Measurements of greenhouse gas (GHG) emissions from paddy fields can often include flux measurement errors due to either instrument errors or unfavorable weather. Therefore, data post-processing, including the gap-filling process, is required to improve data quality and quantify the GHG flux budget. This study applied machine learning (ML) techniques with polynomial and multivariate polynomial regression models for gap-filling methane (CH₄) and carbon dioxide (CO₂) fluxes from closed chamber (CC) method measurements and compared results with mean diurnal variation (MDV) and look-up table (LUT) techniques. The most influential factors affecting methane emissions in the paddy field were used for input variables in the models: air temperature, soil temperature, soil redox potential, soil water content, solar radiation, and days after transplanting. The models' performances were compared using mean absolute error (MAE) and root mean square error (RMSE). The results showed that MAE and RMSE for gap-filling CH₄ fluxes were 1.299-2.984 and 2.499-4.981 mg CH₄ m^-2 h^-1, respectively. Also, the multivariate polynomial regression models performed better for gap-filling CH₄ fluxes (RMSE = 2.499 mg CH₄ m^-2 h^-1) than the polynomial regression models, MDV (RMSE = 3.210 mg CH₄ m^-2 h^-1), and LUT (RMSE = 3.339 mg CH₄ m^-2 h^-1) techniques. The MAE and RMSE for gap-filling CO₂ fluxes were 0.282-0.949 and 0.435-1.078 g CO₂ m^-2 h^-1, respectively. The ML techniques with polynomial regression using solar radiation (RMSE = 0.435 g CO₂ m^-2 h^-1) and multivariate models (RMSE = 0.445 g CO₂ m^-2 h^-1) perform better on gap-filling CO₂ fluxes than MDV (RMSE = 0.544 g CO₂ m^-2 h^-1), and LUT (RMSE = 0.553 g CO₂ m^-2 h^-1) techniques. The gap-filling using the multivariate polynomial regression models used in this study improved the reliability of the diurnal variation in GHG fluxes. Therefore, ML techniques could be a proper alternative for gap-filling GHG fluxes.

　稻田温室气体 (GHG) 排放的测量通常会包含由于仪器误差或不利天气而导致的通量测量误差。因此，需要进行数据后处理，包括间隙填充过程，以提高数据质量并量化温室气体通量预算。本研究应用机器学习 (ML) 技术以及多项式和多元多项式回归模型来填补间隙甲烷 (CH ₄ ) 和二氧化碳 (CO ₂) 密闭室 (CC) 方法测量的通量，并将结果与​​平均日变化 (MDV) 和查找表 (LUT) 技术进行比较。影响稻田甲烷排放最有影响的因素被用作模型中的输入变量：气温、土壤温度、土壤氧化还原电位、土壤含水量、太阳辐射和移栽后天数。使用平均绝对误差 (MAE) 和均方根误差 (RMSE) 比较模型的性能。结果表明，间隙填充CH ₄通量的MAE和RMSE分别为1.299-2.984和2.499-4.981 mg CH ₄ m ^-2 h ^-1。此外，多元多项式回归模型在填补空白方面表现更好 CH ₄通量 (RMSE = 2.499 mg CH ₄ m ^-2 h ^-1 ) 优于多项式回归模型、MDV (RMSE = 3.210 mg CH ₄ m ^-2 h ^-1 ) 和 LUT (RMSE = 3.339 mg CH ₄ m ^-2 h) ^-1）技术。间隙填充CO ₂通量的MAE和RMSE分别为0.282-0.949和0.435-1.078 g CO ₂ m ^-2 h ^-1。_{使用太阳辐射 (RMSE = 0.435 g CO 2} m ^-2 h ^-1 ) 和多元模型 (RMSE = 0.445 g CO _{2 )}进行多项式回归的 ML 技术m ^-2 h ^-1 ) 在间隙填充 CO ₂通量方面比 MDV (RMSE = 0.544 g CO ₂ m ^-2 h ^-1 ) 和 LUT (RMSE = 0.553 g CO ₂ m ^-2 h ^-1 ) 技术表现更好。本研究中使用的多元多项式回归模型的间隙填充提高了温室气体通量日变化的可靠性。因此，机器学习技术可能是填补温室气体通量缺口的合适替代方案。