Paleoenvironment reconstructions and climate simulations of the …fluteau/UIA/articles/9_PERON-bourquin-fluteau.pdf · of characterizing the impact of climate on fluvial sedimentation

© 2005 Lavoisier SAS. All rights reserved.

Geodinamica Acta 18/6 (2005) 431–446

Paleoenvironment reconstructions and climate simulations of the Early Triassic: Impact of the water and sediment supply

on the preservation of fluvial systems

Samuel Péron

a

, Sylvie Bourquin

a*

, Frédéric Fluteau

b

and François Guillocheau

a

a

Géosciences Rennes, UMR 6118 du CNRS, Université de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex, France;

b

Laboratoire de Paléomagnétisme, IPGP, Université Paris VII, 4 place Jussieu, 75252 Paris cedex 05, France

Received: 14/12/2004, accepted: 07/11/2005

Abstract

Paleoenvironmental reconstructions and climatic modelling allow us to investigate the influence of water and sediment supply on thepreservation of fluvial systems within a given geodynamic context. To simulate climate, we need global-scale paleoenvironmental andpaleotopographic reconstructions. However, the present study only covers the West-Tethys domain, where sedimentological and strati-graphic data allow us to check climate simulation results against geological data. We focus our modelling on the Olenekian, with the aimof characterizing the impact of climate on fluvial sedimentation in the West-Tethys domain. The climatic simulations show that paleo-climates differ between Western Europe and North Africa. A more humid climate is simulated over North Africa, whereas a rather aridclimate prevails over Western Europe. In Western Europe, the sediments are preserved for the most part in endoreic basins and the pres-ence of rivers in an arid environment suggests that these rivers are mainly fed by precipitation falling on the North Africa VariscanMountains. In North Africa, sedimentation is exclusively preserved in exoreic basins (coastal plain sediments). Consequently, the lackof preserved fluvial systems in endoreic basins in North Africa either could be due to a shortage of accommodation space in this area, oris linked to the climatic conditions that controlled the water and sediment supply.

© 2005 Lavoisier SAS. All rights reserved.

Keywords:

Fluvial sedimentation; Early Triassic; Paleogeographic maps; West-Tethys domain; Climate simulations

1. Introduction

The deposits of fluvial environments are preserved eitherin exoreic basins, where the fluvial systems are connected tothe sea, or in endoreic basins, without any marine connexion.Fluvial systems in endoreic settings are mostly preservedduring warm arid or during humid climatic phases, which areassociated with aeolian or lacustrine deposits, respectively.This can be observed at the present day [1], [2], [3], [4]… aswell as through geological time [5], [6], [7], [8], [9]…

In the continental realm, the main difficulty is quantifyingthe impact of allocyclic factors (climate, eustatic effects,deformation and sediment supply, Fig. 1) on the preservationof sedimentary systems [10], [11], [12], [13], [14], [15]. Infact, the preservation of stratigraphic cycles is controlled bytwo parameters: the accommodation (A) and the sedimentsupply (S

S

) [16], [17], [18], [19], [20], [21], [22]. Theaccommodation characterizing the space available for sedi-mentation is controlled by the deformation of the basin(subsidence/uplift of the basin, denoted A

SU

, fig. 1), eustatic

* Corresponding author.E-mail address: [email protected]

432

S. Péron et al. / Geodinamica Acta 18/6 (2005) 431–446

(noted E, Fig. 1) and/or lake level variations (noted L

L

,Fig. 1, controlled either by basement deformation, notedA

SU

, or eustatic, noted E, and/or climate, noted A

LL

). Thesediment supply (noted S

S

, Fig. 1) is controlled by the cli-mate (noted S

c

, Fig. 1) and/or by the deformation of the basinbasement (noted S

D

, Fig. 1) or the source area of the silici-clastic sedimentation [23], [24], [25], [26], [27], [28]…Consequently, sedimentation is controlled only by twoexternal factors, climate (noted C, Fig. 1) and deformation(noted D, Fig. 1), while eustatic variations (E) are tectoni-cally (noted A

TE

, Fig. 1) or climatically induced (noted A

GE

,Fig. 1)).

To investigate the relationship between the sedimenta-tion, climate and deformation of basins, we use bothsedimentological analyses and climate modelling to studythe well-preserved Triassic fluvial successions of Peri-Teth-yan basins. The Triassic is a transitional period associatedwith the onset of the break-up of the Pangean supercontinentand the formation of the Mesozoic basins in a globally warmand dry climate. Detailed sedimentological and stratigraphicanalyses of Triassic fluvial deposits within the Europeangeodynamic context allow us to characterize the evolution ofthe fluvial style through time.

Numerical modelling is a powerful tool for simulating cli-mates at both global and regional scales. Climate represents

an equilibrium state between the different components of theglobal climatic system (atmosphere, ocean, biosphere, cryo-sphere and lithosphere surface). Because of the complexityof this system, and despite the development of powerfulcomputers, the entire climatic system cannot be simultane-ously simulated. Nevertheless, present-day or past climatescan be modelled by specifying certain components of the cli-matic system as boundary conditions, such aspaleogeography, sea-surface temperature and cryosphere.These boundary conditions should be reconstructed usingthe best available data. Although this is an easy task for thepresent day, it is much more problematic for the distant past.To overcome this difficulty, the sensitivity of climate isinvestigated with different boundary conditions that takeaccount of these uncertainties. In this study, we test the sen-sitivity of climate to the paleotopography and, consequently,its influence on sediment supply and the preservation of flu-vial systems.

We focus our modelling on the Late Early Triassic(Olenekian), and the impact of climate on fluvial sedimenta-tion in the western Tethys domain. Indeed, we need global-scale reconstructions to simulate climate. In this study, how-ever, we focus on Western Europe, where sedimentologicaland stratigraphic data allow us to check the results of climatesimulation against geological data.

Fig. 1 Influence of allocyclic parameters along longitudinal profiles with multiple local baselevel.


433

2. Paleogeographic reconstructions

2.1. Plate tectonic reconstructions

In the framework of this study, the plate tectonic recon-struction for the Early Triassic makes use of paleomagneticdata while assuming a geocentric axial magnetic field. Thepaleolocations of Laurussia (North America and Baltica)and Siberia have been calculated using apparent polar wan-der paths (APWP) defined by well determined poles [29].Because the Gondwanan poles are relatively sparse, theposition of this continent with respect to Laurussia has beenextensively discussed in the paleomagnetic literature (twoalternative Pangean plate reconstructions, respectivelycalled Pangea A and Pangea B, have been proposed [30],[31], [32]. However, this problem is beyond the scope this

paper (for further discussion see [33], [34]. We here considerthe best and most recent paleomagnetic poles to compute anAPWP for Gondwana [33], [35], [Besse, pers. comm.].Southern Asia is made up of a mosaic of accreted blocks thatrifted away from Gondwana during the Late Paleozoic andMesozoic [36], [37]. The locations of these blocks during theEarly Triassic have been calculated using the paleomagneticpoles from Enkin

et al.

[38].

2.2. Land-sea distribution

Two paleogeographic reconstructions are attempted here:we firstly build up as accurate a map as possible on a globalscale ([39] to [43], Fig. 2A), and, secondly, a map at higherresolution focused on West Tethys ([44] to [67], Fig. 2B). Toestablish this second paleogeographic map, we use pub-

Fig. 2 Palaeogeographic maps for Scythian (245 Ma). (A) Global paleogeographic map representing a synthesis of different reconstructions: (i) Platetectonic kinematic reconstruction (after [30]), ([35]), Besse, pers. comm., (ii) shoreline and palaeoenvironment reconstructions (after [39], [40], [41],[42], [43]), (iii) topography reconstructions (after [34], [40]) (B) detailed palaeogeographic map of the West-Tethys domain: (i) kinematicreconstruction (see above), (ii) reconstructions of palaeoenvironments and fluvial systems: in Germany (after [44], [45], [46], [47]), in England (after([48], [49]), in the Netherlands (after [50], [51], [52]), in the North Sea (after [53]), in the Maghreb (after [54], [55]), in Spain ([56], [57], [58]), in theEifel basin (after [59], [60]), in southern Scandinavia (after [61]), in Sardinia (after [62], [63] [64]) and in France (after [65], [66]), and Gaetani ([67]),(iii) topographic reconstructions (after [34], [40]).

434


lished paleogeographic maps [42], [43], [67], [68] as well assedimentological and sequence stratigraphy analyses of thecontinental environments. During the Early Triassic, theoceanic areas are dominated by the Panthalassa and theTethys Oceans. On the world scale, the land-sea distribution,the continental sedimentation areas and the orogenic beltsare estimated from published data. On the scale of WesternEurope, the sedimentological studies and sequence stratigra-phy analysis of the continental environments allow us toconstrain the river basin size, the areas in erosion, transit andsedimentation, as well as the fluvial style (alluvial fan,braided, meandering or anastomosing) and the associatedpaleoenvironments (aeolian, flood-plain, lacustrine or sab-kha). We consider the paleogeographic maps for the LateScythian (Olenekian).

2.3. Paleotopography

It remains very difficult to estimate the elevation ofmountain ranges in the distant past. There is still no absolutemethod for quantifying paleoelevations at the global scaleand through geological time. Nevertheless, indirect methodscan provide a range of paleoelevations. The estimates ofmountain height are firstly based on the geodynamic con-text. The Early Triassic represents a transition between aperiod of mountain building (Variscan and Urals orogenicbelts) and the initial stages of the break-up of Pangea. How-ever, since there are still many uncertainties aboutpaleoelevations, we propose two reconstructions here: thefirst one corresponds to a low-relief scenario (Fig. 3A), andthe second to a high-relief scenario (Fig. 3B).

In the high-relief scenario (Fig. 3B), the Appalachiansunderwent their final orogenic event during the Early Per-mian. The estimates of elevation of the Appalachians varyfrom 3 [69] to 6 km [70]. According Fluteau

et al.

[34], amoderately elevated mountain range about 4 km high stillexisted during the Late Permian. Thus, an elevation of about3,000 m (slightly less at the model resolution) is attributed tothe highest peaks of the mountain range in the Olenekian.The Hercynian (Variscan) mountain range reached its maxi-mum height during the Late Carboniferous. Fluteau

et al.

[34] suggests that the height of the range could have beenaround 2000-3000 m in the Late Permian. We can thus esti-mate that the elevation of the Hercynian (Variscan)mountain range was about 1,500 m during the Olenekian.This result is confirmed by sedimentological analyses of flu-vial successions in France, which allow us to constrain thepaleoelevation of these series using the topographic slopegradients of each depositional environment [15]. WesternEurope is also characterized by numerous areas of moder-ately high relief that are generally elongated N-S, upliftedduring the onset of the Pangean breakup (Fig. 2B). Thesetopographic areas correspond to the uplifted parts ofgrabens. An elevation of about 500 m is proposed for suchareas. The collision between Laurussia and the Siberia-

Kazakstan block led to the uplift of the Urals during the LatePermian [39], [40], [43]. Large amounts of siliciclastic sedi-ment were deposited farther westward in the Urals seaway,suggesting that this mountain range was strongly eroded bymid-latitude precipitation during the Late Paleozoic andEarly Mesozoic. This sedimentation also suggests that therelief was high enough to trigger precipitation over its flank.Based on these considerations, we estimate that the elevationwas about 3 km during the Early Triassic. The southernlandmasses are characterized by some remnants of oldmountain ranges and by the Gondwanalides along the Pan-thalassa coast [41], [42], [43].

In a lower elevation scenario, we reduce the elevation ofthe Hercynian (Variscan) —Appalachians mountain rangeby a factor of 2. This lowering allows us to test the sensitivityof the climate to the height of the mountain range in theWest-Tethys domain.

Fig. 3 Paleoelevations reconstruction of the Early Triassic (Olenekian)for climate simulations: (A) low-relief scenario and (B) high-reliefscenario for the Hercynian (Variscan) and Appalachians.


435

3. Climate modelling

We used the Laboratoire de Métérologie Dynamique(LMD, France) Atmospheric General Circulation Model(LMDZ) [71]. It involves a 3D climatic model based on theequations of atmospheric dynamics, thermodynamics andthe physics of the atmosphere. This 3-D atmospheric modelhas a standard resolution of 96 points regularly spaced inlongitude, 71 points regularly spaced in latitude and 19 ver-tical layers unevenly spaced from the surface to 40 km high.The equations are solved using a discretized atmosphere.The purpose of each numerical experiment is to calculate theequilibrium of the atmosphere with boundary conditionsrepresenting the other components of the climatic system(paleogeography, surface temperature of the oceans, albedoand roughness of the ground, extension of the ice caps,orbital parameters, chemical composition of the atmosphere,CO2 in particular, and the solar constant). The experimentsare run for 20 years. Then, to remove the effects of the inter-nal variability of the model and obtain the average climate,the mean values are calculated over the final 10 years. Theseexperiments are qualified as “snap-shot experiments”. Thisgrid point model includes a full seasonal cycle as well as adiurnal cycle. This 3D atmospheric model has beenemployed to investigate present and past climates [72, 73,74, 75, 76]. Moreover, the LMDZ model has been validatedby checking its capacity to simulate the current climate.

The two paleogeographic reconstructions discussed hereare used to force the LMDZ Atmospheric General Circula-tion Model. Other boundary conditions are fixed in the set ofexperiments. There is no global dataset of sea-surface tem-peratures (SST) for the Early Triassic. Thus, we have todefine SST on a daily basis for each oceanic grid point. Weapplied the same SST values as in Fluteau

et al.

[34], whichhave been used to simulate the Late Permian climate.

The CO

2

atmospheric content is fixed at 1,200 ppmaccording to the geochemical modelling of the carbon cycledue to Berner [77]. Nevertheless, we know that large uncer-tainties remain. In the present study, we lower the solarconstant by 1% (1,343 Wm

–2

instead of 1,370 Wm

–2

), anduse the present-day orbital parameters. We also assign thevalues of albedo and roughness of continental points accord-ing to vegetation. Because of the lack of any globalbiospheric reconstruction, we use a “transfer” function cali-brated on the present day to assign an albedo and a roughnessat each grid point according to the elevation and the latitude.After the Permo-Triassic ecological crisis, the Early Triassicvegetation could have been relatively sparse or differentfrom that mimicked by the albedo and roughness values pro-posed in this study. Nevertheless, our primary goal is toinvestigate the sensitivity of climate to paleogeography.

4. Sedimentation of the Triassic successions in the western Tethys domain

A sedimentological study was first focused on the ParisBasin, where a complete set of subsurface data (cores andwell-logs) is available [7], [78], [79]. Comparisons and cor-relations with the Germanic Basin [44], [45], [46], [80], [81]allow us to determine the evolution of the river basins fromthe Early to Late Triassic in Western Europe (Fig. 4). At thebeginning of the Early Triassic (Induan), the sedimentationarea was restricted to the central part of the Germanic Basin:the Lower Buntsandstein Formations, which are character-ized by a playa system, are only found in the central basinand do not have any equivalents on the western margin [82].These deposits characterize the regressive tendency of theuppermost Zechstein (“Bröckelschiefer”), which continuesinto the Lower Buntsandstein [46], [83]. The angular uncon-formity at the base of the Upper “Bröckelschiefer” wascaused by non-deposition or erosion of the uppermost Zech-stein at the basin margin [45], [84]. During the Olenekian(Figs. 2B; 4A), the sedimentation is characterized in thewestern Germanic basin by braided fluvial systems evolvinglaterally, toward the central part of the basin, into more orless ephemeral lakes [44], [45], [46], [75], [80], [82] andaeolian deposits [50], [85]. Moreover, in the northern Ger-manic Basin, alluvial fan deposits are described that comefrom the Fennoscandian Massif (Fig. 2B). During the Olene-kian, in the north-western Tethys domain, fluvial depositsare preserved in a large endoreic basin.

Above the major Hardegsen unconformity, dated asOlenekian [45], [51], [86], [87], the Triassic exhibits achange in fluvial system style. This discontinuity is markedby an erosional surface, characterized by the first develop-ment of paleosols (“Zone limite violette”). During theHardegsen Phase, an important structural event occurs inNW Europe [51], leading to the formation of rift systems inNW Germany [45], [88]. Above this surface, the Triassicseries show an onlap relationship, which implies some tec-tonic activity just before the renewal of fluvialsedimentation. The fluvial systems are the lateral equiva-lents of playa and, eventually, sabkha and marine deposits.In Anisian and Ladinian times, the Germanic Basin is con-nected with the Tethys Ocean [89], and the marginal fluvialsystems are developed in an exoreic basin setting (Fig. 4B).In the Paris Basin, the Upper Triassic sedimentation (Fig.4C) represents braided rivers (Carnian and Norian) associ-ated with lacustrine environments [7], [15]. In the centralbasin, the climate is arid and characterized by evaporitic seb-khas [79], [90]. These fluvial systems are located in morenumerous small drainage catchments (around 50 km long)compared with the Early Triassic systems (> 1,000 kmlong). In fact, there is no a single drainage catchment, but asuccession of small catchments that reflect the independenceof the present-day Paris Basin from the Germanic Basinsince the Upper Carnian [78], [79].

436


Fig. 4 Evolution of fluvial systems from the Early to Late Triassic in Western Europe. (A) During the Olenekian (after [66]), at the scale of theGermanic Basin, the rivers sources are mainly located in the present-day Armorican Massif, the paleocurrents indicate a general direction towards theNNE. (B) During the Anisian, the fluvial systems are connected to the Tethys sea. (C) During the Lower Carnian and Norian, fluvio-lacustrinesedimentation takes place (after [7]).


437

This Triassic fluvial evolution (Fig. 4), from large endo-reic river basins to a coastal system and small river basins,could result from the geodynamic evolution of Pangea. Infact, the Early Triassic represents the beginning of the Pan-gea break up, and the formation of large endoreic basinswhere sediments could be preserved. Consequently, thepreservation of fluvial systems and their spatial distributioncan be attributed to a tectonic setting controlling the accom-modation space. On the other hand, the stratigraphicarchitecture observed in the fluvial environments could bethe result of other allocyclic factors.

In fact, the sedimentological studies of Triassic fluvialdeposits located in the western part of the Germanic Basin[7], [66], [91], [92], reveal stratigraphic cycles at differentscales (ranging from genetic units, with a duration of 20 to100 ka, to minor cycles with a duration of 1 to 5 Ma). Thestratigraphic cycles reflect the variations in accommodationspace (A) and sediment supply (S

S

). At the scale of thesmallest stratigraphic cycle, the A and S

S

variations cannotbe tectonically induced. For the Middle Triassic (

i.e.

Ladin-ian to Anisian) and the Upper Triassic (

i.e.

Carnian toNorian), the smallest stratigraphic cycles reflect sedimentsupply and/or water table variations induced by changes insea level [66] or lake level [7], [15]. For the Early Triassic(

i.e.

Olenekian), the sedimentological study of Péron [92]leads to the following reconstruction of the depositionalenvironments (Fig. 4A): (1) braided rivers in an arid context,with very few floodplain deposits, (2) marginal erg (braidedriver and aeolian deposits and (3) playa lake environment(fluvial and ephemeral-lake environment). The river catch-ments are mainly located in the present-day ArmoricanMassif, and the paleocurrent directions are generallytowards the NNE (Fig. 4A). The facies associations areessentially composed of stacked channel-fill facies with veryfew flood plain deposits (< 3%) and without paleosols. Inmore distal areas of the Germanic Basin, the sedimentationcorresponds to ephemeral lakes or aeolian deposits [46],[50], [85], [93]. In the northern part of the basin, alluvial fansediments are present (Fig. 4A). In this arid context, very lit-tle influence is seen from sea-level variations, exceptedlocally in the extreme eastern part of the Germanic Basin[94] that corresponds to the maximum westward extent ofplaya environments [66]. The smallest stratigraphic cycles,which are observed within an environment without commu-nication with the open sea, reflect sediment supply and lake-level variations controlled by climate variations. Within theTriassic series, this could arise from variations of precipita-tion in a warm climate. For the Olenekian, two hypothesescould explain the preservation of fluvial systems in the Euro-pean basins. According to one hypothesis, the siliciclasticsediment supply from the Hercynian (Variscan) range wasconstant, and thus the stratigraphic cycles result solely fromlake-level fluctuations. The lake levels could be controlledby monsoonal activity, under the influence of sea-level fluc-tuations [80]. Alternatively, the basin was situated in an aridcontext and the stratigraphic cycles result from sediment and

water supply variations controlled by precipitation over themountain ranges. Climate simulations are used to investigatethe relationships between sedimentation and climate duringthe Early Triassic (Olenekian).

In North Africa, fluvial sedimentation during theScythian occurs only in the Tunisian and Libyan basins andis characterized by coastal plain deposits [54], [55], [95]. Inthe other Triassic basins (Morocco, Algeria), an angularunconformity at the base of the Triassic represents theboundary between pre-Triassic deposits and the Middle orUpper Triassic sedimentary succession. Fluvial sedimenta-tion without sea connection seems to be absent in thiswestern part of Tethys.

5. Climate simulation results

Two experiments (low topography, Fig. 3A, and hightopography, Fig. 3B) were run to simulate the Olenekian cli-mate and test the impact of orography on climate at that time.Firstly, we discuss a climate simulation in the case of lowtopography (Fig. 3A). Then, we compare these results withsimulations in the case of high topography (Fig. 3B). Bothexperiments are compared with proxy data. Finally, we con-sider the consequences of the simulated climate onsedimentation over Western Europe.

The climate simulated in the low relief experiment is glo-bally warm and mostly dry (Fig. 5). Indeed the global annualmean surface temperature is 17.3°C, about 4°C above that ofthe present-day. The difference in surface temperatureresults from a change in land-sea distribution as well as theabsence of ice sheets and sea ice during the Scythian. Theglobal annual mean precipitation is about 3.2 mm/day(1,150 mm/year). In fact, these first observations maskimportant disparities in temperature and rainfall over theEarth’s surface as well as large seasonal fluctuations.

The continents of Laurussia and Gondwana are separatedby a mountain range (mostly represented by the Appalachi-ans) trending along a SW-NE axis. Because of the high relief(up to 1,800 m, Fig. 3B), the mean annual surface tempera-ture does not exceed 25°C (Fig. 5A). By contrast, the meanannual surface temperature exceeds 30°C over two vastmountainous areas located on either side of the equator, onein Laurussia (including North America, southern Greenland,and western Europe) and the other in northern Gondwana(northern part of South America and North Africa).

When averaged annually, the simulation yields heavyprecipitation over the western part of the Appalachians,whereas the easternmost part close to the Paleotethys Oceanremains dry (Fig. 5B). In the subtropics, the two warmerzones are characterized by weak precipitation, although Lau-russia is much drier than northern Gondwana (Fig. 5B). Avast area extending from the coast of Panthalassa to northernPaleotethys receives less than 0.5 mm/day (180 mm/year) ofrainfall, which is close to the present-day threshold value fora desert environment. In the southern hemisphere, the

438


Fig. 5 Climate simulations for the Early Triassic (Olenekian), assuming low relief of the mountain range separating the continents of Laurussia andGondwana.


439

warmer area (expressed as an annual mean) does not corre-spond to a drier climate. In fact, rainfall in the northern partof Gondwana (North Africa and the Arabian Peninsula,Fig. 5B) ranges from at least 1 mm/day (360 mm/year) up to5 mm/day (1,800 mm/year).

The mean annual precipitation in fact masks seasonalchanges both over the area of equatorial relief and in the sub-tropics. Precipitation occurs during the winter and spring allalong the mountain range, whereas the summer is dry for amajor part of this range (Fig. 6). A dissymmetry in precipi-tation on either side of the equatorial relief is controlled bythe geographic position of the relief and the axial trend of themountain range: moisture mainly arrives from Panthalassaand the simulated precipitation over Laurussia remains weakthroughout the year, whereas a dry/wet season is simulatedover northern Gondwana. The summer overheating favoursthe development of a thermal low pressure cell over Laurus-sia and northern Gondwana (Fig. 6A). Moist air massescoming from Paleotethys cannot penetrate farther inland, so

precipitation is very weak and does not exceed 0.5 mm/dayas observed during the autumn (Fig. 6B). During the winter(austral summer, Fig. 6C), the paleogeographic configura-tion of Gondwana shifts the thermal low pressure cellsouthwards, allowing Panthalassan moist air masses to pen-etrate far inland over the northern part of Gondwana, leadingto mixing with the Paleotethys moist air masses. Thus, theaustral summer is marked by a monsoon-type circulationalong the southern Paleotethys coast bringing moisture andproducing precipitation over north-eastern Gondwana. Dur-ing the spring (austral autumn, Fig. 6D), a rainy season(2-4 mm/day) is simulated over northern Gondwana, whichis due to the onset of the warming of the northern subtropicsand a weakening of the thermal low pressure zone abovenorthern Gondwana. We observe a maximum of radiativesolar flux located over the equatorial mountain range, whichamplifies precipitation over the relief, as well as a low-pres-sure cell, which remains in a southerly position(corresponding more or less to the arid area). As for winter,

Fig. 6 Detail of West-Tethys climate simulations during the Early Triassic (Olenekian), assuming a low-relief scenario and for each season: (A)summer, (B) autumn, (C) winter (austral summer), and (D) spring (austral autumn). See figure 5 for the location.

440


rainfall from Panthalassan moist air masses can penetrate farinland over the northern part of Gondwana.

At high latitudes (Siberia and southern Gondwana), themean annual temperature is close to freezing point, and mayeven fall below this threshold value locally over elevatedareas (Fig. 5A). A strong latitudinal gradient is thus simu-lated in both hemispheres at mid-latitudes. This strongthermal gradient drives the mid-latitude precipitation belts,which could be locally enhanced over mountain ranges suchas the Urals (Fig. 5). A similar intensification is seasonallyobserved over the Fennoscandian Mountain ranges. Thehighest precipitation is generally simulated during the springof both hemispheres when the thermal gradient is strongest.The Olenekian climate clearly depends on the prescribedboundary conditions. However, the elevation of the moun-tain ranges remains a major source of uncertainty. Thus, weperformed a second experiment taking account of Hercynian(Variscan) —Appalachians and Fennoscandian Mountainranges with higher relief (Figs. 3B, 7). The difference in ele-vation (Fig. 7A) between the two experiments reaches about1,000 m for the highest areas of the Appalachians, and about500 m for the eastern part of this range as well as the moun-tains located on both sides of the Arctic Seaway. Figures 7Band 7C present the differences in mean annual temperatureand precipitation in response to the difference in elevation.The change in elevation evidently leads to a cooling of morethan 6°C over the highest summits, along with a warmingover the adjacent lowlands in both hemispheres that exceeds1°C over Gondwana. This warming could exceed 2°C overnorthern Gondwana in winter. A decrease in precipitation issimulated in the lowlands of Gondwana along the mountainrange (Fig. 7C). This drying is more pronounced in winterand spring. No change is simulated during the summer andautumn, which already correspond to the dry season. Thereis no change in precipitation over Laurussia. A change in ele-vation modifies the atmospheric circulation, increasing theadvection of moisture over the relief, whereas the adjacentlowlands become drier. This drying implies a decrease incloudiness and favours climatic warming.

We also compared the climate simulations with proxydata.

6. Comparisons between climate simulations and sedimentological analysis

While the climatic simulations were carried out at a glo-bal scale, our study is nevertheless focused on the westernTethys basins. In this area, we can compare the simulationswith sedimentological data, allowing us to interpret theresults in terms of sediment supply and fluvial system pres-ervation. We calculated monthly mean precipitation (

i.e.

MMP, Fig. 8) and the monthly mean difference betweenevaporation and precipitation (

i.e.

MMEvP, Fig. 9) and themean monthly temperature (

i.e.

MMT, not shown) in differ-ent regions selected according to their respective

paleoenvironments. Three areas belonging to WesternEurope are named “Germanic Basin” “Spain” and “WesternFrance” (Figs. 8A, 9A), while an area located in the northernGondwana continent is named “North Africa”, and anotherarea, named “North Africa Variscan Mountains” corre-sponds to the eastern part of the Hercynian (Variscan) —Appalachians mountain range (Figs. 8A, 9A).

Fig. 7 Detail of West-Tethys simulations during the Early Triassic(Olenekian). (A) The difference in elevation between the twoexperiments reaches about 1,000 m for the highest areas of theAppalachians, and is about 500 m for the eastern part of this range andfor mountain ranges located on both sides of the Arctic Seaway.Difference in mean annual temperature (B) and precipitation (C) inresponse to the difference in elevation for the western Tethys domain.See figure 5 for the location.


441

Fig. 8 Monthly mean precipitation for the low- and the high-relief scenario. (A) Location of the different regions selected, F: FennoscandianMountains, GB: Germanic Basin, W: Western France, S: Spain, NAV: North Africa Variscan Mountains, NA: North Africa (See figure 2B for thelocation). (B to G) Monthly mean precipitation for these different regions.

442


Sedimentological studies of Western Europe show thatpaleoenvironments during the Olenekian were made up oflarge rivers, with their sources mainly located at the marginsof playa systems containing ephemeral lake and/or aeoliandeposits (Figs. 2B, 4A). The simulated climate in this area isin agreement with these field observations. MMT (Fig. 5A)does not fall below 26°C in both experiments, and MMP isextremely weak, not exceeding 0.2 mm/day (Fig. 8B, C). Interms of hydrological budget (MMEvP), we simulate overwestern-Europe (Spain) a near equilibrium between evapo-ration and precipitation throughout the year, due to theabsence of available water in the soil (Fig. 9B), whereasevaporation exceeds largely precipitation in the Germanicbasin (Fig. 9C). This paleoenvironment is compatible with asimulated dry climate not controlled by the paleo-elevationof the North Africa Variscan Mountains. The presence of

rivers in an arid environment suggests that these rivers werefed by precipitation falling on adjacent areas.

The sedimentation in the Germanic basin is mainly repre-sented by ephemeral lakes or aeolian deposits (Fig. 2B, 4A).This is clearly consistent with the simulated climate, whichis warm and arid in this area. Monthly temperature (MMT,not shown) ranges from about 20° in winter to almost 33°Cin summer. Precipitation is as weak as 0.4 mm/day. The highevaporation rate (Fig. 9C) simulated in this area is due to thefact that we assume a large lake in our paleogeographicreconstructions. Thus, evaporation exceeds largely precipi-tation throughout the year. Without a large lake, theevaporation rate would be in equilibrium with precipitation.These results suggest that the lifetime of lake in this areawould have been short because of the high evaporation rate.This is consistent with proxy data. The ephemeral lake

Fig. 9 Monthly mean difference between evaporation and precipitation for the low- and the high-relief scenario for the sedimentation areas. (A)Location of the different sedimentation areas, GB: Germanic Basin, S: Spain, NA: North Africa (See figure 2B for the location). (B to D) Monthly meandifference between evaporation and precipitation for these three sedimentation areas.


443

deposits occurring in the Germanic Basin during this periodcan be attributed to precipitation flowing out of the Fennos-candian Mountain ranges, and/or are associated with watersupply imported by rivers from the North Africa VariscanMountains.

In the northern Germanic basin, the Triassic sedimentsare mostly derived from the Fennoscandian highland as wellas from the remnants of the Bohemian massif (Fig. 2B). Thisis consistent with our simulations, showing that seasonalprecipitation, mainly in winter, falls on the FennoscandianMassif (Fig. 8D). Indeed, precipitation simulated over theFennoscandian Mountain ranges does not vary significantlyin response to an increased paleoelevation of this range.

Precipitation simulated on the North Africa VariscanMountains (Fig. 8E) shows that the mountains produce awater supply able to control the sediment supply (S

S

) of therivers. This scenario is consistent with the sedimentologicalanalysis (rivers in an arid climate context) and the paleo-cur-rent directions. The water supply evidently depends on theelevation of the mountain ranges. The presence of large riv-ers suggests a voluminous water supply, suggesting that theNorth Africa Variscan Mountains were still elevated duringthe Olenekian. If we assume that the relief was already flat-tened at that time, the precipitation would have been tooweak to feed rivers. However, the simulation on WesternFrance shows a very weak precipitation (Fig. 8F). This sce-nario is inconsistent with sedimentological data: rivers comefrom the Armorican Massif, which represent the fluvial tran-sit area. Two hypothesis may arise: (1) either thepaleoelevation of the Armorican Massif has been under-esti-mated or (2) a continuous mountain range, located in thewestern Iberian Peninsula, should be considered betweenArmorican Massif and North Africa Variscan Mountains.The latter hypothesis implies that water was supplieddirectly from the North Africa Variscan Mountains. The cli-mate of the Germanic basin is not sensitive to either an upliftof the North Africa Variscan Mountain or the FennoscandianMountain ranges.

During the Scythian in North Africa, the Triassic basinsmainly exhibit erosion or by-pass sedimentation. Olenekiansedimentation is characterized exclusively by coastal plaindeposits located in the Tunisian and Lybian basins (Fig. 2Band see chapter 4). Simulated temperature is very hotthroughout the year, about 30-33°C. In contrast to the otherareas discussed here, this region is subject to seasonal pre-cipitation occurring mainly in winter and spring (Fig. 8G).Clearly, the evaporation exceeds precipitation throughoutthe year because of the heat (Fig. 9D). However, theseresults are not inconsistent with the presence of rivers. Thislocal precipitation is added to the water supply coming fromrainfall over the North Africa Variscan Mountain. By meansof simulations, we show that the elevation of the NorthAfrica Variscan Mountains controls the aridity of the NorthAfrica basin as well as the water supply to rivers. During, theOlenekian, the accommodation space available for the sedi-mentation (A) is located in Western Europe, and the

sediments are preserved in an endoreic basin where thesmallest stratigraphic cycles, (

i.e.

genetic units) seem to becontrolled by sediment and water supply variations inducedby precipitation over the mountain range. Consequently, inthis area, A and S

S

are in equilibrium (Fig. 10A and a). How-ever, the sediment supply derived from areas of high relief isnot preserved in the endoreic setting of the North Africanbasins, but is preserved only as coastal deposits. Conse-quently, two hypotheses can be considered. In the firsthypothesis, we assume that no accommodation space (A < 0)is created in this area, and that all the sediment supply tran-sits to the sea (Fig. 10B and b). In the second hypothesis,accommodation space is created (A > 0) but, because of theclimatic conditions in a basin subject to seasonal precipita-tion, the sediments are not preserved and they undergo by-pass in continental areas. In fact, at the onset of the fluvialsedimentation, the sediment supply fills in the basin, S

S

< Aand preservation is possible (Fig. 10B and c). When S

S

> A(no subsidence, coupled with a constant sediment supply),the sedimentation area could be in transit during an initialstage and then become subject to erosion. As a result, all thesediments transit to the sea and are preserved in coastal envi-ronments (Fig. 10b).

Fig. 10 Hypothetical longitudinal profiles of the European Triassicendoreic basin (A) and of the North-Africa Triassic exoreic basins (B).When the sediments of an endoreic basin (a) is connected to the sea allthe sediment supply will transit to the sea (b) or, alternatively, endoreicbasin can be preserved (b).

444


7. Conclusion and perspectives

By combining climate simulations with paleoenviron-mental reconstructions based on sedimentological studies onthe scale of West Tethys, we propose a scenario for the pres-ervation of fluvial systems in Triassic times. Thepreservation of these systems is firstly controlled by accom-modation space, which is itself tectonically induced incontinental environments. During the Early Triassic (Olene-kian), fluvial sediments in Western Europe are preserved inendoreic basins, while, in North Africa, they are preserved incoastal environments without the development of endoreicbasins. The simulations demonstrate that climatic conditionsdiffered between these two areas, with North Africa beingmore humid and subject to seasonal precipitation. In NorthAfrica, the absence of fluvial systems preserved in endoreicbasins could either be due to a lack of accommodation spaceor could be linked to the specific climatic conditions of thisarea. In the latter case, we might ask whether the water sup-ply is able to control the preservation of fluvial systems.

To address this question, we propose (1) an analysis of thePermo-Triassic transition in North Africa to show if sedi-ments in this area were actually undergoing transit or beingeroded during the Scythian, (2) a comparison with other flu-vial systems preserved in coastal environments under warmclimatic conditions during other geological periods (Creta-ceous, for example).

Acknowledgements

We are grateful for the financial support of INSU/CNRSprogramme ECLIPSE:

Environnements et Climats du Passé,entitled “Quantification des flux sédimentaires au Trias:réponse des systèmes sédimentaires continentaux aux sollic-itations climatiques et tectoniques”

. We thank the LSCE/DSM/CE at Saclay for providing computing time on super-computers. We gratefully acknowledge Dr J. TotmanParrish and Dr. G. Bachmann for their critical and construc-tive reviews of the paper. Dr. M.S.N. Carpenter post-editedthe English style.

References

[1] Stanistreet I.G., McCarthy T.S., The Okavango Fan and the classifi-cation of subaerial fan systems, Sediment. Geol. 85 (1993) 115-133.

[2] Blair T.C., McPherson J.G., The Trolheim alluvial fan and faciesmodel revisited, Geol. Soc. Am. Bull. 104 (1992) 762-769.

[3] Bull W.B., Discontinuous ephemeral streams, Geomorpholgy 19(1997) 227-276.

[4] Legarreta L., Uliana M.A., Anatomy of Hinterland depositionalsequences: Upper Cretaceous fluvial strata, Neuquen basin, west-central Argentina, in: Shanley K. W., McCabe P. J. (Eds.), RelativeRole of Eustasy, Climate and Tectonism in Continental Rocks, Soc.Ec. Paleontol. Mineral., Sp. Publ 59 (1998) pp. 83-92.

[5] Clemmensen L.B., Hegner J., Eolian sequence and erg dynamics:the Permian Corrie Sandstone, Scotland, J. Sediment. Petrology 61(1991) 768-774.

[6] Ryang W.H., Chough S.K., Sequential development of alluvial/lacustrine system: Southeastern Eumsung Basin (Cretaceous),Korea, J. Sedim. Res. 67 (1997) 274-285.

[7] Bourquin S., Rigollet C., Bourges P., High-resolution sequencestratigraphy of alluvial fan-fan delta environment: stratigraphic andgeodynamic implications - An example from Keuper ChaunoySandstones, Paris Basin, Sediment. Geol. 121 (1998) 207-237.

[8] Anderson D.S., Cross T.A., Large-scale cycle architecture in conti-nental strata, Hornelen Basin (Devonian), Norway, J. Sediment. Res.71 (2001) 255-271.

[9] Veiga G.D., Spalletti L.A., Flint S., Aeolian/fluvial interaction andhigh-resolution sequence stratigraphy of a non-marine lowstandwedge: the Avilé Member of the Agrio formation (Lower Creta-ceous), central Neuquén Basin, Argentina, Sedimentology 49 (2002)1001-1019.

[10] Olsen H., Larsen P.-H., Structural and climatic controls on fluvialdepositional systems: Devonian, North-East Greenland, Int. Ass.Sediment. Spec. Publs. 17 (1993) 401-423.

[11] Postma G., Fortuin A.R., van Wamel W.A., Basin-fill patterns con-trolled by tectonics and climate: the Neogene «fore-arc» basins ofeastern Crete as a case history, Int. Ass. Sediment. Spec. Publs 20(1993) 335-362.

[12] Schumm S.A., River response to baselevel change: implications forsequence stratigraphy, J. Geol 101 (1993) 279-294.

[13] Shanley K.W., McCabe P.J., Relative role of eustasy, climate, andtectonism in continental rocks, Soc. Ec. Paleontol. Mineral. Sp.Publ. 59 Tulsa (1998) 234 p.

[14] Dalrymple M., Prosser J., Williams B., A dynamic system approachto the regional controls on deposition and architecture of alluvialsequences, illustrated in the Statfjord formation (United Kingdom,northern North sea), in: Shanley, K.W., McCabe P.J (Eds.), RelativeRole of Eustasy, Climate and Tectonism in Continental Rocks, Soc.Ec. Paleontol. Mineral Spec. Publ. 591 (998) pp. 65-81.

[15] Bourquin S., Robin C., Guillocheau F., Gaulier J.-M., Three-dimen-sional accommodation analysis of the Keuper of the Paris Basin:discrimination between tectonics, eustasy, and sediment supply inthe stratigraphic record, Mar. Petrol. Geol 19 (2002) 469-498.

[16] Jervey M.T., Quantitative geological modelling of siliciclastic rocksequences and their seismic, in: Wilgus C.K., Hastings B.S., KendallC.G.St.C., Posamentier H., Ross C. R., Van Wagoner J. C. (Eds.),Sea-level changes: an integrated approach, Soc. Econ. Paleontol.Mineral. Sp. Publ. 42 (1988) pp. 47-70.

[17] Cross T.A., Controls on coal distribution in transgressive-regressivecycles, Upper Cretaceous, Western Interior, U.S.A, in: Wilgus C.K.,Hastings B.S., Kendall C.G.St.C., Posamentier H., Ross C. R., VanWagoner J. C. (Eds.), Sea-level changes: an integrated approach,Soc. Ec. Paleontol. Mineral. Sp. Publ. 42 (1988) pp. 371-380.

[18] Galloway W.E, Genetic stratigraphic sequences in basin analysis I:architecture and genesis of flooding-surface bounded depositionalunits, Am. Assoc. Petrol. Geol. Bul. 73 (1989) 125-142.

[19] Galloway W.E., Williams T.A., Sediment accumulation rates in timeand space: Paleogene genetic stratigraphic sequences of the north-western Gulf of Mexico basin, Geology 19 (1991) 986-989.

[20] Cross T.A., Lessenger M.A., Sediment volume partitioning: ration-ale for stratigraphic model evaluation and high-resolutionstratigraphic correlation, in: Gradstein F.M., Sandvik K.O. & MiltonN.J. (Eds.), Sequence stratigraphy. Concepts and applications, Nor-wegian Petroleum Society (NPF) Sp. Publ. Elsevier Science B.V. 8Amsterdam (1998) pp. 171-195.

[21] Muto T., Steel R.J., Principles of regression and transgression: thenature of the interplay between accommodation and sediment sup-ply, J. Sedim. Res 67 (1997) 994-1000.


445

[22] Muto T., Steel R.J., The accommodation concept in sequencestratigraphy: some dimensional problems and possible redefinition,Sediment. Geol. 130 (2000) 1-10.

[23] Paola C., Heller P.L., Angevine C.L., The large-scale dynamics ofgrain-size variation in alluvial basins, 1: theory, Basin Research 4(1992) 73-90.

[24] Blum M.D., Toomey III R.S., Valastro S.J., fluvial response to LateQuaternary climatic and environmental change, Edwards Plateau ofTexas, Palaeogeogr., Palaeoclimatol., Palaeocol 108 (1994) 1-21.

[25] Marr J.G., Swenson J.B., Paola C., Voller V.R., A two-diffusionmodel of fluvial stratigraphy in closed depositional basins, BasinResearch 12 (2000) 381-398.

[26] Miller D.J., Benda L.E., Effects of punctuated sediment supply onvalley-floor landforms and sediment transport, Geol. Soc. Am. Bull112 (2000) 1814-1824.

[27] Paola C., Quantitative models of sedimentary basin filling, Sedi-mentolgy 47 (2000) 121-178.

[28] Pederson J., Pazzaglia F.J., Smith G., Ancient hillslope deposits:missing links in the study of climate controls on sedimentation,Geology 28 (2000) 27-30.

[29] Van der Voo R., Palaeomagnetism of the Atlantic, Tethys and Iape-tus Oceans, Cambridge University Press (1993) 411 p.

[30] Van der Voo R., French R.B., Apparent polar wander for the Atlan-tic-bordering continents: Late Carboniferous to Eocene, EarthPlanet. Rev. 10 (1974) 99-119.

[31] Irving, E., Drift of the major continental blocks since the Devonian,Nature270 (1977) 304-309.

[32] Livermore R.A., Smith A.G., Vine F.J., Late Palaeozoic to earlyMesozoic evolution of Pangea, Nature 322 (1986) 162-165.

[33] Torcq F., Besse J., Vaslet D., Marcoux J., Ricou L.E., Halawani M.,Basahel M., Paleomagnetic results from Saudi Arabia and thePermo-Triassic Pangea configuration, Earth Planet. Sci. Lett. 148(1997) 553-567.

[34] Fluteau F., Besse J., Broutin J., Ramstein G., The Late Permian cli-mate. What can be inferred from climate modelling concerningPangea scenarios and Hercynian range altitude?, Palaeogeogr., Pal-aeoclimatol., Palaeoecol 167 (2001) 33-71.

[35] Besse J., Torcq F., Gallet Y., Ricou L.E., Krystyn L., Saidi A., LatePermian to Late Triassic palaeomagnetic data from Iran: constraintson the migration of the Iranian block through the Tethyan ocean andinitial destruction of Pangaea, Geophys. J. Int. 135 (1998) 77-92.

[36] Sengör A.M.C., Hsu K.J., The Cimmerides of Eastern Asia: historyof the Eastern end of the Paleo-Tethys, Mem. Soc. Geol. France 147(1984) 139-167.

[37] Metcalfe I., Gondwanaland dispersion, Asian accretion and evolu-tion of eastern Tethys, Austr. J. Earth Sci. 43 (1996) 605-623.

[38] Enkin R.J., Courtillot V., Leloup Ph., Yang Z., Xing L., Zhang J.,Zhuang Z., The paleomagnetic record of upper most Permian, lowerTriassic rocks from the south China block, Geophys. Res. Lett. 19(1991) 2147-2150.

[39] Ziegler P.A., 5. Permo-Triassic development of Pangaea, in: ZieglerP.A. (Ed.), Geological Atlas of western and Central Europe, TheHague: Shell International Petroleum Maatsschappij BV. (1990) pp.68-90.

[40] Ziegler A.M., Hulver M.L., Rowley D.B., Permian world topogra-phy and climate, in: Martini I.P. (Ed.), Late Glacial and Post-GlacialEnvironmental Changes - Quaternary, Carboniferous-Permian andProterozoic, Oxford University press, New-York (1997) pp. 111-146.

[41] Wilson K.M., Pollard D., Hay W.W., Thompson S.L., Wold C.N.,General circulation model simulations of Triassic climates: Prelimi-nary results, in: Klein G.D. (Ed.), Pangea: Paleoclimate, Tectonicsand Sedimentation During Accretion, Zenith, and Break-up of aSupercontinent, Geol. Soc. Am. Sp. Pap. 288 (1994) pp. 91-116.

[42] Scotese C.R., Atlas of Earth History, PALEOMAP Project, Arling-ton, Texas, 2001, 52 p.

[43] Golonka J., Ford D., Pangean (Late Carboniferous - MiddleJurassic) paleoenvironment and lithofacies, Palaeogeogr., Palaeocli-matol., Palaeoecol 161 (2000) 1-34.

[44] Richter-Bernburg G., Stratigraphische Synopsis des deutschenBuntsandstein, Geol. Jb. (1974) 127-132.

[45] Röhling H.-G., A lithostratigraphic subdivision of the Lower Trias-sic in the Northwest german lowlands and the german sector of theNorth Sea, based on Gamma-Ray and Sonic logs, Geol. Jb. (1991)3-24.

[46] Aigner T., Bachmann G.H., Sequence-stratigraphic framework ofthe German Triassic, Sediment. Geol. 80 (1992) 115-135.

[47] Wachutka M., Aigner T., Reservoirgeologie und Petrophysik: eineAufschluss-Analogstudie im Buntsandstein (Trias, Schwarzwald,SW-Deutschland), Z. angew. Geol., 47 (2001) 191-198.

[48] Audley-Charles M.G., Triassic, in: Duff P.M.D., Smith A.J. (Eds.),Geology of England and Wales, Geol. Soc. London (1992) pp. 307-324.

[49] Belivanova V., Triassic in the Golo Bardo Mountain - one examplefor the Balkanide facial type of Triassic in Bulgaria, Zbl. Geol.Paläont. Teil I 2 (1998) 1105-1121.

[50] Clemmensen L.B., Triassic lacustrine red-beds and palaeoclimate:The «Buntsandstein» of Helgoland and the Malmros Klint Memberof East Greenland, Geol. Rundsch. 68 (1979) 748-774.

[51] Geluk M.C., Röhling H.-G. High-resolution sequence stratigraphyof the Lower Triassic ’Buntsandstein’ in the Netherlands and north-western Germany. Geologie en Mijnbouw 76 (1997) 227-246.

[52] Geluk M.C., Palaeogeographic and structural development of theTriassic in the netherlands - new insights, Zbl. Geol. Paläont. Teil I1 (1998) 545-570.

[53] Goldsmith P.J., Hudson G., Van Veen P., Chapter 9: Triassic, in:Evans D., Graham C., Armour A., Bathurst P. (Eds.), The Millen-nium Atlas: petroleum geology of the central and northern NorthSea, Geol. Soc. London (2003) pp. 105-127.

[54] Kamoun F., Peybernès B, Ciszak R., Calzada S., Triassic paleogeog-raphy of Tunisia, Palaeogeogr., Palaeoclimatol., Palaeoecol. 172(2001) 223-242.

[55] Hallet T D., Petroleum Geology of Lybia, Elsevier Science B.V.(2002) 503 p.

[56] Lopez-Gomez J., Arche A., Sequence stratigraphic analysis andpaleogeographic interpretation of the Buntsandstein and Muschel-kalk facies (Permo-Triassic) in the SE Iberian Range, E Spain,Palaeogeogr., Palaeoclimatol., Palaeoecol 103 (1993) 179-201.

[57] Lopez-Gomez J., Arche A., El Pérmico y el Triasico del Levanteespañol. Caracteristicas principales y consideraciones paleogeogra-ficas, Cuadernos de Geologia Ibérica (1995) 201-234.

[58] Sanchez-Moya Y., Sopeña A., Ramos A., Infill architecture of anon-marine half-graben Triassic basin (Central Spain), J. Sedim.Res 66 (1996) 1122-1136.

[59] Mader D., Evolution of fluvial sedimentation in the Buntsandstein(Lower Triassic) of the Eifel (Germany), Sediment. Geol. 37 (1983)1-84.

[60] Mader D., Teyssen T., Palaeoenvironmental interpretation of fluvialred beds by statistical analysis of palaeocurrent data: examples fromthe Buntsandstein (Lower Triassic) of the Eifel and Bavaria in theGerman Basin (Middle Europe), Sediment. Geol 41 (1985) 1-74.

[61] Olaussen S., Larsen B.T., Steel R., The Upper Carboniferous-Per-mian Oslo Rift: basin fill in relation to tectonic development., in:Pangea: Global environments and resources, Can. Soc. Petrol. Geol.Mem. 17 (1994) pp. 175-197.

[62] Cassinis G., Durand M., Ronchi A., The Permian and Triassic con-tinental framework of Nurra (NW Sardinia). Rendiconti dellaSocietà Paleontologica Italiana, 1 (2002,) 297-305.

[63] Cassinis G., Durand M., Ronchi A., Permian-Triassic continentalsequences of Northwest Sardinia and South Provence: Stratigraphiccorrelations and palaeogeographic implications. Bollettino dellaSocietà Geologica Italiana, vol. spec. n° 2 (2003) 119-129.

446


[64] Sciunnach D., The Permo-Triassic clastics of Nura: sedimentology,heavy mineral provinces and palaeogeography, Rend. Soc. Paleont.Ital. (2002) 177-184.

[65] Durand M., Paléocourants et reconstitution paléogéographique:l’exemple du Buntsandstein des Vosges méridionales (Triasinférieur et moyen continental), Thèse de Sciences de la TerreNancy 22 4 (1978) 301-390.

[66] Bourquin S., Péron S., Durand M., Lower Triassic sequence stratig-raphy of the western part of the Germanic Basin (west of BlackForest): fluvial system evolution through time and space, Sediment.Geol. (in press)

[67] Gaetani M., Map 4: Olenekian (245-243 Ma), in: Dercourt J., Gae-tani M., Vrielynck B., Barrier E., Biju Duval B., Brunet M.-F., CadteJ.-P., Crasquin S., Sandelulescu M. (Eds,), Atlas Peri-Tethys, Pale-ogeographical maps Maps and Explanatory Notes CCGM/CGMW,Paris (2000) pp. 27-32.

[68] Veevers J.J., Powell C.M., Collinson J.W., Lòpez-GamundÌ O.R.,Synthesis, in: Veevers J.J., Powell C.M. (Eds.), Permian-TriassicPangean Basins and Foldbelts along the Panthalassan Margin ofGondwanaland, Geol. Soc. Am. Mem. 184 (1994) pp. 331-353.

[69] Faill R.T., A geologic history of the North-Central Appalachians.Part 3. The Alleghany orogeny, Am. J. Sci. 298 (1998) 131-179.

[70] Levine J.R., Deep burial of coal-bearing strata, anthracite region,Pennsylvania: sedimentation or tectonics?, Geology 14 (1986) 577-580.

[71] for further information about the LMDz AGCM: http://www.lmd.jussieu.fr/~lmdz

[72] Khodri M., Ramstein G., Paillard D., Duplessy J.C., Kageyama M.,Ganopolski A., 2003. Modelling the climate evolution from the lastinterglacial to the start of the last glaciation: the role of Arctic Oceanfreshwater budget. Geophys. Res. Lett. 30 (2003) 1606, 10.1029/2003GL017108.

[73] Braconnot P., Marti O. Impact of precession on monsoon character-istics from coupled ocean atmosphere experiments: changes inIndian monsoon and Indian Ocean climatology. Mar. Geol.201(2003) 23-34.

[74] Donnadieu Y., Fluteau F., Ramstein G., Ritz C., Besse J. Is there aconflict between the Neoproterozoic glacial deposits and the snowball Earth interpretation: an improved understanding with numericalmodelling. Earth Planet. Sci. Lett. 208 (2003) 101-112.

[75] Donnadieu Y., Ramstein G., Fluteau F., Besse J., Meert J. G., Is highobliquity a plausible cause for Neoproterozoic glaciations, Geophys.Res. Lett., 29 23 (2002) 2127.

[76] Chenet A-L., Fluteau F., Courtillot V., Modelling massive sulphateaerosol pollution following the large 1783 Laki basaltic eruption.Earth Planet. Sci. Lett. (in press)

[77] Berner R.A., Palaeo-CO2 and climate, Nature 358 (1992) 114.[78] Bourquin S., Guillocheau F., Keuper stratigraphic cycles in the Paris

Basin and comparaison with cycles in other Peritethyan basins (Ger-man Basin and Bresse-Jura Basin), Sediment. Geol 105 (1996) 159-182.

[79] Bourquin S., Vairon J., Le Strat P., Three-dimensional evolution ofthe Keuper of the Paris Basin based on detailed isopach maps of thestratigraphic cycles: tectonic influences, Geol Rundsch 86 (1997)670-685.

[80] van der Zwan C.J., Spaak P., Lower to Middle Triassic sequencestratigraphy and climatology of the Netherlands, a model, Palaeoge-ogr., Palaeoclimatol., Palaeoecol. (1992) 277-290.

[81] Bachmann G.H., Lerche I., Epicontinental Triassic, Heft 9-10 2. Zbl.Geol. Paläont. Teil I, Stuttgart (1998) 876 p.

[82] Geluk M., Stratigraphy and Tectonics of Permo-Triassic basins inthe Netherlands and surrounding areas, dissertation Utrech Univer-sity (2005) 171 p.

[83] Aigner T., Hornung J., Junghans W.-D., Pöppelreiter M., Baselevelcycles in the Triassic of the South-German Basin: a short progressreport, in: Bachmann G.H., Lerche I (Eds.), Epicontinental Triassic,Zbl. Geol. Paläont. Teil 1 7-8 (1998) pp. 537-544.

[84] Szurlies M., Bachmann G.H., Menning M., Nowaczyk N. R.,Käding K.-C., Magnetostratigraphy and high-resolution lithostratig-raphy of the Permian-Triassic boundary interval in CentralGermany, Earth Planet. Sci. Lett. 212 (2003) 263-278.

[85] Ulicny D., A drying-upward aeolian system of the Bohdasin Forma-tion (Early Triassic), Sudetes of NE Czech Republic: record ofseasonality and long-term palaeoclimate change, Sed. Geol. 167(2004) 17-39.

[86] Trusheim, F., Zur Gliederung des Buntsandsteins. Erdöl-Zeitschrift79 (1963) 277-292.

[87] Wolburg, J., Vom zyklischen Aufbau des Buntsandsteins. N. Jb.Geol. Paläont. Mh. 9 (1968) 535-559.

[88] Best. G., Kockel F., Schöneich H., Geological history of the south-ern Horn Graben, Geologie en Mijnbouw 62 (1983) 25-34.

[89] Marcoux J., Baud A., Ricou L.-E., Gaetani M., Krystyn L., BellionY., Guiraud R., Moreau C., Besse J., Gallet Y., Theveniaut H., LateAnisian (237 to 234 Ma), in: Dercourt J., Ricou L.-E., Vrielynck B.(Eds.), Atlas Tethys palaeoenvironmental maps, Atlas and Explana-tory Notes, Gauthier-Villard, Diffusion CGMW, Paris (1993) pp.21-33.

[90] Bourquin S., Friedenberg R., Guillocheau F., Depositionalsequences in the Triassic series of the Paris basin: Geodynamicimplications, Cuadernos de Geologia Ibérica 19 (1995) 337-362.

[91] Eschard R., Lemouzy P., Bacchiana C., Désaubliaux G., Parpant J.,Smart B., Combining sequence stratigraphy, geostatistical simula-tions, and production data for modelling a fluvial reservoir in theChaunoy Field (Triassic, France), Am. Assoc. Petrol. Geol. Bull. 82(1998) 545-568.

[92] Péron S., Nature et contrôle des systèmes fluviatiles du domaineOuest Péri-Téthysien au Trias inférieur: sédimentologie de faciès,reconstitutions paléoenvironnementales et simulations climatiques,thèse Université Rennes 1 (2005), 145 p.

[93] Clemmensen L.B., Tirsgaard H., Sand-drift surfaces: a neglectedtype of bounding surface, Geology 18 (1990) 1142-1145.

[94] Roman A, Sequenzstratigraphie und Fazies des Unteren und Mit-tleren Buntsandsteins im östlichen Teil des Germanischen Beckens(Deutschland, Polen), Dissertation University of Wittenberg Halle(2004) 140 p.

[95] Ben Ismail M.-H., Les bassins mésozoïques (Trias-Aptien) du Sudde la Tunisie: stratigraphie intégrée, caractéristiques géophysiqueset évolution géodynamique, Thèse de Doctorat, Univ. Tunis (1991)466 p.

Documents

Paleoenvironment reconstructions and climate simulations of the …fluteau/UIA/articles/9_PERON-bourquin-fluteau.pdf · of characterizing the impact of climate on fluvial sedimentation