We collected microplastics on six beaches on the windward side of Oahu. Sampling was carried out in October 2017 during low tides. At each beach, the high tide line was identified by a strip of debris or discoloration from the previous high tide, and inland and parallel to the high tide line, the storm tide line was identified by a linear concentration of debris such as branches, twigs or seeds. Three 15-m transect lines were haphazardly laid along the high tide line, and storm tide lines, end to end. Random coordinates were generated to locate three 30 cm x 30 cm quadrats along each transect, and in each quadrat the first two cm of sand was collected (Fig 3B). In this study we quantified microplastics 500 µm - 5 mm in size. Larger plastic pieces were separated out. Sand was sieved with a 500 µm mesh, collecting only plastics and other materials larger than that size. In total there were 18 samples from each beach, nine each from high and storm tide lines. Individual pieces of microplastic collected per sample were counted by pouring samples through a 5 mm sieve, discarding larger materials caught in the sieve, and retaining materials that passed through.We used density separation (floating in water) and visual inspection with a stereomicroscope to separate plastics from organic (shells, sand, wood) and inorganic (metal) materials. Previous studies have shown that other i.e. spectroscopic methods can identify small-sized microplastics overlooked using microscopy but by restricting our analysis to particles >500 µm we are unlikely to have significantly undercounted microplastics. The number of plastic pieces was then divided by the quadrat size to find the density of plastic pieces per unit area. Size distribution of microplastics was characterized using stacked sieve shakers with mesh sizes of 4 mm, 2 mm and 1 mm. Samples collected from the same tide line at the same site were combined. The stack was shaken for 30 s and then the number of plastics in each size range were counted. One sample of sand was collected at each beach near the transect lines by collecting the first 2 cm to be analyzed for particle size distribution as a measure of texture. Sand samples from each beach were weighed and distributed through the stacked sieve shaker with mesh sizes 4 mm, 2 mm, 1 mm, 500 mm, 250 mm, 100 mm, 63 mm. The sieve shaker was run for 3 minutes. The portion from each sieve was weighed.

We report microplastic densities on windward beaches of Oahu, Hawai`i, USA. Microplastic densities, surveyed on six Oahu beaches, were highest on the beaches with the coarsest sands, associated with high wave energy. On those beaches, densities were very high (700-1700 particles m-2), as high as those recorded on other remote island beaches worldwide.