Use of sinkhole and specific capacity distributions to assess vertical gradients in a karst aquifer

The carbonate-rock aquifer in the Great Valley, West Virginia, USA, was evaluated using a database of 687 sinkholes and 350 specific capacity tests to assess structural, lithologic, and topographic influences on the groundwater flow system. The enhanced permeability of the aquifer is characterized in part by the many sinkholes, springs, and solutionally enlarged fractures throughout the valley. Yet, vertical components of subsurface flow in this highly heterogeneous aquifer are currently not well understood. To address this problem, this study examines the apparent relation between geologic features of the aquifer and two spatial indices of enhanced permeability attributed to aquifer karstification: (1) the distribution of sinkholes and (2) the occurrence of wells with relatively high specific capacity. Statistical results indicate that sinkholes (funnel and collapse) occur primarily along cleavage and bedding planes parallel to subparallel to strike where lateral or downward vertical gradients are highest. Conversely, high specific capacity values are common along prominent joints perpendicular or oblique to strike. The similarity of the latter distribution to that of springs suggests these fractures are areas of upward-convergent flow. These differences between sinkhole and high specific capacity distributions suggest vertical flow components are primarily controlled by the orientation of geologic structure and associated subsurface fracturing.     

Quantitative sinkhole hazard assessment. A case study from the Ebro Valley evaporite alluvial karst (NE Spain)

Quantitative sinkhole hazard assessments in karst areas allow calculation of the potential sinkhole risk and the performance of cost-benefit analyses. These estimations are of practical interest for planning, engineering, and insurance purposes. The sinkhole hazard assessments should include two components: the probability of occurrence of sinkholes (sinkholes/km² year) and the severity of the sinkholes, which mainly refers to the subsidence mechanisms (progressive passive bending or catastrophic collapse) and the size of the sinkholes at the time of formation; a critical engineering design parameter. This requires the compilation of an exhaustive database on recent sinkholes, including information on the: (1) location, (2) chronology (precise date or age range), (3) size, and (4) subsidence mechanisms and rate. This work presents a hazard assessment from an alluvial evaporite karst area (0.81 km²) located in the periphery of the city of Zaragoza (Ebro River valley, NE Spain). Five sinkholes and four locations with features attributable to karstic subsidence where identified in an initial investigation phase providing a preliminary probability of occurrence of 0.14 sinkholes/km² year (11.34% in annual probability). A trenching program conducted in a subsequent investigation phase allowed us to rule out the four probable sinkholes, reducing the probability of occurrence to 0.079 sinkholes/km² year (6.4% in annual probability). The information on the severity indicates that collapse sinkholes 10–15 m in diameter may occur in the area. A detailed study of the deposits and deformational structures exposed by trenching in one of the sinkholes allowed us to infer a modern collapse sinkhole approximately 12 m in diameter and with a vertical throw of 8 m. This collapse structure is superimposed on a subsidence sinkhole around 80 m across that records at least 1.7 m of synsedimentary subsidence. Trenching, in combination with dating techniques, is proposed as a useful methodology to elucidate the origin of depressions with uncertain diagnosis and to gather practical information with predictive utility about particular sinkholes in alluvial karst settings: precise location, subsidence mechanisms and magnitude, and timing and rate of the subsidence episodes.     

Flooding of Sinking Creek,Garretts Spring karst drainage basin,Jessamine and Woodford counties,Kentucky, USA

Tashamingo Subdivision in Sinking Creek karst valley, a tributary of the Garretts Spring drainage basin in Jessamine and Woodford counties, Kentucky, was flooded in February 1989. To determine the cause of flooding, the groundwater basin boundary was mapped, discharge data were measured to determine intake capacity of swallets, and hydrologic modeling of the basin was conducted. Swallet capacity was determined to be limited by the hydraulic parameters of the conduit, rather than by obstruction by trash. Flooding from a precipitation event is more likely, and will be higher, when antecedent soil moisture conditions in the watershed are near saturation. Hydrologic modeling shows that suburban development of 20 percent of the southeast basin will cause a small increase in flood stage at Tashamingo Subdivision.     

Hydrogeologic role in sinkhole development in the desert of Kuwait

Sinkhole development is unlikely in desert areas with very low precipitation. However, a few cases of land subsidence and sinkhole development took place within the suburbs of Kuwait City. A few sinkholes developed in a sudden and rapid way, leading to great economic losses. In this paper the mechanism and causes of such a land subsidence are described. Decline in groundwater level and downward infiltration of excess irrigation are suggested to be the main factors in the development of the land subsidence in Kuwait. Urbanization and excessive garden irrigation are most probably the triggers of the sudden and rapid land subsidence.     

Application of matrix analysis in delineating sinkhole risk areas along highway (I-70 near Frederick,Maryland)

Sinkhole collapse in the area of Maryland Interstate 70 (I-70) and nearby roadways south of Frederick, Maryland, has been posing a threat to the safety of the highway operation as well as other structures. The occurrence of sinkholes is associated with intensive land development. However, the geological conditions that have been developing over the past 200 million years in the Frederick Valley control the locations of the sinkholes. Within an area of approximately 8 km², 138 sinkholes are recorded and their spatial distribution is irregular, but clustered. The clustering indicates the existence of an interaction between the sinkholes. The point pattern of sinkholes is considered to be a sample of a Gibbsian point process from which the hard-core Strauss Model is developed. The radius of influence is calculated for the recorded sinkholes which are most likely to occur within 30 m of an existing sinkhole. The stochastic analysis of the existing sinkholes is biased toward the areas with intensive land use. This bias is adjusted by considering (1) topography, (2) proximity to topographic depressions, (3) interpreted rock formation, (4) soil type, (5) geophysical anomalies, (6) proximity to geologic structures, and (7) thickness of overburden. Based on the properties of each factor, a scoring system is developed and the average relative risk score for individual 30-m segments of the study area is calculated. The areas designated by higher risk levels would have greater risk of a sinkhole collapse than the areas designated by lower risk levels. This risk assessment approach can be updated as more information becomes available.     

Subbottom acoustic and sedimentary records of past surface water–groundwater exchange through sinkhole lakes in south Georgia,U.S.A

Lake Balboa is a deep sinkhole lake in south Georgia. Subbottom acoustic profiling and long sediment cores reveal four stratigraphic units within >20 m of lacustrine fill above acoustically stratified and faulted Miocene/Pliocene overburden and basement Eocene limestone which hosts the Floridan Aquifer system (FAS). Fill consists of thin, rhythmically bedded, peaty sediments (Unit I), indicative of slow sedimentation in a shallow swamp until 13,110–12,680 cal BP, a mix of desiccated and over-consolidated clay and silty sand (Unit II) implying periodic subaerial exposure of the lake bed until 9470–9025 cal BP, and gyttja associated with a rapid 8 m rise in water level by 9120–9020 cal BP (Unit III), as well as an additional rise of 10 m to modern lake levels (Unit IV). Accumulation of nearly 4.5 m of low permeability gyttja, and higher lake levels have fundamentally changed exchange between surface water and groundwater, reducing by 40% or more the rate of vertical leakage to the FAS and increasing shallow aquifer discharge to the lake.     

Geology and evolution of lakes in north-central Florida

 Fluid exchange between surficial waters and groundwater in karst environments, and the processes that control exchange, are of critical concern to water management districts and planners. High-resolution seismic data were collected from 30 lakes of north-central Florida. In each case study, lake structure and geomorphology were controlled by solution and/or mechanical processes. Processes that control lake development are twofold: (1) karstification or dissolution of the underlying limestone, and (2) the collapse, subsidence, or slumping of overburden to form sinkholes. Initial lake formation is directly related to the karst topography of the underlying host limestone. Case studies have shown that lakes can be divided by geomorphic types into progressive developmental phases: (1) active subsidence or collapse phase (young); (2) transitional phase (middle age); (3) baselevel phase (mature); and (4) polje (drowned prairie) – broad flat-bottom that have one or all phases of sinkhole. Using these criteria, Florida lakes can be classified by size, fill, subsurface features, and geomorphology. Received: 28 July 1998 · Accepted: 9 September 1998     

Sinkhole hazard assessment in Minnesota using a decision tree model

An understanding of what influences sinkhole formation and the ability to accurately predict sinkhole hazards is critical to environmental management efforts in the karst lands of southeastern Minnesota. Based on the distribution of distances to the nearest sinkhole, sinkhole density, bedrock geology and depth to bedrock in southeastern Minnesota and northwestern Iowa, a decision tree model has been developed to construct maps of sinkhole probability in Minnesota. The decision tree model was converted as cartographic models and implemented in ArcGIS to create a preliminary sinkhole probability map in Goodhue, Wabasha, Olmsted, Fillmore, and Mower Counties. This model quantifies bedrock geology, depth to bedrock, sinkhole density, and neighborhood effects in southeastern Minnesota but excludes potential controlling factors such as structural control, topographic settings, human activities and land-use. The sinkhole probability map needs to be verified and updated as more sinkholes are mapped and more information about sinkhole formation is obtained.     

A genetic classification of sinkholes illustrated from evaporite paleokarst exposures in Spain

This contribution analyses the processes involved in the generation of sinkholes from the study of paleokarst features exposed in four Spanish Tertiary basins. Bedrock strata are subhorizontal evaporites, and in three of the basins they include halite and glauberite in the subsurface. Our studies suggest that formation of dolines in these areas results from a wider range of subsidence processes than those included in the most recently published sinkhole classifications; a new genetic classification of sinkholes applicable to both carbonate and evaporite karst areas is thus proposed. With the exception of solution dolines, it defines the main sinkhole types by use of two terms that refer to the material affected by downward gravitational movements (cover, bedrock or caprock) and the main type of process involved (collapse, suffosion or sagging). Sinkholes that result from the combination of several subsidence processes and affect more than one type of material are described by combinations of the different terms with the dominant material or process followed by the secondary one (e.g. bedrock sagging and collapse sinkhole). The mechanism of collapse includes any brittle gravitational deformation of cover and bedrock material, such as upward stoping of cavities by roof failure, development of well-defined failure planes and rock brecciation. Suffosion is the downward migration of cover deposits through dissolutional conduits accompanied with ductile settling. Sagging is the ductile flexure of sediments caused by differential corrosional lowering of the rockhead or interstratal karstification of the soluble bedrock. The paleokarsts we analysed suggest that the sagging mechanism (not included in previous genetic classifications) plays an important role in the generation of sinkholes in evaporites. Moreover, collapse processes are more significant in extent and rate in areas underlain by evaporites than in carbonate karst, primarily due to the greater solubility of the evaporites and the lower mechanical strength and ductile rheology of gypsum and salt rocks.     

Natural and human-induced dissolution and subsidence processes in the salt outcrop of the Cardona Diapir (NE Spain)

The Cardona Diapir in NE Spain, with a salt outcrop about 0.9 km² in area, has a well-developed endokarstic system that used to discharge into the Cardener River. Underground mining for potassium salt carried out from 1930 to 1990 caused significant changes in the topography and hydrology of the diapir. The accumulation of two halite slag heaps, totalling around 10 million tons, locally dammed the surface drainage, creating closed depressions and preferential zones of water recharge. The waters that infiltrated in one of these depressions, largely derived from uncontrolled sewage disposal, led to the generation of a 335-m-long human-induced cave excavated in one of the slag heaps. Moreover, the inflow of freshwater from the surrounding sandstone aquifer, caused by the excavation of a ventilation gallery, resulted in the generation of a 280-m-long cave. In March 1998, the interception of a phreatic conduit by a halite mine gallery 50 m deep caused dramatic changes in the hydrology and geomorphology of the diapir, including: (a) a sudden decline in the piezometric level of the karstic aquifer; (b) the inflow of freshwater and debris from the Cardener River into the endokarstic system and the mine galleries. A tunnel had to be constructed to divert the river flow from the salt outcrop; (c) massive dissolution of salt, creating new cavities and enlarging the pre-existing ones, including both mine galleries or cave passages. The 4,300-m-long Salt Meanders Cave was largely generated by the inrush of water from the Cardener River into the mine galleries; and (d) the generation of a large number of sinkholes in the vicinity of the Cardener River. An inventory of 178 sinkholes has allowed us to estimate minimum probability values of 4.7 and 8 sinkholes/km²·year for time intervals previous and subsequent to the 1998 mine flood event, respectively.