Sinkholes

Of the geological hazards signified by ground subsidence, sinkholes are among the most difficult to identify with remote sensing because of their range of manifestations in size, depth, and timescale of development. The progression of sinkhole-related land subsidence ranges from very slow to very rapid, with some cover collapse occurring in minutes to seconds. On the small end, sinkholes can be less than a meter in diameter and difficult to resolve spatially through remote sensing. In general, sinkholes have several signatures observable with remote sensing, the most notable being elevation difference and subsidence relative to the surrounding area, but including localized or persistent standing water, drainage of surface water without an obvious surface flow pathway, and anomalous vegetation type or condition. Remote sensing methods have been demonstrated for both mapping sinkholes and monitoring sinkhole hazards, including movement precursory to catastrophic collapse. Sensors or methods capable of measuring the abovementioned observables that have been demonstrated for sinkhole hazards include lidar, radar altimetry (single-pass interferometry), synthetic aperture radar differential interferometry (InSAR), hyperspectral and infrared imaging, and a range of geotechnical methods, often combined with modeling of water flow paths. In this chapter, the various methods and their capabilities and limitations are discussed. Additional considerations required for sinkhole hazard assessment or early warning systems are presented.