Bayesian calibration of the ICBM/3 soil organic carbon model constrained by data from long-term experiments and uncertainties of C inputs

Abstract

Models with various complexity can asses soil C sequestration in agriculture. In this study, we updated the Introductory Carbon Balance Model (ICBM) with 28 years of additional data and included multiple long-term bare fallow experiments for old pool kinetics. We validated the model with data from a new long-term sister experiment. The new calibration included uncertainty in the estimation of below-ground C inputs to soil. The model now considers above- and belowground and external C inputs separately (ICBM/3). The underlying mathematical approach is the same, with two state variables and a climatic parameter, and such simple structure remained robust enough to describe soil organic C dynamics over six decades. We tested including an inert soil C pool in the model structure, but it did not decrease the observed variance. Similarly, an intercept in the functions for estimating belowground C input from crop yield was not useful. Results suggest that root C contributes more to the old organic C pool than aboveground C inputs. We also evaluated parameters interactions, in particular between C inputs and their transformation into more stable soil C interacted and decomposition kinetics. We also describe new functions for estimating the ICBM climatic parameter in a more user-friendly way.

Keywords:

• SOC decomposition
• SOC models
• Bayesian uncertainty estimation
• ICBM model

Acknowledgments

We are thankful to our Faculty for providing resources to maintain the Swedish long-term experiments and especially to Per Hillström, who carefully managed the Ultuna frame trial for the three recent decades, and to Karl-Gunnar Söderdahl, who took over the management in recent years. We are also grateful to the persons and institutions who provided data from the long-term bare fallow network sites (see Barré et al. [33]). In particular, we acknowledge the inputs from Vladimir Romanenkov (Kursk site, Lomonosov Moscow State University, All-Russian Institute for Agrochemistry), Folkert van Oort and Sabine Houot (the two French sites, French National Institute for Agricultural Research and Agroparistech), and Bent T. Christensen (Askov Experimental Station, Dept. Agroecology, Aarhus University).

Disclosure statement

During the preparation of this work, the author(s) used Grammarly in order to improve the readability of the manuscript. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author, LM. The data are not publicly available due to being property of the author’s institution, which requires following particular policies for disclosing them.

Additional information

Funding

This work was funded by Swedish Government Research Council for Sustainable Development (FORMAS, grant number 2022-00214) within the project “Carbon sequestration in Swedish cropland soils” and the Swedish Farmers’ Foundation for Agricultural Research (SLF, grant number O-18-23-141). The SMED (Swedish Environmental Emissions Data) consortium also provided funding for this research.