Globorotalia truncatulinoides Trace Element Geochemistry (Barium, Magnesium, Strontium, Manganese, and Calcium) from the Gulf of Mexico Sediment Trap

LA-ICP-MS: Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) on G. truncatulinoides specimens was conducted at The University of California, Davis Stable Isotope Laboratory, using a Photon Machines 193 nanometer (nm) ArF ultraviolet (UV) excimer laser with an ANU HelEx dual-volume laser ablation cell coupled to an Agilent 7700× quadrupole-ICP-MS, please see Data_Dictionary_Trace_Element_Geochemistry.docx for instrument setting details. The G. truncatulinoides shells were placed on double sided carbon tape spiral side up to ensure a horizontal sampling surface for each chamber. Laser spot size of 44 × 44 micrometers (μm) in diameter was used with a repetition rate of 6 hertz, Hz, (non-encrusted forms) and 8 Hz (encrusted forms). Due to the thickness of the calcite, a higher repetition rate was needed to ablate through the encrusted shells. For the smallest foraminifers, the spot size was decreased to 30 × 30 μm to ensure ablation within a single chamber. Masses measured included magnesium-24 (24Mg), magnesium-25 (25Mg), aluminimum-27 (27Al), calcium-44 (44Ca), manganese-55 (55Mn), strontium-88 (88Sr), yttrium-89 (89Y), and barium-138 (138Ba). Standard reference materials from the National Institute of Standards and Technology (SRM NIST) 610, 612, and 614 glass standards were run before and after each batch of samples as an external standard. An O. universa shell, which is demonstrated to have highly reproducible trace element (TE) profiles throughout, was analyzed before and after each run as an internal working standard (7.0 ± 0.7 millimoles per mole [mmol/mol] Mg/Ca, 2 standard deviations [σ], Fehrenbacher and others, 2015). Outliers in the TE/Ca (trace element/calcium) profiles that were greater than ± 6 standard deviations from a 3-point rolling mean were removed from raw LA-ICP-MS signals, then data were reduced using Iolite Software (Paton and others, 2011). Depth profiles were analyzed on each chamber (F–F2 for every specimen, and up to F7 for select specimens). If chambers were large enough, we analyzed up to 3 repeat spots to assess reproducibility and data quality assurance and quality control (QA/QC). The pooled standard deviation among replicate spots was 0.44 micromoles per mole, μmol/mol, (n=71) for encrusted and 1.24 μmol/mol (n=270) in non-encrusted specimens. Individual O. universa specimens were analyzed via LA-CIP-MS at Oregon State University using a Teledyne/Photon Machines 193 nm ArF UV excimer laser ablation system with a HelEx laser ablation cell coupled to a Thermo Scientific X-series II quadrupole ICP-MS, please see Data_Dictionary_Trace_Element_Geochemistry.docx for instrument setting details. Gas composition and flow rate were determined by adjusting the flow of argon (Ar) and helium (He) as necessary to achieve high count rates on the sample/standard while maintaining ThO+/Th+ ratios less than 0.2% (tuned daily). Shell fragments were analyzed from the inside - out. Samples were analyzed using a 50–70 μm diameter spot size at 5 Hz repetition rate and 1.0–1.5 joules per square centimeter (J/cm2) laser fluence. Data acquisition varied between 20–60 seconds (s) per spot analysis. Masses measured included 24Mg, 25Mg, 27Al, 43Ca, 44Ca, 55Mn, 78Sr, 88Sr, 138Ba, and uranium-238 (238U). 43Ca was used as the internal standard and 44Ca was monitored for consistency. Dwell times varied by analyte and ranged from 0.20–0.5 milliseconds (ms). Individual mean elemental concentrations were calculated using Iolite Software (Paton and others, 2011). The mean TE/Ca ratio for each profile was then calculated by normalizing to the known TE concentration in the drift/background-corrected bracketed analyses of the NIST SRM 610 and 612 glass standards at 5 Hz and approximately 4.5 J/cm2 laser fluence (Jochum and others, 2011).

Trace Metal Analysis in Seawater: The Ba, Sr, and Ca concentrations were measured in seawater samples collected from the sediment trap site from the surface to 1100 meters water depth in July 2018 and February 2019. Additional water sampling was carried out on the July 2018 cruise at 2 nearshore sites on the Louisiana shelf in 17 meters water depth (28.87°N, 90.49°W) and 50 meters water depth (28.38°N, 90.47°W). All water samples were collected in a Niskin water sampling rosette, filtered through acid-cleaned Millex® PVDF (0.45 μm pore size) membrane filters into acid-leached 50 milliliter (mL) centrifuge tubes, and acidified with 25 microliters (μl) Optima grade concentrated nitric acid (HNO3). Seawater was diluted 10-fold with 2% nitric acid, spiked with an internal standard containing beryllium (Be), scandium (Sc), Y, and indium (In) to correct for instrumental drift, and analyzed using an Agilent 7500cx ICP-MS equipped with a High Matrix Introduction accessory, which uses aerosol dilution to reduce matrix effects, and an octopole reaction cell, which uses helium gas to reduce polyatomic interferences (Wilschefski and Baxter, 2019). Samples were analyzed for Sr in ultra-robust mode and in the presence of helium, and for Ca and Ba in robust mode and the absence of helium. Two external six-point calibration curves, one for elements with concentrations in the ppm range (sodium [Na], Mg, Ca) and one for elements with concentrations in the ppb range (lithium [Li], Mn, Sr, Ba), were analyzed before and after samples. Individual elemental standards were obtained from SPEX CertiPrep (Metuchen, New Jersey).

Solution Based ICP-MS: The 49 solution-based G. ruber (pink and white) analyses from the 2010 sediment trap samples were cleaned according to the Barker and others (2003) procedure in a laminar flow clean bench and analyzed for Mg/Ca and Ba/Ca using inductively coupled plasma mass spectrometry at Texas A&M, College Station, Texas, United States of America, on a Thermo Scientific Element XR High Resolution ICP-MS. The corresponding Mg/Ca data are published in Richey and others (2022). (2019). All other solution-based trace element analyses (for example, G. truncatulinoides, O. universa, G. menardii, P. obliquiloculata, G. tumida, N. dutertrei and G. ruber collected from 2018 sediment trap samples) were carried out using a Thermo Finnegan Element 2 in the Center for Elemental Mass Spectrometry at the University of South Carolina. Foraminifers were crushed gently in a microcentrifuge tube and cleaned as samples for laser analyses in a 1:1 mixture of H2O2 and NaOH in a warm water bath (approximately 65 °C) for 10 minutes, during which samples were sonicated 4 times for 10s. Samples prepared for solution analyses were dissolved in 1% nitric acid and then diluted in a matrix of 2% nitric acid to a consistent concentration of 20 parts per million (ppm) Ca. Dissolved samples were run along with in house standard solutions, each containing 20 ppm Ca and varying concentrations of other analytes including 25Mg, 43Ca, 55Mn, 87Sr, 137Ba, and 238U.