Subtask 4.1: Present Day Northern Gulf Coast Landscape Structure

Subtask Leader: Amar Nayegandhi - Jacobs Technology, Inc./USGS Florida Integrated Science Center, St. Petersburg

Subtask 4.1 Aim:

Spatial knowledge of the Northern Gulf of Mexico (NGOM) region involves mapping natural habitats, vegetation communities, cultivated lands, residential development and urban zones. These data may be absent, and if previously acquired, rapidly becomes obsolete. In particular, subsequent to the landfalls of intense storms during the summer of 2005, there is a need to create accurate, finely detailed up-to-date maps that capture the present-day biophysical structure of the NGOM terrestrial landscape.

The NOAA Coastal Change Analysis Program (C-CAP) in collaboration with the USGS National Land Cover Dataset (NLCD) have recently created a post-Hurricane Katrina land-cover map using 30-meter resolution Landsat imagery for the areas impacted by the hurricane. This 2006 C-CAP land-cover data product complements the Gulf of Mexico C-CAP land-cover products for 1996, 2001, and 2005. The Coastal NLCD classification scheme, as derived by USGS-EROS and NOAA C-CAP, is the standard scheme used for C-CAP land-cover mapping projects. The methods and procedures for developing the land-cover map are also well documented. 

The existing land-cover maps are created at the effective resolution of the Landsat source imagery.  At 30-meter resolution, some of the wetland and upland classes do not provide the necessary detail, especially in transition zones. High-resolution satellite imagery, obtained from either Quickbird or IKONOS sensors, will provide the necessary pixel resolution to prepare land-cover data sets over selected regions determined to require high spatial resolution. 

Prior work in this area using pixel-based classification with high resolution data show some problems because the level of information in the data produce a lot of incorrect classified pixels. New automated methods need to be investigated to improve land-cover classification using high-resolution satellite imagery. This task will also have to ensure that the classification scheme used for the land-cover analysis can be cross-walked with the existing NOAA C-CAP scheme. 

Subtask 4.1 Activities:

The goal of Subtask 4.1 is to create up-to-date post Hurricanes Katrina and Rita maps that describe the terrestrial landscape of the entire NGOM region at the highest practical spatial resolution.

The specific objectives of this subtask are to:

1. Compile existing information on northern Gulf Coast land-cover,
2. Evaluate the use of high-resolution satellite imagery to conduct landcover classification using two prototype locations with significant diversity in landscape structure.
3. Create a data set of QuickBird and/or IKONOS multispectral satellite images for portions of the coastal plains of Louisiana, Mississippi, and Alabama that require high-resolution landscape structure characterization,
4. Construct, evaluate, and publish a post Hurricanes Katrina-Rita land-cover map for the above locations of the NGOM project study region that underwent change during summer 2005, at a target spatial resolution finer than 5 meters, based on the analysis of commercially available multi-spectral satellite images.