DayCent-CUTE: A global sensitivity, auto-calibration, and uncertainty analysis tool for DayCent

Abstract

Soil organic carbon (SOC) is a crucial metric for mitigating greenhouse gas emissions and developing climate-smart agriculture. DayCent is widely used to simulate SOC dynamics and soil trace gas fluxes in various ecosystems. In this study, we developed DayCent-CUTE (auto-Calibration, sensitivity, and Uncertainty analysis ToolSet) for conducting global sensitivity analysis (GSA), auto-calibration, and uncertainty analysis for the model. The tool encompassed a pair of GSA methods and two distinct parameter optimization methods.

A collection of 30 field experiments, encompassing 212 combinations of management treatments and 581 SOC measurements, was divided into 18 sites for calibration and 12 sites for independent model evaluation. The posterior parameter distribution obtained from the auto-calibration process reduces the model bias and RMSE values, while the Nash-Sutcliffe efficiency and R² values showed improvements. The DayCent-CUTE proves to be an efficient and flexible tool that enhances the applications of the DayCent model.