxies for Ra-226), beryllium-7 (477.6 keV), Bi-214 (609.3 keV; proxy for Ra-226), cesium-137 (Cs-137, 661.6 keV), and potassium-40 (K-40, 1640.8 keV). Sample count rates were corrected for detector efficiency determined with International Atomic Energy Agency RGU-1 reference material, standard photopeak intensity, and self-absorption using an uranium-238 (U-238) sealed source (planar detectors only, Cutshall and others, 1983). All activities, with the exception of short-lived Pb-214 and Bi-214, were decay-corrected to the date of field collection. The radioisotopic activities reported in the Excel spreadsheet include the counting error for all samples, and results from each core are on its own tab. The critical level is reported for each core. A comma-separated values data file containing the tabular data in plain text is aslo included in the download file.

Following collection, foraminiferal sample centrifuge tubes were gently shaken twice a day for two weeks to ensure thorough staining with the rose Bengal (Schönfeld and others, 2012). Following staining, wet sample volumes were recorded using a marked syringe. Once recorded, samples were washed over a 63-, 125- and 850-micron (µm) sieve to remove clay material, and to separate out large organics (Schönfeld and others, 2012). All samples were wet sieved once and subsequently dried in an oven at 40 °C for a minimum of 24 hours. The 125–850 size fraction of each sample was picked to obtain census data (Schönfeld and others, 2012). Dry samples were split into equal parts using a microsplitter and spread evenly over a gridded picking tray. Entire splits were picked until at least 200 identifiable foraminiferal specimens were acquired to enable the calculation of foraminiferal densities (number of specimens per milliliter). Separate counts were made for stained (live) and unstained (dead) specimens to calculate the ratio and percentage of live to dead specimens. Specimens were identified using previously published literature (Brönnimann, 1979; Buzas and others, 1985; Loeblich and Tappan, 1988; Edwards and Wright, 2015; Haller and others, 2019; Rabien and others, 2015; World Register of Marine Species) and open nomenclature was utilized according to Bengtson (1988). Broken specimens were only counted if the umbilicus was preserved; chamber fragments were not counted. A species abbreviation and taxonomic reference table is available for reference. Live and dead foraminiferal census data are reported in the Excel spreadsheet. A comma-separated values data file containing the tabular data in plain text is included in the download file.