Spatial spectral radiometric and temporal dimensions of remotely sensed data for monitoring wetland vegetation in Southern California

Remotely sensed data and analysis techniques previously applied in wetlands applications on the Atlantic and Gulf coasts need to be refined to suit southern California`s wetland characteristics. A scaling framework was used to define the most appropriate dimensions of remotely sensed data for mapping vegetation composition in natural and restored wetlands of the Sweetwater Marsh National Wildlife Reserve. Field and image data were used in the framework to define optimal spatial, spectral, radiometric and temporal dimensions. Spatially, the optimum pixel sizes were between 0.75m - 2.0m. Spectrally, maximum reflectance differences between vegetation cover types were in the red and NIR bandwidths. Temporally, reflectance differences were maximized between a majority of vegetation cover types during the months of May and June. Image data for several monitoring projects continue to be collected using the scaling parameters established in this work. 10 refs., 4 figs., 3 tabs.