-15 
95. 
nal 
phy 
oon 
et 
204. 
the 
SM 
Jers 
vey 
of 
ton, 
95. 
and 
ual 
see, 
1ces 
Info 
ling 
tos. 
atic 
tric 
RS- 
New 
205. 
y of 
tion 
ical 
rate 
-16 
the 
and 
ms, 
the 
tric 
GPS, IMAGE PROCESSING AND GIS TECHNIQUES 
~ FOR COASTAL WETLAND MAPPING APPLICATIONS 
Roy Welch and Marguerite Remillard 
Center for Remote Sensing and Mapping Science (CRMS) 
Department of Geography 
The University of Georgia 
Athens, Georgia 30602 USA 
ISPRS Commission IV, Working Group 1 
KEY WORDS: GIS, GPS, Vegetation Mapping, Database Development 
ABSTRACT: 
Global Positioning System (GPS), image processing and geographic information systems (GIS) techniques were employed to establish a 
database for South Florida wetland environments threatened by urban expansion, agriculture, exotic plant invasion and hurricanes. 
1. INTRODUCTION 
Geographic information systems (GIS) are currently an integral 
component of natural resource monitoring and management 
operations worldwide. Useful not only for data storage, man- 
ipulation and map production, the strength of GIS is spatial 
data analysis and predictive environmental modeling 
(Remillard and Welch, 1993; Cowen, et al., 1995). Remotely 
sensed satellite image data and aerial photographs serve as an 
excellent reference base for GIS data layers, provide ground 
control and up-to-date thematic information. The integration 
of GIS with image processing and Global Positioning System 
(GPS) techniques has proved particularly useful for resource 
mapping applications. 
In compiling resource databases, thoughtful consideration must 
be given to: 1) project and end-product requirements; 2) 
coordinate systems and ground control; 3) GPS; 4) mapping 
and database assembly techniques; and 5) GIS analysis 
procedures. In the following discussion, a project is presented 
that combines GPS, image processing and GIS technologies 
that are particularly useful in preparing databases for coastal 
wetland mapping applications. 
2. DIGITAL DATABASES FOR SOUTH FLORIDA 
NATIONAL PARKS/PRESERVES 
The University of Georgia’s Center for Remote Sensing and 
Mapping Science (CRMS) is working in conjunction with the 
U.S. Department of Interior’s National Park Service (NPS) to 
construct a GIS database and associated detailed vegetation 
maps for ecologically unique wetlands in South Florida, 
including Everglades National Park, Biscayne National Park, 
Big Cypress National Preserve and the Florida Panther Refuge 
(Welch, et al., 1995). This 6,000 mi? area, collectively referred 
to as the "Parks", is threatened by urban expansion, nutrient 
runoff from agricultural lands, invasive exotic plant species, 
increased recreational use and episodic disturbances such as 
931 
hurricanes (Davis and Ogden, 1994). Extending roughly from 
Miami on the east to Naples on the west, southward to Florida 
Bay, the Parks represent the remaining lands of the greater Big 
Cypress Swamp and Everglades ecosystems that once covered 
approximately one-third of the Florida peninsula (Figure 1). 
  
  
    
  
South Florida 
Vegetation Database 
  
  
  
  
  
  
  
[^ National Preserve 
\ 
  
  
  
    
   
   
  
Everglades 
National 
Park 
  
National 
Park 
y — 
  
  
  
  
Figure 1. South Florida vegetation database study area. 
Preservation of the natural vegetation within these parks is of 
national concern and requires detailed information on vegeta- 
tion distributions, along with planimetric features such as 
roads, rivers and canals. In order to meet these objectives and 
construct a GIS database in digital format, it was necessary to: 
1) develop techniques for using control transferred from 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996