129-32 Determining Cross-Shore Variation in Grain Size Using Image Analysis: A Case Study of Sand from Puerto Rico

Abstract:

Various studies have demonstrated a relationship between grain size and depositional environment. For example, fluvial, beach, and dune samples have different grain size distributions as a result of the transport mechanisms and energy in these environments. However, traditional grain size analysis using sieves can be tedious. In contrast, laser granulometry can determine grain size distributions efficiently, but does not provide information on the composition of the grains that would come from visual inspection either with the naked eye during sieve analysis or subsequent microscopy. Image analysis for grain size on the other hand allows the researcher to not only determine grain size but also provides the opportunity to determine grain composition through visual inspection of microscope pictures of sample grains.

This study seeks to develop a method for characterizing sample grain sizes from microscope photos of sand samples collected along cross-shore transects from various beaches in Puerto Rico using the image processing software ImageJ. The primary goal is to differentiate foreshore (intertidal) from backshore (dune) sands based on grain size distribution. Initial attempts to automate the process of measuring grain size dimensions using ImageJ were unsuccessful, as polished biogenic carbonate grains often had a glare and were difficult to distinguish from the image background using image edits of contrast and color. A method of measuring grain dimensions from images manually using ImageJ was therefore developed, with the maximum and minimum dimension of at least 125 grains per sample determined. The minimum dimension of a grain was important in order to process the data in a simulation of traditional grain size analysis with sieves for comparison to previous studies. Not only was the method employed able to produce a grain size distribution that was in agreement with distributions from the expected environments as found in previous studies, but the grain size data collected using this method was also reproducible among two different undergraduate students with different amounts of training. Additionally, the collection of images allowed an efficient visual determination of grain composition, and thereby mineralogy, that is also in agreement with the location of the samples along the cross-shore transect.

© Copyright 2026 The Geological Society of America (GSA), all rights reserved.