Soil accumulation in a Chesapeake Bay salt marsh: Modeling 500 years of global change, vegetation change, and rising atmospheric carbon dioxide

Wetlands store over 20% of terrestrial carbon; therefore, changes in carbon cycling in wetlands may represent an important potential feedback on rising CO₂ and resulting greenhouse warming. The goal of my dissertation is to understand how global changes and vegetation changes alter the accumulation of soil organic matter (SOM) in a Chesapeake Bay salt marsh. I quantified several key linkages between global changes (i.e., CO₂, temperature, sea-level, and salinity), species composition, plant production, and decomposition of SOM to develop five ecosystem models which simulate soil profiles and elevation change on a century time scale. I tested these models with data collected within and adjacent to an ongoing CO₂ enrichment study in a Chesapeake Bay salt marsh.; The most complex of the five models (Full model) assumed that C ₃ and C₄ species differ with respect to plant production, litter chemistry, and belowground structure. The simplest model (Null model) assumed these characteristics were the same for the two species. The Null model reproduces the general data set more accurately than the other models, whereas the F-S model, in which rooting profiles are assumed identical for both species, performs best at reproducing profiles of specific organic chemistry fractions. Although none of the models accurately reproduce the observed values of elevation change, the models which include species-specific root profiles (i.e., the Full and F-C models) correlate the strongest with observed values (r2 = 0.78 and 0.72) compared to the other models (r 2 < 0.19). The Full and F-C model may be useful for examining potential feedbacks between C₄ abundance and elevation change, mediated by species differences in root profiles.; When I incorporated observed effects of CO₂ on key ecosystem functions in the Full and F-S models, I found that CO₂-mediated changes in microbial activity best predict CO₂ effects on soil profiles. Predictions of the Full model also correlate with the observed effect of CO₂ on elevation change more strongly (r2 = 0.78) than the other models (r2 < 0.03). In general, models predict that the effects of CO₂ on SOM and elevation change are likely to be subtle, depending on initial soil chemistry and vegetation type.