Deformed stromatolites in marbles of the Mesoproterozoic Wumishan Formation as evidence for synsedimentary seismic activity

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

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

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

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

1 lntroduction＊

Stromatolites are structures left by masses of microorganisms, living mainly in shallow-marine environments where carbonates precipitated, although they are now also well known from siliciclastic sediments (Noffke, 2010;Noffke and Chafetz, 2012).The oldest stromatolites,which were built by particles that were trapped by what were most probably symbioses of organisms that now are jointly called cyanobacteria, date from the Archean; they represent the oldest macroscopically identif i able traces of life on Earth.The structures that these micro-organisms tend to leave consist, as far as visible in the field, of irregular laminae (e.g.,Horodyski, 1975; Hips and Haas, 2006;Clarke and Stroker, 2013).The irregular surface of the organic material, which may have been deformed by wave action or other processes in shallow-marine environments,has resulted in a mineral fabric that is commonly characterized by the occurrence of flat particles (often micas)that were originally not all orientated parallel to the overall lamination; during burial metamorphism, the original fabric is commonly changed, however, such that (almost)all micas and other platy particles are directed perpendicular to the main stress direction.

Stromatolites have been described mostly from the Phanerozoic (modern examples can be found in, for instance, Shark Bay, Australia).Archean examples are relatively rare but they have been investigated fairly frequently because they represent the earliest macroscopically recognizable forms of life (e.g.,Noffke, 2010, 2012); yet, the organic origin of specif i c stromatolitic structures has been much debated, but geochemical and isotopic investigations supported time and again a biogenic origin (e.g.,Von Blanckenburget al., 2008; Lepotet al., 2009, 2013; Craiget al., 2013).Proterozoic occurrences are more common,but have less frequently been described in detail (e.g.,Bekker and Eriksson, 2003; Bekkeret al., 2003; Van Loon and Mazumder, 2013); this must probably be ascribed to the fact that the volumes of rocks from a specif i c period become less frequent with time, due to ongoing erosion,but also due to processes (like metamorphism)that destroy the original structures in a rock in whole or in part, so that structures such as stromatolites become diff i cult to recognize.

Here we describe an example of well-preserved Mesoproterozoic stromatolites.They are extraordinary in that(1)they are old, (2)they occur in limestones that have turned into dolomites by diagenetic processes that commonly obscure sedimentary structures and fossils, (3)the dolomites have been turned into marbles as a result of thermal metamorphism caused by the nearby intrusion of a granite, and (4)the stromatolites have been deformed in a way that must be ascribed to seismic activity, conf i rming that the area was affected by tectonic processes during the time-span that the stromatolites formed, as must be deduced from the fact that the deformation took place when the sediment had not yet been lithif i ed and was still in a plastic state.

2 Geological setting

The Yan-Liao aulacogen (Fig.1), which is situated in the northern part of the Sino-Korean palaeo-plate (North China Craton), is a vast area where sediments accumulated during the Meso- and Neoproterozoic, building up successions of up to 9200 m thick (Heet al., 2000; Qiao and Gao, 2007).The thickest succession is found in the center of the aulacogen, more precisely in the Jixian section.This area can be considered as part of a rift system where the Columbia supercontinent started break-up between 1600 Ma and 1200 Ma (Rogers and Santosh, 2002; Qiaoet al.,2007).This break-up was accompanied by earthquakes that resulted in seismic shocks that triggered soft-sediment deformation (Qiaoet al., 2007; Ettensohnet al., 2011).The resulting structural elements are prominent: they were possibly reactivated even much later, when folding and faulting in the Beijing area occurred during the Late Jurassic-Early Cretaceous Yanshanian orogeny.

The geology of this area has been studied already some 80 years ago (Kaoet al., 1934).The Meso-Neoproterozoic rocks are now grouped into twelve formations:from bottom to top the Mesoproterozoic Changzhougou,Chuanlinggou, Tuanshanzi, Dahongyu, Gaoyuzhuang,Yangzhuang, Wumishan, Hongshuizhuang, Tieling, and Xiamaling Formations, which are overlain by the Neoproterozoic Longshan and Jingeryu Formations.The Neoproterozoic is unconformably overlain by limestones of Cambrian age.

The ages of the various formations, and thus of the softsediment deformation structures (SSDS)that they contain,have been well established in the past few decades.The Wumishan Formation, in which the deformed stromatolites described below are present, has been dated on the basis of SHRIMP U-Pb zircon ages as between 1550 Ma and 1450 Ma (Gaoet al., 2007; Suet al., 2008; Liet al.,2010; Suet al., 2010),i.e., Early Mesoproterozoic.

Seismic activity

A fault of over 800 km long runs from southwest Shijiazhuang to northeast Lingyuan and Tieling (Fig.1),roughly following the axis of the Yan-Liao aulacogen.This Shijiazhuang-Lingyuan Fault became active during the Early Mesoproterozoic (Heet al., 2000); it probably was the main trigger for regional earthquakes in the geological past (Qiao and Gao, 2007; Su and Sun, 2012).The first earthquake-triggered SSDS that have been reported from China are situated close to this fault (Song, 1988).Since then, earthquake-triggered SSDS have been found at several other sites in the same region (e.g.,Lianget al.,2002; Qiao and Gao, 2007; Ettensohnet al., 2011; Suet al., 2013).

It is for the above reasons commonly assumed now that the SSDS in the Wumishan Formation are controlled or at least in fl uenced by their distance to the Meso- to Neoproterozoic Shijiazhuang-Lingyuan Fault, as most sites exhibiting SSDS are situated less than some 20 km from the fault.

3 The Wumishan Formation

The Wumishan Formation (the thickest and most widely distributed formation of the Meso- and Neoproterozoic on the Yan-Liao aulacogen)crops out over about 12% of the Beijing area (somewhat over 2000 km2).It consists mostly of well-laminated dolostones (approx.89%)and silicif i ed dolostones or siliceous chert (10%).The remaining 1% are thin shale interbeds and terrigenous sediments(sandstones).The total thickness of the Wumishan Formation in the Beijing area is between 1500 m and 3495 m.

The formation is largely composed of stacked peritidal parasequences, each a few meters to tens of meters thick,separated by silty shales (Ettensohnet al., 2011).The parasequences are both fining and shallowing upwards.Subtidal high-energy cross-bedded dolarenites are present at the base.These pass up through lagoonal dolosiltites to low-energy peritidal laminites with mudcracks and stromatolites in the upper part of a parasequence.

On the basis of the stacking patterns, the rock composition, and the stromatolite characteristics, the Wumishan Formation is divided into four members.Member 1 is composed mainly of terrigenous sediments and bituminous dolostone; the thickness of Member 1 is about 450 m.The rock types of Member 2 vary distinctly, from quartz sandstone to sandy dolostone or laminated dolostone, refl ecting dramatic changes of the palaeo-environment; the thickness of Member 2 is about 400 m in the Ming Tombs area.Member 3 is the thickest member in this formation(854 m in the Ming Tombs area).The main rock types are laminated dolostone with abundant pillar/cone-shaped stromatolites.Member 4 is mainly composed of laminated thinly-bedded dolostone with banded cherty dolomite and occasional stromatolites.

The stromatolites indicate that the Wumishan Formation accumulated, apart from its basal part, essentially in a stable, shallow-water, peritidal platform environment where “normal” sedimentary structures (especially interference ripples)also indicate that the dolomite from the Wumishan Formation accumulated in a shallow-water environment.

4 The deformed Wumishan stromatolites

The deformed stromatolites under study occur about 4 km north of Taoyukou Village (GPS 40°15′46.57″N,116°26′21.66″E)in the Changping District (north of the Beijing area).They are exposed in the wall of the valley at the other side than where the road is situated.The site can be reached over some kind of concrete dam built to regulate the ephemeral stream that may be present.In the direct neighborhood of the study site, the Wumishan Formation still shows its usual appearance (Fig.2).The rocks are dolostones with locally levels with stromatolites (Fig.3), but also with levels where the rocks have been deformed, although the under- and overlying layers have not or hardly been affected (Fig.4).

At the study site, the dolostones have been changed into marbles.This must be ascribed to thermometamorphism in a halo around the granite that intruded the Wumishan For-mation during the Indo-Chinese Epoch (～250-200 Ma).

4.1 Description of the deformations

In spite of the metamorphism that they have undergone,the stromatolites in the marbles are still perfectly recognizable (Fig.5).They most commonly show a kind or irregular lamination, locally giving the impression of local upward accumulation of sediment (Fig.6).It appears,however, that some of the stromatolite layers have been deformed (Figs.7, 8).The deformations in many respects closely resemble the SSDS found nearby in the “normal”(= non-stromatolite)layers (Fig.4), but the typical architecture of the stromatolites also results in different forms of SSDS.

Most of the deformations of the stromatolites consist of irregular folds, sometimes accompanied by faults that do not affect the overlying layers; in a few cases, the faults extend downwards to maximally a few centimeters into the underlying, non-deformed layer; the underlying layer may or may not be a stromatolite.The compressional folds and the small upthrusts indicate that there was a lateral pressure, but even in a single deformed stromatolite unit the deformation intensity need not be the same at each level,suggesting that inhomogeneities within the unit absorbed the pressure responsible for the deformation in (slightly)different ways.

The pressure must, considering the shape of the deformations, have acted in a horizontal direction, affecting only the uppermost sediments.As the deformations are restricted to certain vertical intervals, with non-deformed intervals in between, it must be concluded that the stromatolites were still unlithif i ed when the deformation took place.They thus form soft-sediment deformation structures, and must have been deformed while still at the sedimentary surface.

4.2 Interpretation

Numerous processes are capable of deforming unconsolidated layers at the sedimentary surface (Van Loon,2009).The deformed sediments under study were, however, not “normal” deposits, but they must have been still fairly slimy microbialites to which calcareous particles became attached that occurred in suspension.The presence of the microbialites, which during the Mesoproterozoic most probably consisted of cyanobacteria, indicates that the water was shallow, as the cyanobacteria required light.The shallow-marine character, in turn, suggests that the surf i cial sediments, including the microbial mats, may have been affected by wave action and/or currents.

It is well known from present-day microbial (mainly algal)mats in the intertidal zone that they may easily become wrinkled or even folded by wave or current action.This results commonly in local deformations (mostly in the form of a “heap” of complexly deformed microbial material), and such deformations have been found, indeed,in the stromatolites under study (Fig.9).These types of agents seem unlikely for several horizons of the stromatolites under study, however, because their types of deformation are distinctly different from those found in recent microbial mats.Moreover, it seems that the affected layers became deformed over the total length of the exposure,which is uncommon for present-day microbial mats as they become usually broken up into pieces under the inf l uence of current and/or wave activity.

Another signif i cant aspect is, as mentioned above, that the deformation was apparently due to a pressure parallel to the bedding plane, and that this pressure affected only the uppermost decimeters.Also important is that the deformation events happened repeatedly, as proven by the alternation of undeformed and deformed layers.

All the above characteristics are consistent with those that result in shallow-marine sediments as a result of an earthquake-triggered longitudinal shock wave that proceeds through the uppermost sediments.The fact that it is well known that earthquakes that deformed sediments of the Wumishan Formation in the Yong valley occurred on average every 20,000-32,000 years (Su and Sun, 2011),makes it even more likely that the deformations under study are due to such tectonic activity.Consequently, the layers with extended deformations interbedded between undeformed layers should be considered as seismites.

5 Discussion

Two main questions arise from our observations: (1)Are the irregularly laminated sediments under study really stromatolites, and (2)Can processes other than seismic shocks be held responsible for the deformations?

5.1 Stromatolite origin

Stromatolites are remarkable features.They are considered as sedimentary structures rather than as fossils, but they are not sedimentary structures in a narrow sense: it is the type of substrate (a commonly irregular slimy mass)rather than the depositional process (settling from suspension, or moving in suspension)that results in accumulation, but this substrate itself is not preserved in the sedimentary record.This makes it diff i cult to prove that what are considered in the field as stromatolites are stromatolites, indeed, as no recognizable true fossil traces are left.In the field, stromatolites are, however, easily recognized:they look like stromatolites!

In specif i c cases, for instance where stromatolites have formed as a result of upwards growing columns of cyanobacteria, the “sedimentary” structures are convincing.Where the stromatolites formed as a cover on a calcareous or siliciclastic substrate, this is less clear.Nevertheless,the structures under study show all the characteristics that stromatolites tend to show in the field and, moreover, no good physical explanation can be found for the irregular type of lamination inside the stromatolites.Consequently,it seems beyond any reasonable doubt that the laminated structures are stromatolites.

Traces of life from the Mesoproterozoic are scarce.Stromatolites are the only features that are commonly considered to point at organic activity during the Mesoproterozoic.This makes an interpretation of the structures as stromatolites not self-evident.Yet, stromatolites are also known to have built comparable sediments much earlier,viz.during both the Palaeoproterozoic and the Archean.Palaeoproterozoic stromatolites that were deformed by seismic shocks and that show similar SSDS as the structures under study, have been described recently by Van Loon and Mazumder (2013)from the Singhbhum Craton in northeast India (Fig.10).

5.2 Origin of the deformations

The Wumishan Formation was formed mainly under relatively quiet conditions in a peritidal environment.The depth of the water varied, as did the precise environment,which ranged from marine below wave base to lagoonal.Sedimentary structures such as small-scale cross-bedding,wave and interference ripples,etc.record these conditions.Most probably heavy storms affected the sedimentary surface, but no tempestites have been described from the formation thus far.The conditions must consequently for most of the time have been fairly favorable for the development of microbial mats in the near-shore area.Such microbial mats, of which it is unknown whether they extended during the Mesoproterozoic for kilometers or were restricted to local developments of only a few meters wide, must have been affected by wave activity and by tidal currents.

Such agents must have destroyed at least sometimes the original “fabric” of the microbial mats, tearing them apart,dragging them along over the sea f l oor, and concentrating them to heaps that caused folds, wrinkles and possibly faults in the organic masses.Folds and wrinkles are commonly observed in the Wumishan stromatolites, indeed,but it is, as a rule, hardly possible to find out whether such deformations result from wave and/or current activity, or whether they were a result of the irregular surface of the organic masses themselves.

One of the characteristics of layers in which stromatolites occur that have been affected by the above “normal”processes, is that the type and intensity of the deformations are different from place to place, as the wave or current activity varies in space; the differences in deformation type and intensity are, moreover, superimposed on the irregular shapes of the microbial mats themselves.As a consequence, such stromatolite layers are characterized by deformations that are discontinuous and that seem to change laterally in type and intensity in a more or less haphazard way.

The deformed stromatolites under study show, however, a different picture.They form layers that are deformed all over, whereas the under- and overlying layers are undeformed.This implies that some events must have happened that affected the complete sedimentary surface, whereas the underlying layers were not affected.This excludes local tectonics.On the other hand, events that affect a sedimentary surface continuously in such a way that an entire layer becomes deformed, are rare.Even more so if the type of deformation indicates that there was not a vertical pressure (e.g.,by storm-wave loading)but a lateral one.The only process known that can be held responsible for such deformations is a longitudinal shock wave that proceeds through the uppermost sediments parallel to the sedimentary surface.Such shock waves are induced by sudden changes in the stress field, for instance by the impact of a meteorite or an earthquake of suff i cient magnitude.

The repeated occurrence of the deformed layers practically excludes impacts as a cause of the shock wave, as it would imply that, over a time-span of tens or even hundreds of thousands of years impacts must have occurred that all affected the same region.The most plausible explanation therefore seems to be the occurrence of earthquakes.Considering the presence of a large fault that was active during the Mesoproterozoic (see above), this explanation for the deformations seems by far the most plausible.

Interesting in this context is the occurrence of similarly deformed stromatolites in the Palaeoproterozoic Chaibasa Formation of NE India, from a region (the Singhbhum craton)that is also known to have been affected by earthquakes during the accumulation of the sedimentary unit in which stromatolite layers occur (Van Loon and Mazumder, 2013).These seismically deformed stromatolites show the same characteristics as those under study here.It thus seems that stromatolites are relatively easily deformed by seismic shocks, which is understandable because the deformation must have taken place when the microbial mats with the sedimentary particles attached to it, were still in a plastic state.

6 Conclusions

Stromatolites have been found in the Mesoproterozoic Wumishan Formation in the Changping District.They are recognizable in spite of a diagenetic change from limestone to dolostone and a later local thermometamorphisminduced transformation from dolostone into marble caused by the intrusion of a granite.

The stromatolites have partly been deformed.The deformation structures and the intercalation of deformed units between undeformed layers indicate that the deformation process took place before the stromatolites had become lithif i ed (soft-sediment deformation).

The deformation must be ascribed to longitudinal shock waves through the uppermost sediment layers.This can be explained satisfactorily only by the occurrence of earthquakes that triggered seismic shocks.

The occurrence of such earthquake-induced seismic shocks, in combination with the fact that the resulting deformations took place when the sediment had not yet been lithif i ed, proves that synsedimentary seismic activity took place.

Bekker, A., Eriksson, K.A., 2003.Paleoproterozoic drowned carbonate platform on the southeastern margin of the Wyoming Cra-ton: A record of the Kenorland breakup.Precambrian Research,120(3-4): 327-364.

Bekker, A., Karhu, J.A., Eriksson, K.A., Kaufman, A.J., 2003.Chemostratigraphy of Paleoproterozoic carbonate successions of the Wyoming Craton: Tectonic forcing of biogeochemical change?Precambrian Research, 120(3-4): 279-325.

Clarke, J.D.A., Stoker, C.R., 2013.Searching for stromatolites: The 3.4 Ga Strelley Pool Formation (Pilbara region, Western Australia)as a Mars analogue.Icarus, 224(2): 413-423.

Craig, J., Biff i, U., Galimberti, R.F., Ghori, K.A.R., Gorter, J.D.,Hakhoo, N., Le Heron, D.P., Thurow, J., Vecoli, M., 2013.The palaeobiology and geochemistry of Precambrian hydrocarbon source rocks.Marine and Petroleum Geology, 40: 1-47.

Ettensohn, F.R., Zhang, C.H., Gao, L.Z., Lierman, R.T., 2011.Softsediment deformation in epicontinental carbonates as evidence of paleoseismicity with evidence for a possible new seismogenic indicator: accordion folds.Sedimentary Geology, 235: 222-233.

Gao, L.Z., Zhang, C.H., Shi, X.Y., Zhou, H.R., Wang, Z.Q., Song,B., 2007.A new SHRIMP age of the Xiamaling Formation in the North China plate and its geological signif i cance.Acta Geologica Sinica, 81: 1103-1109.

He Zhengjun, Song Tianrui, Ding Xiaozhong, Zhang Qiaoda, Meng Xianghua, Ge Ming, 2000.Early synsedimentary faulting of the Meso-Proterozoic Yanshan rift and its inf l uence on event sedimentation.Journal of Palaeogeography, 2: 83-91 (in Chinese with English abstract).

Hips, K., Haas, J., 2006.Calcimicrobial stromatolites at the Permian-Triassic boundary in a western Tethyan section, Bükk Mountains, Hungary.Sedimentary Geology, 185(3-4): 239-253.

Horodyski, R.J., 1975.Stromatolites of the lower Missoula Group(Middle Proterozoic), Belt Supergroup, Glacier National Park,Montana.Precambrian Research, 2(3): 215-254.

Kao, C.S., Hsiug, Y.H., Kao, P., 1934.Preliminary notes on Sinian stratigraphy of North China.Bulletin of Geological Society of China, 13: 243-288.

Lepot, K., Benzerara, K., Rividi, N., Cotte, M., Brown Jr., G.E.,Philippot, P., 2009.Organic matter heterogeneities in 2.72 Ga stromatolites: Alteration versus preservation by sulfur incorporation.Geochimica et Cosmochimica Acta, 73(21): 6579-6599.

Lepot, K., Williford, K.H., Ushikubo, T., Sugitani, K., Mimura, K.,Spicuzza, M.J., Valley, J.W., 2013.Texture-specif i c isotopic compositions in 3.4 Gyr old organic matter support selective preservation in cell-like structures.Geochimica et Cosmochimica Acta, 112: 66-86.

Li, H.K., Zhu, S.X., Xiang, Z.Q., Su, W.B., Lu, S.N., Zhou, H.Y., Geng, J.Z., Li, S., Yang, F.J., 2010.Zircon U-Pb dating on tuff bed from Gaoyuzhuang Formation in Yanqing, Beijing: Further constraints on the new subdivision of the Mesoproterozoic stratigraphy in the North China Craton.Acta Petrologica Sinica,26: 2131-2140.

Liang, D.Y., Song, Z.M., Zhao, C.H., Nie, Z.T., 2002.Discovery of Mesoproterozoic seismites at Baishi Mountain, Hebei Province and its geological signif i cance.Geological Bulletin of China, 21:625-630.

Noffke, N., 2010.Microbial maps in sandy deposits from the Archean era to today.Berlin: Springer-Verlag, 194.

Noffke, N., Chafetz, H., 2012.Microbial mats in siliciclastic depositional systems through time.SEPM Special Publication, 101:198.

Qiao Xiufu, Gao Linzhi, 2007.Mesoproterozoic palaeoearthquake and palaeogeography in Yan-Liao Aulacogen.Journal of Palaeogeography, 9: 337-352 (in Chinese with English abstract).

Qiao, X.F., Gao, L.Z., Peng, Y., 2007.Mesoproterozoic earthquake events and breakup of the Sino-Korean Plate.Acta Geologica Sinica, 81: 385-397.

Rogers, J.J.W., Santosh, M., 2002.Conf i guration of Columbia, a Mesoproterozoic supercontinent.Gondwana Research, 5: 5-22.

Song, T.R., 1988.A probable earthquake-tsunami sequence in Precambrian carbonate strata of Ming Tombs District, Beijing.Chinese Science Bulletin, 33: 1121-1124.

Su, D., Qiao, X., Sun, A., Li, H., Somerville, I.D., 2013.Large earthquake-triggered liquefaction mounds and a carbonate sand volcano in the Mesoproterozoic Wumishan Formation, Beijing,North China.Geological Journal, doi:10.1002/gj.2501, 21 pp.

Su Dechen, Sun Aiping, 2011.Soft-sediment deformation and occurrence frequency of palaeoearthquake in the Mesoproterozoic Wumishan Formation, Yongding River Valley, Beijing.Journal of Palaeogeography, 13: 591-614 (in Chinese with English abstract).

Su Dechen, Sun Aiping, 2012.Typical earthquake-induced soft-sediment deformation structures in the Mesoproterozoic Wumishan Fm., Yongding River valley (China)and interpreted earthquake frequency.Journal of Palaeogeography, 1: 71-89.

Su, W.B., Zhang, S.H., Huff, W.D., Li, H.K., Ettensohn, F.R.,Chen, X.Y., Yang, H.M., Han, Y.G., Song, B., Santosh, M.,2008.SHRIMP U-Pb ages of K-bentonite beds in the Xiamaling Formation: Implications for revised subdivision of the Meso- to Neoproterozoic history of the North China Craton.Gondwana Research, 14: 543-553.

Su, W.B., Li, H.K., Huff, W.D., Ettensohn, F.R., Zhang, S.H.,Zhou, H.Y., Wan, Y.S., 2010.SHRIMP U-Pb dating for a K-bentonite bed in the Tieling Formation, North China.Chinese Science Bulletin, 55: 3312-3323.

Van Loon, A.J., 2009.Soft-sediment deformational structures in siliciclastic sediments: An overview.Geologos, 15: 3-55.

Van Loon, A.J., Mazumder, R., 2013.First find of biogenic activity in the Paleoproterozoic of the Singhbhum craton (E India).Geologos, 19: 185-192.

Von Blanckenburg, F., Mamberti, M., Schoenberg, R., Kamber, B.S., Webb, G.E., 2008.The iron isotope composition of microbial carbonate.Chemical Geology, 249(1-2): 113-128.