Biogeochemistry-Informed Neural Network (BINN v1.0) for improving accuracy of model prediction and scientific understanding of soil organic carbon storage
- 1Cornell University (Ithaca, United States of America)
- 2Nanyang Technological University (Singapore, Singapore)
Abstract
The increasing availability of large-scale observational data and the rapid development of artificial intelligence (AI) provide unprecedented opportunities to enhance our understanding of the global carbon cycle and other biogeochemical processes. However, retrieving mechanistic knowledge from these large-scale data remains a challenge. Here, we develop a Biogeochemistry-Informed Neural Network (BINN) that seamlessly integrates a vectorized process-based soil carbon cycle model (i.e., Community Land Model version 5, CLM5) into a neural network (NN) structure to examine mechanisms governing soil organic carbon (SOC) storage from big data. BINN demonstrates high accuracy in retrieving biogeochemical parameter values from synthetic data in a parameter recovery experiment. Furthermore, by incorporating Monte Carlo (MC) dropout to generate posterior distributions, we demonstrate that BINN can effectively quantify uncertainty in estimated parameters. We use BINN to predict six major processes (or components in process-based models) regulating the soil carbon cycle from 25 925 observed SOC profiles across the contiguous US and compare them with the same processes previously retrieved by a Bayesian inference-based PROcess-guided deep learning and DAta-driven modeling (PRODA) approach. The good agreement between the spatial patterns retrieved by BINN and PRODA (average correlation coefficient = 0.86) suggests that BINN's ability of capturing mechanistic knowledge is consistent with the established Bayesian-based methods. Additionally, the integration of neural networks and process-based models in BINN improves computational efficiency by more than 50 times over PRODA. We conclude that BINN is an efficient framework that harnesses the power of AI, large-scale data, and process-based modeling to understand large scale soil carbon cycle. Applying BINN to other domains of Earth system science will facilitate new scientific discoveries while improving the predictive performance of Earth system models.
Related
- Cascading impacts of Atlantic meridional overturning circulation weakening and collapse on marine biogeochemistry
- Riverine biogeochemistry from the Andes to the ocean: longitudinal patterns along a Mediterranean Andean catchment in central Chile
- From Trace Element Biogeochemistry to Critical Minerals Resourcing: Lessons Learned and Opportunities for Innovation.
- Three-stream modelling of radiative transfer for the simulation of Black Sea biogeochemistry in a NEMO framework
- The spirit of independent innovation: an exploratory study of Mr. J.S. Lee's scientific research traits
BibTeX
@article{xu2026biogeochemistryinformed,
title = {{Biogeochemistry-Informed Neural Network (BINN v1.0) for improving accuracy of model prediction and scientific understanding of soil organic carbon storage}},
author = {Haodi Xu and Joshua Fan and Feng Tao and Lifen Jiang and Fengqi You and Benjamin Z. Houlton and Ying Sun and Carla P. Gomes and Yiqi Luo},
journal = {Geoscientific model development},
year = {2026},
month = jul,
doi = {10.5194/gmd-19-6777-2026},
url = {https://doi.org/10.5194/gmd-19-6777-2026},
note = {CC BY},
}
RIS
TY - JOUR
TI - Biogeochemistry-Informed Neural Network (BINN v1.0) for improving accuracy of model prediction and scientific understanding of soil organic carbon storage
AU - Haodi Xu
AU - Joshua Fan
AU - Feng Tao
AU - Lifen Jiang
AU - Fengqi You
AU - Benjamin Z. Houlton
AU - Ying Sun
AU - Carla P. Gomes
AU - Yiqi Luo
PY - 2026
DA - 2026/07/01
JO - Geoscientific model development
DO - 10.5194/gmd-19-6777-2026
UR - https://doi.org/10.5194/gmd-19-6777-2026
AB - The increasing availability of large-scale observational data and the rapid development of artificial intelligence (AI) provide unprecedented opportunities to enhance our understanding of the global carbon cycle and other biogeochemical processes. However, retrieving mechanistic knowledge from these large-scale data remains a challenge. Here, we develop a Biogeochemistry-Informed Neural Network (BINN) that seamlessly integrates a vectorized process-based soil carbon cycle model (i.e., Community Land Model version 5, CLM5) into a neural network (NN) structure to examine mechanisms governing soil organic carbon (SOC) storage from big data. BINN demonstrates high accuracy in retrieving biogeochemical parameter values from synthetic data in a parameter recovery experiment. Furthermore, by incorporating Monte Carlo (MC) dropout to generate posterior distributions, we demonstrate that BINN can effectively quantify uncertainty in estimated parameters. We use BINN to predict six major processes (or components in process-based models) regulating the soil carbon cycle from 25 925 observed SOC profiles across the contiguous US and compare them with the same processes previously retrieved by a Bayesian inference-based PROcess-guided deep learning and DAta-driven modeling (PRODA) approach. The good agreement between the spatial patterns retrieved by BINN and PRODA (average correlation coefficient = 0.86) suggests that BINN's ability of capturing mechanistic knowledge is consistent with the established Bayesian-based methods. Additionally, the integration of neural networks and process-based models in BINN improves computational efficiency by more than 50 times over PRODA. We conclude that BINN is an efficient framework that harnesses the power of AI, large-scale data, and process-based modeling to understand large scale soil carbon cycle. Applying BINN to other domains of Earth system science will facilitate new scientific discoveries while improving the predictive performance of Earth system models.
SN - 1991-959X
SN - 1991-9603
ER -