Estimation and impact of surface subsidence in a karst terrain

The lateral and vertical extent of surface subsidence can contribute to the failure of surface structures. One method of empirical subsidence prediction which may have application in karst terrain is the use of profile functions. Application of this method involves fitting a mathematical function to a significant number of actual subsidence basin profiles. The resulting empirical constants describe the shape of the basin.

Ten dolines and one collapse on Chestnut Ridge, near the southeast corner of the Department of Energy's Oak Ridge, Tennessee Reservation, were surveyed. Two profile functions; the hyperbolic tangent function and the negative exponential function were fit to the field data. The negative exponential function is recommended. It provided a better fit to thirty-four of thirty-nine profiles.

The lateral extent of sinkhole subsidence is linearly related to the depth of overburden, while the magnitude of vertical displacement cannot be predicted from field data. Subsidence prediction requires an adequate assumption of subsurface geometry and behavior. Numerical analysis using the finite element method provided a valuable tool to identify significant relationships. Observed values of subsidence appear much greater than that required to cause severe damage to surface structures.