The term aulacogen from the Greek aulax (ανλαξ) ‘furrow,’ introduced by N. S. Shatsky (1960, in Schmidt and Hoppe, 1971 and Dennis et al., 1979, pp. 14–16; see Milanorsky, 1981, p. 214 for reference to the term's first appearance in Soviet Literature 1961), was originally intended to designate narrow, elongate depressions that strike into cratons from reentrants that face an adjoining larger basin or a mountain belt that grew out of a geosyncline. Shatsky had recognized these structures in the 1940s (“transverse basins and transverse fractures”; p. 57) he considered them a subset of his marginal transverse structures of old platforms; Shatsky, 1946), and they appeared to have been genetically related to the larger basin into which they opened. He also noted that they internally segmented the cratons. Shatsky's original concept (Shatsky, 1946, 1955), which has since been modified, at least from the viewpoint of the genesis of these structures, may be summarized as follows: An aulacogen may open into a syneclise (a roughly equant cratonic basin of large dimensions in the Russian tectonic terminology: See Dennis et al., 1979, p. 87), a geosyncline, or an ocean. It contains a sedimentary section at least three times as thick as the section found on the surrounding platform into which it strikes and which it segments. This sedimentary section generally correlates with the even thicker sedimentary content of the larger basin into which the aulacogen opens. Novikova (1964, p. 68) enphasized that aulacogens of the east European platform predate its syneclises. Shatsky was a convinced fixist and believed that the cause of the subsidence that created both the aulacogen and the associated larger basin was a density increase in the lower crust and/or in the upper mantle. The geometry of the aulacogens was ascribed to a pre-existing planetary network of fractures or weak zones that served to localize the subsidence. Due to this predetermined location and orientation, aulacogens may be subjected to repeated rejuvenations (the posthumous movements of Shatsky; see especially 1955), which, in some cases, involve a weak folding and metamorphism of the sedimentary content of the aulacogen, as first emphasized by Borissjak (1903) on the example of the Donetz basin.
Following Shatsky's initial recognition aulacogens were studied extensively and classified according to a variety of criteria, especially by Russian and German geologists (e.g., Bogdanov, 1961; von Gärtner, 1969), although no marked progress was made in our understanding of the origin of these features. Shatsky's hypothesis on the origin of aulacogens was ad hoc and especially failed to explain the nature of the posthumous movements of the structure. After the advent of the theory of plate tectonics, our understanding of rift systems, including the aulacogens that are now known to be rifts, has increased considerably, for good reviews see Burke (1977, 1980), Milanovsky (1981), and Mohr (1982). In this entry I adhere to Shatsky's original definition of aulacogen, with the omission of syneclise because it is a very nebulous term (see Dennis et al., 1979, p. 87) and because many syneclises Shatsky had in mind in relation to the aulacogens with which he worked turned out to be oceans (e.g., the North Caspian Depression; see Burke, 1977) or molasse basins. Therefore, an aulacogen is defined as a rift that originated in relation to an ocean-opening event, opens into the ocean or into a mountain belt (that resulted from the obliteration of a former ocean), and strikes into and internally segments a craton. Aulacogens have formed at all times in Earth's history, the oldest known being the ca. 3 Ga-old Pongola structure in southeastern Africa (Burke et al., 1985).
Origin of Aulacogens
Figs. 1E—G illustrate another mechanism to form aulacogens that was suggested by Grant (1971). The tip of a propagating rift (Figs. 1E and F , segment I) may become dormant as a result of the spreading jumping elsewhere, forming a new rift (Fig. 1F , segment I′). If ocean opening then takes place along the segments I′-II-III, then the original rift tip, I, may be left as an aulacogen striking into the continent from the reentrant formed as a result of transform/ridge (II/III) intersection (Fig. 1G ). The Triassic graben complexes (e.g., the Connecticut Graben) around the New York Bight, an Atlantic reentrant formed by such a ridge/transform intersection (Fig. 2) seem to fit this model well (Dewey, 1977). In this case, pregraben doming is not a necessity.
Fig. 1B indicates that the sequences of events shown in Figs. 1A, C—D and E—G may be initiated by extensional membrane stresses set up in a continent as a result of the change of its radius of curvature due to latitudinal motion of the Earth (Turcotte and Oxburgh, 1973; Turcotte, 1974). However, once a rift or a rift system is established, the evolution that follows is essentially the same as those depicted in Fig. 1. Probably the only difference between the initial configurations 1A and B is that 1B will have no pregraben doming. But to show that rifting is actually the result of membrane stresses, one should prove, using paleomagnetism, e.g., the latitudinal motion of the continent in which rifting has occurred and the sense of this motion (i.e., the motion was from a pole toward the equator and not vice versa).
Transform segments of ridge/transform intersections during the initial breakup of a continental mass may be quite substantial in size (e.g., the DeGeer Line between the Barents Shelf and Greenland; the Agulhas Fracture Zone between the Malvinas Plateau and South Africa), and until the two separating continents completely clear each other, there will be an ever-shortening (along strike) strike-slip regime affecting the transform segments. Grabens may form at this stage at various angles to the transform domain as tension gashes and/or rotated and opened Riedel and anti-Riedel shears (see Wilson, 1960; Tchalenko, 1970) (Fig. 1H ). As the strike-slip fault finally materializes as a very narrow fault zone, along which the two diverging continents will clear, grabens will be cut and left, on the resulting transform margin, as aulacogens opening into the ocean (Fig. 1I ). The Mesozoic grabens on- and offshore in South Africa (Cape Grabens, Fig. 2) (Rigassi and Dixon, 1972; Dingle, 1976) are probably related to shear created by South America sliding past Africa along the Agulhas Fracture Zone. Pregraben stratigraphy here is preserved within the graben trough, indicating that there was no pregraben doming as a prelude to rifting (Burke, 1976).
Yet another way in which an aulacogen may form is depicted schematically in Figs. 1J-K . If a continental piece were rifted and rotated for a limited amount from the main continent (e.g., the Iberian Peninsula's rifting away from Eurasia to open the Bay of Biscay; Van der Voo, 1969), the resulting rift scar (Fig. 1K ) may be left as an aulacogen after this activity becomes dormant. Burke and Dewey (1974) argued that the Benue Trough (Fig. 2) probably opened in response to an aborted attempt, during the Barremian/Aptian of northwestern Africa, to rift and rotate away from the rest of the continent (for a detailed, recent review of the Benue aulacogen, see Ajakaiye and Kogbe, 1981).
Evolution and Significance of Aulacogens
Following initial rifting, graben facies (Bird and Dewey, 1970) rocks, composed predominantly of marginal fanglomerates, arkoses, mudstones, and locally, playa evaporites, begin to be deposited within the main trough of the rift. These overlie, usually unconformably, the prerift rocks. A knowledge of the prerift sections beneath the graben facies, plus the detailed stratigraphy of the marginal fanglomerates, may enable one to see whether any pregraben doming occurred in the area. For example, the occurrence of the Nubian sandstone section beneath the Suez Graben (T. Thompson, pers. comm., in Burke, 1977) and the preservation of the Robberg Formation beneath the Uitenhage Rift in South Africa (Rigassi and Dixon, 1972) show that there was no prerifting up-arching over the sites of the future graben development. If uplift had occurred these thin (<400 m) sedimentary sequences would have been eroded off. In the case of the Upper Rhine Graben, the conglomeratic marginal facies of the Pechelbronn beds, the Küstenkonglomerate (coastal conglomerates) contain Jurassic clasts in their lower parts and increasingly older clasts up the section indicating the progressive uplift and unroofing of the graben shoulders following the medial Eocene rifting (Şengör et al., 1978).
Subsequent to the formation of ocean, the continental margin, on which the aulacogen is located, will subside and shelf sediments will be deposited on it. The aulacogen at this stage may localize continental drainage and nucleate large deltas at its mouth, as has happened to the Mississippi Embayment (Cenozoic deltaic progradation, McGookey, 1975) and the Benue Trough (Niger Delta; Burke, 1972) (Figs. 1D, G, I ), for delta location in relation to rift tectonics see Audley-Charles et al., (1977). Such large, pre-orogenic deltas found in the stratigraphy of an orogenic belt against a rift striking at high angles to it (Fig. 3) may help identify the rift as an aulacogen (e.g., in the Mackenzie Mountains of Canada; Şengör et al., 1978), as opposed to an impactogen, e.g. (see Impactogen).
Burke (1975) has suggested that because ocean opening is a diachronous event, seawater may spill from the more advanced sectors into regions still at the rift stage and form very thick salt deposits (e.g., the Sergipe Basin; Asmus and Ponte, 1973). The most important role of such salt deposits seems to be in trapping oil and gas generated and preserved within the graben facies rocks, as in the Cabinda Rift.
Whether or not doming predates rifting, graben formation is almost always followed by the uplift of graben shoulders due to a rifting-induced pressure drop and consequent partial melting and volumetric expansion in the underlying mantle. After the rift becomes inactive, these shoulders will reverse and subside, much like the young Atlantic-type continental margins. This reversal may induce an axis-perpendicular compression onto the rift contents and result in axis-parallel folding of these (Hoffman et al., 1974) (Fig. 4). This is one way to induce folding onto the contents of the aulacogen, a process originally noted by Borissjak (1903) and Shatsky.
When orogeny begins at the continental margin with which an aulacogen is associated, either because of beginning of subduction or collision of this margin with another continent or island arc, the aulacogen enters a new phase of its evolution that is characterized by renewed tectonic instability. In Fig. 4 it is assumed, for graphic simplification, that this activation is in the form of rerifting. The normal faults associated with this phase were drawn to be precisely the same as those that had led to the initial down-faulting of the aulacogen; this situation is highly unlikely, but it illustrates the principles involved. The new rifting phase will have the same characteristics as the previous one, such as alkaline to peralkaline or even tholeiitic vulcanism, new graben facies development, renewed shoulder uplift, etc., with the important difference that the nearby mountain range will now provide an independent sediment source. In the case of the Athapuscow Aulacogen (Hoffman, 1973) and the Southern Oklahoma Aulacogen (Ham and Wilson, 1967), coarse clastic sediments eroded from the rising collisional orogen have been deposited in a trough larger than the original aulacogen.
Reactivation of an aulacogen may happen by compressional deformation of the rift contents, if the aulacogen strikes at high angles to the direction of convergence of the associated orogen. The Timan Mountains in northwestern Soviet Union, a compressed aulacogen related to the Ural collision, is an example of such compressional reactivation of aulacogens (Siedlecka, 1975). Strike-slip faulting parallel or subparallel with the axis of an aulacogen may also occur during the reactivation of the structure, again depending on the angle between the axis of the aulacogen and the direction of convergence. Wickham et al. (1975) recognized such large-scale strike-slip faulting in the Southern Oklahoma Aulacogen associated with the Ouachita collisional orogenesis.
Finally, one of the more important characteristics of aulacogens that was first noted by Borrisjak (1903) and by Shatsky (1955)—namely, their repeated reactivations—may be ascribed to the observation that oceans tend to open and close roughly along the same places (Wilson, 1966) and that marginal structures such as aulacogens may serve to nucleate new ones as well. This situation also introduces the added complexity of confusing reactivated aulacogens with impactogens, rifts first formed by continental collision (see Impactogen ). Repeated compressional reactivation of aulacogens also frequently occurs if a number of orogenic belts of different ages evolve parallel with one another and with the trend of the aulacogen along the margin of the craton on which the aulacogen is located. For example, the Dnyepr-Donetz aulacogen was compressed at three different times during its history: in the late Paleozoic as a result of the Hercynian collisions along the southern periphery of the East European Platform; in the late Triassic-early Jurassic because of the Cimmeride collisions; and in the late Mesozoic-early Cenozoic owing to the Alpide compressional events (Şengör, 1984). Therefore the repeated compressional reactivations of the aulacogens are also caused by Wilson-cycle-related events along the margins of cratons.
Aulacogens are important structures from a purely scientific viewpoint because they preserve the valuable pre- and syn-rifting stratigraphic record of now vanished oceans. Such early records are usually obliterated beyond recognition by the eventual collisional orogeny that marks the end of the life cycle of an ocean. They are also important economically because they often contain rich resources of petroleum. Today they are being actively studied all over the world, and future studies no doubt will continue to enlarge our data bank and deepen our understanding of these very interesting structures.
Shchipansky A.A., Bogdanova S.V., 1996. The Sarmatian crustal segment: Precambrian correlation between the Voronezh Massif and the Ukrainian Shield across the Dniepr–Donets Aulacogen. Tectonophysics 268 (1–4): 109–125.
Hogan J.P., Gilbert M.C., Price J.D., 2000. Crystallisation of fine– and coarse–grained A–type granite sheets of the Southern Oklahoma Aulacogen, USA. T Roy Soc Edin–Earth 91: 139–150 Part 1–2.
Veis A.F., Larionov N.N., Vorob'eva N.G., et al., 2000. Significance of microfossils for Riphean stratigraphy of the southern Urals (Bashkirian meganticlinorium) and adjacent region (Kama–Belaya aulacogen). Stratigr Geo Correl+ 8 (5): 423–446.