Lower- and middle-triassic palaeomagnetic analysis from southern Tunisia

Lower- and middle-triassic palaeomagnetic analysis from southern Tunisia

JOURNAL OF GEODYNAMICS12, 163-176 (1990) 163 LOWER- A N D MIDDLE-TRIASSIC PALAEOMAGNETIC ANALYSIS FROM S O U T H E R N TUNISIA MONCEF GHORABI Labor...

578KB Sizes 2 Downloads 24 Views

JOURNAL OF GEODYNAMICS12, 163-176 (1990)

163

LOWER- A N D MIDDLE-TRIASSIC PALAEOMAGNETIC ANALYSIS FROM S O U T H E R N TUNISIA

MONCEF GHORABI Laboratoire de G~omagn~tisme de Saint-Maur, C.N.R.S. et Universitd Paris VI, 94107 Saint-Maur des Foss~s Cedex, France. Laboratoire des Ressources Mindrales, Facull~ des Sciences de Tunis, 1060 Belvedere Cedex, Tunis, Tunisia.

(Accepted May 30, 1990)

ABSTRACT Ghorabi, M., 1990. Lower- and Middle-Triassic palaeomagnetic analysis from southern Tunisia. In: J. J. Dafiobeitia and B. Pinet (Editors), Geophysics of the Mediterranean Basin. Journal of Geodynamics, 12: 163-176. From Lower- and Middle-Triassic formations (respectively, the Sidi Stout and the Kirchaou sandstones) in southern Tunisia, 13 sites have been chosen for a palaeomagnetic study. In spite of the presence of two antiparallels, normal and reverse clusters of stable magnetization directions, the magnetization carried by these rocks results from a relatively recent remagnetization.

INTRODUCTION

Ballard et aL (1986) studied that Permo-Triassic sedimentary rocks from southern Africa, and they suggest that widespread remagnetization of these rocks occurred. In southern Tunisia, a relatively important Triassic series can be observed (Busson, 1967; Bouaziz, 1986) in the Dahar area (the northern margin of the Saharan platform), and a palaeomagnetic analysis has been carried out at different levels in this series in order to test if the same remagnetization phemenom exist in northern Africa and in order to precise the Triassic location of Africa. Definition of this position is important for a good understanding of the evolution from Pangea to the Jurassic location of Gondwana and Laurasia. GEOLOGICAL SETTING

Southern Tunisia can be divided into three main regions: - - The Tebaga de Medenine, a monocline formed by Permian formations, 0264-3707/90/$3.00

© 1990 Pergamon Press plc.

164

GHORAB1

....... The Saharan platform (Dahar) with subhorizontal tabular structure, including formations of from Triassic to Cretaceous age, often with relatively reduced thickness, and evaporitic and dolomite facies, - - The Jeffera, a collapse plain with Mio-plio-quaternary deposits covering the oldest levels, which present an increasing thickness towards the NE, with a complicated structure of the same type as the Atlas (Busson, 1967). The Dahar region is limited to the North by the Tegaba de Medenine. This area can itself be divided into different geological parts (Busson, 1967), separated by Oued Tataouine. The province, called "Centrale", is characterised by a thick Mesozoic sequence, and the northern province presents a more condensed Mesozo]'c series (Figs. la and lb). In the northern region (Fig. la), the Mesozoic levels begin with the Sidi Stout red sandstones, attributed to the Lower Triassic by comparison with the Bir el Jaja sandstones in NW Lybia (Adloff and al., 1985) (Fig. la). It was conformably overlying the Cheguimi sandstones, attributed to the Upper Permian (Bouaziz, 1986). The Lower Triassic becomes regularly thinner in the direction of the Saharan platform. On the contrary, in the north-western area (Bir Soltane, Tegaba) it is entirely lacking because of the post-Middle-Triassic erosion. In fact, vertical movements have been individualized in the Tebaga area. The Late-Triassic deposit (the Messaoudi dolomite, Bouaziz and Mello, 1985) is on the Lower Triassic red sandstones and the Upper Permian and it is superimposed on the Sidi Stout unconformity. The Sidi Stout sandstones outcrop only locally under this conformable deposit. They present a monotonous lithology and are azoic, except for some hematized woods. In the "central" province (Fig. la), the lower Triassic is only known from boreholes. The outcropping Triassic is formed at the bottom by: the Kirchaou red sandstones with cross-bedding; they mainly appear near the Jebel Rehach and in the Kirchaou area (they are attributed to the Middle Triassic by comparison with the Ras-Hamia sandstones of NW Lybia, Bouaziz, 1986).

SAMPLING AND M E T H O D S

In the Lower- and Middle-Triassic formations sampling has been carried out at 13 sites (fifty-six hand samples): two in the E1 Hmai'a E1 Kbira region, four in the Sidi Stout region, one in the Mounet Derharhra region, four towards a cross-section at E1 Kronnab and two along two crosssections at Kef et Touareg (Fig. la). From each site at least 3 or 4 specimens were drilled for the palaeomagnetic study. A JR4 magnetometer constructed by Geofyzika-Brno was used for the

2km

.........

caleOiimnm~2kin

IN

III

v P

i



•"

~e

~

3

4

5

6

7

J

Ill

??ii~.

.,

Zig. la Geological map of studied area (Busson, 1967; Bouaziz, 1986) and sampling sites. k: Northern region; B: "Central" region; 1: Lower Triassic (Sidi Stout sandstones); 2: Middle Triassic (Kirchaou andstones); 3: Upper Triassic; 4: Jurassic; 5: Cretaceous; 6: Sampling sites

;©IlI I I i

0

s

A

':i~EiEi!!ii~ :!iiiiil

i' fll I I Kbira

r~ .q

,3

t~

rl

rl

Oued M e s t a o u a (Mhira)

Province Jebel Krepta

Northern

A

I

2

3

( e l el A n e b a

Province

• ~k~--.

8km

.

--..

...:--.

|W

Comparison of Triassic sequences in the two provinces (Northern and "Central"). : Permian; 2: Lower Triassic, Sidi Stout sandstones: 3: Middle Triassic, Kirchaou sandstones: 4: Carnian, Rehach dolomites, ?ouareg sandstones and Mekraneb dolomites; 5: Upper Carnian-Lower Norian. Mhira clay; 6: Norian-Rhetian, Messaoudi lolomites; 7: Upper Triassic to lower Bathonian

fig. lb

ten

s

::::::::::::::::::::::::: ::::::::::;:;;9,"

,,.

, =

~..

.....

/

, : . . . . . :+ .:+ + ,:+ . : + : +

,;,],?;,75:'],~';',',',',

::::!:::i:i:::): ::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

,'/

l, /

,!iii!i!i,!iii!iiiiiiiiiii!ii!i!ili!i'!!i!', ,;

~E

Cen t rale'"

2

LOWER- AND MIDDLE-TRIASSIC PALAEOMAGNETIC ANALYSIS

167

magnetization measurements. Its minimum detection limit is 1.10-10 (A/m). The viscosity coefficient (v) was measured in order to investigate the possible influence of the viscous magnetization in the global magnetization: The magnetization of some specimens has been measured three times: twice after 15 days storage in the earth's magnetic field, with the Z axis of the sample being towards the geomagnetic field direction; in one case, in the normal sense of the field (the Z1 value) and in the other case in the reverse sense (the Z2 value). And once after 15 days storage in a zero field (the M value); v = 100(Z1 - Z2)/2M. The natural remanent magnetization of all specimens was measured only after remaining at least 15 days in a zero field, in order to eliminate the main part of the viscous component acquired after or during the sampling. The alternating fields, heating treatments and the precedent combined methods have been applied to study the evolution of the magnetization of 2 pilot specimens (2 specimens per site). The instruments used for this purpose are a furnace allowing the simultaneous treatment of 15 specimens in a residual field lower than 10 nT, and an alternating-field demagnetizer allowing of reaching 150 mT (Le Goff, 1985). Successive applications of alternating-field treatments with an increasing field, on 2 specimens by site did not allow us to perform the magnetization analysis, the unblocking fields in these samples being very high. Successive heatings at increasing temperatures of up to 400 °, 500 ° or 600°C according to the samples, followed by alternating-field treatment, have revealed that, in spite of the high field used (120 roT), the value of the magnetic moment here also shows almost no change. Then, only successive heatings at increasing temperature were effective for magnetization analysis. This method was applied to 2 specimens from each sample from all the sites. RESULTS

The magnetization intensities (for the Sidi Stout and the Kirchaou sandstone formations) are often relatively important for sandstones, and only 10% of the samples were eliminated because of their very weak magnetization. In 2 sites, the remanent magnetization moments were too weak after heating at temperatures lower than 300 ° C, and these sites have also been eliminated. 1°

The Sidi Stout sandstone

The values of the viscosity coefficient are often lower than 10%. The N R M of samples from this formation have variable intensities, reaching JOG 12/2-4-4



~-..~.

.

\.

/

\

;~.

//

./

Fig. 2 Sidi Stout sandstones (Lower Triassic). Full circles, direct directions; Open circles, reversed directions, Open star, present terrestrial magnetic field (stereographic projection). a: NRM directions before dip correction; b: Stable directions of magnetization (high-temperature) before dip correction; c: Stable directions of magnetization (high-temperature) after dip correction.

/

© >

©

LOWER- AND MIDDLE-TRIASSIC PALAEOMAGNETIC ANALYSIS

169

(A/m). The magnetization vectors are mainly around a direction relatively close to the recent field (Fig. 2a). During thermal treatments, the magnetization sometimes increases strongly up to 150 ° C after heating, but in one site (XXIX), it only goes up to 600°C (Fig. 3). The magnetization always disappears after heating at 675 ° C. This indicates that the magnetization of these red sandstones is probably carried by minerals of the hematite family. In all specimens, the first magnetization component, dissappearing between 150 and 300 °, is probably a viscous magnetization. All specimens from the same site (XXIX) in the Lower Triassic, "memorize" two high-temperature components. They are almost antiparallel in the same sample. In all the other samples, only one high-temperature component has been obtained; all stable components at high temperature are in about two antiparallel directions close to the recent geomagnetic field (Fig. 2b). The hysteresis cycle of samples from these sandstones was analysed. The paramagnetism is largely dominant in regard to a ferromagnetism which is carried by hematite in fine grains. 4 8 0 0 10 - 4

§O0OC ~

EA I T . U P

is

,S

.:

SOUTH

'

500 NRM = 4800.1"04(A/liE)

601

WElT.DOWN

Fig. 3 Zijderveld diagram (thermal treatment) for one specimen from Sidi Stout sandstone (Lower Triassic). Full (open) squares represent projections of the vector end points on the vertical (horizontal) plan.

\

'\



a



o



oo



oo

Fig. 4

o



/

//

/

/

o//

/

i '

"

'

"

"

.o,

ql

+

...,

;

j

'\ ,,

/

/

-\

N

o

.

•- ~

"\

~

Kirchaou sandstones (Middle Triassic)• See Fig. 2•

o

/

..~

.



o

oo o o

db

+

. . . . . . .

.

f----~~

b

a

J I~

/

"j

C



0 >.

O

LOWER- AND MIDDLE-TRIASSIC PALAEOMAGNETIC ANALYSIS

171

2 ° _ The Kirchaou sandstones The viscosity coefficient in this formation is weak (value lower than 2% for 75% of the samples). Compared to those from the Sidi Stout sandstone formation, the samples from the Kirchaou Formation in general present a stronger NRM intensity, reaching 14000.10 -4 SI. The (relatively scattered) N R M vector directions, like those for the Sidi Stout sandstones, are mainly around the direction of the recent geomagnetic field (Fig. 4a). The evolution of magnetization during the thermal analysis is very different from one sample to another. For one third of them, 90% of the magnetization disappears after heating at 150 ° C. For the other samples, only 25 to 50% of the magnetization was eliminated after heating at this temperature. Later, the decrease is progressive and depends on the temperature applied up to 675 ° C. This indicates that hematite is probably the carrier of the magnetization. Two components can be separated (Fig. 5); one, probably having a viscous origin, is eliminated at low temperature at 150 ° C. The second component is stable at high temperature. It has a statistically well-defined

EA,ST.UP

j6sooc W E S T. D O W N

,oo°c

I

I

I

t

I

!

s

BO U TH

Fig. 5 Zijderveld diagram (thermal treatment) for one specimen from Kirchaou sandstones (Middle Triassic). See Fig. 3.

172

GHORABt

direction close to the recent geomagnetic field and, according to the sample, a normal or reverse sense (Fig. 4b). The hysteresis loop analysis shows that paramagnetism is dominant in regard to the ferromagnetism. It probably results from monodomain grains of hematite.

DISCUSSION

Figures 2b and 4b present the stable " high temperature" directions before dip correction (and Fig. 2c and 4c after dip corrections) obtained respectively from Lower (the Sidi Stout sandstones, 2b and 2c) and Middle (the Kirchaou sandstones, 4b and 4c) Triassic levels. The main part of the directions, for the Sidi Stout and the Kirchaou sandstones, are grouped around two antiparallel directions. The presence of such antiparallel directions within a formation is generally considered as an indicator of the primary origin of magnetization. From average directions of each sample, normal and a reverse average directions for each formation were calculated (Tab. 1A and 2A) and the corresponding paleomagnetic virtual poles have been determined (Tab. 1B TABLE I

Mean direction (A) of stable magnetization (high temperature) and corresponding paleomagnetic poles (B). Sidi Stout sandstone formation; Lower Triassic. (Lat., latitude; Long., longitude; Im, mean inclination; Dm, mean declination; M D Nor., normal mean direction; MD. Inv., reverse mean direction; n, number of specimens in (A) and number of samples in (B); Lt. P. and Lg. P., pole latitude and longitude; dp, dm, confidence elliptic semi-axe values; K and ~95, Fischer parameters). (1): before dip correction, (2): after dip correction A:

(1)

M D . Nor. M D . Inv.

Lat.

Long.

lm

Dm

K

a9s

n

33.55 -

10.24 -

46.27 -49.06

353.49 179.77

21.3 21.5

5.4 7.6

32 16

-

-

-49.59 -48.89

354.00 184.39

18.90 24.14

5.7 7.1

32 16

M D . Nor. (2) M D . lnv.

B:

(1) (2)

1.at.

Long.

Im

33.55

10.24 44.75 47.74

Dm

K

1195

Lt.P.

Lg.P.

dp

tim

n

352.76 355.39

15.9 14.6

6.9 7.2

80.47 83.85

.... 126.77 -128.71

5.5 6.l

8.7 9.4

26

LOWER- AND MIDDLE-TRIASSIC

PALAEOMAGNETIC

ANALYSIS

173

and 2B). The normal average direction is slightly better defined than the reverse one (Tab. 1 and 2) for the Sidi Stout Formation but for the Kirchaou Formation the reverse is the better defined. The clear antiparallelism of normal and reverse magnetizations should be then, an indicator of the primary origin of the magnetization. The average differences between normal and reverse directions for these two formations are not notable (only on the order of 8 ° for the Sidi Stout Formation and 4 ° for the Kirchaou Formation). These differences are of the same order as those between Lower and Middle-Triassic mean directions (on the order of 5°). There was apparently a stability of the African platform during this period, and Africa shoud be situated at a relatively high latitude. A comparison of these poles with the apparent polar-wander path for Africa (Fig. 6 after Irving and Irving, 1982), shows that these poles are very different from the other Permo-Triassic African poles, except for a Triassic pole from Lybia (Nairn and Martin, 1978). They are relatively close to the Late-Cretaceous to recent African poles. Two assumptions can therefore be proposed for the magnetization of these Lower- and Middle-Triassic rocks: - - These poles are effectively Triassic; from Permian to early Triassic, the continental drift of Africa, towards the North, has been extremely fast, followed by stable period, then by a drift back southwards. Such a displacement is very difficult to admit, and does not seem to be confirmed by other analyses. T A B L E II M e a n direction (A) of stable m a g n e t i z a t i o n (high t e m p e r a t u r e ) a n d c o r r e s p o n d i n g p a l e o m a g n e t i c poles (B). K i r c h a o u sandstones; M i d d l e Triassic. See T a b l e 1. A~

O) (2)

Lat.

Long.

lm

Dm

K

ags

n

D M . Dir. D M . Inv.

32.98 -

10.80 -

39.55 -43.95

346.34 183.12

10.5 18.2

7.8 6.4

31 26

D M . Dir.

-

-

41.27

7.78

10.50

7.8

31

D M . Inv.

-

-

-44.14

182.65

17.40

6.6

26

B: Lat.

Long.

(1)

32.98

10.80 42.20

(2)

-

-

Im

43.34

Dm

K

{][95 Lt.P.

Lg.P.

dp

dm

n

354.44

12.5

6.9

80.13

-138.20

5.2

8.5

33

353.69

12.1

7.0

80.52

-132.08

5.4

8.7

-

174

GHORABI

--- The magnetization is clearly more recent that the deposition of these beds and the presence of normal and reversed antiparallel directions should not be significant of the magnetization's primarity. The presence of normal and reversed magnetization in the same site, with relatively significant scatter, clearly shows that the magnetization has probably been acquired only irregularly during a long period (local chemical phenomena?).

¢0

% 280

260

o

60S

- -

e loo

100

160 i

20o

220 ~' 230

/

4030~-- t,jOa ~

150"E __~ f Fig. 6 Apparent polar-wander curve of Africa after Irving and Irving, (1982); full circles, Irving and Irving data; full stars, poles after dip correction (this study); open star, Nairn and Martin (1978) Upper Triassic formation.

LOWER- AND MIDDLE-TRIASSIC PALAEOMAGNETIC ANALYSIS

175

Since the Cretaceous, this direction has, moreover, been close to the geomagnetic field directions. In the two formations, the existence of frequent hematite-rich concentrations on the surface clearly shows that hematite has been subjected to recent mobilization. This is another argument in favour of chemical remagnetization in these rocks. Arguments concerning convergence suggest that remagnetization exists, in spite of the presence of antiparallel directions. This remagnetization should then probably be chemical remanent magnetization (CRM) acquired during a relatively recent period with the change of polarity of the earth's magnetic field. The comparison of African paleopositions from the poles of this study, and others from Morocco (At) (Hailwood, 1975) and Lybia (Az) (Nairn and Martin, 1978) (Lower and Middle Triassic, Fig. 7), shows that many among them are near to the present position of Africa. This clearly confirms that many poles actually correspond to recent remagnetization. However, the pole from the Ait-Aadel (Hailwood, 1975) is perhaps more in relation with the complex tectonic history of this area. The pole position from the Lybian formation differs slightly from those obtained for the Lower and Middle Triassic (the Sidi Stout and the Kirchaou formations). It also corresponds to remagnetized rocks.

Fig. 7 Paleopositions of Africa from pa|eomagnetic poles obtained from: Ss, Sidi Stout sandstone Formation; K, Kirchaou sandstone Formation; At, Ait-Aadel Formation (Morocco); Az, Aziza Formation (Lybia), compared to A, the present position (arbitrary longitudes).

176

GHORABI

CONCLUSION

The presence of antiparallel directions within Lower- and Middle-Triassic formations of Southern Tunisia is not an indicator of primary magnetization. Nowadays, these magnetizations correspond to remagnetization in relation with relatively recent chemical phenomena. It is then necessarly to consider with care the presence of antiparallel directions to determine the age of magnetization. We must also treat some earlier works with care because some authors attribute the age of formation to that of the magnetization but they could differ largely.

ACKNOWLEDGEMENTS

The author wishes to thank B. HENRY, Professor S. SASSI, M. LAJMI, Professor R. TRUILLET, Services des Mines de I'O.N.M., the group of the Projet de Cartographie G6ologique et Inventaire des Substances Utiles du Sud tunisien, M. C. BROWNER and BALTI for their help, and the Total Company for the financial support.

REFERENCES Adloff, M. C., Doubinguer, J., Massa, D. and Vachard, D., 1985. Revue de l'lnstitut Franqais du P6trole. Vol. 40, N" 6, Novembre-D,~cembre, 1985. Ballard, M. M., Van der Voo, R. and H~ilbich, I. W., 1986. Remagnetizations m Late Permian and Early Triassic rocks from southern Africa and their implications for Pangea reconstructions. Earth and Planetary Science Letters. 79, 412-418. Bouaziz, S., 1986. La d6formation dans la plateforme du Sud tunisien (Dahar et Jeffaral: Approche multiscalaire et multidisplinaire, Th~se de Doctorat de sp6cialit6. Facult6 des Sciences de Tunis. Bouaziz, S. and Mello, J., 1985. Mise en +vidence de la discordance de Sidi Stout dans la province centrale du Sud tunisien et de son terme transgressif: la dolomie de Messaoudi (Rh6tien). M6moire du Service g~ologique N ° 54. Busson, G., 1967. Le m6sozo'ique saharien, lere partie: l'extr~me Sud tunisien, l~dition du C.NR:S. S6rie geologique N"8. Hailwood, E. A., 1975. The Paleomagnetism of Triassic and Cretaceous Rocks from Morocco, Geophus. .I. Roy. Astron. Soc., 41, 219 235. Irving, E. and Irving, G. A., 1982. Apparent polar wander paths, Geophys. Surveys: 5, p. 141 188. Le Golf, M., 1985. Description d'un appareil ~ d6saimanter par champs alternatifs; 61imination de l'aimantation r6manente anhyst6r&ique parasite; Can. J. Earth Sci. 22, p. 1740-I747. Nairn, A. E. M. and Martin, D. L., 1978. A Paleomagnetic Study of the Triassic {Carnian) Aziza formation, E. O. S. Trans. Amer., Geophy. Union, 59, 272-200.