Digital mapping

Normal color digital ortho-photos with six inch resolution were used to test the
feasibility of mapping the habitat of the Florida apple snail, the exclusive food
source of the Endangered Florida snail kite. Several remote sensing techniques
such as unsupervised and supervised classifications are evaluated and accuracy
assessed. While the overall ability to classify apple snail habitat is significant,
sparse areas of emergent vegetation which are preferred foraging areas for the kite
were excluded by the classification and needed to be re-introduced for further
analysis. The methodology can be repeated to measure change over time since the
NC imagery of Grassy Waters Preserve is acquired frequently. Additionally, three
spectral levels of density abundance of wetland graminoid species that denote apple
snail habitat were examined for mapping potential on the imagery. A site suitability
analysis using a weighed overlay tool in ArcGIS is suggested as a means to include
numerous other key factors influencing habitat selection by the snail kite in a
mapping project.

Digital orthophoto quadrangle quarter (DOQQ) is a new type of imagery. The DOQQ is a product of the United States Geological Survey, which has many uses. The spectral and spatial qualities of a DOQQ are equal to and surpass the qualities of traditional satellite imagery. These qualities can be demonstrated by answering the following questions: (1) Is the quality of the spectral information on a color infrared DOQQ comparable to a SPOT and TM Landsat satellite imagery for the purpose of digital image classification? (2) What are the optimal number of classes for achieving an accuracy supervised classification of a one-meter resolution DOQQ? (3) What is the effect of changing the spatial resolution on image classification of a DOQQ?

Watershed delineations in low topographic relief landscapes with canal networks require hydrography. This hypothesis was tested by utilizing digital elevation models (DEMs). Seven DEMs were processed using two automated delineation methods. Both methods include filling of localized sinks, but an enhanced method incorporated incremental "tipping" to provide flow direction. In two watersheds studied, DEMs with only topography delineated watersheds within 24% and 84% of the manual delineation. With the addition of hydrography, including canals, both watersheds were delineated to within 85% and 92% of the manual delineation. Contrary to the assumption that tipping would improve delineation, tipping decreased the delineated area by removing a water flow path. Based on this research, DEM input data and delineation method strongly influence automated watershed delineations in low relief conditions.

A new technique was devised that produces a LIDAR based predevelopment Digital Elevation Model and offers a more superior visualization, of the natural landscape prior to development, than the Bare Earth Model produced by Florida International University's Hurricane Center. Two locations in southeastern Florida have been chosen for this study. The first site is utilized as a control for testing the model and contains a present day ridge system that runs north and south through the Boca Raton cemetery. The second site is a paleowatershed, the Yamato Marsh which was located in Delray Beach. Modeling of the marsh site will assist in delineating surficial geomorphic features before they were anthropogenically altered. The predevelopment Digital Elevation Model successfully represents human derived land-use changes and defines the watershed boundaries of the historic Yamato Marsh.