Signals of nonlinear, multiscale and stochastic processes in coastal landscapes
Kearney, William Sheppard
MetadataShow full item record
Salt marshes are some of the most productive and valuable landscapes on earth, but they are vulnerable to the effects of sea-level rise, erosion and eutrophication. These processes act on a wide range of temporal and spatial scales, which complicate assessments of the health and stability of marsh ecosystems. High-frequency monitoring using in situ sensors captures the complete range of these dynamics, but extracting meaningful physical and ecological information from these signals requires process-based models coupled with statistical techniques. I develop and apply such methods to study two coastal landscapes, a coastal pine forest on the Eastern Shore of Virginia and a mesotidal salt marsh complex in the Plum Island Estuary, Massachusetts. Observations from groundwater wells in the Virginia pine forest indicate that storms are the dominant controls on the hydrology of the forest and that tidal influence is nonexistent. This forest exhibits a distinct spatial pattern in age structure in which young trees do not grow at low elevations. This pattern can be explained by a model that includes the interaction of sea-level rise, storms and the age-dependent variation in tree stress response, which predicts that the long-term evolution of the boundary is an ecological ratchet. Stresses due to sea-level rise slowly push the boundary at which young trees can survive upslope. Powerful storms then kill the mature, persistent forest at low elevations, which quickly pushes the forest boundary up to the regeneration boundary. Salt marshes need to accumulate sediment to replenish material lost as sea-level rises and creek banks erode. Fluxes of sediment can be monitored with simultaneous high-frequency observations of flow from acoustic Doppler current profilers and turbidity from optical backscattering sensors. I first investigate the relationship between water level and flow in marsh channels and develop predictive stage-discharge models to simplify the monitoring of fluxes. I then construct sediment budgets for eleven salt marshes in the Plum Island Estuary. The observed budgets depend strongly on the unique hydrodynamic conditions of each marsh channel. Variability in these conditions leads to the observed spatial and temporal variability in sediment fluxes from these marshes.