Challenges in Unifying Physically Based and Machine Learning Simulations Through Differentiable Modeling: A Land Surface Case Study
Open Access
- 24 February 2025
- journal article
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 52 (4)
Abstract
Differentiable geoscientific modeling has shown promise for leveraging machine learning (ML) to unify physically based and data‐based modeling. Here, we critically analyze this promise in the context of large‐scale parameter optimization with the Noah‐MP land model as an example. The differentiable parameter learning framework is used to calibrate Noah‐MP soil and vegetation parameters such that the simulated surface soil moisture better matches satellite observations over the contiguous US. We found that the optimized parameters only marginally improved soil moisture (average RMSE = 0.092 m 3 m −3 ) upon uncalibrated Noah‐MP (RMSE = 0.10 m 3 m −3 ). Scaling and bias correction factors, often used in ML approaches for enhancing generalizability, were found to limit the transferability of the optimized physical parameters to the land model. The global objective function further compromises the algorithm's ability to simultaneously capture contrasting moisture regimes. Addressing these challenges is necessary to advance ML‐based calibration frameworks to better learn and represent the constraints of the physical model.Keywords
This publication has 20 references indexed in Scilit:
- From Hydrometeorology to River Water Quality: Can a Deep Learning Model Predict Dissolved Oxygen at the Continental Scale?Environmental Science & Technology, 2021
- A review of surrogate models and their application to groundwater modelingWater Resources Research, 2015
- Numerical assessment of metamodelling strategies in computationally intensive optimizationEnvironmental Modelling & Software, 2012
- A comparison of methods for a priori bias correction in soil moisture data assimilationWater Resources Research, 2012
- The community Noah land surface model with multiparameterization options (Noah-MP): 2. Evaluation over global river basinsJournal of Geophysical Research, 2011
- The community Noah land surface model with multiparameterization options (Noah-MP): 1. Model description and evaluation with local-scale measurementsJournal of Geophysical Research, 2011
- Calibration of complex subsurface reaction models using a surrogate-model approachAdvances in Water Resources, 2008
- Bayesian analysis of computer code outputs: A tutorialReliability Engineering & System Safety, 2006
- Land information system: An interoperable framework for high resolution land surface modelingEnvironmental Modelling & Software, 2006
- Temporal and spatial scales of observed soil moisture variations in the extratropicsJournal of Geophysical Research: Atmospheres, 2000