Metal and other element partitioning in southwest Puerto Rico ultramafic soil, Rio Loco sediment, and Guanica Bay sediment

SOIL AND SEDIMENT COLLECTION: Find-grained stream sediment was collected by hand from stream banks or stream beds June 19-21, 2017, and placed into wire-top polyethylene bags. Seabed sediment was collected using a boat-deployed Petit Ponar benthic grab sampler, and the 0-2 centimeter surface layer was subsampled into wire-top polyethylene bags. Forest soil was collected from Guanica Dry Forest on January 19, 2018, and from two sites in the Bosque Susua on March 27, 2018. A shallow trench was dug in the forest floor at each site and 2-centimeter-thick soil intervals were subsampled into wire-top polyethylene bags. Soils up to 6 centimeters deep were sampled from Guanica Dry Forest and up to 8 centimeters deep from Bosque Susua.

SEDIMENT SAMPLE PREPARATION: Soil and sediment were oven-dried at 60 degrees Celsius, disaggregated gently to avoid changing grain sizes, and dry-sieved through stainless steel sieves to obtain the less than 0.063 millimeter-diameter fine fraction for sequential extractions and analyses of total element contents.

SHORT-LIVED RADIONUCLIDE ACTIVITIES: Bulk sediment was oven-dried at 60 degrees Celsius, homogenized, and packed into 10 milliliter vials with air-tight caps. Activities of short-lived radionuclides were counted for 72 hours in a high-purity low-background germanium (HPGe) well detector at the USGS Pacific Coastal and Marine Science Center and quantified relative to certified reference materials with similar geometries and counting times. Beryllium-7 activities were counted at 477.6 kiloelectron volts (keV) and Cesium-137 activities at 661.6 keV. Unsupported Lead-210 activities were calculated by subtracting Lead-210 activities counted at 46.52 keV from the average of supported Radium-226 activities counted indirectly at 351.9 keV and 609.3 keV. Detector efficiency for Beryllium-7 was interpolated from peaks at 351.9 keV and 609.3 keV using International Atomic Energy Agency (IAEA) reference standard RGU-1. Detector efficiency for Cesium-137 was determined using IAEA Standard Reference Material 300. Detector efficiency for Lead-210 was determined using IAEA standards RGU-1 and RGTh-1. Combined uncertainties include 1-sigma counting errors, detector efficiency, and background corrections.

SEQUENTIAL EXTRACTIONS: Fine-grained (less than 0.063-mm-diameter) soil and sediment were sent to the Analytical Chemistry Project of the USGS Geology, Geophysics, and Geochemistry Science Center (GGGSC), which contracted with nationally accredited AGAT Laboratories for 5-stage sequential extractions. The five phases are operationally defined and include: 1) adsorbed and calcium-bound, 2) hydrous iron-, manganese-oxide-bound, 3) crystalline iron-oxide-bound, 4) sulfide-bound, and 5) residual. Quantification was performed by inductively coupled plasma optical emission spectroscopy (ICP-OES) for major and some minor elements and by inductively coupled plasma mass spectroscopy (ICP-MS) for some minor and all trace elements.

TOTAL ELEMENT CONTENTS: Fine-grained (less than 0.063-mm-diameter) soil and sediment were sent to the Analytical Chemistry Project of the USGS Geology, Geophysics, and Geochemistry Science Center (GGGSC), which contracted with nationally accredited AGAT Laboratories for total element content analyses. Soil and sediment were decomposed with a four-acid digestion at low temperature or a sodium peroxide fusion at 750 degrees Celsius followed by reconstitution in nitric acid for elements hosted in acid-resistant minerals. Quantification was performed by inductively coupled plasma optical emission spectroscopy (ICP-OES) for major and some minor elements and by inductively coupled plasma mass spectroscopy (ICP-MS) for some minor and all trace elements. Total mercury was measured by cold vapor atomic absorption.