Systematic Discrepancies Should Be Addressed Before Effective Parameter Learning: Insights From Soil Moisture Simulations Using Differentiable Process‐Based Models
Open Access
- 11 November 2026
- journal article
- Published by American Geophysical Union (AGU) in Water Resources Research
- Vol. 62 (11)
Abstract
Differentiable hydrologic models combining neural networks with process‐based backbones—fully implemented or approximated by surrogate models—offer highly competitive accuracy and improved interpretability compared to either model type used alone. However, the impacts of surrogate model fidelity (accurate sensitivity of relationships compared to the original model) and systematic discrepancies between models and observations on parameter learning effectiveness remain underexplored. This study evaluates optimization performance of the differentiable HBV model ( δ HBV) and its surrogate variants in simulating soil moisture across the contiguous United States. Results show that high‐fidelity surrogate models can support effective differentiable parameter learning once systematic discrepancies are addressed. In contrast, unresolved systematic discrepancies degrade optimization performance, with particularly severe deterioration when low‐fidelity surrogates are used. These findings provide one possible explanation for the modest improvements reported in recent surrogate‐based differentiable parameter learning (dPL) applications. Linear correction partially mitigates these systematic discrepancies, improves optimization performance, and reduces compensatory parameter adjustment, although it does not resolve the underlying model–observation mismatch. Structural refinement remains a physically grounded long‐term direction, while pragmatic correction modules can provide useful partial mitigation when such refinement is not yet available. In addition, fully implementing differentiable process‐based models or using a rigorously trained surrogate can improve gradient fidelity and reduce amplified deterioration associated with surrogate infidelity.Funding Information
- U.S. Department of Energy (DE‐SC0016605)
This publication has 43 references indexed in Scilit:
- Building Surrogate Models Using Trajectories of Agents Trained by Reinforcement Learning2024
- Reconstructing GRACE-derived terrestrial water storage anomalies with in-situ groundwater level measurements and meteorological forcing dataJournal of Hydrology: Regional Studies, 2023
- Identification of the terrestrial water storage change features in the North China Plain via independent component analysisJournal of Hydrology: Regional Studies, 2021
- Data assimilation of uncalibrated soil moisture measurements from frequency-domain reflectometryGeoderma, 2020
- Comparison of the use of a physical-based model with data assimilation and machine learning methods for simulating soil water dynamicsJournal of Hydrology, 2020
- Prolongation of SMAP to Spatiotemporally Seamless Coverage of Continental U.S. Using a Deep Learning Neural NetworkGeophysical Research Letters, 2017
- Joint Sentinel‐1 and SMAP data assimilation to improve soil moisture estimatesGeophysical Research Letters, 2017
- A review of surrogate models and their application to groundwater modelingWater Resources Research, 2015
- Using a bias aware EnKF to account for unresolved structure in an unsaturated zone modelWater Resources Research, 2013
- Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modellingJournal of Hydrology, 2009