Data and code for "Leveraging Machine Learning to Predict and Quantify Methane Ebullition in Rice Paddy Fields"

This repository contains the processed dataset and the R and Python scripts used to reproduce the analyses presented in the manuscript: Leveraging Machine Learning to Predict and Quantify Methane Ebullition in Rice Paddy Fields The repository includes: the final processed dataset used for all analyses; a data dictionary describing all variables and their units; R scripts for methane ebullition detection and diffusive CO₂ flux estimation; a Python preprocessing script for data preparation and class balance assessment; Python scripts for machine learning classification and regression analyses; an R script for Generalized Additive Model (GAM) analyses. The repository provides all data and code necessary to reproduce the analyses presented in the manuscript.