Research Focus

My research develops AI-supported hydrological models and tracer-aided models, applied across spatial scales from plot to global. 

Under the concept of "Modeling as a learning process", the aim is to gain hydrological insights using AI as leverage based on process-based models as an archive of the domain knowledge. This includes using eXplainable AI (XAI) tools to interpret what AI models learn, whether about hydrological processes themselves or about model errors in process-based models, and to generate testable hypotheses. 

By applying this framework to tracer-aided models, the insight can be extended from celerity (hydrological response) to velocity (mixing dynamics). This is critical for diagnosing whether models are right for the right reason, and for predicting how catchments respond under non-stationary conditions (e.g., climate change).

Education

2012-2016 BSc in Environmental Science and Engineering (Magna Cum Laude), Ewha Womans University, Seoul, Republic of Korea

2016-2018 MSc in Environmental Science and Engineering, Ewha Womans University, Seoul, Republic of Korea

   “Simultaneous mitigation of methane and odor using pilot-scale high compact hybrid biocover "

2019-2023 MSc in Tropical Hydrogeology and Environmental Engineering, Technical University of Darmstadt, Darmstadt, Germany

   “Can XAI offer a new perspective for groundwater recharge estimation? - Global-scale modeling using Neural Network and Random Forest" (Ref: Jung, H., Saynisch-Wagner, J., and Schulz, S. (2024), WWR, https://doi.org/10.1029/2023WR036360)

Since 2023 PhD candidate in Geography, Humboldt-Universität zu Berlin, Berlin, Germany

PhD Topic

Artificial Intelligence for Tracer-Aided Ecohydrological Modelling - using AI to extend process-based ecohydrological models where data are sparse, and to diagnose their structural uncertainties through systematic error analysis

• WP1 Review of tracer-aided modeling: recent advances, limitations, and opportunities for AI integration

  Jung, H., Tetzlaff, D., Birkel, C. and Soulsby, C. (2025), Recent Developments and Emerging Challenges in Tracer-Aided Modeling. WIREs Water, 12: e70015. https://doi.org/10.1002/wat2.70015

• WP2 Development of a sequential AI model for spatio-temporal prediction of soil moisture and soil water isotopes at catchment scale

  Jung, H., Tetzlaff, D., Yordanova, D. and Soulsby, C. (In review), Development of a new generic AI model for spatio-temporal prediction of soil moisture and soil water isotopes

• WP3 AI Error Analyser: diagnosing structural uncertainties in tracer-aided ecohydrological models through systematic analysis of process-based model errors at plot scale

Conferences

EGU General Assembly 2026 (Oral Presentation)

Jung, H., Soulsby, C., Wu, S., Birkel, C., and Tetzlaff, D., 2026. Machine Learning Integration Strategies for Process-based Ecohydrological Modeling: Addressing Epistemic Uncertainties of Water Mixing Dynamics in Tree Water

EGU General Assembly 2024 (Oral Presentation)

Jung, H., Soulsby, C., Tetzlaff, D., 2024. A deep learning approach for spatio-temporal prediction of stable water isotopes in soil moisture