COS 142-2 - Salinization, inundation and tree mortality interact to affect greenhouse gas emissions from stressed coastal forests

Abstract: Sea level rise and intensifying storms cause salinization and freshwater inundation of coastal forest soils which can result in localized tree mortality and altered ecosystem carbon cycling. However, it is not yet clear if salinization and inundation interact with tree mortality to affect greenhouse gas (GHG) emissions and potentially feedback with climate change. To address this gap in knowledge, we investigated the effects of chronic and pulsed salinization and freshwater inundation on GHG emissions from coastal forests experiencing localized tree mortality. First, we collected in situ intact soil cores soils from forest stands experiencing chronic salinization and associated tree mortality (high salinity stands) and adjacent forest stands with lower salinity soils and healthy overstory trees (low salinity stands) in baldcypress and loblolly forest types. Here, we tested the hypothesis that chronic salinization and associated tree mortality decrease CO₂ production by altering the quantity and quality of soil organic matter. Second, we collected in situ, intact soil cores from high and low salinity stands in loblolly forest before and after a hurricane related storm surge to test the hypothesis that underlying exposure to chronic salinization at the stand scale can mediate the effects of pulsed salinity from a storm surge on soil CH₄ fluxes. Finally, in lab mesocosms, we exposed soil cores from high and low salinity stands to three water addition treatments to test the hypothesis that underlying exposure to chronic salinization at the stand scale can mediate the effects of freshwater inundation on CH₄ fluxes. We found that chronic salinization and associated tree mortality decreased CO₂ fluxes in loblolly (p< 0.0001), but not baldcypress forest with in situ soil inundation patterns potentially driving the site effect. Opposite to our expectation, CH₄ fluxes correlated positively with in situ chronic and pulsed salinity in a loblolly forest (R²=0.39, p=0.001), with high organic matter availability potentially decreasing competition between methanogens and sulfate reducers despite increased sulfate concentrations in higher salinity soils. Finally, rainwater addition significantly increased CH₄ fluxes in low mortality stands (loblolly pre-hurricane, low, F_2,15=27.59, p< 0.001; baldcypress, low, F_2,15=9.77, p=0.002), but not high mortality stands, suggesting that stand level salinity and mortality can mediate the effects of extreme precipitation events on GHG fluxes. Overall, we demonstrated that patterns in greenhouse gas emissions from stressed coastal systems can be context dependent and vary based on environmental covariates such as soil inundation, organic matter availability and pH.