Author
Listed:
- Tsai-Ning Weng
(National Central University)
- Ray-Shyan Wu
(National Central University)
- Tzu-Han Weng
(National Central University)
- Yi-Ho Chen
(National Central University)
- Tsai-Chun Lai
(National Central University)
- Jui-Yun Hsieh
(National Central University)
- Chu-Chun Hsu
(National Central University)
- Yuan-Chien Lin
(National Central University)
Abstract
Groundwater is a relatively stable water resource and vital for human existence and economic development. Therefore, properly utilizing groundwater becomes very important when faced with water shortages. However, it is difficult to obtain information about the pumping amount, which has crucial impact on the groundwater hydrological cycle. Most studies use monthly groundwater data and previous time steps of groundwater information to model and predict without considering pumping information. Therefore, this study proposes a wavelet-deep learning model, combining wavelet analysis and deep learning, using the Daliao area of Kaohsiung, Taiwan as an example, which has considerable historical data from a smart groundwater real-time IoT pumping meter system to monitor pumping quantities for the main wells. We extracted relevant features from hourly observation data from 23 August 2017 through 30 January 2020, and investigated each factor’s features relative to groundwater level by wavelet transform, and then used recurrent neural networks (RNN) and long short-term memory (LSTM) deep learning models to summarize and predict multiple factor impacts on groundwater level under different time lags. Hourly prediction models of the LSTM and RNN achieved reliable cross validation performance with 0.81 and 0.78 coefficient of determination ( $${\text{R}}^{2}$$ ), respectively. Especially after the feature extraction by wavelet analysis and the addition of artificial groundwater pumping information, it will greatly increase the prediction accuracy. This study provides a feasible and accurate approach for groundwater level prediction, and hence will be an important reference for groundwater resources management and risk assessment, helping to achieve sustainable groundwater use.
Suggested Citation
Tsai-Ning Weng & Ray-Shyan Wu & Tzu-Han Weng & Yi-Ho Chen & Tsai-Chun Lai & Jui-Yun Hsieh & Chu-Chun Hsu & Yuan-Chien Lin, 2025.
"Groundwater Level Prediction by Wavelet Deep Learning with Smart Pumping Data,"
Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA), Springer;European Water Resources Association (EWRA), vol. 39(6), pages 2717-2742, April.
Handle:
RePEc:spr:waterr:v:39:y:2025:i:6:d:10.1007_s11269-024-04088-0
DOI: 10.1007/s11269-024-04088-0
Download full text from publisher
As the access to this document is restricted, you may want to search for a different version of it.
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:spr:waterr:v:39:y:2025:i:6:d:10.1007_s11269-024-04088-0. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.springer.com .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.