Posters

Poster Session 2, Tuesday,  October  4, 10:40–12:40

Poster 106

Variability of Coastal Plumes on the Inner Shelf of the South Atlantic Ocean based on Artificial Intelligence Methods

Coastal plumes play a key role in coastal dynamics, being associated with productivity in the aquatic environment, and contributing to patterns of deposition and transport of suspended particulate matter. The main objective of this work is to propose an alternative and innovative methodology to evaluate the spatial and temporal variability of coastal plumes formed by suspended particulate matter (SPM) of continental origin. This methodology is based on the combination of remote sensing tools and artificial intelligence methods to identify and monitor the morphology of these surface features. The case study will be the coastal plume of Lagoa dos Patos, but the methodology can be widely applied. In this context, images from the LANDSAT series (5 TM, 7 ETM+, 8 and 9 OLI), for the period from 1984 to 2022, totaling 318 scenes, underwent atmospheric correction in order to mitigate atmospheric effects, and were visually inspected for the purpose of clouds. Then, through the application of generic semianalytical algorithm, the SPM concentration was estimated for the scenes in the red and near infrared (NVI) bands, where by comparisons the band with the best accuracy in the MPS estimation was defined. Subsequently, based on the similarity between pixels used in the artificial intelligence methods, the regions considered as plumes were delimited in the images, making it possible to evaluate the temporal characteristics and evaluate the spatial and temporal variability of the Lagoa dos Patos plume. Therefore, the results obtained prove the efficiency of the method in identifying these features in satellite images.

Yanna Rodrigues, Universidade Federal do Rio Grande, [email protected]

Juliana Tavora, University of Twente, [email protected], 0000-0002-0314-1250

Elisa Helena Hernandes, Universidade Federal do Rio Grande, [email protected], 0000-0003-1869-0233