|
|
|
| Home | Projects | Members | Publications | Committee | Data | RSPA | Meetings | Contact |
| Introduction |
|
| Image Processing | |
| Results | |
| Return to Environmental Assessment Home | |
| Return to NCRST-E Home |
|
The Big Creek watershed is a small basin on the northern fringe of Atlanta, Georgia's vast urban sprawl (Figure 1). The area of the basin is approximately 72 square miles upstream from USGS Gage # 02335700 (Big Creek near Alphareta, GA) and is contained in three counties: Fulton, Forsyth and Cherokee. From 1980 to 2000, the population of the basin increased by 200% and the number of new businesses increased 15-fold. Increases in developed land of this magnitude are associated with concomitant increases in impervious surfaces, which have a significant effect on the hydrologic regime. Whereas "development" is often quantified in terms of population or housing statistics in a geospatial context of counties, zip codes, or other "districts," remote sensing offers an alternative means of quantifying development in a pixel- or segment-based geospatial context. Because of the magnitude of changes within Big Creek, it was selected for study using remote sensing imagery to classify land use and land cover of Big Creek. Changes in the percentage of developed land within the watershed are correlated to changes in streamflow at decadal intervals.
Figure 1. Map showing the location of Big Creek with respect to Atlanta, GA and a "true-color" image of Big Creek watershed as viewed from space in the year 2000. Click on image for a larger version.
|
|
Acquisition
Three Landsat images were acquired for the purpose of classifying land cover in the Big Creek watershed. The watershed is fairly small and is contained well within a single image. Only summertime images were acquired during the time of year when the vegetation is fully leafed out. The images acquired are identified in Table 1. None of the images collected in 1990 were cloud-free so an image from 1991 was acquired instead. Notice that each of the images were obtained from a different generation of sensor. The MSS sensor has the lowest spatial and radiometric resolution and thus limits the overall number of classes that can be satisfactorily achieved.
Table 1: Specifications for images acquired for the Big Creek classification.
Georectification
Atmospheric Correction
Segmentation and Classification
First, the image is segmented. That is, adjacent pixels with similar spectral characteristics are grouped into a segment or image object (Figure 2b). After training by the image analyst, the segments are then clustered into classes (Figure 2c). Thirteen land cover and land use classes were defined for the classification (Figure 3).
|
|
The annual population within Big Creek watershed was obtained from US Census data. The basin's population increased by 104% from 1980 to 1990 and by an additional 105% from 1990 to 1998 (Figure 4). During the period between 1980 and 1998, the number of new business starts increased 15-fold (Figure 4). The estimated 1998 population distribution based on Tiger census blocks is shown in Figure 5.
Classification results for Big Creek watershed are shown in Figure 6. In Figure 7, the classification is collapsed into two principle classes, Developed and Undeveloped, to highlight growth in the watershed as a result of urban sprawl on the northern fringes of Atlanta. Woodland, which comprises about 30% of the watershed, remained fairly constant between 1980 and 2000. Fields, however, either pasture and/or fallow cropland, exhibit the greatest change from about 17.6% in 1980 to 9.4% in 1990 to 3.7% in 2000. Overall, Developed Land comprises about 11.7% of the watershed in 1980 and 17.0% in 1990, but increased to 31.6% 2000. These figure correspond to a 45% increase in Developed Land during the first decade and a 86% increase during the second decade.
|
