The Vilui Basin and the Late Cretaceous Continental Interior Climate Paradox

Summary

PALAEOCLIMATE models (AGCMs) applied to extreme greenhouse worlds such as those of the Mesozoic and Early Tertiary all suggest continental interiors experienced similar climate regimes to similar areas today: large mean annual ranges of temperature and generally dry conditions. For the central Siberian region these predictions are in marked disagreement with preliminary angiosperm foliar physiognomic analyses of museum fossil collections, which indicated wetter, warmer and more equable annual climate. Without knowing the geological context of the fossils the reliability of these determinations was open to doubt. New field evidence from the Vilui Basin which for the first time places several qualitative and quantitative strands of palaeoclimate evidence in their geological context demonstrates that AGCMs indeed fail to reproduce the observed continental interior climate under greenhouse world conditions. The reasons for the model-data mismatch may be due to a lack of a dynamic ocean and appropriate vegetation feedbacks, but the reliability of future climate predictions depends on properly quantifying this paradox.

Objectives and Background

The objectives of the project were as follows:

1) To characterise the nature of the previous (late Mesozoic) warm Arctic Ocean.

2) To characterise Eurasian circum-Arctic vegetation and climate for the late Mesozoic from geologic data.

3) To model its influence on the Eurasian circum-Arctic vegetation, sedimentary facies and climate of that time.

4) To model the climatic and vegetational effects in Europe and Russia caused by warm water circulation in the Arctic Basin.

5) The strategy of the proposed research is to combine different methods of past climate reconstruction (computer modelling, palaeobotany, sedimentology, isotopic study) in order to understand the near-polar climatic conditions that existed in a time of pronounced global warmth.

Meeting these objectives required a combination of climate modelling and geological data collection. The only way that climate models for greenhouse worlds can be evaluated is against geological data. Most geological data is qualitative but some offer quantitative information. The most widely used quantitative temperature proxy for the oceans is oxygen isotopes. These data are often used to specify the sea surface temperatures for GCMs that lack dynamic oceans - such as the models used in this project. In situations where ocean temperatures are provided to the model, or where dynamic oceans are allowed to achieve themal equilibrium through model processes, the model has to be tested against land-based climate proxies.

Land-based proxies can be both qualitative and quantitative. Qualitative data can be obtained from leaf or pollen-based floristic studies (which kinds of taxa occur where), but in some instances faunal remains can be used. Animals have the disadvantage though that many migrate and warm-blooded species have homeostatic meachanisms and fur or feathers that tend to isolate them from the external environments. However plant remains are not without thier problems. Spatial and temporal resolution is easily lost when using pollen grains because they can be easily ransported long distances and reworked, while leaf remains are not so readily preserved as pollen.

Despite their lower preservation potential the most powerful climate proxy on land is that based on leaf physiognomy. Leaves tend to be transported only short distances (<10km) before they are destroyed through decay, being consumed or mechanical fragmentation. After burial they are almost never reworked so both spatial and temporal resolution is retained in fossil assemblages. Decoding the climate signal carried in the leaf physiognomy has become formalised using a multivariate statistical technique known as CLAMP. We have used CLAMP to evaluate climate model performance where the models are least contrained, over land.

In the course of the project it became obvious that a conflict existed between the available geological evidence for the Russian interior and the model predictions. This conflict exists under a wide range of Arctic Ocean conditions. A strong characteristic of all palaeoclimate models is that for greenhouse worlds such as the mid Cretaceous they predict cold dry continental interiors - interiors much more extreme than the geological data suggest occurred. The UGAMP and Hadley Centre models are no exception and the most pronounced continental interior in mid Cretaceous times existed in what is now eastern Siberia. Preliminary examination of museum collections of fossil leaves from the area, now housed in Moscow, indicated a temperate climate with winter temperatures hardly falling below freezing. However the depositional context of the fossils was not clear and this is crucial for reliable interpretation of the assemblages.

INTAS/RFBR Project 95-0949 funded an expedition to this area, specifically the Vilui Basin, to field verify the depositional context of fossil leaf collections made in 1954 by Vakhrameev. What follows is an account of the recent findings.

Sedimentology

In Cretaceous times, the Vilui Basin constituted the most pronounced continental interior region of the Northern Hemisphere (Smith et al., 1981). The basin is situated between the two main subterrains of the Siberian Platform.

Fig. 1 Geological map of the Vilui Basin (afterVachrameev & Pushczarovskii 1954). The Sangar Group includes Lower Cretaceous strata. The Vilui Group includes the latest Albian to Maastrichtian Timerdyakh Formation and the Maastrichtian and possibly early Tertiary Linde Formation. The ages of the formations are constrained by palynological and palaeobotanical data (Vachrameev 1958, Bolkhoditina 1959, Semoylovich 1965)

During field studies in 1999 more than twenty 30 - 80 m high and 100m to several km wide river bank exposures, primarily of the alluvial Timerdyakh Formation, were surveyed during a traverse across the Vilui Basin, following courses of the Lena, Vilui and Tyung Rivers (Fig. 1). Early Cretaceous strata exposed in the basin are dominated by stream channel sandstones interbedded with tabular bodies of floodplain deposits, including autochthonous coal seams and stacked palaeosols. Middle to Late Cretaceous successions show strong and upwardly increasing channel cannibalism and reworking of floodplain and levee deposits.

FIG. 2: A ~30m high river bank outcrop showing amalgamating middle Cretaceous river channels in cross-section (1-5) at the Tyung River locality T1. Plant fossils are abundant in fine grained partings between dune deposits of the channel No. 2, which was seasonally active, capturing plant leaves during seasons of low discharge (autumns).

Due to bank failures at the time of deposition, the basal channel deposits commonly include mud and peat balls, slumped fossil tree bases, drift wood and log jams, and reworked siderite concretions, all reflecting the former presence of interfluve areas. Sites of ancient vegetation are represented by rare immature palaeosols with rooting, destratification and slickensides. The best preserved fossil plant leaf assemblages are found in abandoned channel deposits, and in rhythmical low-discharge interbeds of seasonally active channels (Fig. 2). Preservation of delicate floral components indicate limited downstream transport prior to deposition. In the central parts of the Vilui Basin, the Timerdyakh Formation is overlain by the Linde Formation. This is a gravelly to sandy crossbedded siliciclastic unit dominated by short transported kaolinized basement material reflecting local tectonic rejuvenation.

A representative set of samples from the entire Vilui Basin, of in situ and reworked floodplain mud rocks were analysed with respect to clay mineralogy. In Early Cretaceous samples kaolinite and smectite occur in subequal amounts, whereas kaolinite dominates, sometimes together with illite, without notable smectite content in middle to Late Cretaceous samples (Timerdyach and Linde Formations). This apparent change in smectite/kaolinite ratio probably reflects an increase in humidity (Sellwood & Price, 1994), from the middle Cretaceous and on. The investigated clay minerals revealed no signs of arid climate conditions of the Siberian continental interior in Late Cretaceous times. Palaeomagnetic analysis of the Timerdyakh Formation sediments suggested a palaeolatitude of 72°N.

Palynology Results

Palynological investigations of the Timerdyakh Formation reveal a very high floral diversity, ferns and flowering plants being represented most. Of 190 taxa, 61 are spores (at least 33 of them belong to ferns), 16 are gymnosperm pollen, 10 are monocot pollen of uncertain affinities (including two probable palm species), and 87 are dicot pollen. Not more than 10% of the assemblage shows evidence of having been reworked from pre- middle Cretaceous times. This remarkable diversity of palynomorphs, together with presence of thermophilic taxa, indicates relatively warm and humid climatic conditions, similar to those of warm temperate regions today. Click here to download an Excel spreadsheet of the palynological data,

FIG. 3 Examples of pollen grains from the Timerdyakh Formation. Preservation is generally good.

CLAMP Results

Understanding the disparity between models and observations in greenhouse world continental interiors requires quantification of the mismatch. For non-marine environments this is best achieved using foliar physiognomy as a quantitative climate proxy. A combination of museum and new collections of the Late Cretaceous leaf flora of the Vilui Basin were analysed using the Climate Leaf Analysis Multivariate Program (CLAMP)(Wolfe, 1993; Kovach and Spicer, 1995; Wolfe and Spicer, 1999; Spicer, 2000). Analysis of Timerdyakh Formation leaves placed them within the physiognomic space occupied by modern temperate vegetation, unlike a test sample from the modern Vilui Basin vegetation which plotted in isolation along axis three (Figure 4.).

FIG. 4 CLAMP Axis1 vs 3 plot showing modern (isolated red dot) and fossil (white squares) Vilui samples in respect of the calibration modern vegetation samples. The modern calibration samples are colour coded in relation to mean annual temperature with blue being the coldest to red the warmest. The blue crosses are those samples that belong to the "alpine nest". See the CLAMP website for further details.Click here to download a Quicktime animation of the plot rotating to show the three dimensional relationships of the samples.

This confirms that the modern adaptations to extreme annual temperature range are not seen in the Cretaceous fossils and that the modern dataset used was appropriate for obtaining reliable palaeoclimate data. The CLAMP analysis yielded a mean annual temperature of 13.3±3.4°C (2s) and warm month mean of 21.5±3.8°C. A cold month mean temperature of 5.1±5.2°C was determined assuming seasonal symmetry. The length of the growing season (7.5±1.8 months) is largely a function of the high palaeolatitude light regime rather than temperature. The growing season total precipitation (1005±856 mm), the mean monthly growing season precipitation (149±98 mm), the three consecutive wettest months precipitation (570±378 mm), and the three consecutive driest months precipitation (334±238 mm) all suggest a moderately wet regime year round. This is also reflected in the mean annual relative humidity (76.2±17.6%). When these CLAMP results are compared to either UGAMP or Hadley Centre AGCM predictions there are significant differences between them and the model data. Although quantified uncertainties in model predictions are rarely stated work in progress by us suggests that are of the same order as those presented here for CLAMP. Based on CLAMP the Cretaceous continental interior of Asia (Vilui Basin) appears remarkably equable, with winter temperatures well above freezing for all but the coldest days, and summer warm month mean temperatures around 22°C. This is in marked contrast with the wide annual range of temperatures seen in continental interiors today, and with predictions of AGCMs (Table 1). Another difference is that CLAMP (in agreement with our palynological and clay mineralogical data) predicts a moderately wet regime year round, whereas the AGCMs suggest more or less continuous drought (although precipitation uncertainties are high for both CLAMP and the models).

	CLAMP	UGAMP Model	Hadley Centre Model
MAT (°C)	13.3	4.3	0.5
WMMT (°)	21.5	34.4	27.3
CMMT (°C)	5.7	-17.8	-20.9

Table 1 Comparison of CLAMP temperature data with AGCM predictions for the Vilui Basin

Conclusions

In conclusion, the Vilui Basin AGCM predictions clearly do not match those obtained from foliar physiognomy, palynology and sedimentology, and may indicate that the middle Cretaceous atmospheric dynamics were significantly different to those of the Present and those incorporated into AGCMs. The reasons for the model-data mismatch are still unclear but are probably due to insufficient understanding of ocean and vegetation climate feedbacks. However, this paradox must be understood before we can have full confidence in future global warming predictions.

Methods

Sedimentological, clay mineral, palaeomagnetical and palynological studies were performed according to standard procedures (Miall 1996, Hardy & Tucker 1988; Erdtmann 1954 page 34-35 (Chlorination-Acetolysis Method), Klaus 1987 page 228-234 (anorganic procedure), Tyson 1995 (point counting) page 459-460. Samples were not sieved in order to retain small angiosperm pollen >10µm )). The CLAMP technique used decodes the climatic signal inherent in the physiognomy of leaves of woody dicotyledonous plants. CLAMP calibrates the numerical relations between leaf physiognomy of the woody dicotyledons and meteorological parameters in modern terrestrial environments. Using this calibration, past climatic data are determinable from leaf fossil assemblages. The statistical engine used is the Canonical Correspondence Analysis (CANOCO) which is a direct ordination method that orders samples, in this case vegetation sites, based on a set of attributes. In CLAMP the attributes are the scores of the 31 leaf character states taken from more than 20 species of woody dicots in each vegetation site. The largest calibration dataset used here consists of foliar physiognomic measurements and climate observations from 173 modern vegetation sites, predominantly in Northern Hemisphere, and includes low temperature sites comprising the "alpine nest"8.

References

Bolkhovitina, N.A., 1959. Spore and Pollen Assemblages from the Mesozoic of the Vilui Trough and Their Stratigraphical Significanse. Geological Institute, USSR Acad. Sci., Transactions, vol. 24: 187 pp.(In Russian).

Erdtman, G.,1954 An Introduction to Pollen Analysis, Waltham, Mass, U.S.A, 238pp.

Hardy, R. & Tucker, M.E., 1988: X-ray powder diffraction of sediments. In M. E. Tucker (ed.) Techniques in Sedimentology, 191-228. Blackwell Scientific Publications.

Herman A.B., and Spicer, R.A., 1996, Palaeobotanical evidence for a warm Cretaceous Arctic Ocean. Nature 380: 330-333.

Klaus, W., 1987 Einfuehrung in die Palaeobotanik, Band 1, Grundlagen-Kohlebildung-Arbeitsmethoden-Palynologie, Deuticke, Wien, 314pp.

Kovach, W.L., and Spicer, R.A., 1995, Canonical Correspondence Analysis of Leaf Physiognomy: a Contribution to the Development of a New palaeoclimatological Tool. Palaeoclimates, 1: 125-138.

Miall, A.D., 1996. The geology of fluvial deposits; sedimentary facies, basin analysis and petroleum geology. Springer Verlag, 582 pp.

Samoylovich, S.R., 1965. New species of Upper Cretaceous angiosperm pollen from Yakutia. All-Union Oil Scientific Research Geological Inst., Transactions, vol. 239: 121-141 (In Russian).

Sellwood, B.W. & Price, G.D., 1994 Sedimentary facies as indicators of Mesozoic Climate. In J.R.L. Allen, B.J. Hoskins, B.W.Sellwood, R.A.Spicer, and P.J. Valdes (Editors) Palaeoclimates and Their modelling: With Special reference to the Mesozoic Era. Chapman and Hall. London, pp. 17-26.

Smith, A.C., Hurley, A.M., and Briden, J.C., 1981, Phanerozoic Paleocontinental World Maps. Cambridge University Press, 102pp.

Spicer, R.A., 2000, Leaf Physiognomy and Climate Change, In Biotic Response to Global change: the Last 145 Million Years, Culver, S.J., and Rawson, P., (eds.)

Tyson, R.V., 1995 Sedimentary organic matter ­ Organic facies and palynofacies, Chapmann & Hall, London, 615pp.

Vachrameev, V.A.,1958 Stratigraphy and fossil flora of Jurassic and Cretaceous Deposits of Vilui Trough and adjacent part of Near Verkhoyansk Foredeep. In: Regionalnaya Stratigrafiya SSSR, Vol 3, USSR Acad. Sci. Publ., Moscow, 340 pp. (In Russian).

Vachrameev, V.A., and Pushczarovskii, Yu.M.,1954. On Mesozoic geological history of the Vilui Trough and the adjacent part of the Near Verkhoyansk Foredeep. In: Voprosy Geologii Azii, vol. 1, USSR Acad. Sci. Publ., Moscow: 588-628 (In Russian).

Wolfe, J.A., 1993. A method of obtaining climatic parameters from leaf assemblages. U.S. Geol. Surv. Bull., 2040, 73 pp.

Wolfe, J.A., and Spicer, R.A., 1999, Fossil Leaf Character States: Multivariate Analysis. In Jones, T.P., and Rowe, N.P. (eds.) Fossil Plants and Spores: Modern Techniques. Geological Society, London, 233-239.