Parameter calibration in global soil carbon models using surrogate-based optimization

Abstract

Abstract. Soil organic carbon (SOC) has a significant effect on carbon emissions and climate change. However, the current SOC prediction accuracy of most models is very low. Most evaluation studies indicate that the prediction error mainly comes from parameter uncertainties, which can be improved by parameter calibration. Data assimilation techniques have been successfully employed for the parameter calibration of SOC models. However, data assimilation algorithms, such as the sampling-based Bayesian Markov chain Monte Carlo (MCMC), generally have high computation costs and are not appropriate for complex global land models. This study proposes a new parameter calibration method based on surrogate optimization techniques to improve the prediction accuracy of SOC. Experiments on three types of soil carbon cycle models, including the Community Land Model with the Carnegie-Ames-Stanford Approach biogeochemistry submodel (CLM-CASA') and two microbial models show that the surrogate-based optimization method is effective and efficient in terms of both accuracy and cost. Compared to predictions using the tuned parameter values through Bayesian MCMC, the root mean squared errors (RMSEs) between the predictions using the calibrated parameter values with surrogate-base optimization and the observations could be reduced by up to 12 % for different SOC models. Meanwhile, the corresponding computational cost is lower than other global optimization algorithms.

Modeled Microbial Dynamics Explain the Apparent Temperature Sensitivity of Wetland Methane Emissions

“…The need for including biological processes in global soil models has been gaining more attention recently (Sulman et al., 2014; Walker et al., 2018; Wieder et al., 2013; Xenakis & Williams, 2014) and is a major focus of this study. Microbial dynamics have been included in several models for carbon dioxide production (Huang et al., 2018; Xu et al., 2018) and also in recent models of methane emissions (Oh et al., 2020; Wang et al., 2019; Xu et al., 2015).…”

Estimating Soil Respiration in a Subalpine Landscape Using Point, Terrain, Climate, and Greenness Data

“…Alternatively, simple statistical models have lower computational costs and therefore are more feasible for application across large spatial extents, provided adequate representation of field measurements to inform model development. In fact, such models have already been utilized as surrogates for calibrating process‐based models (Xu et al, ).…”

Artificial Intelligence for Earth System Predictability (AI4ESP) (2021 Workshop Report)

“…AI/ML approaches have shown promise in correcting model biases with respect to observations or a high-fidelity model simulation, optimizing model fidelity Kennedy and O'Hagan 2001;Couvreux et al 2021;Xu et al 2018;. Emulating a complex model's parameter sensitivities following human-constructed trial simulations have been used to aid model calibration and uncertainty quantification.…”