QUANTITATIVE DISTRIBUTION OF GEOMORPHIC FEATURES ACROSS THE GREAT MARSH, MA, VIA INTEGRATION OF MULTI-TEMPORAL MULTISPECTRAL REMOTE SENSING WITH LIDAR AND GIS

Eco-geomorphological modeling in salt marshes faces unique challenges due to tidal oscillation and variability, fieldwork logistics, and the inherent dynamic nature of these environments. Recently developed technologies and methods introduce the capability to create fine-scale, system-wide databases that may be used to better understand the geomorphic evolution of intertidal systems. This study combines the use of 20cm AIMS-1 multispectral imagery flown at spring high and low tides with a LIDAR-derived DEM of a New England estuarine system to quantify relationships between marsh features, their metrics, elevations, and tidal datums. These methods may also be effectively applied to analogous and complementary systems in support of future development of process-driven models to explain field observations.

Geomorphologic analyses of the distribution of marsh features at the Great Marsh, Massachusetts, study area support field observations and qualitative observations in the literature. Water-filled ponds (also called pools) are concentrated in regions of the marsh around and above mean higher high water (covering ~6% of marsh surface area). Drained ponds and pannes are distributed across the marsh platform; drained ponds cover up to 15% of the low marsh surface, whereas pannes are most abundant in the high marsh (~7% total area). The geomorphic distribution of pannes and ponds indicates strong elevational control on feature distribution, most likely as a function of depth and duration of inundation. These analyses are complicated by human interference in the form of ditches, which cover nearly 7% of the marsh surface peaking in the upper low marsh. These ditches have the general effect of both draining ponds and inhibiting pond formation, altering the natural distribution of drained ponds, pannes and water-filled ponds. These preliminary analyses move us closer to understanding and developing process-driven models which will be increasingly important in a regime of accelerating sea-level rise.