Geography
Mapping Phosphorus in the Everglades
In the 1880s, a real-estate developer named Hamilton Disston bought four-million acres in Florida for $1 million and began what would eventually develop into a network of 2400 km of canals and an extensive water management system. The establishment of the canal system, in conjunction with urban and agricultural development, reduced the historic Florida Everglades over time by 50% and introduced a variety of nutrients foreign to the ecosystem, including phosphorus. Large-scale efforts to sustainably manage what’s left of the Everglades involve an amalgamation of efforts from scientists and policymakers to understand, monitor, and direct future drivers of environmental change. One of these drivers is phosphorus, a nutrient that previously did not exist in the region and which has immensely altered the ecosystem.
The Historic Everglades
During my internship with the Everglades Foundation, I worked on a variety of GIS and mapping projects, including a historical ecology analysis of the Florida Everglades. In collaboration with the senior GIS Scientist at the Everglades Foundation, Rosanna Rivero, and Noosha Mahmoudi, a PhD student at FIU, a vegetation map of South Florida produced in 1913 was transformed for GIS compatibility using ArcGIS, Adobe Illustrator, and Photoshop.
Synthetic Aperture Radar and the Florida Everglades
Synthetic Aperture Radar (SAR) is an active remote sensing technique in which microwave signals are emitted from a transmitter toward a subject area, reflected by the features of that area, and returned to a sensor. SAR is recognized as a valuable and effective tool in mapping and monitoring wetland ecosystems. This study examines three different SAR platforms (ERS, Radarsat, and JERS) in terms of their usefulness in monitoring the changing hydrologic conditions in a region within the Florida Everglades known as the Shark River Slough.
Optical Remote Sensing and Land Classification in Maine
This goal of this project was to classify three different types of satellite imagery (ASTER, Landsat ETM+, and SPOT) in terms of deciduous and evergreen forest, water, built structures, and cleared land. The study area is the region surrounding Augusta, Maine, in the north-eastern United States. A soft-classification method known as neural network classification was utilized. Rather than each pixel being assigned to one class, as in hard-classification methods, the soft-classification process assigns a degree of class membership to each pixel for the five land-cover types. The accuracy of the classified images was assessed using a high-resolution aerial photograph of the region in which land classes could be clearly distinguished. The overall agreement between the classified images and the aerial photograph was highest in the Landsat ETM+ classified imagery, with an r-squared value of 0.85. The images in the table below visually represent the degree of class membership (where 0=0% and 1.00=100%) for the three different sensors according to the soft-classification. The results are being analyzed in light of seasonality of different cover types as tracked using multitemporal satellite imagery provided by AVHRR and MODIS.
Interactive Web Mapping and GIS
Web-based cartography and geographic information systems (GIS) have proliferated alongside developments in web application technologies and the integration of these technologies with GIS. Commonly manifested as navigational tools, weather maps, and “infotainment”, the interactive web map has become a widespread, familiar mode of communication for many individuals and organizations with computer and Internet access. These maps are not only user-friendly but also can be easily developed using free open-source mapping application programming interfaces (APIs) made available on the Internet. This project exemplifies one way in which interactive web mapping technologies can be used to communicate geographic research to scientists, professionals, and the general public via the Internet by applying web technologies to the organization and visualization of geographic research involving the use of GIS and remote sensing data in monitoring the hydrology of the Florida Everglades. Three technologies are utilized: XML, Google Maps API for Flash, and ActionScript (Adobe Flash/Flex).
Mapping Bicycle Injuries in Florida
A series of thematic maps of yearly bicycle-related injuries in Florida is presented using various methods of map classification. These maps help to identify areas that could improve their roads for bicycle communities in Florida.