Imaging Spectroscopy Applications for Assessing Wetland Vegetation
Distributions and Coastal Resiliency in Louisiana
Abstract
Coastal wetlands provide a wealth of ecosystem services, including
improved water quality, protection from storm surges, and wildlife
habitat. Louisiana’s wetlands, however, are threatened by development,
pollution, and relative sea level rise (RSLR)—the combination of sea
level rise and subsidence rates. Despite widespread wetland loss, areas
such as the Wax Lake and Atchafalaya river deltas are in fact growing
due to their sediment loads, resulting in a complex of both degradation
and aggradation along the Louisiana coast. In order to understand and
model how coastal wetlands are responding to RSLR, there is a need for
improved vegetation mapping, biomass estimation, and landscape-scale
study of accretionary processes. AVIRIS-NG offers high spatial and
spectral resolution data that can be integrated with external
datasets—including from in situ measurements, monitoring stations, and
other remote sensing platforms—to study these distributions and
processes. Spectra derived from AVIRIS-NG imagery were used to
parameterize Multiple Endmember Spectral Mixture Analysis (MESMA) for
mapping vegetation functional types in addition to partial least squares
regression (PLSR) models for plant aboveground biomass (AGB). The
historical Landsat record complemented this analysis by deriving maps of
change in wetland health and sediment availability through time. Each of
these remotely sensed parameters were investigated to determine their
combined relationship to Louisiana’s coastal accretion rates. In
quantifying landscape-scale processes that impact wetland accretion,
this research aids the assessment of coastal resiliency in the face of
sea level rise. Further, the investigated imaging spectroscopy methods
pertaining to vegetation mapping, biomass estimation, and accretionary
modeling will inform future studies under the global Surface Biology and
Geology mission.