Genesis of an unusual clastic dike in an uncommon braided river deposit

Su Dechen , A. J. (Tom) van Loon , Sun Aiping

1. Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, China

2. State Key Laboratory of Continental Tectonics and Dynamics, Beijing 100037, China

3. Geological Institute, Adam Mickiewicz University, Maków Polnych 16, 61-606 Poznan, Poland

1 Introduction*

Clastic dikes, now also known as “injectites”; see,among others Satur and Hurst (2007), Rodrigueset al.(2009), Scottet al.(2009), Kane (2010), Whitmore and Strom (2010), Hurstet al.(2011) are soft-sediment deformation structures (SSDS) that occur frequently in sediments that have been deposited subaqueously, particularly in successions with a high sedimentation rate. They tend to be massive, without an internal structure (Peterson,1968; Hiscott, 1979, although more or less developed flow structures may be present (Harms, 1965; Scottet al.,2009); some authors claim distinct banding of 1-10 mm(Diggs, 2007), due to differences in grain size (Archer,1984) or grain alignment (Hillier and Cosgrove, 2002).Most commonly they are built of medium and fine sand(Lowe, 1975; Jolly and Lonergan, 2002; Obermeieret al.,2002); they are rarely coarser and truly coarse-grained dikes (with boulder-sized clasts) are exceptional (Hubbardet al., 2007). As a rule, they result from the sudden upward movement of a water/sediment mixture (Van Loon,2009), derived from a pore-water containing layer, through the overlying sediments, into the direction of lowest pressure,i.e. towards the sedimentary surface. At the surface,the water/sediment mixture flows out and the water mixes with the water body in which the sediments are deposited. The expelled sedimentary particles either form mud or sand volcanoes at the sedimentary surface (Van Loon,2009, 2010; Van Loon and Maulik, 2011) or become reworked at the sedimentary surface by currents or waves.

It is also possible (though much rarer), however, that dikes develop in a downward direction, for instance when faulting results in tension fissures that develop normal to the least direction of stress; such fissures may be filled by mobilized sand flows (Pratt, 1988; Schlische and Ackermann, 1995; Matsuda, 2000; Roweet al., 2002; Silver and Pogue, 2002; Le Heron and Etienne, 2005). Large examples have been described by Netoff (2002) from the Jurassic in Utah (USA), by Gozdzik and Van Loon (2007) from the Quaternary in Poland, by Parizeet al. (2007) from the Cretaceous of France, by Whitmore and Strom (2010)from the Permian in Arizona, and by Luninaet al. (2012)from the Holocene of Hungary.

Many clastic dikes are related to seismic activity (Obermeier, 1996, 1998; Obermeieret al., 2002; Luninaet al.,2012). This is because acoustic waves pass through soft rock during an earthquake event; lasts some 0.8-2 s,which is of the same order as the duration of the formation of clastic dykes, which propagate under turbulent flow conditions with velocities of some 4-65 m/s (Leviet al.,2011). The propagation of clastic earthquake-induced clastic dikes thus couples two processes: fracture propagation(see Anderson, 1995) and the flow of clastics in a predominantly fluidized way.

Many studies regarding clastic dikes pay attention to their geometry (Hurstet al., 2011). The dikes may be straight and planar (e.g.Taylor, 1982), curved (Parizeet al., 2007), and the dikes may be tapering (Rodriguez-Pascuaet al., 2000) or bifurcating (Hubbardet al., 2007) or even show ptygmatic folding (Surlyk and Noe-Nygaard,2001). It is remarkable that all these studies consider the dikes as 2-D features, probably because, particularly in lithified sediments, 3-D exposures are scarce. It should be kept in mind, however, that dikes are, by definition, 3-D bodies. In the example described here, we must conclude that the dike is not extended into two directions (i.e., it does not have the form of a sheet), but rather that it has a columnar geometry.

A prerequisite for the formation of upward-directed clastic dikes is that a sufficiently high pressure (an “abnormal” pressure according to Montenatet al., 2007) is built up to be able to make a flow break through the overlying sediment (Jolly and Lonergan, 2002; Hurstet al.,2003; Chenet al., 2009). This happens as soon as the fluid pressure surpasses the lithostatic pressure exerted by the weight of the overburden (Rodrigueset al., 2009). Such a process is not self-evident, as pore water tends, under common conditions, to seep gradually to the sedimentary surface, through the pores of the overlying sediment.However, when the overlying sediment is not sufficiently or not at all-permeable, thus forming some kind of seal,such seepage cannot take place (or only extremely slowly),so that the pore water pressure in the water-containing buried layer (or layers) increases, until it is high enough to make a flow break through the overlying sedimentary seal.This can occur when only a thin sedimentary cover has been built up, and this will result in small sediment/fluidescape structures, which are commonly found in, for instance, shallow-marine sediments where sandy and clayey layers alternate. Such structures are even more common in lacustrine deposits (e.g.Moretti and Sabato, 2007; Tasgin,2011; Luninaet al., 2012), particularly the Pleistocene glaciolacustrine deposits (Brodzikowskiet al., 1987; Chungaet al., 2007).

Much rarer are escape structures-and consequently much larger sedimentary dikes-in coarse-grained, poorly sorted sediments such as those deposited by braided rivers.This type of river is characteristic in areas where the river gradient is relatively high, a situation which tends to result in a high erosion rate; the consequence is a poorly sorted sediment through which pore water from buried layers can easily escape.

Escape structures and clastic dikes are even more rarely found in loesses: the silt-sized material tends to be insufficiently permeable to allow seepage from buried sediment.Loess is principally deposited by wind, and pore water is therefore not necessarily present. Deposition tends to result in an undulating surface, however, so rain water accumulates in the lowermost part and may become covered by new loess layers; such new loess will become wet in its lower part, but if sufficient new loess is deposited immediately afterwards, the upper part will be dry. The consequence is that some moist loess may be included in overall dry loess, but even in such a case escape structures are rare. This may also be due to the fact that much of the rain and run off water that may concentrate in loess depressions will evaporate before new loess deposition prevents further evaporation.

As a result of the processes described above, clastic dikes in braided river deposits are rare, and clastic dikes breaking through a loess body are also rare. The occurrence of such a feature is therefore remarkable.

2 The clastic dike

The dike under study is not only remarkable because it penetrates loess-like silty sediment (see Section 2.2), but it is also uncommon in that it contains large clasts (the largest clast measured had a visible length of some 30 cm). The dike was found in August 2011 along a newly constructed section of road G108 (near 75 km) in the Fangshan District of the Beijing area, China. Its position is 39°47’37”N and 115°46’31”E at an altitude of 280 m above sea level (Fig.1). The base of the dike is, unfortunately, no longer exposed, because a wall has been constructed (Fig. 2).

Fig. 1 Satellite and geological map of the Fangshan area with the clastic dike. 1-Mesoproterozoic (mainly dolomites of the Wumishan Formation); 2-Lower Paleoproterozoic limestones; 3-Upper Paleoproterozoic; 4-Mesozoic volcaniclastics; 5-Pleistocene;6-Holocene; 7-Mesozoic intrusion; 8-Fault.

2.1 Description of the dike

The dike, which is now partly hidden behind a wall that hides its lower part (Fig. 2) extended before the wall was constructed to the level of the road and must therefore be at least 8.5 m height; it is about 60 cm wide at its base and some 40 cm in its top part. The boundaries between the dike and the surrounding loess-like material (for the sake of simplicity called “l(fā)oess” in the following text) and gravel are very clear and almost vertical.

From bottom (Fig. 3) to top (Fig. 4), the dike gradually contains less large clasts, until only a discontinuous“vertical string” of clasts remains; it cannot be excluded,however, that this apparent upward diminishing amount of clasts should be ascribed to the columnar character of the dike, which seems not entirely straight but a bit irregular. The size of the gravel clasts ranges from a few millimeters to 30 centimeters; the clasts do not show any sorting and their shapes range from very angular to subrounded. The gravel in the dike consists entirely of dolomite, indicating that the clasts are derived from the Mesoproterozoic Wumishan Formation, which is the only rock unit under the loess, and is exposed where the loess has been eroded away.

The dike remained practically unchanged after it had been found in August 2011 until a heavy flood occurred on July 21, 2012 (Fig. 2 A, 2B). It appeared in September 2012 that the upper part of the dike had, probably as a consequence of the flood, been largely washed away (Fig.2C), leaving only the already mentioned vertical string of stones (Fig. 4).

2.2 Description of the host sediment

The dike breaks through a unit that consists largely of well sorted silt-sized material that shows all the characteristics (including the formation of vertical cliffs) of true loess deposits. It can not, however, be a primary, purely aeolian,loess, as strings (Fig. 5), horizons and discontinuous layers of pebble-to cobble-sized clasts occur, as well as isolated clasts. The clasts were eroded from the tilted Mesoproterozoic, Wumishan Formation, which underlies the loess-like host sediment.

The clast-rich layers in the silty host sediment may be considered as breccias or conglomerates, as some of the clasts are rounded to subrounded; whereas most clasts are subangular to (mainly) very angular. For the sake of simplicity, we will call them “conglomerates”. The conglomerates are poorly sorted,i.e.they consist of clasts of all sizes (up to a maximum of 30 cm), but they contain relatively little sand-sized material; the matrix consists mainly of the same loess that the dike intruded, and in which the conglomerate layers and lenses are embedded.

The lack of rounding of the clasts indicates that they were not transported over long distances. This is supported by the anomalously low amount of sand in the conglomerate layers: there was apparently insufficient transport to abrade the clasts to a considerable degree.

The uniform grain-size of the silty material obscures any original sedimentary structures that may be present.Some faint indications of lamination are thought to occur locally, commonly in the form of parallel lamination,exceptionally as cross-bedding; this almost exclusively occurs within the fine-grained parts of the conglomerates(Fig. 6), and extremely rarely in the silty material. It can,however, not be excluded that the search for such structures in the “l(fā)oess” has biased the researchers. One might therefore describe the fine-grained material most properly as massive.

Fig. 2 Overview of the clastic dike. A- The dike in its geological context, as present on June 12, 2012. B- Closer view on August 19,2011; C- Closer view on July 3, 2012; D- Closer view on September 6, 2012.

3 Interpretation

A sound interpretation of the genesis of the clastic dike requires insight into the conditions under which it was formed. For this reason, the age and the origin of the host sediment is interpreted first.

Fig. 3 The lower part of the dike, as far as still visible on September 6, 2012, with very densely packed clasts of different(mostly large) sizes, showing no preferred orientation. All clasts are dolomites of the Wumishan Formation.

Fig. 4 The top part of the dike on September 6, 2012, consisting almost exclusively of a vertical string of clasts that are still of decimeter size, but distinctly smaller than those in the bottom part. All clasts are the Wumishan dolomites.

3.1 Age of the host sediment and the clastic dike

The loess was deposited during the Malan interval,which belongs to the Late Pleistocene (Q3 in Fig. 1B) (Lü Jinbo and Li Wei, 2000). The thermoluminescence age of 33-35.6 ka and the carbon age (of the calcium carbonate)of 23 ka (Lu Yanchouet al., 1987) of the Malan loess at the Zhaitang section indicate that the host material is very young. Obviously, the dike must be even younger, but no precise dating is possible.

3.2 Genesis of the host sediment

The joint, common occurrence of clasts of several decimeters large in layers of only centimeters to a few decimeters thick, alternating with decimeter to meter thick layers(Fig. 7) of silt requires specific depositional conditions.Most similar occurrences are interpreted as glacial diamictons and as mass-flow deposits. Neither interpretation applies here.

A genesis as a glacially deposited diamicton (till) can be excluded for several reasons. The most important is that diamictons do not show alternating layers of gravel and silt, but true mixtures of the various particle sizes. Another argument is that the clasts do not show glacial striae.Moreover, a land-ice cover tends to bring material from more or less remote areas, but all large clasts are of local nature. Finally, no other morphological or sedimentological signs of a glaciation are present, nor has this type of sediment (here or elsewhere in the Fangshan District) been described as glacial.

A genesis as a mass-flow deposit can also be excluded. Low-viscosity flows such as turbidites and mudflows destroy primary sedimentary structures (and sometimes produce new structures) because of complete mixing of the original sediment, but cannot form new gravel layers in fine-grained sediment. High-viscosity flows such as slumps can preserve some of the original layering, but results-due to the processes involved-in bent, commonly broken up layers, if any layering is preserved at all.

Fig. 5 Thin strings of pebbles are common in the silty host rock, excluding an exclusively aeolian depositional environment.

Fig. 6 Vague parallel lamination in a relatively coarse part of the fine-grained host material.

The succession must therefore be explained in another way. The normal stratification of the gravelly layers and lenses indicates that “normal” transport media (water, ice,wind) must have been involved. The size of the clasts excludes wind, and the absence of a glacial context and of glacial characteristics excludes ice. The material thus must have been transported and deposited by water. The presence of large clasts indicates that this must have been running water. Running water capable of transporting large clasts may be shallow marine (e.g.deposition by longshore current in the neighborhood of a hard-rock cliff).Marine transport may also have taken place by waves,forming a pebble beach in a bay surrounded by hard-rock units. However, a marine origin must be excluded because of the position of the study site 280 m above level, since this would imply that the area would have been invaded by the sea before or during the Quaternary, and was uplifted at least 280 m during the past 35,000 years. No evidence is available for such a dramatic event. Moreover, the angular shape of the gravel clasts is inconsistent with a transgression conglomerate or a pebble beach. The conclusion must therefore be that the sediments have a continental origin.

Fig. 7 Characteristic facies of the host rock: pebble strings and pebble layers a few centimeters thick to a few decimeters thick, intercalated with decimeter to meter-thick silty layers.

Continental environments where large clasts are transported by water currents, while also transporting large amounts of silt-sized material, are scarce. They include fans, rivers and deltas. Fans are characterized by numerous channels that may be incised deeply, by a dominance of mass-transported sediments, and by a distinctly inclined surface. The sediments under study do not fulfill any of these requirements, so such a fan origin must be ruled out.More or less the same holds for a (lacustrine) delta. This leaves a river as the only possibility.

Meandering rivers are, as a rule, characterized by deep channels (with a gravel lag) and floodplains with finegrained overbank deposits. These subenvironments are situated beside each other, but ongoing accumulation of fluvial deposits and shifting of meanders may also result in vertical successions with alternating channel and floodplain deposits. However, in the sediments under study, the channels are far too small and shallow to be ascribed to a meandering river. Moreover, the small channels at the study site are incised in the silts that should represent the floodplain deposits, but in this case they should represent crevassesplay deposits. They are not, however, as crevasse-splay deposits are continuous and show a clear gradual decrease in thickness away from the main channel. The deposits thus cannot be explained as deposited by a meandering river,which leads inevitably to the conclusion that they must represent a braided river system.

If the silts at the study site were more poorly sorted sediments, the overall impression from the sedimentary succession would be that of a braided river deposit, indeed(Fig. 7). This leaves the question of how such an uncommon braided river deposit can have been formed. The answer cannot be found in the transport or depositional conditions, as these are similar for braided rivers all over the world. The solution must therefore be the lithological nature of the sediments in the source area, and this source area cannot have been far away, as indicated by the angular clasts.

As mentioned above, the clasts consist of local Mesoproterozoic material, mainly dolomites. These do,obviously, not provide quartz silt (even though some of the dolomites may contain minor amounts of fine siliciclastic particles). There is, however, one specific source that covers older rocks over extensive areas: true loess.It must thus be concluded that such loess deposits, which occur frequently in the neighborhood and which tend to be many meters thick, constitute the source of the siltsized material in the Quaternary deposits under study,although it cannot be excluded that the loess was deposited, entirely or partly, at or nearby the study site by wind.As sandstones or quartzites do not occur in the neighborhood, it is logical that quartz sand is not present in the deposits. The bimodal grain-size distribution of these uncommon Quaternary braided river deposits can thus be well explained.

3.3 Genesis of the clastic dike

The uncommon grain-size distribution of the host rock forms the clue for the genesis of the clastic dike. As mentioned above, braided river deposits are rarely intruded by clastic dikes, because pore water from buried layers can seep through the commonly poorly sorted and fairly permeable overlying braided river deposits to the sedimentary surface.

In the present case, this was, however, impossible because a pore-water containing gravel layer was apparently rapidly covered by an impermeable loess layer. Thus, pressure could gradually build up. The hydrostatic pressure became consequently so high that a sediment/water mixture was able to find a way upward through what may have been a zone of weakness. Because overpressure in the buried gravel was exceptionally high, large clasts could not only be transported laterally to the place where the break-through took place, but also upwards through the tunnel just formed through the silt-sized material. It is interesting in this context that the dike also cuts through a few gravelly layers,which may also have supplied both clasts and water, thus intensifying the process.

When the sediment/water mixture moved upwards,the hydrostatic pressure gradually decreased, resulting in a diminishing transport capacity. This is reflected by the smaller size of the clasts (although still up to over one decimeter; see Fig. 8) in the middle part of the dike, and even smaller in the upper part, which seems to narrow upwards, where the boundaries with the host rock also becomes less clear.The trigger for the sudden upward escape of the overpressurized sediment/water mixture cannot be reconstructed with any certainty. Possibly shock waves released during an earthquake were responsible, numerous researchers have not found that dikes are one type of soft-sediment deformation that commonly originates as a result of earthquakes. It should be noted in this context that 67 earthquakes have been documented for the Beijing area during recorded human history with magnitudes that are now interpreted to have exceeded a magnitude of 7.0.

4 Discussion of the sedimentary envi?ronment

We interpret the sedimentary environment as a braided river system (see Section 3.2). This system could develop because rain water ran periodically off the hills and mountains that surround the study site, and formed sheet floods in the more level parts of the area. These sheet floods carried clasts that were eroded from the Mesoproterozoic rocks,as well as loess that was encountered along the way. If this is the case, a characteristic braided river environment existed (Fig. 7).

It is also possible that the area was actually a more or less level place where aeolian deposition of loess prevailed. In that case, it might be considered as an aeolian environment occasionally reached by sheet floods resulting from drainage of rain water that fell in the surrounding hills and mountains. These sheet floods reworked the previously deposited loess. This only underlines that nature does not draw distinct boundaries between different environments, so it seems a matter of semantics is terms of how the environment is classified. Whatever choice may be made in this respect, it also underlines that only a more complete palaeogeographic picture might indicate whether the aeolian or the fluvial character prevailed most of the time.

5 Conclusions

Analysis of the environmental conditions of a succession built of loess-sized material with intercalations of gravel layers and horizons indicates that it must have accumulated in a braided river environment where aeolian activity was strong and may or may not have prevailed. A dominance of loess-like silty sediments in a braided river system is uncommon. This silty material must at least have been partly derived from elsewhere, but may partly also have been deposited at or close to the study locations.This implies an uncommon sedimentary environment that might be characterized as fluvio-aeolian.

Fig. 8 Clasts in the middle part of the dike.

The alternation of poorly sorted and highly permeable conglomerates and impermeable loess-like material led to increasing hydrostatic pressure in water-saturated sheet flood deposits, eventually resulting in the upward escape of pore water (with clasts), thus forming in the Fangshan District a large dike with exceptionally large clasts, which must also be considered as an uncommon phenomenon for a braided river succession.

Acknowledgements

The authors are indebted to Prof. Lü Jinbo, Mr. Duan Xu, and Mr. Guo Rongtao for valuable suggestions and for their help in the field. The second author is indebted to Prof. Feng Zengzhao and Prof. Bao Zhidong, who provided the financial means to come to China to carry out the fieldwork.

Anderson, T. L., 1995. Fracture mechanics (2nded.). Boca Raton;CRC Press, 688.

Archer, J. B., 1984. Clastic intrusions in deep-sea fan deposits of the Rosroe formation, lower Ordovician, western Ireland. Journal of Sedimentary Petrology, 54: 1197-1205.

Brodzikowski, K., Gotowala, R., Haluszczak, A., Krzyszkowski,D., Van Loon, A. J., 1987. Soft-sediment deformations from glaciodeltaic, glaciolacustrine and fluviolacustrine sediments in the Kleszczów Graben (central Poland). In: Jones, M. E., and Preston, R. M. F. (eds). Deformation of sediments and sedimentary rocks. Geological Society, London, Special Publication, 29:255-267.

Chen, J., Van Loon, A. J., Han, Z., Chough, S. K., 2009. Funnelshaped, breccia-filled clastic dykes in the Late Cambrian Chaomidian Formation (Shandong Province, China). Sedimentary Geology, 221: 1-6.

Chunga, K., Livio, F., Michetti, A. M., Serva, L., 2007. Synsedimentary deformation of Pleistocene glaciolacustrine deposits in the Albese con Cassano area (southern Alps, northern Italy), and possible implications for paleoseismicity. Sedimentary Geology,196: 59-80.

Diggs, T. N., 2007. An outcrop study of clastic-injection structures in the Carboniferous Tesnus Formation, Marathon Basin, Trans-Pecos Texas. In: Hurst, A., Cartwright, J. (eds). Sand injectites:implications for hydrocarbon exploration and production. AAPG Memoir: 209-219.

Gozdzik, J., Van Loon, A. J., 2007. The origin of a giant downward directed clastic dyke in a kame (Belchatów mine, central Poland).In: Gruszka, B., Van Loon, A. J., Zielinski, T. (eds). Quaternary geology - bridging the gap between East and West. Sedimentary Geology, 193: 71-79.

Harms, J. C., 1965. Sandstone dikes in relation to Laramide faults and stress distribution in the southern Front Range, Colorado.Geological Society of America Bulletin, 76: 981-1002.

Hillier, R. D., Cosgrove, J. W., 2002. Core and seismic observations of overpressure-related deformation within Eocene sediments of the Outer Moray Firth, UKCS. Petroleum Geoscience, 8: 141-149.

Hiscott, R. N., 1979. Clastic sills and dikes associated with deep-water sandstones, Tourelle Formation, Ordovician, Quebec. Journal of Sedimentary Petrology, 49: 1-10.

Hurst, A., Scott, A., Vigorito, M., 2011. Physical characteristics of sand injectites. Earth-Science Reviews, 106: 215-246.

Jolly, R. J. H., Lonergan, L., 2002. Mechanisms and controls on the formation of sand intrusions. Journal of the Geological Society,London, 159: 605-617.

Kane, I. A., 2010. Development and flow structures of sand injectites: the Hind Sandstone Member injectite complex, Carboniferous, UK. Marine and Petroleum Geology, 27: 1200-1215.

Le Heron, D. P., Etienne, J. L., 2005. A complex subglacial clastic dyke swarm, Sylheimaj?kull, southern Iceland. Sedimentary Geology, 181: 25-37.

Levi, T., Weinberger, R., Yyal, Y., 2011. A coupled fluid-fracture approach to propagation of clastic dykes during earthquakes. Tectonophysics, 498: 35-44.

Lowe, D. R., 1975. Water escape structures in coarse-grained sediments. Sedimentology, 22: 157-204.

Lu, Y. C., Mortlock, A. J., Price, D. M., Readhead, M. L., 1987. Thermoluminescence dating of coarse-grain quartz from the Malan loess at Zhaitang section, China. Quaternary Research, 28: 356-363.

Lü, J. B., Li, W., 2000. Research history of Cenozoic erathem, Beijing. Beijing Geology, 1: 1-7 (in Chinese with English abstract).

Lunina, O.V., Andreev, A.V., Gladkov, A. S., 2012. The Tsagan earthquake of 1862 on Lake Baikal revisted: a study of secondary coseismic soft-sediment deformation. Russian Geology and Geophysics, 53: 594-610.

Matsuda, J.-I., 2000. Seismic deformation structures of the post-2300 a BP muddy sediments in Kawachi lowland plain, Osaka,Japan. Sedimentary Geology, 135: 99-116.

Montenat, C., Barrier, P., d’Estevou, P. O., Hibsch, C., 2007. Seismites: an attempt at critical analysis and classification. Sedimentary Geology, 196: 5-30.

Moretti, M., Sabato, L., 2007. Recognition of trigger mechanisms for soft-sediment deformation in the Pleistocene lacustrine deposits of the Sant’Arcangelo Basin (southern Italy): seismic shock vs.overloading. Sedimentary Geology, 196: 5-30.

Netoff, D., 2002. Seismogenically induced fluidization of Jurassic erg sands, southcentral Utah. Sedimentology, 49: 65-80.

Obermeier, S. F., 1996. Use of liquefaction-induced features for palaeoseismic analysis—an overview of how seismic liquefaction features can be distinguished from other features and how their regional distribution and properties of source sediment can be used to infer the location and strength of Holocene Paleo-earthquakes. Engineering Geology, 44: 1-76.

Obermeier, S. F., 1998. Liquefaction evidence for strong earthquakes of Holocene and latest Pleistocene ages in the states of Indiana and Illinois, USA. Engineering Geology, 50: 227-254.

Obermeier, S. F., Pond, E. C., Olson, S. M., Green, R. A., 2002. Paleoliquefaction studies in continental settings. In: Ettensohn, F. R.,Rast, N., Brett and C. E. (eds). Ancient seismites. Geological Society of America Special Paper, 359: 13-27.

Parize, O., Beaudoin, B., Champanhet, J.-M., Friès, G., Imbert, P.,Labourdette, R., Paternoster, B., Rubino, J.-L., Schneider, F.,2007. A methodological approach to clastic injectites: from field analysis to seismic modeling—examples of the Vocontian Aptian and Albian injectites (southeast France). In: Hurst, A., Cartwright, J. (eds). Sand injectites: implications for hydrocarbon exploration and production. AAPG Memoir, 87: 173-183.

Peterson, G. L., 1968. Flow structures in sandstone dikes. Sedimentary Geology, 2: 177-190.

Pratt, B. R., 1998. Molar-tooth structure in Proterozoic carbonate rocks: origin from synsedimentary earthquakes, and implications for the nature and evolution of basins and marine sediment. Geological Society of America Bulletin, 110: 1028-1045.

Rodrigues, N., Conbbold, P. R., L?seth, 2009. Physical modelling of sand injectites. Tectonophysics, 474: 610-632.

Rodriguez-Pascua, M. A., Calvo, J. P., Vicente, G. D., Gomez-Gras,D., 2000. Soft-sediment deformation structures interpreted as seismites in lacustrine sediments of the Prebetic Zone, SE Spain,and their potential use as indicators of earthquakes magnitudes during the Late Miocene. Sedimentary Geology, 135: 117-135.

Rowe, C. A., Mustard, P. S., Mahoney, J. B., Katnick, D. C., 2002.Oriented clastic dike swarms as indicators of paleoslope? An example from the upper Cretaceous Nanaimo Group, Canada. Journal of Sedimentary Research, 72: 192-200.

Satur, N., Hurst, A., 2007. Sand-injection structures in deep-water sandstones from the Ty formation (Paleocene), Sleipner ?st Field, Norwegian North Sea. In: Hurst, A., Cartwright, J. (eds).Sand injectites: implications for hydrocarbon exploration and production. AAPG Memoir: 113-117.

Schlische, R. W., Ackermann, R. V., 1995. Kinematic significance of sediment-filled fissures in the North Mountain Basalt, Fundy Rift Basin, Nova Scotia, Canada. Journal of Structural Geology,17: 987-996.

Scott, A., Vigorito, M., Hurst, A., 2009. The process of sand injection: internal structures and relationships with host strata (Yellowbank Creek Injectite Complex, California, USA). Journal of Sedimentary Research, 79: 568-583.

Silver, M. H., Pogue, K. R., 2002. Analysis of plan-view geometry of clastic dike networks in Missoula flood slackwater sediments(touchet beds), southeastern Washington. Geological Society of America Abstracts with Programs, 34: 24.

Surlyk, F., Noe-Nygaard, N., 2001. Sand remobilisation and intrusion in the Upper Jurassic Hareelv Formation of East Greenland.Geological Society of Denmark Bulletin, 48, 169-188.

Taylor, B. J., 1982. Sedimentary dykes, pipes and related structures in the Mesozoic sediments of south-eastern Alexander Island.British Antarctic Survey Bulletin, 51: 1-42.

Tasgin, C. K., 2011. Seismically-generated hydroplastic deformation structures in the Late Miocene lacustrine deposits of the Malatya Basin, eastern Turkey. Sedimentary Geology, 235: 264-276.

Van Loon, A. J., 2009. Soft-sediment deformation structures in siliciclastic sediments: a review. Geologos, 15: 3-55.

Van Loon, A. J., 2010. Sedimentary volcanoes: overview and implications for the definition of a “volcano” on Earth. In: Ca?ón-Tapia, E. and Szakacs, A. (eds). What is a volcano? New answers to an old question. Geological Society of America Special Paper,470: 31-41.

Van Loon, A. J., Maulik, P., 2011. Abraded sand volcanoes as a tool for recognizing paleo-earthquakes, with examples from the Cisuralian Talchir Formation near Angul (Orissa, eastern India).Sedimentary Geology, 238: 145-155.

Whitmore, J. W., Strom, R., 2010. Sand injectites at the base of the Coconino Sandstone, Grand Canyon, Arizona (USA). Sedimentary Geology, 230: 46-59.