UC Berkeley

Wetlands and peatlands represent only 3% of the world's soils but account for approximately 21% of the global soil organic carbon (C) stock. Due to their high nutrient availability, more than 10% of wetlands worldwide have been drained for agriculture and further drainage is expected to meet growing agricultural demands. Conversion of wetlands to croplands is likely to contribute to climate change through the loss of soil C and increased the emissions of nitrous oxide (N2O) emissions from agricultural practices. The magnitude and drivers of C and nitrogen (N) emissions are poorly understood for drained wetland soils. For my dissertation, I first conducted a large-scale sampling campaign across a range of soil C concentrations in drained and restored wetland soils to quantify the importance of soil mineralogy in soil C storage and loss. I found that reactive iron (Fe) was negatively correlated to soil C across soils, suggesting reactive Fe pools may drive additional C losses in drained soils and limit C sequestration in flooded soils. Reactive organo-aluminum (Al) complexes were strongly positively correlated to soil C concentrations in drained soils, suggesting organo-Al complexes facilitate aggregation and anaerobic (micro)sites that may protect residual soil C from oxidation. Using automated chambers and cavity ring-down spectroscopy, I also conducted some of the longest continuous greenhouse gas flux measurements from organic-rich maize and mineral-rich alfalfa ecosystems, two dominant land uses in drained peatland soils. Flux measurements were conducted alongside continuous soil sensor monitoring, weekly soil N sampling, and satellite vegetation imagery to explore the biogeochemical and phenological drivers of CO2, methane (CH4), and N2O emissions.

In the maize ecosystem, I found that soils were significant N2O sources, contributing to 16-35% of annual CO2e emissions from these already high emitting ecosystems. Using over 70,000 flux measurements I determined that hot moments of N2O and CH4 emissions represented 1.1 ± 0.2 and 1.3 ± 0.2% of measurements, respectively, but contributed to 45 ± 1% of mean annual N2O fluxes and to 140 ± 9% of mean annual CH4 fluxes. Soil moisture, soil temperature, and bulk soil oxygen (O2) concentrations were strongly positively correlated with both soil N2O and CH4 emissions and soil nitrate (NO3-) concentrations were also correlated with N2O emissions. To further explore the production and consumption pathways of N2O and CH4 I conducted 15N-N2O and 13C-CH4 stable isotope pool dilution experiments under contrasting drained and flooded conditions. Gross N2O production was strongly positively correlated to soil moisture. However, gross N2O consumption rates were highest in drained subsoils and were increased with increasing NO3- concentrations, suggesting N2O consumption was indirectly controlled by substrate availability for denitrifiers. Combined with a decline in net N2O fluxes observed under drained conditions, our results suggest organic-Al complexes may facilitate anaerobic hotspots of N2O consumption. For CH4, gross consumption was generally greater than gross production. Gross CH4 production increased with soil depth, likely driven by both increased soil moisture and anaerobic soil conditions. Gross CH4 consumption was negatively correlated with soil moisture and positively correlated with soil pH, indicating low pH and O2 availability directly limit CH4 consumption. Gross CH4 consumption also accounted for up to 25% of net CO2 production (mean ± SE: 7% ± 3%) in drained soils. Our results suggest that gross N2O and CH4 production were temporally decoupled from gross N2O and CH4 consumption and are driven by soil moisture status and its effects on NO3- and pH. Continuous flux measurements from the alfalfa ecosystem showed a smaller consistent source of N2O and an overall small net CH4 sink. Using more than 100,000 flux measurements over four years from agricultural alfalfa, I also found that hot moments of N2O emissions were only 0.2% to 1.1% of annual measurements but were 31.6% to 56.8% of the annual flux. We found that both the magnitude and contribution of hot N2O moments to overall emissions decreased over time. Normalized difference vegetation index (NDVI), soil temperature, moisture, and oxygen (O2) were all significantly correlated with soil CO2, N2O, and CH4 fluxes, although associations varied across both soil depth and timescales. Our combined observations suggest that plant productivity regulates background N2O emissions but soil moisture and soil O2 availability are the dominant controls on the net greenhouse gas budget of alfalfa agroecosystems. Together, the finding in this dissertation contribute to a better understanding of the biogeochemical pathways of soil CO2, N2O, and CH4 fluxes in managed peatland soils. They also highlight the importance of land management decisions in stimulating soil greenhouse gas emissions and improve our understanding of the processes, controls, and distribution of hot spots and hot moments of greenhouse gas emissions.