Main content area

Hyperspectral discrimination of floating mats of seagrass wrack and the macroalgae Sargassum in coastal waters of Greater Florida Bay using airborne remote sensing

Dierssen, H.M., Chlus, A., Russell, B.
Remote sensing of environment 2015 v.167 pp. 247-258
Sargassum, absorption, biomass, coastal water, energy, habitats, hyperspectral imagery, image analysis, leaves, macroalgae, normalized difference vegetation index, nutrients, reflectance, remote sensing, seagrasses, spectrometers, vegetation, water content, Florida
Floating mats of vegetation serve to transfer biomass, nutrients and energy across marine habitats and alter the spectral properties of the sea surface. Here, spectral measurements from the airborne Portable Remote Imaging Spectrometer (PRISM) imagery at 1-m resolution and experimental mesocosms were used to assess the hyperspectral properties of the macroalgae Sargassum and aggregations of the seagrass Syringodium filiforme wrack in Greater Florida Bay. A simple Normalized Difference Vegetative Index (NDVI) effectively discriminated the presence of vegetation floating on the sea surface. The Sargassum Index derived from reflectance ratios at 650 and 630nm was used to effectively discriminate Sargassum from Syringodium wrack. Mesocosm spectral measurements revealed an initial lowering of wrack reflectance over the first 3days followed by a subsequent increase in reflectance over the next 8days. The age of the wrack estimated from 2 to 5days was best characterized using narrowband indices of the water absorption feature at 930 and 990nm potentially from increasing water content in wrack leaves over time. Hyperspectral imagery (<10nm) was necessary to differentiate between these two types of floating vegetation and assess age of the wrack. PRISM imagery revealed seagrass wrack organized in 5–35m spaced windrows caused by Langmuir circulation. Wrack was only detectable at 60mpixel resolution when densities were high and individual windrows were in close proximity.