Towards sustainable coastal management: aerial imagery and deep learning for high-resolution <i>Sargassum</i> mapping

The massive arrival of pelagic <i>Sargassum</i> on the coasts of several countries of the Atlantic Ocean began in 2011 and to date continues to generate social and environmental challenges for the region. Therefore, knowing the distribution and quantity of <i>Sargassum</i> in the ocean, coasts, and beaches is necessary to understand the phenomenon and develop protocols for its management, use, and final disposal. In this context, the present study proposes a methodology to calculate the area <i>Sargassum</i> occupies on beaches in square meters, based on the semantic segmentation of aerial images using the pix2pix architecture. For training and testing the algorithm, a unique dataset was built from scratch, consisting of 15,268 aerial images segmented into three classes. The images correspond to beaches in the cities of Mahahual and Puerto Morelos, located in Quintana Roo, Mexico. To analyze the results the fβ-score metric was used. The results for the <i>Sargassum</i> class indicate that there is a balance between false positives and false negatives, with a slight bias towards false negatives, which means that the algorithm tends to underestimate the <i>Sargassum</i> pixels in the images. To know the confidence intervals within which the algorithm performs better, the results of the f0.5-score metric were resampled by bootstrapping considering all classes and considering only the <i>Sargassum</i> class. From the above, we found that the algorithm offers better performance when segmenting <i>Sargassum</i> images on the sand. From the results, maps showing the <i>Sargassum</i> coverage area along the beach were designed to complement the previous ones and provide insight into the field of study.