Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine Learning
Climate change is having a significant impact on groundwater storage, affecting water resources in many parts of the world. To characterize this impact, remote sensing and machine learning are essential tools to analyze the data accurately and efficiently. This study aims to predicting the variation...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/16/1/50 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589118077403136 |
|---|---|
| author | Mohamed Hamdi Anas El Alem Kalifa Goita |
| author_facet | Mohamed Hamdi Anas El Alem Kalifa Goita |
| author_sort | Mohamed Hamdi |
| collection | DOAJ |
| description | Climate change is having a significant impact on groundwater storage, affecting water resources in many parts of the world. To characterize this impact, remote sensing and machine learning are essential tools to analyze the data accurately and efficiently. This study aims to predicting the variations of groundwater storage (GWS) using GRACE/GRACE-FO and multi-source remote sensing data, combined with machine learning techniques. The approach was applied over the Canadian Prairies region. The study area was classified into three zones of different aquifer potentials (low, medium, and high) using a combination of remote sensing data and the Classification and Regression Trees (CART) approach. The prediction model was developed using a machine-learning approach based on multiple linear regression to estimate GWS variations as a function of various environmental parameters. The results showed that the developed model was able to predict GWS variations with satisfactory accuracy (up to 95% of the explained variance) and good robustness (96% success rate). They also provided a better understanding of the variations in groundwater storage in the Canadian Prairies. Therefore, this work provides a promising method for predicting GWS, which could eventually be applied to other similar environmental conditions. |
| format | Article |
| id | doaj-art-637fcbe4b45c4d09b7749aee3f38f2a1 |
| institution | Kabale University |
| issn | 2073-4433 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj-art-637fcbe4b45c4d09b7749aee3f38f2a12025-01-24T13:21:51ZengMDPI AGAtmosphere2073-44332025-01-011615010.3390/atmos16010050Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine LearningMohamed Hamdi0Anas El Alem1Kalifa Goita2Centre d’Applications et de Recherches en Télédétection (CARTEL), Département de Géomatique Appliquée, Université de Sherbrooke, Sherbrooke, QC J1N 3C6, CanadaInstitut National de la Recherche Scientifique, Université de Québec, Québec City, QC G1K 9A9, CanadaCentre d’Applications et de Recherches en Télédétection (CARTEL), Département de Géomatique Appliquée, Université de Sherbrooke, Sherbrooke, QC J1N 3C6, CanadaClimate change is having a significant impact on groundwater storage, affecting water resources in many parts of the world. To characterize this impact, remote sensing and machine learning are essential tools to analyze the data accurately and efficiently. This study aims to predicting the variations of groundwater storage (GWS) using GRACE/GRACE-FO and multi-source remote sensing data, combined with machine learning techniques. The approach was applied over the Canadian Prairies region. The study area was classified into three zones of different aquifer potentials (low, medium, and high) using a combination of remote sensing data and the Classification and Regression Trees (CART) approach. The prediction model was developed using a machine-learning approach based on multiple linear regression to estimate GWS variations as a function of various environmental parameters. The results showed that the developed model was able to predict GWS variations with satisfactory accuracy (up to 95% of the explained variance) and good robustness (96% success rate). They also provided a better understanding of the variations in groundwater storage in the Canadian Prairies. Therefore, this work provides a promising method for predicting GWS, which could eventually be applied to other similar environmental conditions.https://www.mdpi.com/2073-4433/16/1/50climate changegroundwater storageremote sensingmachine learninggroundwater potential mapSaskatchewan River Basin |
| spellingShingle | Mohamed Hamdi Anas El Alem Kalifa Goita Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine Learning Atmosphere climate change groundwater storage remote sensing machine learning groundwater potential map Saskatchewan River Basin |
| title | Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine Learning |
| title_full | Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine Learning |
| title_fullStr | Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine Learning |
| title_full_unstemmed | Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine Learning |
| title_short | Groundwater Storage Estimation in the Saskatchewan River Basin Using GRACE/GRACE-FO Gravimetric Data and Machine Learning |
| title_sort | groundwater storage estimation in the saskatchewan river basin using grace grace fo gravimetric data and machine learning |
| topic | climate change groundwater storage remote sensing machine learning groundwater potential map Saskatchewan River Basin |
| url | https://www.mdpi.com/2073-4433/16/1/50 |
| work_keys_str_mv | AT mohamedhamdi groundwaterstorageestimationinthesaskatchewanriverbasinusinggracegracefogravimetricdataandmachinelearning AT anaselalem groundwaterstorageestimationinthesaskatchewanriverbasinusinggracegracefogravimetricdataandmachinelearning AT kalifagoita groundwaterstorageestimationinthesaskatchewanriverbasinusinggracegracefogravimetricdataandmachinelearning |