Identification, prediction and mitigation of sinkhole hazards in evaporite karst
areas
F. Gutiérrez (1), A.H. Cooper (2) and K.S. Johnson (3)
(1) Corresponding author: Edificio Geológicas; University of Zaragoza; C/. Pedro
Cerbuna, 12; 50009 Zaragoza; Spain; Phone: 34 976 761090; Fax: 34 976 761106; E-
mail: fgutier@unizar.es
(2) British Geological Survey; Keyworth; Nottingham; NG12 5GG; UK
(3) Oklahoma Geological Survey; Energy Center; 100 E. Boyd; Norman; Oklahoma
73019-0628; USA
Copyright the authors and their respective institutions, please respect copyright;
for private study use only. Paper published in Environmental Geology. 2008. Vol 53.
1007-1022 and as DOI 10.1007/s00254-007-0728-4
Abstract Sinkholes usually have a higher probability of occurrence and a greater
genetic diversity in evaporite terrains than in carbonate karst areas. This is because
evaporites have a higher solubility, and commonly a lower mechanical strength.
Subsidence damage resulting from evaporite dissolution generates substantial losses
throughout the world, but the causes are only well-understood in a few areas. To deal
with these hazards, a phased approach is needed for sinkhole identification,
investigation, prediction, and mitigation. Identification techniques include field surveys,
and geomorphological mapping combined with accounts from local people and
historical sources. Detailed sinkhole maps can be constructed from sequential historical
maps, recent topographical maps and digital elevation models (DEMs) complemented
with building-damage surveying, remote sensing, and high-resolution geodetic surveys.
On a more detailed level, information from exposed paleosubsidence features
(paleokarst), speleological explorations, geophysical investigations, trenching, dating
techniques, and boreholes, may help to recognize dissolution and subsidence features.
Information on the hydrogeological pathways including caves, springs and swallow
holes, are particularly important especially when corroborated by tracer tests. These
diverse data sources make a valuable database - the karst inventory. From this dataset,
sinkhole susceptibility zonations (relative probability) may be produced based on the
spatial and temporal distribution of the features and good knowledge of the local
geology. Sinkhole distribution can be investigated by spatial distribution analysis
techniques including studies of preferential elongation, alignment and nearest neighbor
analysis. More objective susceptibility models may be obtained by analyzing the
statistical relationships between the known sinkholes and the conditioning factors, such
as weather conditions. Chronological information on sinkhole formation is required to
estimate the probability of occurrence of sinkholes (number of sinkholes/km² year).
Such spatial and temporal predictions, derived from limited records and based on the
assumption that past sinkhole activity may be extrapolated to the future, are non-
corroborated hypotheses. Validation methods allow us to assess the predictive capability
of the susceptibility maps and to transform them into probability maps. Avoiding the
most hazardous areas by preventive planning is the safest strategy for development in
sinkhole-prone areas. Corrective measures could be to reduce the dissolution activity
and subsidence processes, but these are difficult. A more practical solution for safe
development is to reduce the vulnerability of the structures by using subsidence-proof
designs.
Key words: sinkholes, evaporite karst, hazard assessment, mitigation
Introduction
The dissolution of soluble rocks and deposits at the surface, or in the subsurface
combined with internal erosion and deformational processes, can produce closed
depressions called sinkholes or dolines. These hollows characterize karst landscapes and
are usually sub-circular in plan varying in size up to hundreds of meters across, and
typically from a few meters to tens of meters deep (Williams 2003). The word doline,
derived from the Slavic word dolina, is a term mainly used by European
geomorphologists. The term sinkhole is most commonly used in the international
literature when dealing with engineering and environmental issues. The generation of
these karstic depressions is related to the dissolution of carbonate and evaporitic rocks.
Sinkholes in evaporite karst areas occur worldwide (Klimchouk et al. 1996), and pose
numerous practical problems, but compared with sinkholes in carbonate karst terrains
they have received relatively scarce attention. Evaporite karst sinkholes also commonly
show a greater genetic diversity (Gutierrez et al. 2008b). Because of the higher
solubility and lower mechanical strength of evaporites, their susceptibility to sinkhole
formation is greater than that of carbonate karst terrains. The solubilities of gypsum
(CaSO
4
2H
2
O) and halite (NaCl) in distilled water are 2.4 and 360 gr/l, respectively. By
comparison, the solubilities of calcite and dolomite minerals in natural environments are
commonly lower than 0.5 gr/l, depending on the pH, which is largely controlled by the
CO
2
partial pressure (Ford and Williams 1989). Gypsum dissolution rates as high as 29
mm/year have been measured in unconfined hydrogeological conditions in western
Ukraine (Klimchouk and Aksem 2005). In addition, the evaporites tend to have a more
ductile rheology than carbonate rocks, and their commonly lower strength may be
reduced substantially on a human time scale by dissolution processes. Another
peculiarity of evaporite karst is that subjacent dissolution may cause ground subsidence
on a regional scale. When these subsidence phenomena operate over long time periods,
they produce gravitational morphostructures, which may be up to several hundred
kilometers in extent and hundreds of meters in structural relief. These include
depositional basins that may have geomorphic expression (Christiansen 1967; Johnson
1989; Hill 1996), large collapse depressions (Gutiérrez 1996), concordant synclinal
valleys (Gustavson 1986), monoclinal flexures (Anderson and Hinds 1997; Warren
1999; Cooper 2002; Kirkham et al. 2002), and grabens (Cater 1970; Doelling 2000;
Gutiérrez, 2004). Additionally, where large-scale synsedimentary subsidence affects
valley reaches, it may generate dissolution-induced basins more than 100 m deep and
several kilometers long filled with alluvial deposits (Gutiérrez 1996; Benito et al. 2000;
Guerrero et al.2007).
In evaporite karst areas, gravitational deformation of the ground during sinkhole
development may cause severe damage to buildings and other man-made structures
(Cooper and Waltham 1999; Gutiérrez and Cooper 2002), including roads (Benson and
Kaufman 2001), railways (Guerrero et al. 2004; Gutiérrez et al. 2007a), dams (Gutierrez
et al. 2002; Johnson 2008b), canals and ditches (Gutiérrez et al. 2007a); even nuclear
power stations like Neckarwestheim in Germany have been affected (Prof. H. Behmel,
pers. comm.) (Fig. 1). Subsidence phenomena caused by evaporite dissolution have a
substantial detrimental effect on development in numerous regions of the world (Cooper
and Calow, 1998; Gutiérrez et al. 2008a; Johnson, 2008a), and individual sinkhole
events may have a large financial impact. For example, in the Spanish cities of Oviedo
and Calatayud situated on cavernous gypsum, the direct economic losses caused by
single collapse events that affected buildings in 1998 and 2003 were estimated to be 18
and 4.8 million euros, respectively (M. Gutiérrez-Claverol pers. comm. and Gutiérrez et
al. 2004). Sinkholes may also cause the loss of human lives when they occur in a
catastrophic way. Thirty four people have been killed by sudden collapses in the
dolomite karst of the Far West Rand of South Africa (Bezuidenhout and Enslin 1970).
Several people have been swallowed and injured by sinkholes resulting from halite
dissolution on the Dead Sea coast of Israel (Frumkin and Raz 2001). Other sinkhole
problems are related to hydrogeology and hydraulic structures. Sinkholes can act as
water-inlets connected to high-transmissivity karstic aquifers and cave systems making
the impoundment of water in reservoirs difficult (Pearson 1999; Milanovic 2000;
Johnson 2008b). They can facilitate the rapid pollution of the groundwater (Paukstys
and Narbutas 1996), and in places might affect the safety of sensitive structures such as
the radioactive waste WIPP repository in New Mexico (Hill 2003). Moreover, these
topographic depressions are frequently prone to flooding either by the concentration of
surface runoff or by groundwater flooding when the water table rising above their
ground level. This paper presents a basic methodological review of the assessment and
mitigation of sinkhole hazards in evaporite karst areas, contrasting them with the
differences these phenomena show in carbonate karst terrains.
Processes, factors and the impact of human activity
Several relatively similar genetic classifications of sinkholes have been recently
published (Williams 2003; Beck 2004; Waltham et al. 2005). However, the study of
paleokarst reveals that the development of sinkholes in evaporite karst terrains involves
a wider range of processes than those used by the aforementioned classifications.
Guerrero et al. (2008b) proposed a new genetic classification of sinkholes applicable to
evaporite karst areas. It has similarities to Beck’s (2004) sinkhole classification and the
most widely used landslide classifications, such as the one proposed by Cruden and
Varnes (1996). With the exception of solution dolines, the scheme describes the
sinkholes with compound terms: the first descriptor refers to the material affected by
internal erosion and deformational processes (cover, bedrock or caprock), and the
second indicates the main type of process involved (collapse, suffosion or bending)
(Fig. 2). In practice, more than one material type and several processes can be involved
in the generation of many sinkholes. These complex sinkholes, classified as polygenetic
by Williams (2003) and Beck (2004), could be described using combinations of the
proposed terms with the dominant material or process followed by the secondary one
(e.g. cover and bedrock collapse sinkhole, cover suffosion and bending sinkhole).
Two main genetic groups of sinkholes may be recognized: the solution sinkholes,
generated by the differential dissolutional lowering of the ground in areas where the
evaporites are exposed at the surface (bare or uncovered karst), and the different types
of sinkholes resulting from both, subsurface dissolution and downward gravitational
movement (internal erosion and deformation) of the overlying material. Solution
sinkholes are generally shallow depressions that may reach up to several hundred meters
across. The second group is obviously the most important from a ground stability and
engineering perspective. The sinkholes generated over dissolutional voids by the
upward propagation (stoping) caused by collapse of the cavity roofs are designated as
bedrock collapse or caprock collapse sinkholes, depending on whether the cavity
migrates through karst or non-karst lithologies, respectively (Fig. 2). The formation of
these sinkholes may be related to deep-seated dissolutional voids involving the
generation of breccia pipes that may reach several hundred meters in height (Johnson
1989; Ford 1997; Yarou and Cooper 1997; Warren 1999). These sinkholes commonly
show a low probability of occurrence (Beck, 2004; Waltham et al. 2005), and are
generally sharp-edged depressions up to a few tens of meters in diameter. The sinkholes
generated by the progressive interstratal dissolution of the evaporitic bedrock and the
concurrent gradual bending of the overlying evaporitic or non-karstic bedrock may be
termed bedrock bending or caprock bending sinkholes, respectively (Fig. 2). This type
of subsidence, which is particularly frequent in sequences with halite beds, may result in
depressions and troughs several kilometers in length (Kirkham et al. 2002).
Three main end members can be differentiated in areas where the evaporitic bedrock is
mantled by a cover of allogenic sediments or residual soils (Fig. 2): (1) Cover bending
sinkholes are caused by the differential lowering of the rockhead, which may lead to the
gradual bending of the overlying mantle. These are commonly shallow depressions that
may reach several hundred meters in length. In this case, a thick karstic residue may
form between the cover and the “unweathered” evaporitic bedrock. (2) Cover suffosion
sinkholes result from the downward migration of the cover through dissolutional voids
(raveling) and its ductile sagging. A wide range of processes may be involved in the
downward transport of the particles, including down-washing and viscous or
