Soil oxygen (O2) is a fundamental control on terrestrial biogeochemical cycles including processes producing and consuming greenhouse gases (GHG), yet it is rarely measured. Instead, soil O2 is assumed to be proportional to soil moisture and physical soil properties. For example, soil O2 is often inferred from a 25-year old steady-state diffusion model; however, few data exist to test this model in stochastic systems. The variability of soil O2 may be particularly important to GHG emissions from aquatic-terrestrial interface zones because of the convergence of variable hydrology and rapid biogeochemical processing. Our objective is to gain a better understanding of soil O2 variation and its role in controlling GHG emissions across aquatic-terrestrial interface zones. Specifically, we hypothesize that in aquatic-terrestrial interface ecosystems, soil moisture predicts O2 concentration under stable conditions where diffusion dominates, but fails under dynamic conditions (e.g., water table fluctuations or precipitation) due to advective flux during soil drainage and lags in biological O2 consumption. Furthermore, we hypothesize that GHG emissions will correspond to variation in soil O2.
Results/Conclusions
Twenty-four near-continuous (hourly readings) soil O2 and moisture sensors were installed across an aquatic-terrestrial interface zone of a constructed wetland in April 2012. Drought conditions (2012) resulted in minimal soil O2 variation. Despite dry conditions a diurnal pattern of lower soil O2 during the day was observed. When precipitation increased in September (due to the remnants of Hurricane Isaac), soil O2 variation increased substantially with 20 of the 24 sensors recording soil O2 concentrations below 5%. The relationship between soil moisture and soil O2 was non-linear during periods of soil drainage and precipitation. A rapid (change of x% over <24 hours) increase in soil O2 occurred at ~ 40% water filled soil volume during soil drainage. This rapid increased appears related to drainage of soil macropores ahead of changes in bulk soil moisture. As soil moisture increased due to precipitation, soil O2 decreased slower than predicted by simple diffusion models. A lag in O2 consumption is hypothesized to explain this divergence from the diffusion model. Weekly methane and nitrous oxide emissions corresponded to variation in soil O2. Future research will explore the importance of temporal soil O2 and moisture variation in driving GHG emissions.