Identifying Seasonal Water Patterns in Coastal Mangroves from Space

The Florida Coastal Everglades is a transition zone between the marine waters of the Gulf of Mexico and Florida Bay, and the freshwater of the “River of Grass”. With concerns over future climatic uncertainty and increasing demands for fresh drinking water, the ability to monitor and measure changes in water chemistry and quantity over the coastal landscape is crucial to understanding how the ecosystem responds with respect to restoration, conservation and sea-level rise. The large spatial expanse and remoteness of the coastal Everglades can result in high costs related to site-specific research; however, satellite remote sensing can help to mitigate this pecuniary problem. Furthermore, changes in leaf optical reflectance properties brought upon by environmental biophysical stressors can be used as an indicator of surface and subsurface water chemistry and can be useful in calculating evapotranspiration (ET). The objectives of this research are two-fold; 1) to identify relationships between spectra-derived biophysical indices of several mangrove species with the ionic and nutrient concentrations in the porewater, surface water, and groundwater of the mangrove ecotone, and 2) to estimate evapotranspiration along the coastal everglades. This method may provide a means to monitor water chemistry, quality, and quantity along the coastal zone on monthly, seasonal, and decadal time scales using various satellite platforms. Measurements of water chemistry, field-level leaf spectra and satellite data were used to develop a linear model to quantitatively predict water chemistry on the landscape scale within the various coastal mangrove communities of south Florida. Regional scale ET estimates were determined using a satellite-based energy balance approach, and accounted for the spatial variability in surface temperature, albedo, and vegetation spectra. Water samples were collected from surface water, groundwater and porewater and analyzed for ionic (e.g., Cl^-, SO₄^2-, Na²⁺, Mg²⁺, K⁺, and Ca²⁺) and nutrient (e.g., TOC, TN and TP) concentrations. Leaf-level spectra-derived biophysical indices were calculated from field spectrometer readings and used to identify changes to the mangrove spectral properties caused by environmental stressors. Landsat 5TM images from 1993 to 2009 were used to explore regional scale trends on annual and decadal timescales. Satellite spectra were extracted from five sites, spanning variety of mangrove communities with a range of biomass concentrations. Seasonal patterns were identified for both leaf-level and satellite reflectance data which correspond to the wet and dry seasons in south Florida. Strong correlations are exhibited between particular biophysical indices (e.g., EVI, NDVI) and ion concentrations at the field and satellite level. These relationships are primarily dependant on wavelengths in the green (550nm), red (680nm), and near-infrared (780-900nm) wavelengths. Stronger correlations exist when mangrove species are separated by type, indicating different mechanisms associated with water stress for the individual species. Correlations between nutrients and mangrove spectra demonstrate complex relationships, suggesting physiological differences of nutrient uptake caused by salinity-related stress. Ultimately, this research could provide a method to remotely monitor and measure the water chemistry and water budget changes to the environment; natural or man-made.