Monitoring Vegetation Dynamics at a Tidal Marsh Restoration Site: Integrating Field Methods, Remote Sensing and Modeling

Abstract

AbstractSea level rise threatens coastal wetlands worldwide, and restoration projects are implementing strategies that decrease vulnerability to this threat. Vegetation monitoring at sites employing new restoration strategies and determination of appropriate monitoring techniques improve understanding of factors leading to restoration success. In Central California, soil addition raised a degraded marsh plain to a high elevation expected to be resilient to sea level rise over the next century. We monitored plant survival and recruitment using area searches, transect surveys, and unoccupied aircraft systems (UAS) imagery. We used random forest modeling to examine the influence of nine environmental variables on vegetation colonization and conducted targeted soil sampling to examine additional factors contributing to vegetation patterns. Limited pre-construction vegetation survived soil addition, likely due to the sediment thickness (mean = 69 cm) and placement method. After 1 year, about 10% of the initially bare area saw vegetation reestablishment. Elevation and inundation frequency were particularly critical to understanding restoration success, with greatest vegetation cover in high-elevation areas tidally inundated < 0.85% of the time. Soil analysis suggested greater salinity stress and ammonium levels in poorly-vegetated compared to well-vegetated areas at the same elevation. We found that both transect and UAS methods were suitable for monitoring vegetation colonization. Field transects may provide the best approach for tracking early vegetation colonization at moderate-sized sites under resource limitations, but UAS provide a complementary landscape perspective. Beyond elucidating patterns and drivers of marsh dynamics at a newly restored site, our investigation informs monitoring of marsh restoration projects globally.