| Processing Steps |
- Parameter or Variable: microplastic concentration (measured); Units: counts per individual; Observation Category: in situ; Sampling Instrument: hand collection; Sampling and Analyzing Method: San Francisco Bay is the largest estuary on the west coast of North America. It spans approximately 2575 kilometers and is surrounded by continuous, dense urban areas. The study sampled resident mussels and clams from five sites within San Francisco Bay for microplastics. Cages of depurated mussels (denoted transplants) were also deployed at four sites in the Bay for 90 days to investigate temporal uptake of microplastics and microparticles. Resident California Mytilus spp. (mussels) were sampled from three margin sites within the Bay in July 2018. The resident mussels are likely a hybrid of Mytilus galloprovincialis and Mytilus trossulus. Additionally, Corbicula fluminea (clams) were collected at two river sites, as the water at these sites are not saline enough for Mytilus spp. to inhabit. Mytilus californianus were collected from the Bodega Marine Reserve (Bodega Bay, CA) and depurated in seawater tanks for approximately one month. During this time fouling organisms were removed to minimize the potential transfer of non-resident species to the Bay. The depurated mussels were then deployed at four sites in the Bay. At each of four transplant sites, two cages each with 100-150 depurated mussels were deployed. The cages were deployed at the sites between July 15th and 20th in 2018 for approximately 90 days. The transplants were retrieved via satellite GPS between October 23rd and November 2nd in 2018. A subset of depurated Bodega Head mussels was not deployed (time zero samples) and were analyzed for baseline microplastic contamination. At each site, four to ten individual bivalves were collected of similar size and weight and were composited into one sample. Three composite samples were collected at each site (with the exception of Sacramento River and Pinole Point sites where n ¼ 2). All bivalve samples were maintained below freezing until time of microplastic analyses at the University of Toronto (Toronto, Canada). In addition, approximately 100 individual bivalves from the transplant and river sites were collected and analyzed for PAHs at SGS AXYS (Sidney, Canada). The length and width of each bivalve, in the shell, was measured. The shells were then rinsed with reverse osmosis (RO) filtered water before opening. The soft tissue was removed from the shell, rinsed with RO water, and placed in a clean labelled polypropylene composite sample jar. Polypropylene jars are used instead of glass as KOH etches the glass introducing glass particles to our samples. KOH does not etch polypropylene, and we do not find polypropylene particles contaminating our samples or our laboratory blanks. To digest the tissue, 200 g/L of KOH solution (20% KOH in RO water) was added to each composite sample jar at a volume of three times the sample volume (minimum 15 mL). Each sample jar was loosely capped and left for approximately 14 days at room temperature in a fume hood until all organic matter had digested. Each digested sample was then poured through a 125 mm stainless steel sieve and the contents were transferred into a clean polypropylene jar. The sieve was sufficiently rinsed with RO water to maximize recovery. The sample was filtered in a clean cabinet onto a 10 mm polycarbonate filter (Sterlitech) using vacuum filtration. All parts of the vacuum filtration apparatus were rinsed with RO water to ensure all microparticles were transferred onto the polycarbonate filter. The filter was removed from the filtration device using clean stainless-steel tweezers and placed into a clean, glass Petri dish. Using methodology employed in previous studies, suspected microplastic particles were extracted from the polycarbonate filter. The filter (containing particles >125 mm) was analyzed for anthropogenic microparticles under a dissecting microscope. The filter was scanned square by square by following the lines on a grid sticker placed under the Petri dish. Suspected anthropogenic microparticles observed were removed from the filter and placed onto double-sided sticky tape mounted on transparent paper. The suspected anthropogenic microparticles were characterized by shape (e.g., fragment, fiber, film, sphere, pellet, foam) and by color. Only the first 10 particles of each color-category combination (e.g., blue fiber, black fragment) were removed using tweezers and placed on the double-sided sticky tape. The remaining were tallied, but not removed from the filter. To avoid double counting, clear fibers were counted using the black background on the microscope, while all other colored microparticles were counted using the white background. All microparticles picked from each filter were photographed and measured using ImageJ. The longest dimension of each particle was recorded. Raman spectroscopy was used to chemically identify suspected microplastics (Xplora Plus; Horiba Scientific with LabSpec 6 software). For each sample, including blanks, roughly 10% of microparticles from each color-category combination were randomly selected for chemical identification. Because we round up to the nearest whole number, this resulted in chemical analysis of 193 particles out of 772 particles in total (25% of all particles). Using KnowItAll software, spectra were matched to commercial libraries from Bio-Rad, HORIBA and Sigma Aldrich, and in-house libraries ‘SLoPP’ and ‘SLoPP-E’ (Munno et al., 2020). Microparticles were categorized based upon chemical identification results. Categories include plastic (specific polymer type is identified, e.g., polypropylene, polyester, acrylic). Confirmed microplastics only include particles from which spectra matched a specific polymer type. In this study, we report suspected anthropogenic microparticles, confirmed anthropogenic microparticles, and confirmed microplastic counts.; Data Quality Method: In the laboratory, procedural contamination was minimized during extraction by wearing cotton laboratory coats and nitrile gloves. The lab also has a 24-h HEPA filter system to reduce the potential for cross contamination from the deposition of airborne microparticles. All bivalves, glassware, and tools used in extraction were rinsed with RO water three times. The area used for dissection was cleaned prior to use, and glassware was covered with aluminum foil when not in use. The filtration process occurred in a clean cabinet where contamination is minimized. To account for procedural contamination, one blank (RO water) was run every 10 samples and treated as a normal sample. Lab blanks were collected three times during mussel extraction and once during clam extraction. Clam extraction occurred several months after mussel extraction and therefore had its own blank. The mussel samples were blank corrected by color-category combination, which was an average of the three lab blanks collected during mussel extraction. The clam samples were blank corrected by color-category combination from the one lab blank taken during clam extraction. Moreover, transplant mussels were also corrected to account for initial microparticle contamination by subtracting the time zero sample average, also by color-category combination. The time zero sample was first blank subtracted with the lab blank to control for procedural lab contamination, to ensure the count only reflected any initial particles in the mussels that may not have been removed through depuration. A subset of particles from the blanks and time zero samples were analyzed for chemical identity using Raman spectroscopy..
|
