Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine

The East Kolkata Wetlands (EKW) is a globally recognized Ramsar site, where municipal wastewater is sustainably recycled through several aquaculture ponds. This study focuses on a prominent aquaculture pond, Boro Gopeshwar Bheri, within EKW, to assess surface water turbidity, a critical indicator fo...

Full description

Bibliographic Details
Main Authors: Mullick, A., Ghosh, Surajit, Chowdhury, A., Bhattacharyya, S.
Format: Preprint
Language:Inglés
Published: 2025
Subjects:
Online Access:https://hdl.handle.net/10568/180499
_version_ 1855524414426185728
author Mullick, A.
Ghosh, Surajit
Chowdhury, A.
Bhattacharyya, S.
author_browse Bhattacharyya, S.
Chowdhury, A.
Ghosh, Surajit
Mullick, A.
author_facet Mullick, A.
Ghosh, Surajit
Chowdhury, A.
Bhattacharyya, S.
author_sort Mullick, A.
collection Repository of Agricultural Research Outputs (CGSpace)
description The East Kolkata Wetlands (EKW) is a globally recognized Ramsar site, where municipal wastewater is sustainably recycled through several aquaculture ponds. This study focuses on a prominent aquaculture pond, Boro Gopeshwar Bheri, within EKW, to assess surface water turbidity, a critical indicator for the health and functioning of that ecosystem. The study aims to develop a multivariate regression model (MRM) to predict turbidity using satellite-derived spectral reflectance and indices as independent variables. High-resolution imagery from Sentinel-2A MSI was correlated with in-situ turbidity measurements collected using a calibrated turbidimeter. A combination of the key spectral bands was used to understand their strong sensitivity to suspended particulate matter. The model was statistically calibrated using ground-truth turbidity and showed satisfactory predictive performance (R² =0.99; RMSE < 10 NTU), validating it for estimating turbidity in wetland waters. The MRM generates a turbidity image as a continuous raster layer showing predicted turbidity levels (NTU) across the area of interest, enabling spatial and temporal comparisons and helping to identify consistent turbidity patterns and potential zones of elevated turbidity. This hybrid approach of combining remote sensing data and geostatistical analysis offers a high-throughput, cost-effective monitoring framework and a decision-support tool for sustainable wetland management.
format Preprint
id CGSpace180499
institution CGIAR Consortium
language Inglés
publishDate 2025
publishDateRange 2025
publishDateSort 2025
record_format dspace
spelling CGSpace1804992026-01-23T10:10:54Z Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine Mullick, A. Ghosh, Surajit Chowdhury, A. Bhattacharyya, S. aquaculture ponds turbidity modelling wetlands satellite imagery The East Kolkata Wetlands (EKW) is a globally recognized Ramsar site, where municipal wastewater is sustainably recycled through several aquaculture ponds. This study focuses on a prominent aquaculture pond, Boro Gopeshwar Bheri, within EKW, to assess surface water turbidity, a critical indicator for the health and functioning of that ecosystem. The study aims to develop a multivariate regression model (MRM) to predict turbidity using satellite-derived spectral reflectance and indices as independent variables. High-resolution imagery from Sentinel-2A MSI was correlated with in-situ turbidity measurements collected using a calibrated turbidimeter. A combination of the key spectral bands was used to understand their strong sensitivity to suspended particulate matter. The model was statistically calibrated using ground-truth turbidity and showed satisfactory predictive performance (R² =0.99; RMSE < 10 NTU), validating it for estimating turbidity in wetland waters. The MRM generates a turbidity image as a continuous raster layer showing predicted turbidity levels (NTU) across the area of interest, enabling spatial and temporal comparisons and helping to identify consistent turbidity patterns and potential zones of elevated turbidity. This hybrid approach of combining remote sensing data and geostatistical analysis offers a high-throughput, cost-effective monitoring framework and a decision-support tool for sustainable wetland management. 2025-12-30 2026-01-23T10:09:34Z 2026-01-23T10:09:34Z Preprint https://hdl.handle.net/10568/180499 en Open Access Mullick, A.; Ghosh, S.; Chowdhury, A.; Bhattacharyya, S. 2025. Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine. ResearchGate, 13p. doi: https://doi.org/10.13140/RG.2.2.35618.70088
spellingShingle aquaculture
ponds
turbidity
modelling
wetlands
satellite imagery
Mullick, A.
Ghosh, Surajit
Chowdhury, A.
Bhattacharyya, S.
Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine
title Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine
title_full Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine
title_fullStr Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine
title_full_unstemmed Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine
title_short Modeling turbidity of aquaculture pond in the East Kolkata Wetlands using Sentinel-2 bands and ground data through multivariate regression on Google Earth Engine
title_sort modeling turbidity of aquaculture pond in the east kolkata wetlands using sentinel 2 bands and ground data through multivariate regression on google earth engine
topic aquaculture
ponds
turbidity
modelling
wetlands
satellite imagery
url https://hdl.handle.net/10568/180499
work_keys_str_mv AT mullicka modelingturbidityofaquaculturepondintheeastkolkatawetlandsusingsentinel2bandsandgrounddatathroughmultivariateregressionongoogleearthengine
AT ghoshsurajit modelingturbidityofaquaculturepondintheeastkolkatawetlandsusingsentinel2bandsandgrounddatathroughmultivariateregressionongoogleearthengine
AT chowdhurya modelingturbidityofaquaculturepondintheeastkolkatawetlandsusingsentinel2bandsandgrounddatathroughmultivariateregressionongoogleearthengine
AT bhattacharyyas modelingturbidityofaquaculturepondintheeastkolkatawetlandsusingsentinel2bandsandgrounddatathroughmultivariateregressionongoogleearthengine