SANDSNAP: CREATING A NATIONWIDE BEACH GRAIN SIZE DATABASE BY ENGAGING CITIZEN SCIENTISTS

This study assesses the effectiveness of SandSnap, a research initiative that engages citizen scientists in amassing the first public nationwide database of beach sand grain sizes on U.S. coastlines. Citizen scientists can contribute to the database by taking a picture of sand at a beach with a U.S. coin, uploading the image to the SandSnap website, and recording the location using the phone’s built-in GPS. Sediment gradation is returned to the user within 2 minutes of image upload and the results are stored on a public database. Image processing and outreach initiatives are detailed. Results from an experiment where 31 participants with varying phones took a SandSnap of the same sand show the median grain size had a mean percent error of 21.3%. Image processing techniques are continuing to be improved and the supporting neural networks are regularly retrained with more data to improve the accuracy and robustness of the SandSnap results. The spatially and temporally robust beach grain size database being developed by SandSnap will help to improve numerous coastal engineering analyses including coastal resilience and vulnerability quantification, beach nourishment life cycle and uncertainty analysis, beach compatibility for beneficial uses of dredged sediment, and large-scale coastal morphology modeling. Crowdsourcing beach sand grain size data collection provides a cost-effective way to accumulate unique sand grain sizes on a large scale while increasing public engagement and providing a better understanding of sand on U.S. coasts.