![]() ![]() |
|||
|
Landuse and Water
Quality Threats to the Mammoth Cave Karst Aquifer, Kentucky
|
Alan Glennon Chris Groves Michael Anderson Hoffman Environmental Research Institute Western Kentucky University Geography and Geology Bowling Green, Kentucky 42101 (270) 745-4169
John Fry |
|||||||||||||||||||||||||||||||||||||
|
Kentucky’s Green River Basin contains some of the world’s most highly developed karst landscapes. The Mammoth Cave Karst Aquifer includes the world's longest known cave, the 500+km long Mammoth Cave System. Due to the globally significant nature of both the cave and its aquatic ecosystem, the area has been designated as a national park, a World Heritage Site, and an International Biosphere Reserve. In addition, south-central Kentucky was recently selected by the Karst Waters Institute as one of the world's ten karst ecosystems in most critical need of protection (Springston, 1998). The park is home to at least nine federally endangered species, including Kentucky Cave Shrimp (Palaemonias ganteri) and eyeless cavefish (Tryphlichthys subterraneus and Amblyopsis spelaea), as well as numerous other sensitive species. Two species of endangered bats also live within the vadose, or air-filled part of the aquifer. As the aquifer's waters ultimately discharge at major springs along the south side of the Green River, its quality also impacts the river ecosystem that includes six endangered freshwater mussels. The intimate relationship between landuse and groundwater quality in the well-developed karst areas of the Green River Basin occurs due to rapid groundwater velocities. The water can travel as fast as several kilometers per day. This can allow contaminants to travel into, and through, the aquifer in a short time. A toxic spill along I-65 or the CSX railroad in the recharge area, for example, could potentially have rapid and devastating impacts on the habitats of numerous species. Threats to the health of karst ecosystems, including the Mammoth Cave Karst Aquifer, come from a variety of agricultural, urban, and transportation landuse practices. For the Mammoth Cave Karst Aquifer, the US Fish and Wildlife Service has funded the Hoffman Environmental Research Institute to: 1) classify land use, 2) perform a dye-trace investigation to determine if the hydrologic network of the eastern end of the Mammoth Cave System extends into a fourth major drainage basin, 3) develop a Geographic Information System (GIS) as a data storage and retrieval tool, and 4) investigate potential protection strategies for these areas. Anderson landuse classification can be divided into three levels, with a higher level representing further specificity. James Anderson, et al, introduced the standard classification system in the 1976 USGS Professional Paper #964, entitled A Land Use and Land Cover Classification System for Use with Remote Sensor Data. A one digit number represents seven broad categories of Level I landuse, including agriculture, urban development, rangeland, water, wetlands, barren land, and forestland. Level II is broken into at least 29 categories that are depicted by a two-digit number. Level III is most detailed, with a three or more digit number representing over 100 landuse types. The Mammoth Cave Karst Aquifer and adjacent study areas (Suds Spring Basin, Mile 205.7 Spring Basin, and several intermediate and undelineated minor basins) have 94 different Level III landuse types in 7,931 landuse parcels (Table 1). The total area of the Mammoth Cave Karst Aquifer and adjacent study area covers 382km2. Anderson Level III landuse classification at 1:24,000 scale was conducted for the 382km2 comprising the Turnhole Bend Basin, Echo River Basin, Pike Spring Basin, Mile 205.7 Spring Basin, Suds Basin, and intermediate drainage areas. Landuse classification of the study areas was partially completed by the Science and Resource Management staff at Mammoth Cave National Park. Following their methods, the remainder of the study area was classified and placed into digital form by Western Kentucky University. Initial parcels and landuse were identified from 1:24,000 color infrared aerial photographs taken by the Tennessee Valley Authority on March 20, 1990. The images consisted of nine-inch color positive film, which were viewed under magnification on a light table. Most polygons were drawn at 1:4,000, unless one particular landuse type covered a large area. This mainly occurred in the case of a continuous forest of the same type and thickness. Observing color patterns on the color infrared images and transposing the patterns into polygons in Environmental Systems Research Institute’s (ESRI) ArcView GIS program created parcel polygons. In infrared photography analysis, reds and greens are types of vegetation and farmland; purple trees are pines or evergreens; and reddish brown trees are cedars. These color patterns were then translated into Anderson Level III landuse. A base map for landuse parcels was constructed in an ArcView project. The base map consisted of 1:12,000 digital orthophoto quarter quadrangles in the Universal Transmercator North American Datum 1927, Zone 16 coordinate system. Parcel shapes and landuse codes interpreted from the color infrared images were then transposed to the base map. For some of the study area, Mammoth Cave National Park had classified landuse and placed parcels and codes on mylar sheets. As necessary, areas previously classified and stored on the mylar sheets were digitized and placed into the primary ArcView landuse database. Using the query ability of the ArcView GIS digital database, an Anderson Level I landuse table was generated for both the Mammoth Cave Karst Aquifer and total study area. Table 1 presents this Level I landuse for the 336.4km2 of the Mammoth Cave Karst Aquifer. Suds and Mile 205.7 Spring Basin were included in the study area because
they were potential outlets for the dye trace. The results from the Pike
Spring dye trace, in the northeast portion of Mammoth Cave, showed positive.
A dye trace near Candlelight River conducted within this study emerged
at Floating Mill Hollow Spring, thus identifying a new spring basin that
communicates with known passages in the Mammoth Cave System. Additionally,
field checking of 84 randomly selected parcels was conducted to classify
vague areas in more detail and verify quality. Quality control checks found
that classification for this project was above Anderson’s standard 85%
accuracy guideline.
Table 1: Anderson Level I landuse within the Mammoth Cave Karst Aquifer and the total study area
|
||||||||||||||||||||||||||||||||||||||
|
WORKS CITED Anderson, James R., Ernest E. Hardy, John T. Roach, and Richard E. Witmer. "A Land Use and Land Cover Classification System for Use with Remote Sensor Data. USGS Professional Paper #964: 1976 Anderson, James R. Land Use and Land Cover Maps and Statistics from Remotely Sensed Data. Omaha, Neb.: Remote Sensing Committee, Association of American Geographers, Remote Sensing Application Laboratory, University of Nebraska at Omaha, 1977. Lillesand, Thomas M., and Ralph W. Kiefer. Remote Sensing and Image Interpretation. New York: Wiley & Sons, 1994. National Research Council Committee on Remote Sensing for Agricultural Purposes. Remote Sensing with Special Reference to Agriculture and Forestry. Washington: National Academy of Sciences, 1970. Springston, Bob. "Karst Waters Institute Lists Top 10 Endangered Karst Ecosystems." American Caves. 11 (1998): 7. Star, Jeffery L., John E. Estes, Kenneth C. McGwire. Integration of Geographic Information Systems and Remote Sensing. Cambridge; New York: Cambridge University Press, 1997. |
||||||||||||||||||||||||||||||||||||||
| -Back- to Hoffman GIS
Hoffman
Institute |