Estimating the methane flux of the Dajiuhu subalpine peatland using machine learning algorithms and the maximal information coefficient technique.

The eddy covariance (EC) technique has emerged as the most widely used method for long-term continuous methane flux (FCH) observations. However, the completeness of the FCH time series is limited by instrumental failures and data quality issues, resulting in missing data gaps ranging from 20 % to 90 %. In this situation, the excellent performance of machine learning (ML) algorithms in filling missing FCH data has provided a foundation for developing regional-scale FCH models. In this study, we established estimation models for FCH utilizing random forest (RF), support vector machine (SVM), back propagation (BP) and nonlinear multiple regression (MLR) algorithms. The maximal information coefficient (MIC) technique was employed to identify and rank the environmental factors that were correlated with FCH. Our findings revealed that soil temperature (Ts), soil water content (SWC) and air temperature (Ta) were the primary environmental factors influencing FCH. Among the four algorithms, from perspectives of model accuracy and relatively small number of driving factors, the RF models exhibited the best performance, followed by BP and SVM, whereas MLR demonstrated the lowest performance. Among the 144 RF models established using nine datasets, RF model with 8 driving factors in all-year (RF) could capture seasonal variations. Ultimately, we recommend (RF as the optimal model for estimating FCH in the Dajiuhu subalpine peatland.