Hypogene Karst Regions and Caves of the World

The UK has a wide variety of hypogenic caves, including transverse maze caves, relict hydrothermal caves, gypsum mazes and hypogenic karst associated with water rising up deep thrust faults. While few of these are particularly extensive, they offer insights into the mechanisms of speleogenesis and mineralisation. The best developed hypogenic caves in the UK are in the Carboniferous limestones of northern England where at least nine network maze caves with plan lengths exceeding 1 km are known, almost all of which are accessible only via disused mine workings. In South Wales, relict hypogenic cave networks have been documented from cave systems surrounding the South Wales Coalfield. Hydrothermal cave systems are also known in the Bristol region in southern England, the Derbyshire Peak District and North Wales where they are associated with Pb–Zn mineralisation. In all three of these areas, active deep phreatic groundwater circulation is ongoing. Elsewhere in the UK, transverse artesian groundwater flow through Permo-Triassic gypsum in the Ripon area has led to the development of hypogenic cave systems and numerous sinkholes. This chapter documents for the first time all the known hypogenic cave systems found in the UK, discusses their modes of formation and outlines the potential for future discoveries.

Hypogene speleogenesis (HS) describes cave development by rising fluids independent of recharge from the overlying or adjacent areas. Dissolution includes deep-seated acidic sources (CO₂/H₂S), “hydrothermal” cooling, mixing corrosion, and Sulfuric Acid Speleogenesis (SAS) at and above the water table. HS localizes in basin at sites of upward flows and convergences, in deformed strata at structural highs and major faults. In disrupted basins, geothermal gradient “pumps” downward meteoric water. Volcanism and magmatism produce “hyperkarst” combining CO₂, H₂S, thermalism, and microbial activity. Hypogene caves in France are presented according to a conceptual model. Isolated geodes are lined by calcite spar. 3D multistory caves (Pigette, Adaouste) show CO₂ degassing morphologies (bubble trails, folia, tower cones, coral towers) and “hydrothermal” minerals (Fe–Mn oxi-hydroxides, barite, celadonite), with condensation–corrosion cupolas above the water Table. 2D maze caves develop below less permeable strata. Giant ascending shafts (Salins shaft), collapse shafts, and breccia pipes (called “organs” in the N.-French coal basin) correspond to “hyperkarst” phenomena. Caves with ore deposits are associated with sulfurs (Fe, Pb, Zn) or oxidized Fe. They often record past positions of base level and harbor unique features (black tubes, ghost-rock) or rare minerals. Along the water table, thermal gradient and CO₂/H₂S degassing make condensation–corrosion and HS above the water table. Chevalley Aven shows upward dendritic pattern. Eventually, cupolas expand to large isolated chamber. Water table sulfuric caves develop headward with typical morphologies (corrosion tables, condensation domes, sulfuric karrens). Vapor shafts form by warm air convections and condensation–corrosion.

Signs of hypogenic speleogenesis have been detected in a number of caves of the Murcia Region (SE Spain), in some cases revealing active speleogenetic mechanisms rarely observed in hypogene cavities elsewhere in the world. Here, we investigate the hypogenic morphologies and speleothems of four caves in this region, namely Sima de la Higuera, Sima Destapada, Cueva del Agua and Cueva del Puerto. Also, other ten caves showing evidence for hypogenic speleogenesis has been preliminary described. Processes related to ancient and current hydrothermal activity, the discordance of permeability structures in the adjacent beds and the spatial arrangement of the regional hydrogeology have given rise to maze patterns and typical subaqueous hypogenic morphologies. These include spongework mazes, rising wall channels and shafts, feeders, bubble trails, solution pockets, megascallops and rising of chains cupolas, among others. Carbonic acid speleogenesis is responsible for the formation of most of these cave features; however, evidence of sulfuric acid speleogenesis (SAS) has been observed in Cueva del Puerto and Sima del Pulpo, which host massive secondary gypsum deposits. Speleothems typically linked to hydrothermal water upwelling and CO₂ degassing close to the water table are present in most of these cavities, including folia, calcite spar crystals, cave clouds, calcite rafts deposits and several types of cave raft cones. The wide variety of hypogenic speleogenesis indicators and speleothems whose genesis is unconnected to meteoric water seepage reveals that the hydrothermal field of the Murcia Region hosts one of the densest active hypogenic subterranean networks in the world.

The southern and eastern coasts of Mallorca Island (western Mediterranean) host abundant karst caves, some of them containing solutional features and deposits presumably related to hypogene basal recharge. The caves were formed in calcarenites whose ages range from Upper Miocene (reef deposits) to Middle Pleistocene (eolianites), which form a fringing postorogenic belt deposited over Mesozoic folded and thrusted carbonate deposits. The hydrogeological setting corresponds to an unconfined coastal aquifer in very porous eogenetic rocks, but showing important lateral and vertical permeability variations related to different sedimentary facies. Six caves containing hypogene features are distributed in three main coastal areas: the Llucmajor Upper Miocene platform, the Campos Plio-Pleistocene basin, and the Portocristo Upper Miocene littoral fringe. The first two areas are spatially coincident with low-grade geothermal anomalies reported in southern Mallorca, which are associated to important SW–NE faults. The observed cave features include a suite of solutional rising forms embracing, among others, subvertical feeder-like conduits and small ascending wall channels of variegated morphologies and dimensions. Sediments and black crusts enriched in Fe and Mn oxides are frequent along with some uncommon minerals, documenting a deep recharge rising into the unconfined littoral groundwater. Regarding their speleogenesis, the studied sites must be considered as complex littoral caves in which the imprints of hypogene processes are evident, but intermingled with other morphogenetic vectors like coastal mixing processes and a substantial meteoric recharge. The lithofacies variability of the carbonate rocks introduces additional complexity, affecting the hydrological behavior of the coastal aquifers as well as the pattern and morphology of the resulting caves.

Despite its small size Austria is home to about 16,000 caves. The vast majority is epigene in origin and occurs in uplifted karst plateaus. Only about hundred caves of hypogene origin are known, whose features and modes of origin are reviewed in this article. Most currently known hypogene caves cluster along the rim of the southern Vienna Basin. Some of these caves, including Eisensteinhöhle, are characterized by a slight thermal anomaly associated with warm springs. A second province with hypogene karst features is also structurally controlled and occurs south of the Salzach fault in the Central Alps, where caves show morphologies, mineralization and geochemical patterns diagnostic of hypogene origin. In the same area lukewarm springs emerge at lower elevations. A third region where hypogene processes resulted in cave development is the area of the Periadriatic Lineament in Carinthia. Caves such as Kozakhöhle show Laughöhle morphologies and are perched above the valleys, but the presence of CO₂-rich springs suggests that speleogenesis may still be ongoing in the subsurface. Only very few hypogene caves are currently known from the Northern Calcareous Alps (NCA). Most of those seem again to be related to upwelling of aggressive waters enriched in CO₂, while only one cave in this part of the Eastern Alps shows evidence of sulfuric acid speleogenesis (Kraushöhle). Radiometric dates constrain the final stage of the process to between about 160 and 86 ka before present, while the presence of a lukewarm H₂S-bearing spring in a gorge below the cave provides evidence of minor water–rock interactions involving dissolved sulfur species still operating today at this site.

Hypogene speleogenesis is important for karstification in the Apennines because of widespread, aggressive groundwaters containing endogenic H₂S and CO₂. Evidence for karst processes can be found throughout the region, from solutional limestone caves to deposition of carbonate travertine. The caves are up to few tens of kilometers long and show different patterns, ranging from phreatic to almost ideal water table caves. The caves often have a maze pattern and contain large rooms, cupola roofs, blind pits, and anastomotic zones. Solutional processes prevailed in the phreatic zone, also in deep-seated conditions, but subaerial corrosion is developed as well and can play an important morphogenetic role. The main cave-forming processes are related to the H₂S oxidation to sulfuric acid, a process that involves chemotrophic microbial activity, and are responsible for the diffuse deposition of replacement gypsum in subaerial conditions. The great variety of geological settings makes the Apennines a good place to study hypogene karst, and the existence of active branches in several caves also permits a direct study of the cave-forming processes and the comparison with the resultant morphologies.

The Frasassi caves present a unique opportunity to study sulfuric acid speleogenesis in a large karst system that contains active sulfidic processes as well as relict features produced by past speleogenetic episodes. The caves consist of a network of ramifying, mainly subhorizontal passages that reach more than 30 km in total length, and are organized in superimposed and interconnected levels. Sulfidic groundwaters are accessible in the lowermost levels. The water chemistry in the shallow phreatic zone is influenced by mixing phenomena between the sulfidic groundwaters and descending oxygenated water and, to some extent, gas exchange with the cave atmosphere. The sulfidic waters are corrosive with respect to limestone, with average limestone dissolution rates around 100 mm 10−3 year−1 for submerged tablets. The release of gases to the air also causes wall corrosion (linear corrosion rate up to 85 mm 10−3 year−1), which produces gypsum replacement crusts above the water table. Morphological and isotopic analyses show that large relict gypsum deposits in the upper dry levels were created by the same mechanism. Sulfide-oxidizing chemosynthetic bacteria living in the cave produce organic matter and support a rich and diverse ecosystem in the sulfidic branches, and these microorganisms contribute to sulfur cycling and cave formation processes. The evolution of the older cave levels during the Pleistocene was a complex result of changing regional geomorphological and hydrogeological regimes as well as local factors that affect water movement and gas exchange in the shallow phreatic zone within the cave.

Santa Cesarea Terme in Salento is the only area in which hypogenic caves have been recognized in the Apulia region. In this spa area, the rising of sulfidic thermal waters that mix with both recent fresh infiltration waters and coastal salt water has formed four active sulfuric acid speleogenesis (SAS) caves. These caves are characterized by the typical set of sulfuric acid meso- and micromorphologies, and also by the presence of both gypsum and native sulfur. In all caves, biofilms are visible in the sulfidic thermal waters and on the cave walls.

The karst area of Acquasanta Terme is located in the eastern side of central Apennine. The thermal sulfidic springs reach the surface in the core of the anticline where the river valley reaches the top of a Meso-Cenozoic limestone succession, capped by less permeable rocks. An important terraced travertine deposit was produced by the thermal water throughout the Pleistocene. The most interesting caves are in the valley of the Rio Garrafo, a tributary of the Tronto River. These caves developed mainly below the valley bottom, and a part of the stream water sinks through preexisting karst passages mixing with the thermal groundwater ~50 m below. Water temperature in the caves reaches 42 °C, and the water has high salinity (up to > 6 g L−1) with elevated levels of chloride, sulfates and carbonates. Sinking water from the caves and the travertine plate is responsible for the significant dilution of thermal water at the main spring, located 2 km downstream, in the Tronto Valley. The speleogenetic process is mainly related to H₂S oxidation, which is enhanced by the recharge of oxygenated fresh water from the surface. Cave development occurred in different phases. Early passages formed mainly in the phreatic zone, due to the rising thermal water. Deepening of the main surface stream (Tronto River) caused lowering of the water table. As a result, the Rio Garrafo caves were partly dried and invaded by sinking water, while rising gases activated condensation corrosion and the growth of replacement gypsum crusts.

In Sardinia, no active hypogenic caves have yet been discovered or described. Although there are a few thermal springs, mostly correlated to Quaternary volcanic activity, none of these thermal waters have interacted with carbonate rocks. Nevertheless, in the SW of the Island many metal ore deposits hosted in Cambrian limestones have been exploited over the last two centuries, allowing the discovery of so-called mine caves, some of which are clearly of hypogenic origin. These caves formed by thermal waters in a phreatic setting and are now located far above the water table and are no longer active, apart from some recent dripstone formation. The mine tunnels in Mount San Giovanni, near Iglesias and Gonnesa towns, have cut most of these caves: among them the well-known Santa Barbara cave, covered with barite crystals, Santa Barbara 2 cave, with its unique oxidation vents, and Crovassa Ricchi in Argento. Other hypogenic caves have been discovered in the mines of Campo Pisano and Monteponi (Iglesias), Mount Onixeddu (Gonnesa), and especially Masua (Iglesias). A very special case of hypogenic cave is the Corona ’e Sa Craba quartzite system, known for its barite crystals and rich in many mineral species. This chapter summarizes these known inactive hydrothermal and sulfuric acid caves.

Karst in Sicily develops in both Messinian gypsum and Mesozoic or Tertiary limestone rocks. Caves are also found in the basalts of Mount Etna. Except for some rare cases, until recently most caves developed in limestone were considered to be of epigenetic origin. The discovery of gypsum in some of these caves, and especially detailed morphological studies, have allowed defining a hypogenic origin for a dozen of caves up to now. In some of these, the hypogenic evidences are very clear, while others remain in doubt because of the widespread presence of well-developed condensation-corrosion morphologies not necessarily related to hydrothermal fluids. This paper reports the present knowledge of hypogenic caves in the Island of Sicily.

Piedmont (N-Italy) is one of Italy’s regions where Messinian gypsum outcrops most extensively. Some gypsum caves have been explored in the past, but no detailed speleogenetic studies have been carried out. The discovery of unexpected large underground voids in a gypsum quarry near Moncalvo has demonstrated the presence of intrastratal karst systems with an origin at least partly linked to rising groundwater flow. A detailed look at one of the gypsum caves near Alba (Monticello cave) has revealed this labyrinth to be another classical example of intrastratal gypsum karst, clearly formed by the rising of undersaturated groundwater through a confined Messinian gypsum bed. This chapter deals with the typical hypogene morphologies in these caves and also reports some new results on the geochemistry of the waters of the Moncalvo area.

Many hypogene caves have been recognized in several locations in Greece in the last decade. Their identification is mostly based on morphological criteria and in some cases on geochemical evidence, including mineral and water chemistry. Several patterns of cave development have been recorded along with various small-scale morphological features. In this chapter, hypogene caves are presented per region in terms of different geotectonic units. These caves cover the whole Greek mainland and some islands of the Aegean archipelago. Their altitudinal distribution ranges from the sea level up to about 2000 m elevation. Nowadays, most caves are uplifted relicts of hydrothermal systems. The dissolution was caused mainly due to CO₂-rich water, and in a few cases speleogenesis was driven by H₂S. Speleogenesis in most cases took place below the water table in a deep phreatic setting, by slowly convecting waters. However, there are water-table caves, commonly related to H₂S speleogenesis. It is demonstrated that hypogene speleogenesis in Greece is strongly related to the hydrothermal regime. Cave morphology, stratigraphy, hydrogeology, volcanism, tectonics, mineralogy and geochemistry are briefly discussed with respect to the structure of the Hellenic alpine orogene and its evolution.

Research on hypogene karst in Macedonia is quite a recent topic, with systematic study of some areas starting only few years ago, although certain caves with genetic characteristics attributable to hypogene speleogenesis have been described before. Hypogene karst is presently known from three areas: Mariovo, Katlanovo and Krčin. All of the registered examples are connected to hydrothermal karstification in pre-Cenozoic rocks, along the boundaries of Neogene basins formed within the South Balkan extensional system (northern part of the Aegean extensional regime). Deep circulating meteoric waters were heated due to the increased geothermal gradient, discharging mostly along faults at basin boundaries, with subsequent evolution of the hypogene karst systems governed by the lowering of the base level due to valley incision, thus controlling spring positions. Best studied is the hypogene karst in Mariovo, where hydrothermal karstification is locally combined with ghost-rock weathering (in dolomitic marble and dolomite, e.g., Karši Podot Cave and Allchar ore deposit), and sulfuric acid speleogenesis (e.g., Provalata Cave, likely connected to the nearby coal basin; or due to oxidation of metal sulfides, i.e., Allchar ore deposit/Kožuf thermal springs). At Katlanovo Spa, where hypogene karst is developed in marble, thermal springs are found at the bottom of the valley discharging water from deep faults, with thick travertine deposits found also at higher elevation. A similar situation is found in the southern parts of Krčin Mountain, with thermal springs discharging from both limestone (Banjište Spa) and evaporites (Kosovrasti Spa), and a number of caves with hot air circulation located above the Banjište Spa.

Few caves exist in Romania that have provided clear evidences of their hypogene origin; these were described mainly in connection with investigations conducted on ore deposits or thermal water resources used in spas. Genetically, they are grouped in three categories: (1) caves related to upwelling of H₂S-rich thermal waters along Cerna Valley (SW Romania) and around the town of Mangalia (SE Romania), (2) voids created in skarns and limestones during ascending flow of hot metasomatic and hydrothermal ore fluids (caves located mainly in the northern and western part of the country), and (3) mixing of salt and fresh water near the present Black Sea shore (e.g., Piatra Cave). Among all caves in these categories, Movile, a world-renowned biospeleological hot spot, the skarn-hosted caves from Băița, and Valea Rea, an underground mineralogical treasure, are the most representative.

Carbonate regions have great economic importance for water supply, oil and gas reservoirs, geothermal fluids and also Mississippi Valley-type ore deposits. Therefore, the understanding and consequences of flow pattern in carbonates require special interest. The hypogene and epigene karst areas of carbonate sequences were distinguished and associated with different orders of groundwater flow. However, the effect of confinement on flow pattern of carbonate aquifers was not fully considered in previous studies. We demonstrated the most important prerequisites and consequences of the application of the gravity-driven regional groundwater flow concept for carbonate sequences at different degrees of confinement. The results put into a frame the distribution of different springs and caves (epigene and hypogene) of the carbonate system of the Transdanubian Range, Hungary, and provide insights for better understanding of the hydrogeology of areas with similar unconfined and confined settings. Relationship among different flow regimes, distribution and character of springs and hypogene karstification processes, in addition to natural discharge-related phenomena, such as mineral and microbial precipitates, were recognized in the area of Buda Thermal Karst. This area is a natural laboratory where the connection between groundwater flow and karstification processes can be studied.

The Buda thermal karst is one of the most characteristic hypogene karst systems of the world developed in Triassic and Eocene limestone and marl in Budapest, on the right side of the Danube, under 300–400 m high hills. Due to the mixing corrosion enhanced by CO₂ and perhaps sulfuric acid, the water circulation formed multi-storey cave systems with different morphologies and complicated layout along the tectonic fissures. The arrangement of caves is independent of surface topography. The world-famous medicinal waters of Budapest come to the surface through cave passages. Due to the incision of the Danube and the uplift of the area in the Pleistocene, most passages are presently dry; now they lie 100 m above the Danube, though there are still passages below the water-table. The length of the five largest known caves and about 100 smaller ones reaches 55 km, and it is reasonable to estimate the length of unknown underground passages to be comparable. Caves are characterized by complex morphology, suggesting the formation by rising flow, and by rich carbonate-sulfuric mineral precipitations. Some of them (e.g. barite and calcite veins) are associated with bedrock formation and thus predate the cave forming process. There are minerals that precipitated from the warm water that dissolved the cave (Fe- and Mn-containing minerals formed by bacterial colonies, or cave rafts and folia). Evaporitic precipitations (e.g. popcorns, and frostworks) are remarkable. There are frequent precipitations of gypsum (chandeliers, gypsum flowers, etc.) due to the pyrite content of the marl above the passages. Discovering the concealed and hidden caves since the early twentieth-century was helped by quarries and earthworks followed by systematic and deliberate exploration.

During the Alpine orogeny of the Western Carpathians, hypogene caves have originated in different settings and epochs. Several caves of hydrothermal origin in crystalline limestones and metasomatic secondary quartzites (metasomatic silicites) are known in the central zone of the Štiavnica stratovolcano, Štiavnické vrchy Mountains. The early phases of speleogenesis in the crystalline limestone near Sklené Teplice Spa were caused by post-magmatic dissolution linked either to the emplacement of subvolcanic granodiorite intrusions during Late Badenian time or to the spatially associated Late Sarmatian epithermal system. Speleogenesis in metasomatic secondary quartzites in the Šobov quarry is related to deep-seated hydrothermal processes associated with a diorite intrusion in the northern part of the central zone of the Štiavnica stratovolcano during its pre-caldera evolution phase in Upper Badenian. Unusual caves in metasomatic magnesite, with mineralogical evidences of their hydrothermal origin, were investigated in the Revúcka vrchovina Mountains. Paleokarst geoda-like cavities and small hydrothermal caves occur in nappe structures of Triassic carbonates that were uplifted within the Tatra-Fatra belt of core mountains (many of these cavities are integrated into younger and larger non-hypogene caves). Cavities with thermal water in underlying and faulted Mesozoic carbonates (covered by Neogene or Paleogene sedimentary rocks) were identified by boreholes in some intermontane basins (Rimavská kotlina Basin, Liptovská kotlina Basin). A phreatic shaft formed by progradational collapse of non-carbonate beds disrupted by artesian karstification of underlying carbonates occurs at Tornaľa Town in the Rimavská kotlina Basin. Several caves with hypogene morphologies have originated along marginal faults of horst structures or fault edges of horst–graben structures (speleogenesis by thermal or slightly heated ascending waters, but mineralogical evidences of hydrothermal karstification were not found).

Concave and cavernous forms including rising wall channels, rising sets of coalesced copula, ceiling half-tube channels, separate ceiling copula, ceiling chimneys, and half-spherical upward-convex arches locally occur in surface outcrops of Carboniferous arkose sandstones in central and western Bohemia. Many of these negative forms conventionally described as tafoni and/or honeycombs have been traditionally interpreted as products of various exogenous weathering processes. Based on the line of indirect evidence, we propose an alternative interpretation in which these features represent transitional and outlet members of the morphologic suite of rising flow (MSRF), indicative of their subsurface hypogene origin. The negative forms are commonly associated with bedding planes and subvertical fractures mineralized with goethite and jarosite. The reflectance of coal particles embedded in sandstone along mineralized bedding planes (0.91–1.03% R _r ) is appreciably higher with respect to those of adjacent unaltered arkose host rocks (0.61–0.85% R _r ), pointing to the thermal overprint by hot fluids. Moreover, the walls of many cavities are covered by sandy-disintegrated alterite locally mineralized with gypsum, dickite, goethite, authigenic quartz, pickeringite, and bischofite. We suggest that these phenomena, including the origin of characteristic concave forms and mineralogical alterations of arkose host rocks, may have been due to warm, CO₂-saturated and possibly H₂S-rich brines that ascended from the deepest stratigraphic units of the Carboniferous succession via the network of subvertical tectonic fractures and migrated laterally outward along permeable bedding planes. As indicated by the apatite fission track analysis and wider geological observations, the alteration of arkose sandstones probably occurred at relatively shallow depth of burial, during the Tertiary uplift of the Bohemian Massif 15–20 Ma ago. In this environment, the alteration may have been accelerated by the effects of mixing corrosion where heated deep basinal fluids interacted with shallower interstratal waters. When the uplifted sandstone sequences eventually reached the surface, the hypogene cavities and altered cliff walls were subjected to subaerial weathering and fluvial erosion processes the effects of which were superimposed on older hypogene features.

Germany exhibits a very diverse geological history. Thus, a large number of stratigraphically, petrographically and tectonically different carbonate and sulfate rock deposits exist that have been subject to karstification. First, we discuss the possible “agents” (sensu Klimchouk) of hypogene karstification. Three principally different processes are identified: water rising because of buoyancy (either thermally or concentration induced), in situ oxidation of siderite or rising gases (CO₂, CH₄ or H₂S). Next, German caves and karst are discussed in order of stratigraphy. Applying the most pertinent epigene versus hypogene morphological characteristics, it becomes evident that hypogene caves occur in many different areas, often side by side with clearly epigene caves. For many areas, the agents of hypogene speleogenesis still remain unclear. This applies for most caves in the Paleozoic limestones of the Rhenish Schist Massif. The largest of these caves is the >700 × 200 × 20 m measuring cavity at the bottom of the Wülfrath quarry, filled with Lower Cretaceous sand. Only the Iberg/Harz caves seem to be a clear case; here, the worldwide highest concentrations of siderite weathering-induced caves occur. In the South Harz, large cavities were discovered in anhydrite below the water level by mining for the underlying upper Permian Copper Shale. Apparently these were formed by water rising from underlying carbonate aquifers. The large cavities discovered recently in the Blauhöhlen System and some of the deep pit-caves in Upper Jurassic limestone of the Swabian Alb may have their explanation in volcanic CO₂, having emanated from some of the 355 pipes of the Swabian volcanic field. Most striking is the high concentration of hypogene caves in the central Franconian Alb that occur in a small area while the surrounding areas are almost devoid of larger caves. In the geological past, this region formed the center of a continental basin where artesian water was able to ascend along major fractures. This could explain the strange geographic cave distribution as well as suggest that the ascending waters carried sulfide or methane from below.

Karstified rocks are widely distributed in Poland; most of them are covered by unconsolidated Cenozoic deposits. Paleozoic and Mesozoic carbonates and Neogene evaporites host numerous paleokarstic features and solution caves, being of polygenetic and multi-stage origin. Works on hydrothermal karst and sulfide ores emplaced in caves of the Silesian-Kraków region, the large paleokarst caves filled with Permian internal sediment or thermal convection model for caves in the Polish Jura date back to the 1970s. Quite well-documented hydrothermal karst is preserved as paleokarstic features in several localities in the Silesian-Kraków region, Holly Cross Mts. and Sudetes. Characteristic cave sediments including crystalline calcite spar, sulfide ores emplaced within hydrothermal cave sediments and solution-collapse breccias are associated with these features. Travertines, which are surface manifestation of hypogene karst, dated through their paleontological assemblage, support their post-Variscan—Permian and Late Triassic age. Their hydrothermal origin was also determined by detailed studies of sediments and fluid inclusions. Less certain, based mainly on studies of relict cave morphology and limited calcite spar, are the Cenozoic—Paleogene and Neogene hypogene events in the Carpathians (Tatra Mts.) and their foreland (Polish Jura). Review of published references and concepts related to hypogene origin of caves and karst in Poland is presented in this contribution.

The gypsum karst in the Western Ukraine occurs in the southwestern outskirt of the Eastern European Craton and is a model example of evolution of intrastratal karst and of artesian transverse speleogenesis in this lithology. Speleogenesis commenced in the regionally extensive confined Miocene aquifer system, in which the Middle Miocene (Badenian) gypsum bed is sandwiched between two “normal” aquifers with diffused permeability. Speleogenesis developed due to ascending leakage across the gypsum induced by valleys incising into the upper confining sequence. Large gypsum caves in the region (including the five longest gypsum caves in the world) are generally stratiform multi-story mazes of passages developed along vertical and steeply inclined fissures. Cave patterns in plan and the cross section are variable depending on fracture stratigraphy, patterns and connectivity, areal distribution of recharge (at the lower contact) and discharge (at the upper contact) points, and local hydrodynamic conditions and history. Patterns and meso-morphology of the caves clearly demonstrate functional organization that evolved in response to ascending hydraulic communication between the aquifers in the Miocene aquifer system, reflecting the complex pattern of transverse flow across the gypsum. This study demonstrates that artesian transverse speleogenesis is of fundamental importance to groundwater circulation in the region-wide Miocene aquifer system. It has important implications for regional and site-specific hydrogeology, the origin of large bioepigenetic sulfur deposits, and assessment of collapse/subsidence hazard. The style and principal characteristics of the large maze caves in the gypsum karst in the Western Ukraine are not specific to dissolution of sulfate rocks but exemplify general regularities of artesian transverse speleogenesis that are applicable to other lithologies including carbonates.

Zoloushka Cave, the third longest gypsum cave in the world (92 km) and the world’s largest gypsum cave by volume (>0.7 million cubic meter), is an outstanding example of a large maze cave formed in a multi-story artesian aquifer system due to the upward leakage between the aquifers across a soluble unit. Although local settings and evolution caused some peculiar features in speleogenesis, as compared to other large maze caves in Western Ukraine, these singularities provide further insights on variants and mechanisms of both regional and general models of hypogene artesian speleogenesis. The cave is located in the area where the host aquifer system is generally confined until present, but in one of the more uplifted tectonic blocks where the gypsum was partially incised by the nearby valley of the Prut River during the Holocene and rapidly drained due to the quarry operations since 1946. The cave experienced considerable modifications and transformations in the course of the recent transition to unconfined conditions and subsequent artificial dewatering. The geomorphological breaching of the confined aquifer system in the cave area occurred in the Holocene, i.e., later than in the area where most other large maze caves in the region are located. A relatively stable position of the water table in the upper part of the gypsum during most of the Holocene caused pronounced widening of passages, which determined the anomalously large size of passages and high indices of areal karstification in some regions of the cave. The most intense development of conduits occurred in areas where the input of freshwater from the lower aquifer was more abundant. Quarry operation and accompanying groundwater withdrawal during the last 70 years caused drastic and rapid transformations in the cave environment.

The chapter demonstrates, based on multiple lines of evidence, that hypogene speleogenesis was responsible for the origin of caves and conspicuous landform features in the Crimean Piedmont, a part of the Crimean Mountains located in the south of Crimea, a large peninsula in the North Black Sea. The region is located at the margin of the Prichernomorsky artesian basin, along a geodynamically active regional suture zone which separates the fold-thrust structure of the Crimea Mountains from the Scythian plate. Speleogenesis occurred in confined conditions by flow that rose across the Upper Cretaceous–Paleogene–Neogene clay–carbonate succession along cross-formational tectonic discontinuities and interacted with lateral flow in stratabound aquifers. A conceptual model of hypogene speleogenesis in the region links different types of cavities occurring in different lithostratigraphic units into integral, complexly structured, but functionally united void-conduit systems. These systems are shown to play an important role in the formation of cuesta escarpments in the Inner Range of the Crimean Mountains. The Crimean Piedmont is an outstanding example of a region where the recognition of hypogene speleogenesis entails a fundamentally new interpretation of the regional geomorphological evolution and the origin of remarkable landscape features, previously attributed to weathering processes.

This chapter describes the Ordinskaya Cave in the Fore-Urals region, Russia, which is the largest underwater cave of sulfate rocks in the world. The explored length of the cave is about 4900 m. The regional distribution of karst features indicated that a large amount of recharge entered the lower passages during all stages of development. The groundwater in the cave is aggressive with respect to sulfate. Discharge of water with higher mineralization was documented during the spring floods. During summer low-flow periods, subaqueous springs discharge waters under artesian conditions with a lower solute content. In the cave, the degree of saturation of water increases from the bottom to the top in the spring season and is the reverse in the summer. Seasonal variations in the groundwater chemical composition reflect the contribution from the artesian system. The geological data indicate a strong relationship between the karst features and the regional fault network. The characteristic features of the Ordinskaya Cave make it a model object of artesian hypogene speleogenesis.

Hypogene karst development in central Anatolia, Turkey is represented by unique collapse dolines (obruks) developed mainly in Neogene lacustrine limestone formations. Many of these obruks are located in two separate rectilinear zones, one of which appears to mark the suture zone between Tauride-Anatolian and Sakarya Zone tectonic blocks of the Anatolian plate. The other zone coincides with the alignment of three dormant volcanoes. Formation of obruks seems to be associated with upwelling of carbon dioxide released from deep-rooted igneous activity sources along these zones since Late Miocene. Formation of obruks still continues today, whereas the size of recent collapses is much smaller than ancient examples probably due to weakening volcanism. Recent obruk formations are observed in an area where the youngest volcanic activity seems to have occurred. Frequency of obruk formation increased in recent years because of the groundwater’s accelerating piezometric head decline. The stable carbon and noble gas isotope data previously obtained from regional groundwater samples suggested both crustal and mantle sources for the carbon dioxide required for obruk formation. It seems likely that many of the large-scale karst cavity-collapse structures in the world are linked with excessive carbon dioxide release from mantle in orogenic plateaus like central Anatolia.

The Judean Desert in eastern Israel consists of a Late Cretaceous epicontinental carbonate terrain formed at the interface between the Neotethys Ocean and the Gondwanian African-Arabian plate. Various rock types were deposited through fluctuations between marine and continental environments due to ingressions and regressions. Orogenic folding of the Syrian Arc fold system occurred between the Late Cretaceous and the Neogene. The folds formed a barrier for deep flowing groundwater, which upwelled along the SE flank of the folds. Possibly mixing with local confined circulation, the hypogene flow created maze caves at the top of the Late Cretaceous limestone, under the confinement of thick chalk, and marl successions. The larger caves are up to 3.5-km-long 2D mazes, less commonly with some additional tiers. Speleogenesis occurred most likely during the Oligocene, when far-field recharge could reach the Judean Desert. During the Neogene, the Dead Sea transform has dissected the region, forming a deep endorheic depression at the eastern border of the Judean Desert. This was followed by the lowering of the water table and related dewatering of the caves. Fault escarpments and downcutting canyons have dissected the caves, forming the present entrances. The caves are mostly dry, with rare speleothems. Gypsum rinds indicate that hydrogen sulfide and sulfuric acid took part in speleogenesis.

Iran is divided into five major geologic zones—the Zagros, Sanandaj-Sirjan, Central Iran, the East and Southeast, and the Alborz. About 10.5% of the country is underlain by carbonate rock formations, chiefly in a broad western strip, in the northeast, and central parts. Geomorphologic investigations have found clear differences between the extent and intensity of karst surface landform development, karst being more abundant in wetter regions such as the Zagros and very limited in arid Central Iran. Iran is within a very active tectonic region, with some sporadically active volcanoes, strong and widespread seismic activity, numerous active faults, and thermal springs. The hypogene caves that have been studied to date are found chiefly in just three areas: the North West and the Central and Eastern zone where they are associated with the igneous activity, and the Zagros Mountains. Although there is no modern volcanic or other igneous activity in the Zagros, the occurrence of oil fluids is considered to be a potential cause for hypogene cave development. The development of the two largest karst caves in Iran, Katalekhor and Alisadr, is attributed to ascending CO₂ due to volcanic activity. Other large caves such Karaftoo and Kangohar were created by magmatic fluids. Excavations for dams and water transmission tunnels in the non-volcanic Zagros zone have discovered some remarkable, active hypogene systems. Oil brines are considered to be the most important contributor to their development.

Two small limestone ridges, Tyuya-Muyun and Kara-Tash, located in the southern fringe of the intermontane Fergana Valley in SW Kyrgyzstan, host a suite of caves, some of which previously contained Ra, U, and Cu ores that were emptied by mining in the first half of the twentieth century. In early 1920’s, mining geologists used the concept of “classical” karst to predict the morphology of ore lodes. These predictions proved largely unsuccessful, because the origin of karst was more complex than first predicted. Speleogenetic history of the massifs comprises of early epigene, hypogene (thermal, ore bearing), and late epigene karst stages. Some caves bear traces of all three stages, whereas others are monogenetic (hypogene).

The Guadalupe Mountains consist of an uplift of Permian carbonate shelf deposits in a semiarid landscape. A variety of speleogenetic processes, mostly hypogene, have made them one of the world’s best-known cave regions. The most notable caves are Carlsbad Cavern, which contains the largest known cave room in the USA, and Lechuguilla Cave, now the world’s 7th longest. Because the caves are no longer active, there was early confusion about their origin. This was resolved when long-dormant sulfuric acid processes were recognized, with H₂S supplied by nearby oil fields. Potassium-argon dating of the by-product mineral alunite in the Guadalupes indicates speleogenetic ages from 12 to 4 million years, decreasing with lower elevation. Caves show abundant evidence for subaerial corrosion, both by sulfuric acid and carbonic acid in water films. Many seemingly phreatic features have resulted from this subaerial process. Microbial alteration of bedrock has contributed to weathering. There is evidence that isolated caves of greater age, lined by large scalenohedral calcite, were formed by supercritical CO₂ in deep thermal water.

The Permian Basin of west Texas and southeastern New Mexico is host to extensive karst development in carbonate and evaporite strata. The Delaware Basin and associated Northwestern Shelf include both Guadalupian and Ochoan evaporites with diverse hypogene karst features of four general types: fluvial-induced; pressure gradient-dominated; density convection-dominated; and hydrocarbon-enhanced. Although many classic definitions of hypogene speleogenesis emphasize confined or semi-confined hydrogeologic conditions, analyses of Permian evaporites of this region show that hypogene karst can develop in near-unconfined conditions as long as local meteoric influences are minimal. Since the early Paleogene, eastern migration of the Pecos River in New Mexico has created a persistent potentiometric low as a target for upward migration of fluids, producing multi-story rectilinear maze caves as well as large cenote-like collapse structures in the Seven Rivers Formation, with less dramatic examples also common throughout the back-reef facies. Within the Castile Formation, evidence of hypogene speleogenesis is common, ranging from isolated rise structures nearly 100 m deep to multi-story maze caves and intrastratal brecciation. Evaporite calcitization and associated sulfur ores are common as the result of sulfate reduction in the presence of ascending hydrocarbon-rich fluids. Overlying strata of Salado and Rustler formations also host hypogene karst, but to a much lower documented degree and largely associated only with cross-formational breccia pipes and intrastratal brecciation.

Chiricahua Crystal Cave may be the only known limestone cave in the world that intersects room-sized quartz geodes (up to 16 m across). The cave is a relatively deep, laterally confined network maze, with zones of breakdown and vertical shafts. Large euhedral quartz crystals are found at all levels of the cave, lining vugs that range in size from centimeters to tens of meters. These geodes are crosscut by a network of phreatic tubes and abruptly terminating fissures that show little evidence of top-down flow or connection with surface hydrology. All surfaces of the cave below a certain elevation are covered with a thick coating of calcite mammillaries. The cave reflects a series of distinct geochemical environments. We propose that deep-seated hydrothermal silicic acid initiated the formation of the vugs and deposited a quartz rind as it cooled. Hypogenic (and likely hydrothermal) carbonic acid dissolved the subsequent passages and then vacated the cave. Subaerial conditions prevailed for a time before fluid once again entered the cave. A cooler, calcite-rich pool filled the lower reaches of the cave, depositing mammillaries in a stable environment with slow, steady atmospheric and hydrological mixing. Additional features, such as cupolas, solution pockets, phreatic tubes, corroded and altered surfaces, and vertical pits at the bottom of the cave (possible feeders), are also compatible with bottom-up thermal processes.

Although Grand Canyon hosts exceptionally well‐formed vadose caves and exemplary paleokarst, it also contains hypogene caves that provide important information on the canyon’s age and evolution. These caves exhibit speleogenetic materials, passage morphologies, and locations near the top of the Mississippian Redwall Limestone that represent major hypogene dissolution phases. Cave origin resulted from CO₂ and H₂S in solutions that upwelled from depth and mixed with Redwall–Muav aquifer water. Lack of abundant speleogenetic gypsum suggests that CO₂ was the primary solutional agent, while upwelling H₂S likely played only a minor role. Each phase of hypogene speleogenesis in our model encompasses the following sub‐events, from deep to shallow: (1) dissolution of cave passages 500 ± 250 m below the water table or potentiometric surface, sometimes with Fe‐ and Mn‐oxide by-products; (2) deposition of calcite spar linings (~50–100 m below the water table); (3) deposition of calcite mammillary coatings (1–20 m below the water table); (4) deposition of calcite folia at the water table; and (5) deposition of gypsum rinds a few meters above the water table. Controls on the amount of cave dissolution and speleogenetic by-products probably include regional water table fluctuations during the Miocene and Pliocene, in combination with magmatic/tectonic pulses that pumped CO₂ and H₂S from below. The complete cycle of Grand Canyon hypogene speleogenesis includes a largely dissolutional phase under confined conditions and a later (mostly by-product) phase taking place under unconfined conditions. The process terminates when the water table descends below the cave.

We summarize observations on condensation corrosion caves above a regional, moderately thermal aquifer in the Devils Hole Ridge near Ash Meadows (Nevada) as well as at a site near Shoshone (California). At Devils Hole, water of the regional aquifer has been saturated with respect to calcite (i.e., non-aggressive) for the last 2–3 million years. Caves at both sites have been initially formed by extensional tectonics and further enlarged by condensation corrosion.

Colorado, a state with diverse geology and high topographic relief, contains several significant hypogene cave systems. Three are described here: the Orient Mine Cave System, Cave of the Winds, and Glenwood Caverns. All are located near present thermal springs, which help to determine the chemistry of the cave-forming water. The Orient Mine Caves were encountered during mining operations to extract limonite ore for the manufacture of steel. The ore follows the bedding, but hypogene cave galleries also cut across bedding and extend upward as domes. The galleries were formed by warm water rising along the San Luis fault and mixing with cool recharge from nearby mountains. Cave of the Winds follows the plunge of an anticline and probably formed when water with high CO₂ rose along faults and mixed with shallow meteoric water. At Glenwood Caverns, carbonic acid in rising thermal water is responsible for most speleogenetic dissolution, but some dry parts of the cave have a sulfuric acid overprint.

The Bighorn Basin of Wyoming is a region of thermal springs and caves, some with lethal levels of H₂S and CO₂. It also contains many productive oil wells. In one of these caves, “sulfuric acid speleogenesis” was first recognized and documented in North America by Egemeier in 1973. He proposed that most of the cave dissolution was subaerial and the result of H₂S oxidation to H₂SO₄. Later studies by microbiologists have refined his measurements and show that much of the H₂SO₄ is generated in the stream by sulfur-oxidizing bacteria, by the process of “microbial sulfuric acid speleogenesis.” Other caves with similar chemistry are known in the region, but they have only been partly explored because of the locally high sulfide concentrations in their atmospheres.

The Black Hills contain several extensive maze caves in the early Carboniferous Madison Limestone. They include Wind Cave and Jewel Cave, which are among the world’s longest and most complex. Their origin is debated, with diverse hypotheses ranging from artesian conditions to rising thermal water. Recent evidence indicates a polygenetic origin including early Carboniferous diagenesis and paleokarst; deep burial by Carboniferous–Cretaceous strata; re-exposure of the limestone by the Laramide Orogeny (early Paleogene); and major cave enlargement in the late Paleogene along old paleokarst zones. Cave enlargement depended mainly on diffuse recharge through overlying sandstone, mixing with lateral inflow through carbonate outcrops. Only a few of these processes were hypogenic, but recognizing them all helps to clarify the limits of that process.

Regional dissolution trends configured the Middle Devonian Prairie Evaporite basin across Western Canada, resulting in the largest known collapse of a hypogene evaporite karst dominated by halite beds. Dissolution-collapse subsidence resulted from aquifer flows to the north-east within the Alberta and Williston basins, up-structure towards the eastern margin of the salt basin. A 1000-km-long and 150-km-wide salt dissolution trend developed along the eastern evaporite basin margin. The flows occurred as the Western Canada foreland basin deepened in response to the Columbian (Middle Jurassic-Early Cretaceous) tectonism. A second dissolution pattern developed across the southern Saskatchewan area of the Williston basin. Dissolution trends removed up to 250-m-thick beds across south-central Saskatchewan and 100–150-m-thick beds in north-eastern Alberta. These salt dissolution patterns advanced along sets of fault-fracture lineaments that widened and coalesced into larger salt removal areas. The dissolution trends configured the overlying strata as subsidence-collapse troughs tens of km long. In northern Alberta, troughs up to 50 km long configured the Upper Devonian limestone karst palaeotopography, subsequently covered by Lower Cretaceous Athabasca Oil Sands. This bitumen deposit, the largest known, overlies a 300-km-long segment of the dissolution trend that extends along the eastern margin of the Middle Devonian evaporite basin. Syndepositional sand trends up to 25 km long accumulated over collapsed chains of Devonian fault blocks that responded to salt dissolution patterns in the substrate. These sand reservoirs trapped Late Cretaceous-Early Palaeogene oil migrations into the area.

A large lead–zinc ore body formed along the Presqu’ile barrier reef at the northern margin of the Prairie Evaporite (Middle Devonian) salt basin in northern Canada. An early stage of finely crystalline dolomitization of the Pine Point Formation limestone was patchy. A second dolomitization stage resulted in the coarsely crystalline massive Presqu’ile dolomite. Deep-seated hydrothermal brines flowed eastward up-structure, across a segment of the McDonald basement fault, into the eastern end of the barrier reef complex along earlier multi-km-long karst dissolution trends. These metalliferous hydrothermal brines resulted in 30–50-km-long and 2–3-km-wide Mississippi Valley-type Pb–Zn sulphide trends along the base of the Presqu’ile dolomite. The sulphide mineralization was distributed as tabular stratabound ore pods, 100–250 m-long and 10–20 m-thick, and ovoid vertical bodies up to 60 m, cross-cutting the Presqu’ile dolomite and extending into the overlying beds. Mineralizing thermal fluids directed along earlier karstic dissolution trends filled available porosity, including sediment filled cave sites, or hydrothermal brines extensively expanded the pre-existing dissolution trends by solution-replacement of the surrounding host dolomite, resulting in sulphide deposition where chemical conditions were favourable. Spatiotemporal relationships are uncertain between karstification stages, depths of burial, and distribution of these sulphide ore pods. Early stage near-surface karstification may have occurred as early as Devonian, followed by deeper burial during the late Carboniferous Antler Orogeny. Alternatively, the near-surface karstification was Antler, followed by hydrothermal karstification with deep burial during Middle Jurassic-Early Cretaceous (Columbian) to Late Cretaceous (Laramide) tectonism.

The Balcones Fault Zone segment of the Edwards Aquifer of south-central Texas is one of the most important and prolific karst aquifers in the United States. It is formed within the lower Cretaceous Edwards Group Limestone. Since deposition, it has undergone subaerial exposure, burial in the upper Cretaceous, faulting, igneous intrusion, uplift in the Miocene, and weathering processes. The Balcones Fault Zone consists of mostly normal en-echelon faults with as much as 300 m of displacement striking northeast–southwest and dipping down toward the Gulf of Mexico. It forms the Contributing, Recharge, and Artesian zones of the aquifer. Karst characteristics of the Edwards Aquifer are the result of both epigene and hypogene processes, which continue today. The subaerial exposure of the Edwards Limestone (Recharge Zone) contains epigene karst features typical of a karst landscape that contains remnants of relict hypogene processes. Many relict caves in the Edwards Limestone outcrop show evidence of being formed by ascending water. Some appear to be associated with paleo-springs that were abandoned as water levels in the aquifer declined. Evidence of current hypogene processes is found in the saline water zone, which is part of the Artesian Zone. Extremely high permeabilities have been developed by dissolution at depth and driven by a number of processes including artesian hydraulic heads, mixing corrosion, and biogenic acids. As a result, well production in the Artesian Zone is commonly limited only by the size of the pump.

The Arbuckle Mountains form a small but complex geologic province characterized by thick sequences of intensely folded and faulted carbonates, sandstones, and shales of the Late Cambrian through Pennsylvanian (Late Carboniferous). Caves, karst, and paleokarst features are abundant in several limestone and dolomite formations. The region hosts a diverse range of settings and mechanisms capable of forming both epigenic and hypogenic caves. Hypogenic features are found in caves throughout the Arbuckles, but most commonly where deformation has been severe. Carbonate dissolution in these areas is aided by the mixing of fresh and saline waters, and by microbial interaction with hydrocarbons. It is probable that more than one process has been involved in forming hypogenic caves and paleokarst.

Giant spar crystals were discovered in Chilly Bowl Cave in northern Arkansas in an area of deep faults between the Ouachita orogenic province and the Ozark uplands. The crystals provide data that support hypogene speleogenesis, including fluid inclusion paleotemperatures, 13C and 18O, and U–Pb dates that are inconsistent with an epigenetic water source. During the Ouachita orogeny, thermal brines entered the Paleozoic carbonates of the Ozarks to form spar-lined caves (U/Pb spar dates = 52 ± 2 million years). This age significantly extends the range of cave ages in the region. The chemistry of the early waters was derived from the deeply buried Ouachita foreland basin. The spar crystals are intersected by more recent caves related to the present topography which experience upward flow below confining units and by definition are also hypogenic.

The Mammoth Cave area is best known for its epigenic caves. However, the area also contains examples formed or modified by sulfide-rich brines, some associated with petroleum. These include hydrogen sulfide in shallow aquifers generated by reduction in gypsum and anhydrite beds and nodules a few tens of meters beneath the lowest cave passages in Mammoth Cave; sulfides generated in oil reservoirs located in structural traps; and rare sulfide seeps intruding into the cave system that cause local dissolution and a proliferation of cave biota. A highly acidic stream enters at least one cave in the region, and although its effect is subdued by nearby sources of meteoric water, it demonstrates the potential for significant hypogene cave development in the region. Oxidation of iron sulfide inclusions in the host limestones has produced weathering rinds and minor solutional effects on cave passages.

By any measure, both in the classical (Crawford in groundwater as a geomorphic agent. Allen and Unwin, Boston, pp 294–338, 1984; White in geomorphology and hydrology of karst terrains. Oxford University Press, New York, 1988; Sasowsky and White in Water Resour Res 30(12):3523–3530, 1994) and the present framework (Anthony and Granger in J Cave Karst Stud 66(2):46–55, 2004; Simpson and Florea in Caves and karst of America. National Speleological Society, Huntsville, pp 70–79, 2009), the karst landscapes of the Cumberland Plateau (western margin of the Appalachian Basin, eastern USA) are epigenic in nature. Discrete meteoric recharge is conveyed through sinking streams, sinkholes, and an epikarst reservoir into an integrated conduit system and toward topographically lower springs (Florea in J Hydrol 489:201–213, 2013a). The purpose of this brief manuscript is not to redefine that context, but to summarize a suite of data (Florea in investigations into the potential for hypogene speleogenesis in the Cumberland Plateau of southeast Kentucky, U.S.A. Brno, Czeck Republic, pp 356–361, 2013b; Florea in Acta Carsologica 42(2):277–289, 2013c; Florea in Earth Surf Proc Land, 2015) that suggest that geochemical reactions involving sulfur associated with shallow petroleum reservoirs may have played a role in a polygenetic evolution of some Cumberland Plateau caves.

Several caves in the Shenandoah Valley in Virginia show evidence for early hypogenic conduit development with later-enhanced solution under partly confined phreatic conditions guided by geologic structures. Many (but not all) of these caves have been subsequently invaded by surface waters as a result of erosion and exhumation. Those not so affected are relict phreatic caves, bearing no relation to modern drainage patterns. Field and petrographic evidence shows that carbonate rocks hosting certain relict phreatic caves were dolomitized and/or silicified by early hydrothermal fluid migration in zones that served to locally enhance rock porosity, thus providing preferential pathways for later solution by groundwater flow, and making the surrounding bedrock more resistant to surficial weathering to result in caves that reside within isolated hills on the land surface. Features suggesting that deep phreatic processes dominated the development of these relict caves include (1) cave passage morphologies indicative of ascending fluids, (2) cave plans of irregular pattern, reflecting early maze or anastomosing development, (3) a general lack of cave breakdown and cave streams or cave stream deposits, and (4) calcite wall and pool coatings within isolated caves intersecting the local water table, and within unroofed caves at topographic locations elevated well above the local base level. Episodes of deep karstification were likely separated by long periods of geologic time, encompassing multiple phases of sedimentary fill and excavation within caves, and reflect a complex history of deep fluid migration that set the stage for later shallow speleogenesis that continues today.

The northern border of the Appalachian Plateaus in New York State is a limestone–dolomite escarpment with sulfate rocks at depth and hypogene karst springs at its base. All springs contain dissolved carbonates, but many are also exceptionally rich in sulfate, sulfide, or CO₂. None connect to traversable caves, but their chemistry provides clues to their internal character, flow depth, and underground processes. Many show evidence for dedolomitization driven by sulfate dissolution, which forces calcite travertine to precipitate at the surface. Isotopic variation and radium content reveal groundwater flow patterns. Chemical contrasts with nearby epigenic caves highlight the nature of the hypogene springs. A nearby but separate spring area, fed by deep flow along faults, illustrates the effects of high-pressure CO₂ on carbonate groundwater. Despite the lack of underground access, these karst areas give much insight into hypogene processes.

Nanisivik zinc/lead mine is located at Lat. 73°N in northwestern Baffin Island. The host rock is a pervasively dolomitized Proterozoic platform carbonate overlain by a shale aquitard. The principal deposit, the Main Ore, is of zinc, lead, and iron sulfide plus gangue minerals, chiefly secondary dolomite. It extends for 3 km along a horst and is terminated at both ends by modern valley entrenchments. The Main Ore body is consistently ~100 m wide and five to seven m thick. The ceiling is a wide, horizontal corrosion bevel. Within the Main Ore are two or more generations of tapered fins of dolomite that extend into the cavity. The fin edges are sinuous, some meandering. Sharp, horizontal corrosion notches 20–30 cm high extend into the dolomite walls. They are filled with layered pyrites which continue into the ore body, truncating earlier, dipping mineral bodies. Sedimentary textures suggest that the ore cavity formed by paragenesis in a channel-flow mode, with ore and gangue deposition on the floor taking place in tandem with dissolutional cavity creation upward. Fluid inclusions indicate derivation of the metals from exchange reactions with metalliferous sediments (the underlying shales), indicating low water/rock ratios and moderate temperatures. The ore fluids were similar to oilfield brines. Sulfur isotope fractionations indicate temperatures of 90–150 ± 40 °C, suggesting that the Main Ore formed along a gas/brine interface at a depth of at least 1600 m. Latest studies suggest a Grenville age (~1.1 Ga) for the Main Ore.

The Florida Platform includes many examples of suspected hypogene karst, on- and offshore. These features occur in carbonate strata of Early Cretaceous to Late Pleistocene age. Transition zones between fresh and saline groundwater systems can be shown thermodynamically to be developing hypogene karst by groundwater mixing. Although the resulting flank-margin caves have not been confirmed on the basis of cave morphology, several probable examples exist. Carbonate rocks in deeper horizons of Paleocene/Eocene and Cretaceous age are bounded by gypsum/anhydrite and contain strata with unusual porosity development and high hydraulic conductivities. These strata appear to contain hypogene porosity created either along a deep or ancient saline water/freshwater transition zone or by interaction with H₂SO₄. There is evidence that collapse zones from these deeper strata have migrated upward to the modern land surface. The even deeper, Cretaceous system appears to contain H₂SO₄-driven hypogene porosity.

Dissolution of eogenetic carbonates in the Upper Floridan aquifer has produced the world’s densest assemblage of first-magnitude cave springs. Conceptual and numerical models of cave origin in the aquifer have emphasized epigenic and mixing-dissolution processes. We draw upon recent research concerning phreatic caves in the Suwannee River Basin, and dry caves in the west and central Florida, to suggest instead that many caves in the aquifer formed by hypogenic processes. Formerly, undersaturation generated in the subsurface has usually been ascribed to the mixing of, or temperature changes in, subsurface fluids. Here, we describe an alternate process. Cave formation at modern water tables in the aquifer has been linked to respiration of CO₂ in the deep vadose zone and at water tables. Respired CO₂ generates carbonate mineral undersaturation when it hydrates to carbonic acid at water tables. Because undersaturated waters are created at the water table, caves form as isolated macropores. Caves at modern water tables in the aquifer lack initial connections to the surface (e.g., entrances) and have morphologies that are unrelated to surface drainage, making them similar in many respects to flank margin caves (see Mylroie and Mylroie, Chap. 51). Since many caves that are below the modern water table have similar morphologies to caves that are at the modern water table, it is likely that they formed by similar processes operating at water tables associated with lower sea levels. These caves became sources of springs and flooded sinkholes when Holocene sea-level rise elevated water tables close to, and above, the land surface.

Fieldwork in the Bahamian Archipelago in the 1970s and 1980s identified a new cave type, the flank margin cave, as macroscopic dissolutional voids developed in the margin of a freshwater lens, under the flank of the enclosing landmass. These voids are produced by three conditions that exist at the lens margin: mixing dissolution, organic decay horizons, and the increase in freshwater flow rate. The water flow enters flank margin caves as diffuse flow and exits as diffuse flow, a flow regime that produces dissolutional sculpture lacking turbulent flow features, such as asymmetric scallops. The caves are tied to sea level, which controls the freshwater lens position, and as such are excellent indicators of past sea-level position. The caves form without entrances and become accessible only after subaerial erosion has breached their ceilings or walls. Flank margin caves initiate as individual globular dissolutional voids that then intersect as the voids enlarge, increasing cave size in a sudden stepwise manner. As cave development is restricted to the lens margin, the largest flank margin caves acquire a linear shape as voids interconnect parallel to the lens margin. Flank margin caves are hypogene caves based on their diffuse, slow-flow regimes, because the dissolutional aggressiveness is generated below the surface by mixing, and the ascending marine water following the base of the lens to the site of dissolution at the lens margin. Because these caves form rapidly at shallow depths, there is a debate as to their hypogene classification, but they meet all criteria for hypogene speleogenesis.

Sistema Zacatón includes the second deepest underwater cave in the world. It is hypothesized to have formed by volcanogenic karstification, a process that relies on four components to initiate and develop deep subsurface voids: a carbonate matrix, a system of preferential flow paths (e.g., fractures), volcanic activity that increases groundwater acidity, and groundwater flux through the system. Results are compiled into a multiphase speleogenetic model, most phases of which are of Late Pleistocene age. Surface rocks consist of carbonate travertine with Pleistocene mammoth fossils found within the rock matrix. The rocks are interpreted as a hydrothermal travertine terrace formed as nearby volcanic activity peaked, and thus representing the end member of a carbonate mass transfer system originating deep in the subsurface. The modern karst system includes a dynamic array of deep, phreatic sinkholes (also called cenotes) propagated upward through the travertine and exposing hydrothermal water supersaturated with carbon dioxide to the atmosphere. In some cenotes, seals of a second stage of travertine formed as CO₂ degassed, capping the sinkholes with hydrologic barriers. Volcanogenic karstification worldwide is not limited to Sistema Zacatón, although the localized nature, coupled with the extreme degree of karstification, makes it an ideal modern analog for classifying certain other karst systems as volcanogenic.

The caves of Naica (Chihuahua, Mexico) have been known since the beginning of the nineteenth century, when they were intersected by mines. They became world famous in 2000 when the mining activities intersected three small caves, which hosted gigantic gypsum crystals. Their genesis and evolution is strictly related to rising thermal fluids induced by Tertiary intrusive magmatic activity, giving rise to polysulfide deposits. Speleogenesis began about 1.5 Myr BP, induced by rising thermal water. This process lasted for a relatively short time. Later evolutionary stages varied from cave to cave, reflecting alternately deep-seated phreatic, epiphreatic, and vadose environments. Deposition of the giant gypsum crystals began ~400–350 Kyr BP, when the water temperature dropped below 58 °C. Their growth suddenly ceased in 1985, when mine dewatering dried up the caves. Besides the giant gypsum crystals, Naica caves contain a wide variety of other cave minerals (over 40), ten of which are new for the cavern environment. In addition, they include fossil pollen trapped in the gypsum lattice, which allow reconstruction of the external climate of 350 Kyr BP, and microorganisms new to science, which were preserved in the fluid inclusions of the gypsum crystals for tens or even hundreds of thousands of years. In 2015, mining intercepted a major thermal water source, which flooded the mine, and the Cueva de los Cristales and other caves at −290 m were flooded by more than 160 m of hot water.

Sulfur-rich karst springs that drain middle to Late Cretaceous carbonates in southern Tabasco, Mexico, contain elevated total dissolved solids with high chloride and either high sulfide or sulfate. The mostly meteoric waters circulate along intermediate-depth flow paths, interacting with Jurassic evaporites and then mixing with shallow-circulating rainwater to produce brackish, sulfur- and NaCl-rich spring waters. Several caves, notably Cueva de Villa Luz and Cueva de Luna Azufre near Tapijulapa, are developing from these waters. Both are fed by small, warm subterranean springs that release hydrogen sulfide and carbon dioxide at the water–air interface. These gases combine with water vapor in the air and moisture on cave walls to form acidic solutions, which enlarge the cave passages. Speleogenesis is aided by abundant chemoautotrophic microbes that facilitate oxidation and the production of acid solutions. Cave walls and ceilings display several distinctive colonies of these microbes, notably filamentous bundles (snottites) and biovermiculations. The organisms also reside in unusual mineral deposits, including sulfur folia, sulfur crusts, and gypsum paste. Pore fluids in these deposits and growths have very low pH despite their proximity to limestone walls. Speleogenesis appears to follow a three-step process: (1) an early sulfide-rich phreatic phase; (2) a sulfide-rich, vadose–phreatic phase of chemical alteration and mechanical stoping; and (3) a tertiary sulfide-depleted phase of epigenic speleogenesis. The second phase of development dominates speleogenesis in Cuevas de Villa Luz and Luna Azufre.

Knowledge about karst in South America is still limited due to the paucity of cave exploration and the lack of scientific assessment of most cave regions. However, the South American continent presents favorable conditions for the development of hypogene cave systems, especially ancient ones, now decoupled from routes of cave-forming fluids. The majority of recognized hypogene caves occur near the borders of cratonic areas, in Precambrian carbonates. The best-known example comprises the large maze caves (TBV-TBR) near the village of Laje dos Negros, northeastern Brazil, and the hypogene karst of the Vazante area, south-central Brazil. Although other areas have not been subjected to detailed speleogenetic studies, several other sites present caves with hypogene characteristics, such as in the Iron Quadrangle area (state of Minas Gerais), northern state of Bahia, and western states of Mato Grosso and Mato Grosso do Sul, in Brazil. In the Argentinian Patagonia, associated with the Andes Mountain chain, the Las Brujas Cave has been interpreted as a hypogene system. An isolated occurrence of possible quartzite hypogene cave is also described.

Toca da Boa Vista and Toca da Barriguda (TBV-TBR) are two of the longest caves in South America and, although not physically connected, share a common origin and evolution. Together, they comprise over 145 km of passages that are mostly developed in five lithological units in a Precambrian dolomite host rock. The now relict caves display a mixed ramiform/spongework/network pattern that shows clear hypogene morphology. The inception of speleogenesis cannot be precisely determined but likely occurred several tens (if not hundreds) of millions of years ago, placing this system among the oldest caves in the world. The TBV-TBR system is controlled by a NE–SW-oriented fracture corridor; passages follow anticlines and troughs, showing morphological variations depending on the vertical position in relation to the carbonate units. The cave system is represented by a laterally extensive maze controlled by a highly fractured unit in which ascending flow was restricted by an upper less fractured unit (hydraulic seal). The long evolutionary history of the cave left numerous imprints, including widespread and remarkable condensation-corrosion processes that dissolved the bedrock and speleothems, abundant chemical and clastic sedimentation phases and flooded lower passages. Following the interception of the cave by surface denudation, a remarkably rich late Quaternary vertebrate fossil assemblage accumulated.

The Proterozoic carbonates of the Vazante Group, located in southeastern Brazil, host important zinc deposits related to emplacement of hydrothermal metal-rich solutions along a fault zone. Karst landscape comprises a series of elongated residual carbonate hills containing both epigene and hypogene caves. In the vicinity of the mineralized zone, several caves display hypogene morphology, especially in the surrounding areas of the town of Vazante. The caves predominantly have a network pattern controlled by N60W- and N45E-oriented joints and display cupolas, feeders, rising channels, and outlets. Some of the major caves occur close to the Vazante Fault zone. The intercalation of layers of distinct permeability, due to both the stratigraphic sequence and by the fault-oriented mineralized zone, creates aquifer compartments with differing hydraulic heads. Breaching of the aquitard represented by the zinc-rich lens allows rising flow along the fault zone, leading to lateral flow in favorable bedding plane horizons overlaid by impermeable beds. This flow results in both active and now relict hypogene caves. Fragments of hypogene caves have been recorded at depth (in excess of 600 m below the surface) both through tunnels in the underground mine and in deep boreholes.

This chapter aims to describe the hypogene caves of North Africa. A few hypogene caves are known in Morocco, Tunisia, Libya, and Egypt. Algeria hosts most of hypogene caves known in this region. This includes hydrothermal cave springs (Aïn Ouarka, Hassi Dermam, Aïdour), mazes related to CO₂ degassing (Kef el Kaous), iron-ore related caves (Rouïna) and sulfuric acid speleogenesis (SAS) caves especially in the Azerou massif. This last area displays numerous caves recording the geomorphic evolution that follows the gradual exhumation of the thermal aquifer. Preliminary analytical data (geomorphology, mineralogy, isotopes, and U/Th dating) are presented.

In this region, the most typical hypogene caves are located in the arid (10–300 mm/year) NW of Namibia, developed in relatively pure dolostone and calcitic marble. Their origin is mainly evidenced by past hydrothermal activity, which formed chimneys terminating upward into bell-shaped ceilings and alveolar niches, as well as regular and extensive dogtooth calcite coatings on the walls. Barite occurs in one cave. They seem to have formed by low salinity solutions at temperatures not exceeding 70 °C. These caves are widely dispersed in the carbonate-rock areas, and surficial karst morphology is absent or very poorly developed. In this dry area, caves not displaying typical hydrothermal characteristics are also known. The passage morphology is phreatic, forming complex networks, but encompassing only very small areas. They are often reduced to single voluminous chambers. This limited lateral extension might possibly suggest localized upwelling from deep aquifers. Further East in Namibia, in the Otavi Mountains and in the northern part of the Kalahari region, under a higher rainfall regime (500–600 mm/year), the karst is better developed, but less evidences of hydrothermal activity have been observed. Only in one cave unusual chemical composition of groundwater is suggesting deep leaching of silicate rocks. Although the caves are more frequent and better developed, forming complex phreatic mazes and voluminous chambers, they still do not extend widely. Thus, it is possible that hydrothermal activity did occur more commonly, but is less obvious and obscured by surficial water dilution. In NW South Africa, the karst is developed in thick, very cherty dolostone, and due to a wetter climate, the intensity of dissolution is fairly intense. Probably in part due to this impure nature, however, groundwater flow is sluggish, but caves are frequent, mainly forming phreatic mazes of restricted extension and no long interconnected systems. No direct hydrothermal evidences have been reported. Nevertheless, a fair permeability has been proved to be present down to great depth, a condition favorable for the formation of upwelling plumes of moderately warm water in the aquifer. This suggests hypogene processes to be responsible for the caves developed under thick shale cover.

The Sterkfontein Caves, located in the southwest of the Cradle of Humankind, Gauteng, is the world’s richest Australopithecus-bearing locality and has yielded iconic fossils like Mrs Ples (StS 5) and Little Foot (StW 573), as well as Paranthropus robustus, Homo ergaster and Homo habilis fossils and large Earlier Stone Age lithic assemblages. The cave deposits have also yielded enormous assemblages of associated fossil fauna and document over three and a half million years of landscape, environmental, faunal and hominid evolution. These fossiliferous cave deposits represent a more recent episode of a geological history spanning 2.6 Ga years, beginning with the deposition of the dolomites, to the commercial exploitation of the caves by lime miners in the early twentieth century. The location and morphology of the karst caves is a result of a combination of factors including lithological variation within the two host dolomite formations, an early karstification and infilling of the dolomites over two billion years ago and local dolomite fracturing. Vadose zone collapse in densely fractured areas has enlarged chambers and passages, and played a major role in the location and nature of the openings to the landscape. When open to the landscape, a broad range of geomorphological processes, including re-dissolution of interred deposits, creates dynamic sedimentary environments with complex stratigraphic histories. This article reviews the geomorphological history of the Sterkfontein Caves in an effort to consolidate this information as we press forward with new stratigraphic and geomorphological work at the site.

The Tasmanic karsts of eastern Australia extend from the south-eastern tip of Tasmania to the far north of Queensland. The mainland section incorporating the eastern parts of Victoria, New South Wales and Queensland contains more than one hundred and twenty bodies of cavernous Palaeozoic limestone, most of which form impounded karsts. Many of these karst areas are small, difficult to access, contain a small number of caves and have received little attention from cavers and even less attention from scientists. Some 21 karst areas stand out for their number of caves and their diversity of underground features. Almost all of the Tasmanic caves in mainland Australia show a high degree of structural guidance and are developed in massive high-purity limestone. Solution dolines are rare or absent in the surrounding karst. Some caves appear to have been formed entirely by a single hypogenic event while others have been modified by several hypogenic events (multiphase caves). In other caves, hypogene forms have been overprinted by later epiphreatic, paragenetic, fluvial and breakdown events (multiphase/multiprocess caves). Where there are active rivers in the caves, the rivers are often much smaller than the cavities they flow through, take complex paths through the rock and are captured into pre-existing cavities. The evidence for a hypogene origin of these caves is largely morphological, but is supported by circumstantial evidence such as intersection of palaeokarst, proximity to regional faults and tenuous relationships with the surrounding geomorphology and hydrology.