KARST GEOMORPHOLOGY RESEARCH IN SOUTH GEORGIA
my page
Since 1993, I have conducted collaborative studies that examine the geomorphology of sinkholes on the southeast coastal plain of Georgia. The first project, coauthored with Peter Jacobs, examines the morphology, distribution, and geomorphic significance of  >300 new sinkholes that formed in response to flooding  in 1994 at Albany, Ga. The second paper in this series (coauthored with Holly Wilkes, a VSU undergraduate student, and Peter Jacobs) examines spatial relationships between the new sinkholes and pre-flood (old) sinkholes. This paper, which is currently in review, uses digital elevation modeling techniques to evaluate the influence that topography has had on sinkhole development. Finally, in collaboration with Robert Gilbert (Queen's University, Kingston, Ontario, Canada), we have recently conducted several subottom acoustic surveys of sinkhole lakes near Valdosta. Although this project has only just begun, Bob and I are excited by the results that we have obtained from surveys at Lake Balboa which indicates in excess of 25 m of sediment infill beneat the lake.  
 

 
  
 
 
 
 

 
 
 
Spatial Relationships Between New and Old Sinkholes in Covered Karst, Albany, Georgia, U.S.A. 

In Dougherty County, Georgia, thousands of cover subsidence and cover collapse sinkholes have formed by the piping of overburden into cavities in the Ocala Limestone. Although sinkholes may form slowly, sudden collapse is often triggered by heavy rainfall and/or flooding that saturates and weakens soil arches that overly cavities. A spectacular example of this occurred in July 1994 when flooding from Tropical Storm Alberto (TSA) triggered the collapse of >300 sinkholes (Hyatt and Jacobs, Geomorphology, 17:305-316). Here we examine relationships between the July, 1994 (new) sinkholes and pre-existing (old) sinkholes within 71 km2 of suburban Albany. Our purposes are to contrast the dimensions, spatial distributions, and overburden characteristics for these two populations, and to test whether elevation, and the locations of old sinkholes have influenced the locations of new sinkholes. 

The dimensions (length, width, depth, perimeter) and locations of 311 new sinkholes were determined using survey equipment, a global positioning system, and reports from local residents. Similar measures for 329 old sinkholes were derived from digitized USGS 1:24000 map sheets. New sinkholes cluster within TSA flood limits, and are found preferentially in sandy alluvial lowlands as identified by digital elevation modeling. Old sinkholes are significantly larger, more pan-like in cross-sectional form, and group dimensionally into three statistically significant elevation classes. These classes are more widespread topographically than are new sinkholes, probably reflecting the influence of frequent flooding in lowlands, and enhanced piping in uplands. Clustering of new sinkholes within old sinkholes does occur for a few selected sites. Nearest neighbor analysis, however, indicates that statistically new sinkholes have not clustered near old sinkholes. This implies that the locations of old sinkholes by themselves have limited predictive utility to identifying sites for new sinkholes.  

Reference: Hyatt, J. A., Wilkes, H. P., and Jacobs, P. M. In review for inclusion in the Proceedings of the Seventh Multidisciplinary Conference on Sinkholes and the Engineering and Environmental Impacts of Karst, Harrisburg, Pennsylvania.