Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine
Groundwater (GW) is the unique water source for more than one third of the world's populations. GW quality is under serious threat due to the recent rapid urbanization and industrialization. GW contamination is influenced by various interrelated variables, leading to high complexity in the GW q...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | http://eprints.usm.my/46284/ http://eprints.usm.my/46284/1/Jawad%20S.%20I.%20Alagha24.pdf |
| _version_ | 1848880565673000960 |
|---|---|
| author | Alagha, Jawad S. I. |
| author_facet | Alagha, Jawad S. I. |
| author_sort | Alagha, Jawad S. I. |
| building | USM Institutional Repository |
| collection | Online Access |
| description | Groundwater (GW) is the unique water source for more than one third of the world's populations. GW quality is under serious threat due to the recent rapid urbanization and industrialization. GW contamination is influenced by various interrelated variables, leading to high complexity in the GW quality modelling process. Statistical and artificial intelligence (AI) techniques have recently become common GW modelling tools due to their high performance. In this research, hybrid systems composed of two AI techniques namely artificial neural networks (ANNs) and support vector machine (SVM) in addition to various multivariate statistical techniques, were utilized to simulate the concentrations of two GW quality parameters particularly nitrate (NO3-) and chloride (Cl-) in complex aquifers. The models were trained using limited and irregular monitoring data from 22 municipal wells from 1998 to 2010 in Gaza Coastal Aquifer (GCA) which is a complex and highly heterogeneous aquifer. Results of the statistical analyses deepened the understanding of the GCA influencing variables and GW quality trends. Both ANNs and SVM techniques showed very satisfactory simulation performance with comparable results. The correlation coefficient (r) and mean average percentage error (MAPE) for NO3- simulation model were 0.996 and 7% respectively. Meanwhile r and MAPE for Cl- simulation model were 0.998 and 3.7% respectively. The results demonstrated also the merit of performing clustering of input data into consistent clusters prior to separate application of AI techniques for each cluster. Given their high performance and simplicity, the developed models were effectively utilized as GW quality management decision support tools by assessing the effects of various management scenarios on NO3- and Cl- concentration in GCA for 2020 and 2030. Evaluation of GW quality management scenarios indicated that NO3- and Cl- concentrations in the study area municipal wells would noticeably increase if the situation remained without any immediate intervention. On the other hand, GW quality levels in most study area wells would be highly improved if a combination of management scenarios was adopted. |
| first_indexed | 2025-11-15T18:05:09Z |
| format | Thesis |
| id | usm-46284 |
| institution | Universiti Sains Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T18:05:09Z |
| publishDate | 2013 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | usm-462842020-02-20T02:53:53Z http://eprints.usm.my/46284/ Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine Alagha, Jawad S. I. TA1-2040 Engineering (General). Civil engineering (General) Groundwater (GW) is the unique water source for more than one third of the world's populations. GW quality is under serious threat due to the recent rapid urbanization and industrialization. GW contamination is influenced by various interrelated variables, leading to high complexity in the GW quality modelling process. Statistical and artificial intelligence (AI) techniques have recently become common GW modelling tools due to their high performance. In this research, hybrid systems composed of two AI techniques namely artificial neural networks (ANNs) and support vector machine (SVM) in addition to various multivariate statistical techniques, were utilized to simulate the concentrations of two GW quality parameters particularly nitrate (NO3-) and chloride (Cl-) in complex aquifers. The models were trained using limited and irregular monitoring data from 22 municipal wells from 1998 to 2010 in Gaza Coastal Aquifer (GCA) which is a complex and highly heterogeneous aquifer. Results of the statistical analyses deepened the understanding of the GCA influencing variables and GW quality trends. Both ANNs and SVM techniques showed very satisfactory simulation performance with comparable results. The correlation coefficient (r) and mean average percentage error (MAPE) for NO3- simulation model were 0.996 and 7% respectively. Meanwhile r and MAPE for Cl- simulation model were 0.998 and 3.7% respectively. The results demonstrated also the merit of performing clustering of input data into consistent clusters prior to separate application of AI techniques for each cluster. Given their high performance and simplicity, the developed models were effectively utilized as GW quality management decision support tools by assessing the effects of various management scenarios on NO3- and Cl- concentration in GCA for 2020 and 2030. Evaluation of GW quality management scenarios indicated that NO3- and Cl- concentrations in the study area municipal wells would noticeably increase if the situation remained without any immediate intervention. On the other hand, GW quality levels in most study area wells would be highly improved if a combination of management scenarios was adopted. 2013-12 Thesis NonPeerReviewed application/pdf en http://eprints.usm.my/46284/1/Jawad%20S.%20I.%20Alagha24.pdf Alagha, Jawad S. I. (2013) Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine. PhD thesis, Universiti Sains Malaysia. |
| spellingShingle | TA1-2040 Engineering (General). Civil engineering (General) Alagha, Jawad S. I. Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine |
| title | Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine |
| title_full | Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine |
| title_fullStr | Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine |
| title_full_unstemmed | Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine |
| title_short | Development Of Groundwater Quality Management Models Using Artificial Intelligence (Ai) And Statistical Approaches – Case Study – Khanyounis Governorate – Gaza Strip – Palestine |
| title_sort | development of groundwater quality management models using artificial intelligence (ai) and statistical approaches – case study – khanyounis governorate – gaza strip – palestine |
| topic | TA1-2040 Engineering (General). Civil engineering (General) |
| url | http://eprints.usm.my/46284/ http://eprints.usm.my/46284/1/Jawad%20S.%20I.%20Alagha24.pdf |