Quantifying the effects of European beach grass on aeolian sand transport over the last century: Bodega Marine Reserve, California

Introduction of European beach grass (Ammophila arenaria) to coastal dune systems of western North America induced significant changes to the transport and storage of sediment, and consequently the nesting habitat of the western snowy plover (Charadrius alexandrinus nivosus). At the Bodega Marine Reserve and Sonoma Coast State Park, Ammophila was introduced within the ~0.5 km2 dune area in the 1920's to limit the flux of sand through Bodega Harbor and agricultural land. To assess the potential impact of restoration efforts (Ammophila removal) on aeolian sediment flux, we measured sediment flux as a function of wind speeds and ground cover, and used these measurements to parameterize a spatial model for historical sand deposition Fine- to coarse-grained lithic to sub-lithic sand is delivered to the Bodega dune system from Salmon Creek beach, the down-shore terminus of a littoral system fed by the 3846 km2 Russian River catchment, several small (<100 km2) coastal catchments, and seacliff erosion. Littoral sediment traverses the 1.8 km wide dune system from NW to SE via aeolian transport. Ammophila colonization occurred initially adjacent to the shoreface, inducing deposition of a ~10 meter-high foredune and has subsequently encroached the ~0.5 km2 region between the foredune and Bodega Harbor. Comparison of historical topographic maps via raster subtraction indicates rapid construction of both the foredune and a ~15 meter-high transverse dune (Gaffney ridge) at the edge of the planted region. An average accumulation rate of ~4,000 m3/yr is indicated within the study swath by the preserved sediment volumes. Within the modern dune system, unvegetated areas exhibit 2-3 meter wavelength, ~1/2 meter amplitude mega-ripples, and the uppermost 2-10 cm consists of coarse-sand to granule-sized armor layer. In contrast, grain-sizes in vegetated areas are largely vertically homogenous. Open areas are typically 2-8 meters lower than adjacent vegetated areas, and show evidence for net lowering of the land surface (i.e., exposed fence posts, roots). Conversely, vegetated areas appear prone to sediment accumulation, particularly downwind of unvegetated areas. We measured sand transport using 0.5 m high traps deployed at 18 sites throughout the dune field, and used a linear mixed effects model to predict transport rate as a function of wind and ground cover class, taking into account random effects of sampling date and repeated measurements at each site. The analysis indicates up to 450-times higher transport rates in unvegetated areas relative to vegetated areas at peak wind conditions. We then used these results to parameterize a simple raster-based sediment flux model for the 0.5 km wide study area, using LIDAR-based topography and aerial orthophotography to classify ground cover. Due to the nearly complete compartmentalization of sediment flux by vegetative baffling, the model suggests that proposed restoration (removal of vegetative cover) of the seaward 1 km of the dune system would lead to significant increases in sediment transport in the treated area accompanied by accumulation along its vegetated downwind edge, but little to no change in sand flux within Ammophila-covered areas >0.2 km downwind of restored areas.