Communication of Surface Water Turbidity Estimation via Segmented PCA and Machine Learning on Landsat 8 Data

International Journal of Electronics and Communication Engineering

© 2025 by SSRG - IJECE Journal

Volume 12 Issue 12

Year of Publication : 2025

Authors : Mule Abhi Roop, D. Gowri Sankar Reddy

10.14445/23488549/IJECE-V12I12P102

How to Cite?

Mule Abhi Roop, D. Gowri Sankar Reddy, "Communication of Surface Water Turbidity Estimation via Segmented PCA and Machine Learning on Landsat 8 Data," SSRG International Journal of Electronics and Communication Engineering, vol. 12,  no. 12, pp. 11-21, 2025. Crossref, https://doi.org/10.14445/23488549/IJECE-V12I12P102

Abstract:

The second-largest brackish water Lagoon in India is Pulicat Lake, with an area of approximately 759 km². It is being supported by inland river systems and opened to the Bay of Bengal, creating a dynamic interface between freshwater and marine environments. The lake also hosts an abundant flora and fauna; this is very crucial in ensuring the balance of the ecosystem and is also a critical habitat for the birds. The use of turbidity as a measure of water quality, ecosystem, and land cover dynamics is relevant to monitoring turbidity in Pulicat Lake. Remote sensing would be an efficient technique for elucidating both spatial and temporal changes in turbidity. This research indicates that a hybrid mix of Segmented Principal Component Analysis (SPCA) and the machine learning process will be used to enhance the performance of predicting the level of turbidity by applying Landsat 8 images. The suggested SPCA-based system considerably improves the classification accuracy, reaching the general accuracy of 99.2664 percent and a Kappa coefficient of 0.9896, which is higher than the traditional approaches. Precisely, the NDTI/Random Forest model gave an overall accuracy of 88.0189% and a Kappa coefficient of 0.8313, whereas the Band 4/Random Forest model gave an overall accuracy of 98.8656% and a Kappa coefficient of 0.9839. All these findings indicate the efficiency and durability of the SPCA-bound method of monitoring turbidity in ecologically very sensitive areas such as the Pulicat Lake.

Keywords:

Pulicat Lake, Landsat – 8, Turbidity, Principal Component Analysis (PCA), NDTI, Random Forest Algorithm.

References:

[1] Kannan Vaithianathan, Vulnerable Ecosystem: The Pulicat Lake Needs Government's Attention, Earthy Worthy, 2022. [Online]. Available: https://www.researchgate.net/publication/358888459_Vulnerable_ecosystem_the_Pulicat_lake_needs_government's_attention
[2] S. Rajakumari et al., “Study of the Pulicat Lagoon on the basis of Deprived Vegetation and Water Area against Increased Land Surface Temperature,” Research Square, pp. 1-31, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[3] R. Illangovan, Restoration of Polluted Lakes Using Structural and Non Structural Approches, Scribd, 2016. [Online]. Available: https://www.scribd.com/presentation/595382568/Restoration-of-Polluted-Lakes-29-12-2016-Ppt-Autosaved
[4] P.J. Sanjeeva Raj, Macro Fauna of Pulicat Lake, National Biodiversity Authority, pp. 1-67, 2006. [Online]. Available: http://nbaindia.org/uploaded/docs/bulletin6-pulicatlake.pdf
[5] R. Syamala, and E. Hemavathy, “Physico-Chemical Parameters and Land Use Patterns of Pulicat Lake, Tamil Nadu, India,” International Journal of Advanced Scientific and Technical Research, vol. 8, no. 6, pp. 10-37, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Wei Jiang et al., “Detecting Water Bodies in Landsat 8 OLI Image Using Deep Learning,” International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 42, no. 3, pp. 669-672, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Palash Uddin, Md. Al Mamun, and Md. Ali Hossain, “PCA-based Feature Reduction for Hyperspectral Remote Sensing Image Classification,” IETE Technical Review, vol. 38, no. 4, pp. 377-396, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Earth Explorer, USGS. [Online]. Available: https://earthexplorer.usgs.gov/
[9] Ahmed Mohsen, Mohamed Elshemy, and Bakenaz Zeidan, “Water Quality Monitoring of Lake Burullus (Egypt) using Landsat Satellite Imageries,” Environmental Science and Pollution Research, vol. 28, pp. 15687-15700, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[10] R.S. Makar et al., “Development of a PCA-based Land Use/Land Cover Classification Utilizing Sentinel-2 Time Series,” Middle East Journal of Agricultural Research, vol. 11, no. 2, pp. 630-637, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[11] S. Vasavi, Venkata Kalyan Chintalapudi, and Akhila Sree Rajeswari Vuppuluri, “Classification of Water Bodies using Ensemble of U-Net and Random Forest Algorithm,” Journal of Image and Graphics, vol. 12, no. 1, pp. 76-89, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Gáspár Albert, and Seif Ammar, “Application of Random Forest Classification and Remotely Sensed Data in Geological Mapping on the Jeebel Meloussi Area (Tunisia),” Arabian Journal of Geosciences, vol. 14, pp. 1-13, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Raj Singh, Vara Saritha, and Chaitanya B. Pande, “Monitoring of Wetland Turbidity using Multi-Temporal Landsat-8 and Landsat-9 Satellite Imagery in the Bisalpur Wetland, Rajasthan, India,” Environmental Research, vol. 241, pp. 1-36, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Sumanta Bid, and Giyasuddin Siddique, “Identification of Seasonal Variation of Water Turbidity using NDTI Method in Panchet Hill Dam, India,” Modeling Earth Systems and Environment, vol. 5, pp. 1179-1200, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Hasti Shwan Abdullah, Mahmoud S. Mahdi, and Hekmat M. Ibrahim, “Water Quality Assessment Models for Dokan Lake using Landsat 8 OLI Satellite Images,” Journal of Zankoy Sulaimani, vol. 19, pp. 3-4, pp. 25-42, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Özlem Akar, and Oğuz Güngör, “Classification of Multispectral Images using Random Forest Algorithm,” Journal of Geodesy and Geoinformation, vol. 1, no. 2, pp. 105-112, 2012.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Kim-Anh Nguyen et al., “Soil Salinity Assessment by using Near-Infrared Channel and Vegetation Soil Salinity Index Derived from Landsat 8 OLI Data: A Case Study in the Tra Vinh Province, Mekong Delta, Vietnam,” Progress in Earth and Planetary Science, vol. 7, pp. 1-16, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Komeli Rokni et al., “Water Feature Extraction and Change Detection using Multitemporal Landsat Imagery,” Remote Sensing, vol. 6, no. 5, pp. 4173-4189, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[19] N. Thirunavukkarasu et al., “Need of Coastal Resource Management in Pulicat Lake–Challenges Ahead,” Indian Journal of Science and Technology, vol. 4, no. 3, pp. 322-326, 2011.
[Google Scholar] [Publisher Link]
[20] Mohammad Haji Gholizadeh, Assefa M. Melesse, and Lakshmi Reddi, “A Comprehensive Review on Water Quality Parameters Estimation using Remote Sensing Techniques,” Sensors, vol. 16, no. 8, pp. 1-43, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Spatial without Compromise, QGIS. [Online]. Available: https://qgis.org/
[22] Luca Congedo, “Semi-Automatic Classification Plugin: A Python tool for the Download and Processing of Remote Sensing Images in QGIS,” Journal of Open Source Software, vol. 6, no. 64, pp. 1-6, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[23] O. Conrad et al., “System for Automated Geoscientific Analyses (SAGA) v. 2.1.4,” Geoscientific Model Development, vol. 8, pp. 1991-2007, 2015. 
[CrossRef] [Google Scholar] [Publisher Link]