Prediction of Groundwater Level for Sustainable Water Management in an Arid Basin Using Data-driven Models
- Mutao Huang, Yong Tian
- Corresponding Author
- Mutao Huang
Available Online October 2015.
- https://doi.org/10.2991/seee-15.2015.33How to use a DOI?
- data-driven; groundwater level forecasting; ANN; SVM; model tree
- Arid and semi-arid regions face major challenges in the management of scarce freshwater resources under economic development and climate change. Groundwater is commonly the most important water resource in these areas. Accurate prediction of groundwater level is an essential component of suitable water resources management. Physically based model are often employed to perform groundwater simulation and predications. However, they are not applicable in many arid and semi-arid regions due to data limitations. Data-driven methods have proven their applicability in modeling complex and non-linear hydrological processes. The focus of this study is the application and comparison of three data-driven models for forecasting short-term groundwater levels. The purpose is to develop a new data-based method for highly accurate groundwater level forecasting that can be used to help water managers, engineers, and stake-holders manage groundwater in a more effective and sustainable manner. A set of popular data-driven models are evaluated and compared, including Artificial Neuron Networks (ANNs), Support Vector Machines (SVMs), and M5 Model Tree. The feasibility and capability of these models are demonstrated through a case study of forecasting five-days ahead groundwater level in an arid and semi-arid basin located in northwestern China. The encouraging simulation results show that the methodologies can simplify and improve the procedure of groundwater level forecast.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Mutao Huang AU - Yong Tian PY - 2015/10 DA - 2015/10 TI - Prediction of Groundwater Level for Sustainable Water Management in an Arid Basin Using Data-driven Models BT - 2015 International Conference on Sustainable Energy and Environmental Engineering PB - Atlantis Press SN - 2352-5401 UR - https://doi.org/10.2991/seee-15.2015.33 DO - https://doi.org/10.2991/seee-15.2015.33 ID - Huang2015/10 ER -