UC Berkeley

Upland humid tropical forest soils experience fluctuations in oxygen (O2) availability and redox potential as a consequence of high rainfall, clay content, and respiration rates. Research in wetland ecosystems suggests that spatial and temporal variation in redox reactions strongly affect the biogeochemical cycling of carbon (C) and nitrogen (N). Here, I explored the impact of soil redox dynamics on decomposition and soil-atmosphere greenhouse gas fluxes in humid tropical ecosystems of the Luquillo Experimental Forest (LEF), Puerto Rico. Traditional theory and ecosystem models predict that elevated soil moisture leads to O2 limitation, constraining the enzymatic processes that mediate organic matter decomposition, and promoting the accumulation of soil C. Testing these hypotheses in upland humid tropical soils revealed the need for a more nuanced conceptual framework. In short: variation in moisture alone did not determine redox dynamics, hydrolytic enzymes activities persisted under reducing conditions, and redox fluctuations promoted decomposition on short (days) and long-term (decades) timescales.

In Chapter One, I showed a relative decoupling between the temporal dynamics of soil moisture, soil redox reactions, and greenhouse gas fluxes over scales of days to weeks, using a field moisture manipulation experiment. Anaerobic biogeochemical processes such as iron (Fe) reduction and methanogenesis co-occurred in proximity to a well-aerated soil atmosphere and were little affected by fluctuations in soil moisture. Instead, redox reactions and gas fluxes appeared to vary constitutively according to differences in microtopography. In Chapter Two, I further explored relationships between reducing conditions and organic matter decomposition, by analyzing extracellular hydrolytic enzyme activities within and among sites differing in topography and rainfall. The enzymatic latch hypothesis proposes that reducing conditions inhibit hydrolytic enzymes via an accumulation of phenolic substances. I found little evidence for an enzymatic latch, and instead documented a strong positive relationship between reducing conditions, using reduced Fe (Fe(II)) as a proxy, and hydrolytic enzyme activities in a subset of sites. Furthermore, enzyme activities generally did not decline in an anaerobic incubation relative to aerobic controls. The assumption that reducing conditions constrain the decomposition activities of hydrolytic enzymes does not appear generally applicable in humid tropical forests.

Next, in Chapter Three I examined the influence of temporal redox fluctuations on decomposition. Anaerobic conditions by definition limit the activity of oxidative enzymes, which require O2. The redox cycling of Fe, however, can potentially generate reactive oxygen species that mimic the function of oxidative enzymes. We demonstrated that concentrations of Fe(II) explained most of the variation in phenol oxidative activity within and among several sites in the LEF. Furthermore, Fe(II) oxidation stimulated short-term respiration, likely via a pH-mediated increase in dissolved organic C. Thus, stimulatory effects of redox fluctuations on oxidative decomposition processes might partially counteract short-term effects of O2 limitation.

Finally, in Chapter Four I examined the overall impact of reducing conditions in comparison with other variables as they related to spatial patterns in soil C concentrations and turnover across the LEF. Soil C increased with Fe(II), an index of reducing conditions, but C tended to decline with increasing concentrations of reducible Fe oxides. Furthermore, the residence time of mineral-associated C (modeled using measurements of bomb radiocarbon) declined with Fe(II) concentrations. Together, the findings from these studies suggest a complex relationship between moisture, redox dynamics, and decomposition. First, short-term fluctuations in rainfall may have little overall impact on redox dynamics and the overall decomposition process, but longer-term differences in moisture among sites are associated with characteristic differences in redox reactions and greenhouse gas fluxes. Second, portions of the decomposition process mediated by hydrolytic enzymes appear resistant to periodic O2 deprivation and chronic reducing conditions, as well as the accumulation of phenolic substances. Third, redox cycling may give rise to important emergent mechanisms not evident under static aerobic conditions, mediated by coupled biotic and abiotic reactions with Fe oxides. Fourth, reducing conditions are associated with elevated soil C concentrations at the landscape scale, although the presence of reducible Fe oxides constrains C accumulation, and redox cycling might accelerate the turnover of mineral C over decadal scales. Together, these findings have implications for understanding the biogeochemical function of humid tropical soils, and their response to altered precipitation regimes and feedbacks to climate change. Two mechanisms thought to underlie the persistence of C in soils--reducing conditions induced by high soil moisture and the presence of reactive Fe minerals--may actually play unexpected roles in the decomposition of soil organic matter, a finding with potentially broad application across terrestrial and aquatic ecosystems.