Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island
Barbados is heavily reliant on groundwater resources for its potable water supply, with over 80% of the island’s water sourced from aquifers. The ability to meet demand will become even more challenging due to the continuing climate crisis. The consequences of climate change within the Caribbean reg...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Hydrology |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5338/10/1/2 |
| _version_ | 1797441782728884224 |
|---|---|
| author | Karl Payne Peter Chami Ivanna Odle David Oscar Yawson Jaime Paul Anuradha Maharaj-Jagdip Adrian Cashman |
| author_facet | Karl Payne Peter Chami Ivanna Odle David Oscar Yawson Jaime Paul Anuradha Maharaj-Jagdip Adrian Cashman |
| author_sort | Karl Payne |
| collection | DOAJ |
| description | Barbados is heavily reliant on groundwater resources for its potable water supply, with over 80% of the island’s water sourced from aquifers. The ability to meet demand will become even more challenging due to the continuing climate crisis. The consequences of climate change within the Caribbean region include sea level rise, as well as hydrometeorological effects such as increased rainfall intensity, and declines in average annual rainfall. Scientifically sound approaches are becoming increasingly important to understand projected changes in supply and demand while concurrently minimizing deleterious impacts on the island’s aquifers. Therefore, the objective of this paper is to develop a physics-based groundwater model and surrogate models using machine learning (ML), which provide decision support to assist with groundwater resources management in Barbados. Results from the study show that a single continuum conceptualization is adequate for representing the island’s hydrogeology as demonstrated by a root mean squared error and mean absolute error of 2.7 m and 2.08 m between the model and observed steady-state hydraulic head. In addition, we show that data-driven surrogates using deep neural networks, elastic networks, and generative adversarial networks are capable of approximating the physics-based model with a high degree of accuracy as shown by R-squared values of 0.96, 0.95, and 0.95, respectively. The framework and tools developed are a critical step towards a digital twin that provides stakeholders with a quantitative tool for optimal management of groundwater under a changing climate in Barbados. These outputs will provide sound evidence-based solutions to aid long-term economic and social development on the island. |
| first_indexed | 2024-03-09T12:29:05Z |
| format | Article |
| id | doaj.art-cc7720c5fc3d4530a21b1d5cc27c7e64 |
| institution | Directory Open Access Journal |
| issn | 2306-5338 |
| language | English |
| last_indexed | 2024-03-09T12:29:05Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Hydrology |
| spelling | doaj.art-cc7720c5fc3d4530a21b1d5cc27c7e642023-11-30T22:31:18ZengMDPI AGHydrology2306-53382022-12-01101210.3390/hydrology10010002Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate IslandKarl Payne0Peter Chami1Ivanna Odle2David Oscar Yawson3Jaime Paul4Anuradha Maharaj-Jagdip5Adrian Cashman6Centre for Resource Management and Environmental Studies, Cave Hill Campus, The University of the West Indies, Bridgetown BB11000, BarbadosFaculty of Science and Technology, Cave Hill Campus, The University of the West Indies, Bridgetown BB11000, BarbadosCentre for Resource Management and Environmental Studies, Cave Hill Campus, The University of the West Indies, Bridgetown BB11000, BarbadosCentre for Resource Management and Environmental Studies, Cave Hill Campus, The University of the West Indies, Bridgetown BB11000, BarbadosBarbados Water Authority, The Pine, St. Michael, Bridgetown BB11000, BarbadosIndependent Hydrological Consultant, Port of Spain 150123, Trinidad and TobagoAkwatix Water Resources Management, Ashford, St. John BB20000, BarbadosBarbados is heavily reliant on groundwater resources for its potable water supply, with over 80% of the island’s water sourced from aquifers. The ability to meet demand will become even more challenging due to the continuing climate crisis. The consequences of climate change within the Caribbean region include sea level rise, as well as hydrometeorological effects such as increased rainfall intensity, and declines in average annual rainfall. Scientifically sound approaches are becoming increasingly important to understand projected changes in supply and demand while concurrently minimizing deleterious impacts on the island’s aquifers. Therefore, the objective of this paper is to develop a physics-based groundwater model and surrogate models using machine learning (ML), which provide decision support to assist with groundwater resources management in Barbados. Results from the study show that a single continuum conceptualization is adequate for representing the island’s hydrogeology as demonstrated by a root mean squared error and mean absolute error of 2.7 m and 2.08 m between the model and observed steady-state hydraulic head. In addition, we show that data-driven surrogates using deep neural networks, elastic networks, and generative adversarial networks are capable of approximating the physics-based model with a high degree of accuracy as shown by R-squared values of 0.96, 0.95, and 0.95, respectively. The framework and tools developed are a critical step towards a digital twin that provides stakeholders with a quantitative tool for optimal management of groundwater under a changing climate in Barbados. These outputs will provide sound evidence-based solutions to aid long-term economic and social development on the island.https://www.mdpi.com/2306-5338/10/1/2deep neural networkselastic networksBarbadosclimate-water nexusgroundwater modellingFEFLOW |
| spellingShingle | Karl Payne Peter Chami Ivanna Odle David Oscar Yawson Jaime Paul Anuradha Maharaj-Jagdip Adrian Cashman Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island Hydrology deep neural networks elastic networks Barbados climate-water nexus groundwater modelling FEFLOW |
| title | Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island |
| title_full | Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island |
| title_fullStr | Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island |
| title_full_unstemmed | Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island |
| title_short | Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island |
| title_sort | machine learning for surrogate groundwater modelling of a small carbonate island |
| topic | deep neural networks elastic networks Barbados climate-water nexus groundwater modelling FEFLOW |
| url | https://www.mdpi.com/2306-5338/10/1/2 |
| work_keys_str_mv | AT karlpayne machinelearningforsurrogategroundwatermodellingofasmallcarbonateisland AT peterchami machinelearningforsurrogategroundwatermodellingofasmallcarbonateisland AT ivannaodle machinelearningforsurrogategroundwatermodellingofasmallcarbonateisland AT davidoscaryawson machinelearningforsurrogategroundwatermodellingofasmallcarbonateisland AT jaimepaul machinelearningforsurrogategroundwatermodellingofasmallcarbonateisland AT anuradhamaharajjagdip machinelearningforsurrogategroundwatermodellingofasmallcarbonateisland AT adriancashman machinelearningforsurrogategroundwatermodellingofasmallcarbonateisland |