Geology for Global Development

Heather Britton: Sinkhole Occurrence and Mitigation

Sinkholes are often overlooked geohazards which, although far less destructive in the short-term than earthquakes and landslides, can be catastrophic to life and severely impact the built environment. This post will explore how these features form and the strategies that have been adopted to predict their appearance. It will also consider how urbanisation in karstic areas is accelerating sinkhole formation and what can be done to mitigate these effects.

Sinkholes form most commonly in karstic terrain – topography shaped by the dissolution of soluble rocks such as limestone, gypsum and dolomite. This dissolution creates cavities beneath the ground surface, and the subsidence of the land into these cavities creates sinkholes. Karstic regions are generally associated with three varieties of sinkhole, displayed in the figures below.

  • Figure 1 – Rainfall and surface water percolate through joints in the limestone. Dissolved carbonate rock is carried away from the surface and a small depression gradually forms. On exposed carbonate surfaces, a depression may focus surface drainage, accelerating the dissolution process. Debris carried into the developing sinkhole may plug the outflow, ponding water and creating wetlands. (Source: USGS Water Science School)

    The first are dissolution sinkholes, which form when dissolution is concentrated in a particular area, often along pre-existing joints or fractures within the rock where groundwater flow preferentially occurs. This causes the land to sink in this area and the result is a sinkhole.

  • Cover-subsidence sinkholes are characterised by a thick overburden of granular sediment which gradually falls into an underlying cavity.
  • The final variety, cover-collapse sinkholes, are undoubtedly the most dangerous as they can develop over a period of hours. Cover-collapse sinkholes occur where cohesive material overlies a soluble bedrock. When dissolution occurs in the carbonate/evaporite the overlying cohesive substance will form an arch over the cavity, making it very vulnerable to collapse. In 2013 Jeff Bush disappeared into one such sinkhole as he slept in his home in Seffner, Florida, demonstrating just how catastrophic these geohazards can be.


Figure 2 – Cover-subsidence sinkholes, where a thick, overlying, sandy sediment layer fills an underlying cavity, usually produced through carbonate/evaporite dissolution. Source: USGS Water Science School

Figure 3 – Cover-collapse sinkhole formation. Due to the cohesive nature of clay and similar sediment, these sinkholes often do not form gradually and instead tend to appear very suddenly, creating the greatest risk to human life and property. This is usually as a result of an influx of water, causing the layers in the clay to slide over one another. Source: USGS Water Science School

Worryingly, the appearance of sinkholes seems to be on the increase. Urbanisation is accelerating the rate at which sinkholes form, as it is intrinsically linked with processes such as construction and mining. Groundwater pumping associated with construction work changes the natural drainage patterns of the land, leading to dissolution in regions where it has not been seen before. On top of this, increased agriculture to feed the growing population involves the drainage of organic soils, leading to the runoff of organic carbon and the production of highly acidic water sources (pH 3.4-4 recorded in some instances) which inevitably accelerates the dissolution of soluble bedrock. Groundwater pumping in particular was responsible for 80% of identified subsidences in the US where sinkholes are a problem across many states. And the issue is not limited to the US – karstic regions around the world are seeing an increase in the number of human induced sinkholes, for example those which effecting the Madrid-Barcelona High Speed railway. Human activities are undoubtedly impacting sinkhole formation, be it through mining, agriculture or construction.

Aerial view of a large collapse structure, the Tres Pueblos sink, along the Rio Camuy, which exits on the far side of the sinkhole. Solution of underlying rock removed the support and the roof of soil and thin bedrock collapsed into the void. This produces what is known as karst topography. (Courtesy United States Geological Survey)

So what is the best way of allowing development to occur without exacerbating the anthropogenic effect that urbanisation can have on sinkhole formation? One of the simplest and most frequently implemented solutions is to avoid building in regions that may develop sinkholes, or have been shown to be prone to sinkholes in the past. This planned development can be implemented on a small scale, but it is unrealistic to avoid the development of large karstic regions altogether. Carefully considering the drainage networks of new builds and infrastructural projects may help to minimise the effects of development on sinkhole evolution – but sometimes the development pressure is much greater than any impetus to properly consider the state of the land which is being built on.

Currently our best solution to urbanisation in karstic areas is to monitor sinkhole development and attempt to detect them as soon as they appear. Monitoring is possible through geomorphological mapping and the use of GIS and DEM (Digital Elevation Mapping), whilst detection can be achieved with relative ease using geophysical methods (e.g., seismic data and radar). Although not preventative, such techniques do allow sinkholes to be identified early in their development, therefore they can be either avoided or filled before any serious damage is attained.

Even knowing where developing sinkholes lie does not stop construction occurring over them. A short term solution to a developing sinkhole is to fill it, often with concrete, as was done in Zaragoza, Spain. Here the sunken region was waterproofed before being injected with concrete. Whilst it might be thought the initial waterproofing step would end the cycle of dissolution and sinkhole formation, this action was actually seen to increase karstic activity. Filling sinkholes with concrete is not a sustainable solution – it does not prevent sinkholes from continuing to subside (although it may prevent catastrophic collapse) and too much concrete would be required to fill all sinkholes which pose a threat to human property. Other materials are in greater abundance, however, for example rubbish. It is highly unlikely that we will grow short of this ‘resource’ and the planet is currently in need of more landfill sites. The primary risk of this technique is the contamination of groundwater and other water sources – even if rubbish was only used in regions where there was a low risk of water contamination, the danger that groundwater flow would change and begin to poison sources of drinking water would always exist, making this a risky strategy, particularly in regions undergoing heavy urban development or in tectonically active areas. Geologists are working towards solutions – This student paper looks into how sinkholes can be stabilised, removing the danger of collapse – but many sinkholes have unique features which set them apart from the rest, therefore one solution may not be appropriate in all instances.

As the number of sinkholes grows, it is becoming more and more important that we develop better ways of dealing with and preventing this geohazard. Currently our efforts are limited to planning around sinkholes and detecting and monitoring their evolution early, but with urbanisation spreading into karstic regions around the world further work must be done to reduce the risk that this hazard holds for communities. Research is ongoing, and hopefully in the near future will yield not only better detection and mitigation technologies, but more effective and sustainable methods of dealing with sinkholes once they form.

This guest post was contributed by a scientist, student or a professional in the Earth, planetary or space sciences. The EGU blogs welcome guest contributions, so if you've got a great idea for a post or fancy trying your hand at science communication, please contact the blog editor or the EGU Communications Officer to pitch your idea.