Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma River
This study aims to establish the optimal regression model for predicting total suspended solids (TSS) and Turbidity based on in situ data and spectral regions of Sentinel-2 images. Various machine learning models were evaluated, including Multilayer Perceptron Regression (MLPR), Random Forest Regre...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Environmental Research Institute, Chulalongkorn University
2024-11-01
|
| Series: | Applied Environmental Research |
| Subjects: | |
| Online Access: | https://ph01.tci-thaijo.org/index.php/aer/article/view/257524 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846162505573859328 |
|---|---|
| author | Ngo Thanh Son Nguyen Duc Loc |
| author_facet | Ngo Thanh Son Nguyen Duc Loc |
| author_sort | Ngo Thanh Son |
| collection | DOAJ |
| description |
This study aims to establish the optimal regression model for predicting total suspended solids (TSS) and Turbidity based on in situ data and spectral regions of Sentinel-2 images. Various machine learning models were evaluated, including Multilayer Perceptron Regression (MLPR), Random Forest Regression (RFR), AdaBoost Regression (ABR), Multiple Linear Regression (MLR), and K-Nearest Neighbors Regression (KNNR). These models were applied to different band combinations of spectral regions: visible (VIS), near-infrared (NIR), shortwave-infrared (SWIR), VIS+NIR (VNIR), and VIS+NIR+SWIR (VNIR+SWIR). The study results revealed that the MLR model, while not the best performer during training (R2 = 0.89 for TSS and R2 = 0.66 for turbidity), did not exhibit overfitting, with corresponding R² values in testing being 0.80 and 0.42, respectively. Variable selection for MLR models identified optimal spectral bands: B3, B5, B6, B8, B11, and B12 for TSS, and B4, B8, B11, and B12 for Turbidity. The final no-intercept multiple linear regression models achieved R2 = 0.88 for TSS and R2 = 0.62 for turbidity. Performance metrics for TSS were superior, with lower MAE, MSE, and RMSE compared to Turbidity. This study underscores the efficacy of using MLR models with selected spectral bands for accurate and generalizable predictions of TSS and turbidity.
|
| format | Article |
| id | doaj-art-d7aaf0f27af54c36a3e0786b77d8b061 |
| institution | Kabale University |
| issn | 2287-075X |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Environmental Research Institute, Chulalongkorn University |
| record_format | Article |
| series | Applied Environmental Research |
| spelling | doaj-art-d7aaf0f27af54c36a3e0786b77d8b0612024-11-20T11:05:09ZengEnvironmental Research Institute, Chulalongkorn UniversityApplied Environmental Research2287-075X2024-11-01464Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma RiverNgo Thanh Son0Nguyen Duc Loc1Faculty of Natural Resources and Environment, Vietnam National University of Agriculture, Hanoi, VietnamFaculty of Natural Resources and Environment, Vietnam National University of Agriculture, Hanoi, Vietnam This study aims to establish the optimal regression model for predicting total suspended solids (TSS) and Turbidity based on in situ data and spectral regions of Sentinel-2 images. Various machine learning models were evaluated, including Multilayer Perceptron Regression (MLPR), Random Forest Regression (RFR), AdaBoost Regression (ABR), Multiple Linear Regression (MLR), and K-Nearest Neighbors Regression (KNNR). These models were applied to different band combinations of spectral regions: visible (VIS), near-infrared (NIR), shortwave-infrared (SWIR), VIS+NIR (VNIR), and VIS+NIR+SWIR (VNIR+SWIR). The study results revealed that the MLR model, while not the best performer during training (R2 = 0.89 for TSS and R2 = 0.66 for turbidity), did not exhibit overfitting, with corresponding R² values in testing being 0.80 and 0.42, respectively. Variable selection for MLR models identified optimal spectral bands: B3, B5, B6, B8, B11, and B12 for TSS, and B4, B8, B11, and B12 for Turbidity. The final no-intercept multiple linear regression models achieved R2 = 0.88 for TSS and R2 = 0.62 for turbidity. Performance metrics for TSS were superior, with lower MAE, MSE, and RMSE compared to Turbidity. This study underscores the efficacy of using MLR models with selected spectral bands for accurate and generalizable predictions of TSS and turbidity. https://ph01.tci-thaijo.org/index.php/aer/article/view/257524Water quality monitoringMachine learning modelSentinel-2 imageryTurbidityTotal suspended solidsUpper Ma river |
| spellingShingle | Ngo Thanh Son Nguyen Duc Loc Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma River Applied Environmental Research Water quality monitoring Machine learning model Sentinel-2 imagery Turbidity Total suspended solids Upper Ma river |
| title | Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma River |
| title_full | Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma River |
| title_fullStr | Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma River |
| title_full_unstemmed | Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma River |
| title_short | Establishing Optimal Machine Learning Models for Monitoring Water Quality in Vietnam’s Upper Ma River |
| title_sort | establishing optimal machine learning models for monitoring water quality in vietnam s upper ma river |
| topic | Water quality monitoring Machine learning model Sentinel-2 imagery Turbidity Total suspended solids Upper Ma river |
| url | https://ph01.tci-thaijo.org/index.php/aer/article/view/257524 |
| work_keys_str_mv | AT ngothanhson establishingoptimalmachinelearningmodelsformonitoringwaterqualityinvietnamsuppermariver AT nguyenducloc establishingoptimalmachinelearningmodelsformonitoringwaterqualityinvietnamsuppermariver |