Methane cycling in humid tropical forests: Stable isotope geochemistry and effects of oxygen dynamics

This dissertation explored the methane (CH4) dynamics of humid tropical forests, focusing on the isotope geochemistry of CH4 and the effects of soil oxygen (O2) variations on CH4 production, oxidation, and emission. Methanogenesis was less sensitive to O2 than methanotrophy. High rates of CH4 production were observed in soils irrespective of O2 content due to the occurrence of methanogenesis in anoxic microsites. In contrast, CH4 oxidation declined below 3% O2. Methanotrophic bacteria were insensitive to O2 above this threshold. Methane production was greater at the soil surface than deeper in the profile, probably because of greater labile carbon content and heterotrophic activity. The functional composition of the methanogenic community varied with depth. Acetate-utilizing methanogens were more abundant at the soil surface, with approximately 58% of the CH4 in the 0--15 cm depth derived from acetate. Hydrogenotrophic methanogens predominated deeper in the profile. More than 70% of the CH4 in the 40--60 cm depth was derived from carbon dioxide (CO2). Ferric iron-reducing bacteria influenced methanogenic activity by suppressing aceticlastic methanogenesis and lowering CH4 production. Methane emissions in the more oxic wet forest were regulated by CH4 oxidation, while emissions from the rain forest were driven by methanogenesis. Fluxes from the rain forest were comparable to that of freshwater wetlands, ranging from 22 +/- 7 to 101 +/- 19 nmol m-2 s-1. Methane emissions from the rain forest were regulated by methanogenesis because methanotrophic bacteria were CH4-saturated. Methane concentrations exceeded the half-saturation constant for CH4 oxidation (K m = 1370 ppmv) by three to thirty-five times. Furthermore, advective transport caused CH4 produced deeper in the soil to bypass surface-dwelling methanotrophs. Surface emissions were strongly depleted in 13C (-70.4 +/- 0.95‰) due to the upward advection of CO 2-derived CH4. Methane availability and methanotrophic activity influenced the isotope fractionation factor for CH4 oxidation (alpha). Incubation experiments showed that variations in alpha were dependent on CH4 oxidation rate and CH4 concentration. This suggests that it is possible to predict alpha empirically rather than relying solely on model estimates, allowing for more precise calculations of CH4 oxidation in situ by isotope mass balance.