Banca de DEFESA: RICARDO CARLOS CARVALHO

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
STUDENT : RICARDO CARLOS CARVALHO
DATE: 26/02/2026
TIME: 14:00
LOCAL: Vídeoconferencia
TITLE:

APPLICATION OF SATELLITE-DERIVED BATHYMETRY TECHNIQUES IN TROPICAL ENVIRONMENTS: A CASE STUDY IN AN AREA OF THE EASTERN CONTINENTAL SHELF OF RIO GRANDE DO NORTE

KEY WORDS:

Band ratio, remote sensing, median filters, atmospheric correction, ACOLITE

PAGES: 102
BIG AREA: Ciências Exatas e da Terra
AREA: Geociências
SUBÁREA: Geologia
SUMMARY:

The morphological characterization of shallow continental shelves is fundamental for understanding sedimentary processes and supporting coastal management. However, a significant portion of these areas has historically lacked high-resolution bathymetric data due to financial and logistical limitations—particularly in stretches where oceanographic conditions hinder vessel-based surveys—associated with traditional hydroacoustic methods. This doctoral thesis proposes and validates an integrated methodological workflow for the application of Satellite-Derived Bathymetry (SDB) in tropical environments. The work is developed in two complementary stages that sequentially address the empirical modeling of submarine relief and the enhancement of radiometric data quality, utilizing a sector of the eastern shelf of Rio Grande do Norte as the study area. This region is characterized by physical obstacles, such as outcrops and reef banks, which impede navigation and compromise the quality of hydroacoustic records. In the first stage, the study focused on evaluating the applicability of SDB through the empirical band ratio model proposed by Stumpf et al. (2003), analyzing its feasibility for preliminary surveys in areas lacking bathymetric information. A relevant methodological contribution of this phase was the incorporation of median filters in post-processing, a technique that effectively mitigated spectral noise amplified by the logarithmic transformations of the model. The second stage, aimed at maximizing the accuracy of previously tested bathymetric models, delved into the analysis of signal radiometric quality, considering the challenges imposed by atmospheric interference and specular reflection (Sunglint). Different processing workflows were compared, contrasting operational Sentinel-2 Level L2A products (corrected by the Sen2Cor algorithm) with approaches applied to Level L1C images using the ACOLITE processor and Deglint algorithm. The results demonstrated that processing via ACOLITE, employing the Dark Spectrum Fitting (DSF) algorithm and SWIR bands to estimate atmospheric reflectance, statistically outperformed L2A products. Furthermore, it was shown that the direct use of L2A images may result in the loss of spectral information relevant to aquatic environments, whereas the correction performed by ACOLITE significantly increased the consistency of data employed in SDB.

COMMITTEE MEMBERS:
Externo ao Programa - 3342435 - ANTONIO RODRIGUES XIMENES NETO - nullExterna ao Programa - 1226434 - MARY LUCIA DA SILVA NOGUEIRA - nullPresidente - 2042405 - MOAB PRAXEDES GOMES
Externo à Instituição - PAULO VICTOR DO NASCIMENTO ARAÚJO - IFRN
Externa à Instituição - TEREZA CRISTINA MEDEIROS DE ARAÚJO - UFPE