Evaluation and Local Calibration of Sentinel-2 Chlorophyll-a Algorithms in Kendal Coastal Waters, Indonesia.

Saved in:
Bibliographic Details
Title: Evaluation and Local Calibration of Sentinel-2 Chlorophyll-a Algorithms in Kendal Coastal Waters, Indonesia.
Authors: Maslukah, L.1 lilik_masluka@yahoo.com, Indrayanti, E.1 elisindrayanti@lecturer.undip.ac.id, Widada, S.1 sugengwidada@lecturer.undip.ac.id, Wirasatriya, A.1 anindyawirasatriya@lecturer.undip.ac.id, Krisna, H. N.2 herunur.krisno@gmail.com, Zainuri, M.1 muhammadzainuri@lecturer.undip.ac.id, Wisha, U. J.3 ulun002@brin.go.id
Source: International Journal of Geoinformatics. Apr2026, Vol. 22 Issue 4, p32-43. 12p.
Subject Terms: *Chlorophyll, *Coasts, *Artificial satellites, *Optimization algorithms, *Indonesians, *Empirical research, *Environmental monitoring, *Remote sensing
Geographic Terms: Indonesia
Abstract: Monitoring chlorophyll-a (Chl-a) concentrations in coastal waters is crucial due to its role as a biogeochemical indicator sensitive to environmental changes. Remote sensing techniques have been widely utilized for Chl-a estimation; however, the precision and relevance of algorithms developed in other regions require comprehensive evaluation, validation, and calibration against in situ Chl-a data. This study evaluated a Sentinel-2A Chl-a algorithm using 10m blue, green, red, and near-infrared (NIR) resolution bands (hereafter, MR4B) and an algorithm based on the green-red (GR) ratio band from another region. The performance of both was compared against an algorithm generated through in situ Chl-a calibration. Calibration was performed on the green-blue, green-red ratio bands, and a single band, using in situ Chl-a data collected on April 24, 2025, coinciding with the S2A satellite's passing time. The results showed that the performance of MR4B and GR was outperformed by the algorithms generated through the calibration process, where the SB algorithm showed superior performance, followed by the green-blue ratio and the green-red ratio, with root mean square error (RMSE) of 0.74 μg/L, 0.89 μg/L, and 0.93 μg/L, respectively. This study showed that the single band algorithm, demonstrated in the use of the green band (SB) provides a more practical and robust approach for Chl-a monitoring in this coastal system, with its simpler structure compared to other algorithms. However, further research is needed to examine the algorithm in the different season. [ABSTRACT FROM AUTHOR]
Database: Energy & Power Source
Description
Abstract:Monitoring chlorophyll-a (Chl-a) concentrations in coastal waters is crucial due to its role as a biogeochemical indicator sensitive to environmental changes. Remote sensing techniques have been widely utilized for Chl-a estimation; however, the precision and relevance of algorithms developed in other regions require comprehensive evaluation, validation, and calibration against in situ Chl-a data. This study evaluated a Sentinel-2A Chl-a algorithm using 10m blue, green, red, and near-infrared (NIR) resolution bands (hereafter, MR4B) and an algorithm based on the green-red (GR) ratio band from another region. The performance of both was compared against an algorithm generated through in situ Chl-a calibration. Calibration was performed on the green-blue, green-red ratio bands, and a single band, using in situ Chl-a data collected on April 24, 2025, coinciding with the S2A satellite's passing time. The results showed that the performance of MR4B and GR was outperformed by the algorithms generated through the calibration process, where the SB algorithm showed superior performance, followed by the green-blue ratio and the green-red ratio, with root mean square error (RMSE) of 0.74 μg/L, 0.89 μg/L, and 0.93 μg/L, respectively. This study showed that the single band algorithm, demonstrated in the use of the green band (SB) provides a more practical and robust approach for Chl-a monitoring in this coastal system, with its simpler structure compared to other algorithms. However, further research is needed to examine the algorithm in the different season. [ABSTRACT FROM AUTHOR]
ISSN:16866576
DOI:10.52939/ijg.v22i4.4937