研究領域

Sustainable Subsurface Resources Laboratory (SSRL)

SSRL focuses on exploring and conserving subsurface resources, such as groundwater resources and geothermal energy. The research topics include hardware and software developments and can be classified into five categories:

Advancing subsurface intelligence to drive innovation in energy and environmental

systems

Our research group is dedicated to developing high-resolution 3D hydraulic and thermal parameter fields—critical foundations for groundwater heat transport and multiphase flow research. By integrating multi-source observational data, such as groundwater head and temperature, we reconstruct heterogeneous geological fields to enable smart solutions for groundwater and energy systems.

Starting with geostatistical methods, our team has progressively advanced toward AI-driven data assimilation. We have successfully developed an encoder-decoder deep learning framework to simultaneously estimate hydraulic and thermal parameter fields. These innovations have been applied to optimize energy efficiency and performance in low-enthalpy geothermal systems and groundwater heat utilization, contributing to the transition toward sustainable energy and net-zero carbon goals.

Low-Temperature Geothermal × AI × Energy Efficiency: Shaping the Next Generation of

Smart Energy Solutions

Our team is dedicated to advancing the application and smart technology development of Low-Temperature Geothermal Systems (LTGS) in Taiwan. We have established two closed-loop geothermal wells, each reaching a depth of 54 meters, at the NTU Experimental Farm. These wells are equipped with fiber-optic sensors for high-resolution subsurface temperature profiling and are successfully integrated with a Ground Source Heat Pump (GSHP) system, forming the foundation of our experimental site.

Building on this infrastructure, we actively collaborate with international partners, including Professor Yu-Feng Forrest Lin (UIUC) and Professor Hikari Fujii, Vice President of Akita University, Japan, to drive research and innovation in LTGS technology. Currently, our team is developing localized LTGS design and testing methodologies for Taiwan by integrating AI, numerical modeling, analytical solutions, and optimization techniques, alongside data from field experiments. These efforts aim to accelerate the deployment of smart and sustainable energy solutions in Taiwan.

FBG × Multilevel Monitoring × Smart Hydraulic Tomography: Advancing Next-Generation 

Subsurface Sensing Technologies

Our team is developing an advanced Fiber Bragg Grating (FBG)--based Multilevel Monitoring System to revolutionize groundwater observation and hydraulic tomography applications. Unlike traditional single-depth monitoring wells, multilevel systems enable the simultaneous measurement of groundwater heads and temperatures at various depths within a single well, significantly improving monitoring efficiency and reducing environmental disturbance.

Facing the challenge of high costs and limited accessibility of commercial multilevel monitoring systems globally, we have collaborated with a leading local technology company, Chi-Bo Technologies, to develop Taiwan’s first indigenized FBG-based subsurface monitoring platform. This system has been successfully tested at multiple contaminated sites across Taiwan, demonstrating robust performance in field applications.

We have further applied this technology to conduct hydraulic tomography (HT) at contaminated sites, enabling the reconstruction of 3D heterogeneous hydraulic parameter fields. These results directly contribute to optimizing in-situ remediation strategies such as injection planning. Our research outcomes have been published in Groundwater, a flagship journal in hydrogeology, and Optics Express, a leading journal in optical science.

Groundwater Level × Spatiotemporal Signal Analysis × AI Forecasting: Decoding Subsurface Dynamics with Precision

Our team is committed to advancing spatiotemporal analysis techniques for groundwater level data, aiming to uncover how external forcing factors—such as rainfall, river stage variations, and anthropogenic pumping—impact groundwater systems over time and space. However, groundwater level data are inherently mixed signals, making it difficult to directly identify the influence of individual forcing sources.

To address this challenge, we have developed a systematic workflow integrating Independent Component Analysis (ICA), Fast Fourier Transform (FFT), and Wavelet Transform techniques. This framework successfully extracts the unique signal patterns of individual forcing sources from multi-well groundwater level observations.

In addition, we have incorporated deep neural network models into our approach, enabling the simultaneous prediction of groundwater level dynamics across multiple observation wells. This system is now being expanded for watershed-scale groundwater behavior analysis. Multiple outcomes of this research have been published in the leading journal Journal of Hydrology, contributing to the practical implementation of smart groundwater management technologies.

Temperature Profiling × Infiltration Estimation × IoT Monitoring: Reconstructing the 

Temporal Dynamics of Groundwater Infiltration

Our team focuses on developing advanced methods to estimate long-term, time-series infiltration rates using subsurface temperature profile analysis, addressing the critical gap in Taiwan’s direct monitoring tools for temporal infiltration dynamics. As surface water typically differs in subsurface temperature, rainfall and surface water infiltration induce notable changes in the subsurface thermal profile. Leveraging this phenomenon, we have established an inversion framework based on temperature profiles to quantify groundwater infiltration processes.

The model integrates a transient analytical solution coupling groundwater flow and heat transport, enabling the inversion of time-series infiltration rates while considering dynamic surface boundary conditions (e.g., surface temperature) and semi-infinite lower boundaries. It is adaptable to various initial subsurface temperature scenarios, significantly enhancing its field applicability and accuracy.

This technology has been validated through numerical simulations and multiple field experiments and has been successfully applied to infiltration monitoring in paddy fields and natural recharge sites, providing valuable groundwater infiltration data to support smart groundwater management. Our research outcomes have been published in the leading journal , the International Journal of Heat and Mass Transfer, and the prestigious Journal of Hydrology, contributing to advancements in intelligent hydrological monitoring and sustainable infiltration studies.

We sincerely invite students and research collaborators who are passionate about AI, energy technologies, and groundwater environmental sciences to join our team. Together, we aim to drive technological development and field applications, promoting digital innovation in the energy and environmental sectors. Our mission is to build an interdisciplinary international collaboration platform focused on developing cutting-edge smart monitoring and data analysis technologies to support the sustainable development of water and energy systems.

By joining us, you will have the opportunity to engage in cross-disciplinary projects, international collaboration, and field-based experimentation. Through comprehensive technical training, you will strengthen your expertise in smart energy and environmental engineering and become a key contributor to the next generation of sustainable technologies

The details for applying to NTU:    

https://admissions.ntu.edu.tw/apply/international-applications/

Scholarship:

https://admissions.ntu.edu.tw/apply/scholarships/

For more detailed information, don't hesitate to get in touch with professor Tsai at jptsai@ntu.edu.tw