How did Jawed Vertebrates Originate and Rise?

INTRODUCTION

Situated in the low latitudes and the extension of the West Pacific Worm Pool, the Bay of Bengal and its neighboring water hosted a great deal of terrigenous sediments denuded from the Tibetan plateau and transported by the Gange-Brahmapoutra fluvial system. Although most parts of the sea floor in the area are occupied by the Bengal deep sea fan, the greatest turbidite body in the world, turbidity current sedimentation does not serve as the only one dynamics to constitute the sedimentary framework. The hemipelagic sedimentation system consisting of planktonic skeleton, eolian dust, tephra and nepheloid layer plays an important role in many locations of the fan (Fang, 1992), which amplify the oceanographic and climatic signals recorded in the region under study. By that advantageous ground the North Indian Ocean containing the Bengal and Indus fans can be regarded as an ideal site to investigate the rapid climatic variability in the low latitudes and its tele-connection to the high latitudes (Keigwin and Jones, 1994).

The analysis materials reported in the paper were collected by N/O Marion Dufresgne and sampled in the Laboratoire des Sciences du Climat et de l'Environnement, which display high-quality millennium environmental signals. The situation, depth and length of the cores under study are shown in Table 1. All of them cover a temporal scale of over 200 ka. We focus our discussion on those events younger than 60 ka except for some special needs.

Table  1.  SITUATION AND LENGTH OF CORES STUDIED IN WORK

 | Show Table

DownLoad: CSV

Quite a few classical researches of the tectono-sedimentary activities related to the Hymalaya lift have been made in the Bay of Bengal (Colin, 1997; France-Lanord et al., 1992; Bouquillon et al., 1990; Stow and Wetzel, 1990; Curry et al., 1982; Moore et al., 1974; Curry and Moore, 1974, 1971). Paleoceanographically, however, it seems more or less ignored to study the sedimentary records in this area, especially when compared with the Arabian Sea, Somali basin and Oman Ridge lying in the North Indian Ocean as well.Duplessy (1982) once studied, using the foraminiferal δ (¹⁸O) records from the bay, the variability of salinity gradient associated with the Indian monsoon. Fontugne and Duplessy (1986) made an approach of transport of surface current in the bay driven by the monsoon by means of the δ (¹³C) data. Foucault and Fang (1987) and Fang (1990) indicated an orbital control of sedimentary process in the distal fan. Ding et al. (1999) and Fang et al. (1999) published their results of paleoproductivity and CaCO₃ dissolution from a rhythmical sequence tele-connecting to the Heinrich and Dansgaard-Oescheger events. A major motivation for this work is to summarize the climatic phases mirrored by the records from the Bay of Bengal and their correspondence from the South China Land.

BASICS OF MARINE SEDIMENT

First of all, we slightly expand the timescale to examine the variability of the sediments with the typical climatic cycles since the last interglaciation.

On the basis of the δ (¹⁸O) curves of Globigerinoides ruber measured by the Centre des Faibles Radioactivités (Gif sur Yvette, France), which can be correlated with SPECMAP (Martinson et al., 1987), a timeframe of alternate (sub) glaciations and (sub) interglaciations can be established in the region studied (Colin, 1997; Fang, 1992, 1990). It has been proved by the records from the different parts of the Bengal deep sea fan that a remarkable contrast of sedimentary flux between glaciation and interglaciation occurred in the fan (Colin, 1997; Foucault and Fang, 1990; Curray et al., 1980). In general the sediments settling during oxygen isotope stages 2, 3 and 4 are constituted by olivine gray (5Y4/1 or 5Y5/2) and CaCO₃-poor (2%-10%) silty clay while those during stages 1 and 5 by yellowish brown (10YR4/2) or grayish white (10YR6/2) and CaCO₃-augmented (15%-52%) clay and even calcareous ooze. Particularly, MD77190 cored at the lower fan and colored by more pelagic feature assumes, in stage 3 and early stage 2, a rhythmical sequence constructed by CaCO₃-rich (14%-32%) clay alternating with CaCO₃-poor (3%-12%) one, whose other characteristics have been reported (Fang et al., 1999).

The fine silt and clay play a dominant role in the all examined cores. Although a few thin (1-5 cm) interbeds of fine quartz sand can be occasionally seen there, we believe they are resulted only from the intermittent overflow of the channels which rarely took place and could give neither destruction nor distortion to the normal carriers of environmental signal. In so far as disturbance element is concerned, what warrants careful consideration consists in remodeling deposits of bottom current. The state shown by the central section of core MD77183 will serve as an illustration.

Farther from the radial resource of turbidity current than MD77181, the core MD77183 has finer-sized sediments, upon the whole, than the former one. There are more silty intercalations with oxidation color, however, in its middle part. The grain size distribution of them indicates their hydrodynamics of traction current origin (Fang, 1992). The strongest impact on the sedimentary record that the bottom current brought about is a severe dissolution of calcareous shell. Within a period spanning near 40 ka, from late oxygen isotope stage 5 to early stage 3 (about 75-37 ka BP), the foraminiferal skeletons were practically wiped out and CaCO₃ content reduced to its minimum (0%-3%) by dissolution. We can not judge whether all the dissolution was generated from the bottom current, but claim, on the basis of the sedimentary texture, that type of current was very stimulated in the past 65 to 37 ka. Because of partial but heavy damage to the foraminiferal proxies in MD77183, we are constrained to show only the result from other two cores though an equally interesting δ (¹⁸O) curve has been found in them.

RAPID CHANGE SIGNALS IN MD77181 AND MD77190

Sampled every 5 or 10 cm, cores MD77181 and MD77190 with a sedimentation rate of 5-15 cm/ka can obviously give us a resolution superior to that in the millenarian scale.

Without dating radioactively the sediments, we subdivide them into the glacial and interglacial stages and calibrate chronologically the borders of the stages by means of correlating the δ (¹⁸O) curve measured by SPECMAP (Martinson et al, 1987). The subdivision and calibration has been further reported by the references from volcanochronology, micropaleontology, carbonate stratigraphy and paleomagnetism (Fang, 1990). The ages for the other positions are interpolated and extrapolated by the separate sedimentation rate in each stage. The precise time graduation can not be expected in terms of such methods. However, thereby we shall be able to set up a solid frame to make an observation of evolutionary tendency in climatology and a comparison between different events.

We choose some effective and sensitive proxies from those measured ones for arguing that significant rapid change occurs in the sedimentary records at the region under study (Fig. 1). Among them the oxygen isotope was measured in the Centre des Faibles Radioactivités as stated above, the calcimetry and clay mineral diffractometry were made in the Laboratoire de Stratigraphie, l'Univercité PCM (Paris Ⅵ) while the granulometry, measuring magnetic susceptibility and identifying foraminifera made in the Institute of Marine Geology and Geophysics, China University of Geosciences.

Figure  1.  Climatic evolution during past 60 ka outlined by various proxies in piston cores MD77181 and MD77190, NE Indian Ocean

DownLoad: Full-Size Img PowerPoint

The δ (¹⁸O) curve constitutes a base to date the sediments and to attach an environmental feature to them in our work. Macrostructurally, there is a pretty correspondence between the δ (¹⁸O) curves from the Bengal fan and the normalized ones (Martinson et al., 1987; Imbrie et al., 1984). In more detailed fashion, stacking those measured curves, we can find all of them show an oscillatory pattern with a surprising similarity, especially for stage 3. It can be reasoned that the similarity is controlled by a mechanism alternating between warm episodes and cold episodes that appears notably active. According to the oscillation from the δ (¹⁸O) curve of MD77181, all Heinrich events during stages 2 and 3 including a special one between H4 and H5 marked with H4-5 may be recorded in their proper place.

Susceptibility is a very useful index in paleoceanographic research though we just bring it into use. Near the continental margin where our materials accumulate, its variability should be closely related to terrigenous ferromagnetic input and redox of sedimentary setting. In MD77181 it varies inversely as the δ (¹⁸O) fluctuating form and responds to the cold time with its peaks except that it runs inexplicably around its minimum (near nil) at all epoch ranging between 230 ka BP and 150 ka BP. Displayed in this work, besides reigning over stages 2 and 4, which represent maximum glacial times in the youngest climatic cycle, the high values of susceptibility seem to cluster around the places where Heinrich events may occur. In MD77190 this index is overall compatible with the alternation of glacial epochs as well. It shows also some details corresponding to the events mentioned above and, more significantly, an abrupt rise at about 36 ka BP when there may be a very important transformation from warm time to cold time during stage 3 (Fang et al., 1999).

Carbonate content in the sediment is finally decided by the linkage of the three factors: calcareous bioproductivity, dissolution of carbonate material and dilution of non-carbonate material (Volat et al., 1980). Because of a great input of terrigenous detritus, the Atlantic-type dilution cycle for carbonate is dominant in the region under study. Although the peaks in the carbonate content curve are mainly distributed by appearances over the warm times, that has nothing to do with the real sedimentary quantity of carbonate. The opposite is the case: the quantity of carbonate was expanded during the glacial times as a result of the promotive productivity energized by accelerated ocean circulation and increased terrigenous nutrients. For this reason, on the assumption that sedimentation rate is relatively constant at a given stage and thus the dilution agency can be removed, it is comprehensible why in MD77181 the clusters of higher CaCO₃ contents overprint themselves on those of higher susceptibility and indicate the cooling events likewise. It should be pointed out that the mechanism stimulating the ocean productive elements does not have a linear relationship to decreasing temperature. We proved that those CaCO₃-rich sediments in the rhythmical sequence of MD77190, with more pelagic and tropical color than MD77181, are the outcome of combining the warming water with the enhanced Indian summer monsoon during stage 3 (Fang et al., 1999).

Additionally the curve of grain-size (> 50 μm) and that of clay mineral (smectite/ (illite+chlorite)) from MD77190 are shown in Fig. 1. The former is, like CaCO₃ content, the result of balancing productivity with dissolution while the latter may contribute to appraising the influence of terrigenous input. Their relatively high values correspond to the rise of temperature or to that of sea level. Besides serving as the sign to recognize H3, H4-5 and H5, they show in the similar way a sharp drop of value demonstrating the climatic transformation around 36 ka BP (H4). The rhythmical sequence characterized by alternating CaCO₃-rich sediment with CaCO₃-poor one in MD77190 is a special and valuable proxy to reveal the climatic variability with high-frequency in the studied region (Fang et al., 1999). We have made a subdivision to the sequence on the basis of examining the thickness, texture (changing frequency) and foraminiferal feature of the two sediments. Phases Ⅰ, Ⅱ and their subordinate parts Ⅰa, Ⅰb, Ⅰc, Ⅱa and Ⅱb, temporally covering both stage 3 and early stage 2 (59-21 ka BP), are reasonably created for describing the evolutionary process of climate (Fang et al., 1999). Each of the divisions displays its own property formed under a certain climatic and oceanographic condition. When Indian summer monsoon is prevailing, abundant rainfall and active drainage from the land stimulate the fertility of the surface water mass and contribute to the formation of the CaCO₃-rich sediment with a great quantity of foraminifera, radiolaria, opal and organic carbon. The other sediment with few organisms formed under a severe environment and its mass occurrence was roughly correspondent to the Heinrich events. The boundaries between the climatic divisions we make lie at 59, 53, 42, 37, 29 and 21 ka BP, respectively. Among these points, the equivalent of H4 event dating from about 37-36 kaBP is a turning point for the transformation of the summer monsoon from rising to falling, which seems to have the greatest impact on regional climate during stage 3 (Fig. 1)

DISCUSSION

Placed in the climatologically transitional and sensitive zone, the oxygen isotopic stage 3 can be described as an eventful period. This stage is sandwiched between two severe glacial maxima (stage 2 and stage 4) and displays most evidences of the high-frequency instability, which includes at least 4 Heinrich cooling events and nearly 20 Dansgaard-Oeschger (D-O) short warming events observed in the North Atlantic and Greenland in the last four glacial cycles. The records from the Northeastern Indian Ocean announced in this paper show a tele-connection between the low latitudes and the high latitudes.

We can follow the tracks of the climatic evolution by analyzing our records (Fig. 1) and compare them with the result of Greenland ice core (Fig. 2).

Figure  2.  Diagram showing a tele-connection between GISP2 (Greenland ice core) and MD77190 (deep water core from the Bay of Bengal)

DownLoad: Full-Size Img PowerPoint

In section Ⅰa (59-52 ka BP), we can find a spell of variable climate closely following the severe cold time (stage 4). As a whole, it may represent an episode warming up but often punctuated by short cooling events. In the meantime, the superficial water temperature calculated by foraminiferal assemblage can sometimes drop by 2-3 ℃ compared with its average and neither cold time nor warm time sustain for over 800 years (400-500 a in general). To sum up, high frequency and big amplitude in the variability of climate during this episode. D-O events 15 to 17 may be regarded as an analogue of the variability in the studied region.

In section Ⅰb (52-43 ka BP), the elements representing warm climate seem to be in the limelight. Our evidences for illustrating that lie in the heavy oxygen isotopic composition and weak susceptibility in MD77181 and the high ratio of smectite to the sum of illite and chlorite, weak susceptibility and especially mass occurrence of the CaCO₃-rich sediment. Despite a very strong dissolution of carbonate (Ding et al., 1999), we can find there a concentration of the materials indicating high productivity. According to our previous result, the promotion of the bio-productivity is caused by the greatly increased Indian summer monsoon rather than cool upwelling (Fang et al., 1999). D-O events 12, 13 and 14 that lasted for a long time but were increasingly cooled, can be correlated to this episode. However, two important cold events named H5 and H4-5 by analogy with the Heinrich series manifest themselves at about 50 ka BP and 43 ka BP to 45 ka BP, respectively. Their appearances inject unharmonious notes into this "warm episode". The H5 may be an event with the greatest variation, judging from the δ (¹⁸O) curve pattern, in the overall stage 3 while the H4-5 appears to be one of the events qualified as the most influential in the region under study (Fang et al., 1999).

In section Ⅰc (43-37 ka BP), most proxies support a conclusion that the warmth still lingered on during that episode (Fig. 1). The entire episode is still under strong dissolution but more of coarse fossil foraminifera than before can be seen in the record, which announce an occurrence of high productivity at that epoch. The other evidences for this include (1) ¹³C heavy anomaly, (2) significant increase of Uvigerina and (3) growth of radiolaria and opal. The high productivity favorably grew during the episode with the persevering summer monsoon and that with the upwelling driven by the increasingly intensified winter monsoon. The latter point can be annotated by a downgraded δ (¹⁸O) curve, promoted susceptibility and developed Globigerina bulloides at the late episode. D-O events 8-11 fading somewhat in warmth corresponded to this episode.

As stated above, we put a dividing line at 36-37 ka BP, which is chronologically analogous to H4, on account of a dramatic turn occurring in all environmental indexes. The part above the line is marked with Ⅱ.

In section Ⅱa (37-29 ka BP), all the indexes show much less warm components than in phase Ⅰ (Fig. 1). According to the record from MD77190, there is still a certain balance between CaCO₃-rich and CaCO₃-poor sediments alternated each other, which means the warm climate stubbornly entrenched itself during this episode. A considerable quantity of both Uvigerina sp. and G. bulloides favors the high productivity and active upwelling triggered off by the prevailing northeastwardly winter monsoon during a relatively cold episode. D-O events 4-7 listlessly occurring in the Greenland ice core can be also found in this section.

In section Ⅱb (29-21 ka BP) all the warm signals are in decay. Though there is no good correlation between them and D-O events, a common ground lying in both can be summarized as follows: some solitary "warm peaks" are surrounded by a wilderness of "cold elements".

After 21 ka BP, our marine records are permeated with a simple and clear variability of climate. The cold glacial maximum is in heavy contrast with the warm postglaciation. A combination of relatively unvaried weather, low sedimentary rate and top beds pistoned prevents us from furthering our work.

Those records from the South China Land can remedy in some degree defects, and more interestingly, offer a coupling pattern for the reference of the marine records.

Cohabiting with the North Indian oceanic sediments under the Indian monsoon regime, the sedimentary record, drilled in the Napahai Lake, Zhongdian Plateau, Northwestern Yunnan, plays an important role in examining the coupling between the ocean and land records. Moreover, the stalagmite record, collected in a cave of the Shennongjia Mountain, Western Hubei, is joined to the regional correlation. Some of their key points have been dated by means of ¹⁴C and U-series methods respectively and the ages of other parts deduced by interpolation.

Quite a few chemical, sedimentary and biological proxies such as δ (¹³C), δ (¹⁸C), percentages of carbonate and of organic carbon, grain-size, magnetic susceptibility and pollen have been used for measuring the climatic change in the studied area. Most Heinrich events, somewhat transformed by some local ones, can be identified by variation pattern of the curves of δ (¹³C) and granulometry (Fig. 3a). In addition, these curves reflect a high frequency variability of climate since the last glacial maximum. The discussion about the climatic meaning of the curves' trend and details of the variation will be done in a subsequent publication. At issue with our marine records, H4 taking place at 36-37 ka BP was only fleetingly touched in the lacustrine record and its importance seems to be replaced by another cooling event presented earlier (about 40 ka BP). More significantly, a strong rebound to the warmth next to the transient H4 lasting until 32 ka BP is against what appears in the marine records. Yao et al. (1997) stressed also this event leaping warmthwards in the record from the Tibetan Guliya ice core. The harmony between the two plateau records implies that this warming event is controlled in a considerable degree by the local solar radiation rather than the weakened Indian summer monsoon.

Figure  3.  δ (¹³C) and granulometric curves showing climatic changes from the lacustrine core, Zhongdian, NW Yunnan. (a) andδ (¹⁸O) curve showing climatic changes from the stalagmite, Shennongjia, West Hubei (b)

DownLoad: Full-Size Img PowerPoint

The stalagmite record, being farther from the Indian monsoon regime but in the front precipitation zone under an interaction among several air circulation regimes shared by the Indian summer monsoon, shows a result closer to the marine records (Fig. 3b). By analyzing the inclusion containing original water in the stalagmite, we conclude that the δ (¹⁸O) value recorded in the rings of stalagmite increases versus cooling and moistening. As a result, the variation embodied by the δ (¹⁸O) curve is climatologically consistent with that we outlined in the light of the marine data. Not only two or three cold events emerging at 37 ka BP and 43 ka BP (or with one at 46 ka BP) but also H4 serving as a dividing line to distinguish the upper and colder part from the lower and warmer part are found in the stalagmite material. We conjecture that the better correlation between ocean and stalagmite records is caused by a smoother geographic passage to link the two sedimentary settings.

Moreover, we can find a correlation between the cold events from the different records more easily than that between the warm events. This unequality may be attributed the fact to be stated as follows. There are many obstacles in geomorphology on the land to throw the warm and wet air from the North Indian Ocean into confusion. In contrast, a compound of cold element bred by the polar source and that by the plateau source occupied a commanding height to bring the relevant signal to the ocean.

CONCLUSION

A tele-connection between the climatic record of the high latitudes and that of the low latitudes can be convincingly established by making in paleoceanography a preliminary but detailed study on the piston cores from the Bay of Bengal. Most Heinrich cold events and D-O warm ones have been located in the marine sediments. By comparison, the records with a high variability from the land show paleoclimatically a similar outline, and in particular a good correlation of the cold events, corresponding essentially to the high latitudes but bear considerably a local color in displaying the warm events. We present in this paper an assumption for explaining the phenomenon that the linkage of cold events between the pole and the tropical ocean is mainly realized by continental transmission while that of warm events between them controlled by oceanic heat transportation. The further research on this interesting subject will hopefully be done later.

ACKNOWLEDGMENTS

This work was sponsored by the National Key Project (G1998040800) and National Natural Science Foundation of China under the auspices of grants (Nos. 49672135 and 49772171). Special thanks are due to our French colleagues of the Laboratoire des Sciences du Climat et de l'Environnement (Gif sur Yvette) who provide us with the oceanic samples, excellent experimental assistance and helpful advice.