Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural Networks
Inland water bodies are critical ecosystems that serve several functions including the provision of freshwater, regulation of climate and hydrological flows, and pollution control. Therefore, effective monitoring and management of these water resources is critical for sustainable water supply syste...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
UJ Press
2024-06-01
|
| Series: | Journal of Digital Food, Energy & Water Systems |
| Subjects: | |
| Online Access: | https://journals.uj.ac.za/index.php/DigitalFoodEnergy_WaterSystems/article/view/3208 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841554999698522112 |
|---|---|
| author | Alice Nureen Adhiambo Omondi Yashon Ouma Simon Njoroge Mburu Cleophas Mecha Achisa |
| author_facet | Alice Nureen Adhiambo Omondi Yashon Ouma Simon Njoroge Mburu Cleophas Mecha Achisa |
| author_sort | Alice Nureen Adhiambo Omondi |
| collection | DOAJ |
| description |
Inland water bodies are critical ecosystems that serve several functions including the provision of freshwater, regulation of climate and hydrological flows, and pollution control. Therefore, effective monitoring and management of these water resources is critical for sustainable water supply systems. This study evaluated the possible use of satellite data to estimate water quality parameters (WQPs) in an inland water body. The study also used artificial neural network (ANN) models in addition to the satellite data to determine the optimum coagulant dose for water treatment. The use of earth observations and machine learning methods has not been done extensively in developing countries, specifically, in water quality monitoring and management. The study utilized empirical multivariate regression modelling (EMRM) of the spectral reflectances from satellite data for the retrieval of Chla-a, Turbidity, and total suspended solids (TSS) concentrations in an inland water body. Using MLP-ANN modelling, the extracted spectral reflectance values from the selected sampling points in the reservoir were used as model inputs for the prediction of treated WQPs. A second MLP-ANN model was developed to predict the optimum coagulant dose required for raw water treatment. The R2 values achieved with AN model 1 were 0.81, 0.76, and 0.81 respectively for TSS, turbidity, and Chl-a, and 0.99 for the optimum coagulant dose. The study concluded that spectral reflectance from medium resolution satellite data products can be used to estimate WQPs from inland water bodies. Further, the ANN models demonstrate that extracted water quality data from satellite images can be used to inform ANN models for water quality predictions, and for the optimization of water treatment plant operations.
|
| format | Article |
| id | doaj-art-3cda4cf5d199459f84315f847b237a79 |
| institution | Kabale University |
| issn | 2709-4510 2709-4529 |
| language | English |
| publishDate | 2024-06-01 |
| publisher | UJ Press |
| record_format | Article |
| series | Journal of Digital Food, Energy & Water Systems |
| spelling | doaj-art-3cda4cf5d199459f84315f847b237a792025-01-08T06:18:53ZengUJ PressJournal of Digital Food, Energy & Water Systems2709-45102709-45292024-06-015110.36615/2x3qd014Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural NetworksAlice Nureen Adhiambo Omondi0Yashon Ouma1Simon Njoroge Mburu2Cleophas Mecha Achisa3Renewable Energy, Environment, Nanomaterials and Water research group, Department of Civil and Structural Engineering, Moi UniversityDepartment of Civil and Structural Engineering, Moi UniversityDepartment of Civil and Structural Engineering, Moi UniversityRenewable Energy, Environment, Nanomaterials and Water research group, Department of Chemical and Processing Engineering, Moi University Inland water bodies are critical ecosystems that serve several functions including the provision of freshwater, regulation of climate and hydrological flows, and pollution control. Therefore, effective monitoring and management of these water resources is critical for sustainable water supply systems. This study evaluated the possible use of satellite data to estimate water quality parameters (WQPs) in an inland water body. The study also used artificial neural network (ANN) models in addition to the satellite data to determine the optimum coagulant dose for water treatment. The use of earth observations and machine learning methods has not been done extensively in developing countries, specifically, in water quality monitoring and management. The study utilized empirical multivariate regression modelling (EMRM) of the spectral reflectances from satellite data for the retrieval of Chla-a, Turbidity, and total suspended solids (TSS) concentrations in an inland water body. Using MLP-ANN modelling, the extracted spectral reflectance values from the selected sampling points in the reservoir were used as model inputs for the prediction of treated WQPs. A second MLP-ANN model was developed to predict the optimum coagulant dose required for raw water treatment. The R2 values achieved with AN model 1 were 0.81, 0.76, and 0.81 respectively for TSS, turbidity, and Chl-a, and 0.99 for the optimum coagulant dose. The study concluded that spectral reflectance from medium resolution satellite data products can be used to estimate WQPs from inland water bodies. Further, the ANN models demonstrate that extracted water quality data from satellite images can be used to inform ANN models for water quality predictions, and for the optimization of water treatment plant operations. https://journals.uj.ac.za/index.php/DigitalFoodEnergy_WaterSystems/article/view/3208Artificial Neural Network, Remote sensing, Water quality monitoring, water treatment |
| spellingShingle | Alice Nureen Adhiambo Omondi Yashon Ouma Simon Njoroge Mburu Cleophas Mecha Achisa Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural Networks Journal of Digital Food, Energy & Water Systems Artificial Neural Network, Remote sensing, Water quality monitoring, water treatment |
| title | Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural Networks |
| title_full | Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural Networks |
| title_fullStr | Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural Networks |
| title_full_unstemmed | Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural Networks |
| title_short | Optimization of Reservoir Water Quality Parameters Retrieval and Treatment Using Remote Sensing and Artificial Neural Networks |
| title_sort | optimization of reservoir water quality parameters retrieval and treatment using remote sensing and artificial neural networks |
| topic | Artificial Neural Network, Remote sensing, Water quality monitoring, water treatment |
| url | https://journals.uj.ac.za/index.php/DigitalFoodEnergy_WaterSystems/article/view/3208 |
| work_keys_str_mv | AT alicenureenadhiamboomondi optimizationofreservoirwaterqualityparametersretrievalandtreatmentusingremotesensingandartificialneuralnetworks AT yashonouma optimizationofreservoirwaterqualityparametersretrievalandtreatmentusingremotesensingandartificialneuralnetworks AT simonnjorogemburu optimizationofreservoirwaterqualityparametersretrievalandtreatmentusingremotesensingandartificialneuralnetworks AT cleophasmechaachisa optimizationofreservoirwaterqualityparametersretrievalandtreatmentusingremotesensingandartificialneuralnetworks |