Significant Global Methane from Tree Stems

Understanding the global patterns and magnitudes of methane (CH₄) fluxes in terrestrial ecosystems and the underlying processing driving these fluxes is crucial for climate change mitigation. However, knowledge of the global CH₄ emissions from tree stems and soil, as well as the influences of various biotic and abiotic conditions remain limited. Here, we compiled a comprehensive dataset consisting of 2,373 in situ observations of CH₄ flux, including 2,255 from soil and 118 from tree stems. Using this dataset, we evaluated the environmental controls of soil and tree stem CH₄ flux, and utilized machine learning techniques to upscale the global distribution and estimate global emissions. The results indicated that CH₄ fluxes in terrestrial ecosystems were largely driven by vegetation characteristics and their interaction with soil moisture and water table dynamics. The estimated annual emissions of terrestrial ecosystems were 173.81 Tg CH₄ yr−1 (152.46–203.02 Tg CH₄ yr⁻¹), with 139.85–176.21 Tg CH₄ yr⁻¹ emitted by soil and 12.61–26.81 Tg CH₄ yr⁻¹ emitted by tree stems. The findings underscored the unignorable CH₄ emissions from tree stems.The full dataset for publications used in our machine learning model and the global CH₄ budget in terrestrial ecosystems from soils and tree stems at a 10 km resolution generated using machine learning (Fig. 5 and Fig. 6) are available. The data that support the findings of this study are openly available in there.