Evolution of Brasiliano-age granitoid types in a shear-zone environment, Umarizal-Caraubas region, Rio Grande do Norte, northeast Brazil

Journal of South American Earth Sciences, Vol. 8, No. I, pp. 79-95, 1995 Copyright © 1995 Elsevier Science Ltd& Earth Sciences & Resources Institute Printed in Great Britain. All fights reserved 0895-9811D5 $9.50 + 0.00

Pergamon 0895-9811(94)00043-3

Evolution of Brasiliano-age granitoid types in a shear-zone environment, Umarizal-Caradbas region, Rio Grande do Norte, northeast Brazil A . C . G A L I N D O 1, R . D A L L ' A G N O L 2, I. M C R E A T H 3, J . M . L A F O N 2 a n d N. T E I X E I R A 2 1Curso de P6s-Graduag~o e m Geoci6ncias, Universidade Federal do Patti, Brazil and Departamento de Geologia, Universidade Federal do Rio Grande do Norte, C P 1639, 59072-970 Natal RN, Brazil 2Centro de Geoci~ncias, Universidade Federal do Parfi, C P 1611, 66075-900 B el6m PA, Brazil 3Departamento de Geologia Geral, Universidade de S~o Paulo, C P 20899, 01498-970 S~o Paulo SP, Brazil

(Received August 1993; RevisionsAccepted March 1994)

Abstract--A sequence of Brasiliano-age granitoid types is exposed in a small area near the cities of Umarizal and Caratlbas in Rio Grande do Norte State, Northeast Brazil. Porphyritic K-alkali-calcic monzogranite is an important facies of the oldest Carafibas intrusion (Rb-Sr whole rock isochron age of ca. 630 Ma), which suffered solid-state deformation due to movements on a major NE-trending shear zone. The intrusion of the Prado and part of the Quixaba bodies was probably controlled by the shear zone. These two bodies include mafic/intermediate rocks, some of which contain two pyroxenes, and have hybrid, partly alkaline and partly shoshonitic geochemical characteristics. Rock types and ages are similar to those of some Pan-African occurrences in southwestern Nigeria. The Tour~o body, intruded at ca. 590 Ma, presents preferred mineral orientations which are probably largely magmatic, since little evidence is found for widespread solid-state deformation. On the other hand, its intrusion may have been facilitated by the presence of the shear-zone faults. The rocks form a monomodal felsic K-alkali-calcic suite. With the exception of the Quixaba body, all these earlier granitoids are magrnatic epidote- and magnetite-bearing porphyritic monzogranites with trace element geochemical characteristics of modern syn-collisional granites. The latest intrusion at ca. 545 Ma is mainly represented by potassic quartz syenites and related rocks, some of which contain fayalite or ferrohypersthene. These rocks possess neither well developed mineral orientations of magmatic origin nor signs of solid-state deformation. They are mineralogieally similar to, but younger than some of the "bauchites" of central Nigeria. Geochemical signatures are comparable with those of modern within-plate granites. All granitoids present high (87Sr]86Sr)i ratios which range from 0.708 to 0.712, and increase with decreasing age. Such ratios are compatible with important or dominant crustal contributions. On the other hand, the more mafic rocks are likely to have formed from enriched mantle.

Resumo--Numa regi~o pouca extensa nas proximidades das cidades de Umarizal e Carafibas no Estado do Rio Grande do Norte no Nordeste Brasileiro, exp6e-se urea seqii~ncia de granit6ides de idade brasiliana. Monzogranitos porfirfticos de afinidade filcalicfilcica pot~issiea compSem a intrus~o mais antiga (Caratibas), que forneceu urea idade de ca. de 630 Ma por meio de urea isocrona Rb-Sr em rocha total. Esse corpo sofreu deforma~ao no estado s61ido devida a movimentos que ocorreram ao longo de urea zona de cisalhamento de grande porte corn direg~o NE. t~ prov~vel que a intrus~o de outros dois corpos (Prado e parte de Quixaba) tenha sido controlada estruturalmente pela mesma zona de cisalhamento. Esses dois corpos apresentam rochas mfificas/intermedifirias. Algumas amostras contain dois pirox~nios, e suas caraterfsticas qufmicas s~o h%ridas, ora shoshonfticas, ora mais alcalinas. As rochas e suas idades semelham-se hs de corpos pan-africanos no SO da Nigdria. A intrnsfio do granit6ide Tour~o ocorreu por volta dos 590 Ma. I~ provfivel que as orienta95es dos minerais sejam dominantemente da fase magmfitica, visto que ha pouca evid~ncia para deformag~o no estado s61ido. Entretanto, a intrusfio desse corpo deve ter sida facilitada pela presen~a de fal, has associadas ~ zona de cisalhamento. As rochas comp6em uma sfiite monomodal f61sica de afinidade alcali-c~ilcica pot~issica. Em todos os granitos mais antigos, s~o importantes monzogranitos porlirfticos. Essas rochas portam magnetita e epidoto magm~itico, e possuem caraterlsticas geoqufmicas de granitos modernos do tipo sin-colisao. O granit6ide mais recente, Umarizal, tern ca. 545 Ma, e comp0e-se por quartzo-sienitos potfissicos e rochas afins, algumas das qnais, portadoras de falalita on ferrohiperst~nio. As rochas est~o desprovidas tanto de orientag6es de origem magm~itica como de deformafes no estado s61ido. Assemelham-se aos "bauchitos"do centro-norte da Nig6ria, embora sejam mais novas do que estes. Suas assinaturas geoquimicas s~o as de granitos intraplacas modernas. Todos os granit6ides apresentam altos valores para as raz6es (87Sr/86Sr)i, desde 0,708 a 0.712. As razfes aumentam corn descrescimento da idade. Tals razzes acusam urea participa~o crustal importante on predominante na g~nese das rochas. Por outro lado, as rochas mals mLficas talvez tivessem sua origem no manto enriquecido.

A d d r e s s all c o r r e s p o n d e n c e and reprint requests to Dr. Ian McReath, Instituto de Geoci~ncias, Universidade de S~o Paulo, P.O. Box 20899, CEP 01498-970, S~o Paulo, SP Brazil 79

80

A. GALINDO, R. DALL'AGNOL, I. MCREATH, et al. INTRODUCTION

THE GEOLOGY OF THE BORBOREMA PROVINCE of Northeast Brazil (Almeida et al., 1977; Fig. la, lb) is characterized by medium to high grade, predominantly early Proterozoic gneiss-migmatite terrains and folded, low grade metasupracrustal sequences. The province suffered strong deformation during the late Proterozoic Brasiliano (equivalent to Pan-African) cycle. Regional scale shear zones are important elements of the present structure, and are commonly believed to be late-Brasiliano features.

/ / --'~ ~ORTAI.I[ZA I

°o

/

Near the towns of Umarizal and Carafbas in western Rio Grande do Norte State (Fig. lc, 2) various Brasiliano granites occur in a restricted region with usually excellent exposure, which simplifies the understanding of the geological relationships. The granites include not only types which are common in the rest of the province, but also varieties which were previously unknown there, and which are uncommon in the rest of the world. This is the case of the fayalite-hedenbergite granites of Umarizal, similar to the Pan-African bauchites of Nigeria (Oyawoye, 1965). This article describes the field relationships, petrographical and geochemical characteristics, and presents Rb-Sr whole rock isochron studies of the granites in the Umarizal-Caradbas region. G E O L O G Y OF T H E UMARIZALCARAIJBAS REGION

NAiL

\ \

O,

~

IO0. Km

"

~ros

___

.,.tA.g.r

,.

UCo L.INE

/

.z/

/

/

~

/\

_/~'A

Fig. la. Location of NE Brazil; b: the Borborema Province of NE Brazil with major lineaments in the basement; (1) Dom Pedro IISobral: (2) Quixada-Quixeramobim; full square- UmarizalCaradbas area; c: Extreme NE Brazil, showing location of Umarizal-Caradbas area (hollow square); += major granite bodies (after Sial, 1987); Potigu~r basin = Mesozoic sediments; S=Serid6 region; CSF=Cachoeirinha-Salguerio foldbelt. Most of the granites cited in Table 2 are localized north of the Patos lineament or its geometric continuation. Within the province, intense felsic magmatism occurred from late Archaean to early Phanerozoic times. Brasilianoage granites are very abundant and widespread, and are often associated with the regional shears. They have c o n s i d e r a b l e p e t r o g r a p h i c a l , m i n e r a l o g i c a l and geochemical diversity (Table 1) and were intruded over a ca. 200 Ma time interval, with peak activity between ca. 650 and ca. 550 Ma (Table 2). In some regions there is a gross time-dependent evolution of granite types (Leterrier et al., 1990) but some of these are recurrent in space and time.

Earlier regional mapping programs (Dantas, 1974; Campos, et al., 1979; Gomes et aI., 1981) identified only large batholithic bodies in the region, but detailed mapping (Arafjo, 1985; Morais Neto, 1987; Curioso, 1987) proved the presence of a number of separate granite bodies. Lins (!987) presented the results of a ground gravimetric and magnetometric study of the region. This revealed important Bouguer anomalies associated with two of the bodies (discussed later in this paper). Using the results of these studies as a base, the region was mapped at the 1:100,000 scale. Detailed studies of the extreme western part of Rio Grande do Norte (Jardim de S~ et al., 1981a) and of the better known Serid6 region to the east (e.g., Jardim de S~ 1984; Hackspacher and S~, 1984) permitted the integration of the geology of the Umarizal-Cara6bas region into the regional picture. Basement rocks correlated with the polydeformed, Late Archaean or Early Proterozoic Caic6 Group of the Serid6 include fine to medium-grained, banded granodioritic orthogneisses, nebulitic or banded migmatites and dikes or bands of amphibolites. Biotite gneisses with subordinate marbles and calc-silicate rocks very similar to the Jucurutu Formation of the Serid6 overlie this basement. The Jucurutu gneisses are medium grade rocks with superimposed retrograde mineral assemblages. Both basement and Jucurutu Formation rocks are cut by porphyritic orthogneisses, probably equivalent to the Transamazonian (ca. 2.0 Ga) G2 granitoids (Jardim de S~ et al., 1981b; Macedo et al., 1990). These granitoids form bodies of variable dimensions and commonly contain xenoliths of the Caic6 Group and Jucurutu Formation. Rocks equivalent to the Serid6 schists are absent from the Umarizal-Caradbas region. Deformed and undeformed Brasiliano granitoids cut these earlier sequences. Six large bodies and some satellites were separated on the basis of intrusion relationships and structural features, although this sometimes leads to the inclusion within the same body of petrographically contrasted facies, which may not be petrogenetically related. These bodies are referred to toponimically as the Umarizal, Quixaba, Tour~o, Caratbas and Prado bodies,

Evolution of Brasiliano-age granitoid types in a shear-zone environment, NE Brazil

81

Table 1: Main Granite Types in the Northern Part of the Borborema Province Type

Lithological Association

Geochemical Type

Type Name, Example, Domain

1

Biot, + hble. tonalite/granodiorite, locally porphyfitic; + amphibolite (often xenoliths); Trondhjemite

Calc-alkaline, low to medium K

Conceic~o (PB): CSF Serdta (PE): CSF

2

Biot. +_hble. monzogranite/granodiorite, usually porphyritic; + K-diorite; rare examples with two pyroxenes; sometimes accompanied by syenogranite

Calc-alkaline to transitional (a); (some shoshonitic features)

Itaporanga (PB): CSF Itapetim (PB): CSF

3

Syenite, qtz-syenite, alkali-felds, granite; sometimes with pyroxenite

Oversaturated alkaline to peralkaline, Saturated perakalineoften perpotassic

Cantingueira (PB): CSE Meruoca (CE): NB, Triunfo (PE): CSF

4

Biot.-leucogranite, sometimes with muscovite and/or garnet

Peraluminous to metaluminous

S5o Miguel (RN): NB Dona Inas (PB): NB

Table 1. From Almeida, et al., 1967; Jardim de $4 et al., 1986, 1987; Sia1,1987; Sial and Ferreira, 1988, 1990; McMurray et al., 1987ab; Leterrier et al., 1990; Domains: NB=Northern Borborema Province, to the north of Patos linement or geometric continuation; CSF=Cachoeirinha-Salgueiro Fold-belt (see Fig. lc). A better term would be K-alkali-calcic: see text.

5Kin POT

I GU

A R

BASIN

BASIN

g b 0 O

p

O

r

Bb

O

. . . .

,.,.~ 0

0

6 b

Fa r m ÷

÷

b

÷

,{.;

4-

•

"- - ' - - ~ "

.

I •

[1

•

•

° e

•

e

PAT'U ; . •

Q •

•

Pe

•

__,.,

.

"~0

00 •

•

:...: . . . .

•

•

•

-

' ' '

II

12

~

13

~

14

Fig. 2. The Umarizal-Carafibas region: l=Caic6 Group; 2=Jucurutu Formation; 3=G2 granitoids; Brasiliano-age granitoids: 4=Prado; 5=Carafibas; 6=Quixaba; 7=Serra do Lima Complex; 8=Tour~o; 9=Umarizal; 10=Mesozoic Gangorra basin; 1 l=Normal Faults; 12=Undivided Faults; 13=Shear zones with sense of movement; 14=Contacts.

82

A. GALINDO, R. DALL'AGNOL, I. MCREATH, e t al.

Table 2: Ages of the Brasiliano Granitoids in the NE Borborema Province, NE Brazil No.

[

Body, Rock Type

Age, method

Reference

NW CEARA Diques, Meruoca (R)

562 ± 10 (1)

Tavares et. al (1990)

Mocambo (G)(3)

548 ± 24 (1)

Sial, et al. (1981)

Paj6 (G)(3)

537 ± 21 (1)

Gorayeb et al. (1991)

Meruoca (G)(3) Serra da Barriga (G)(3)

507 ± 36 (1) (a)

Sial et aL (1981)

482± 8 (1)

Tavares et al. (1991)

665 + 40 (1)

S~ et al. (1988)

617±39(1)

Jardim de S~iet aL (1987)

Serra Negra (D)

838 ± 52 (1)

Jardim de S~iet al. (1987)

9

Serra Negra (G)(2)

659 ± 41 (1)

±bid

10

S~o Jo~o do Sabuji (D)

764± 73 (1)

±bid

11

Acarf (G)(2)

574 ± 25 (1) (b)

±bid

12

idem

547 ± 25 (1) (b)

±bid

13

idem

555 ± 5 (2) (c)

Legrand et al. (1991)

CENTRAL CEARA Or6s Belt (MAG)(2) RIO GRANDE DO NORTE - West of the Set±d6 Region 7

S~o Miguel (G)(4)

RIO GRANDE DO NORTE - Serid6 Region

RIO GRANDE DO NORTE - East of the Set±d6 Region 14

Taipu -Cardoso (G)(2)

666 ± 16 (1)

Hackspacher et al. (1987)

15

Monte das Gamaleiras (G)(2)

512± 36 (1)

McMurry et aL (1987a)

544± 16 (1) (b)

McMurry et al. (1987a)

475 ± 102 (1) (b)

±bid

PARAIBA 16

Dona Inez (G)(4)

17

idem

OTHER IMPORTANT BODIES 18

Itaporanga, PB (G)(2)

625±24(1)

McMurry et al. (1987b)

19

Fazenda Nova, PE (G)(2)

630 + 24 (1)

±bid

Table 2. Errrors reported in this table are 1 CY. Symbols are as follows: R=rhyolite; G=Granite; MAG = micro-augen granite; D=Diorite. Number code refers to granite type in Table 1; in areas where a number of different types are present (e.g., Taipu-Po~o Branco), the code refers to the dated material; l=Rb/Sr whole rock geochemistry; 2 = U/Pb zircon; a=minimum age, Rb/Sr isochron probably reflecting hydrothermal setting; b=different facies of the same body; c=same facies as 11. and the Serra do Lima complex (Fig. 2). Small stocks and other satellite bodies are geographically close to the Umarizal, Carafibas and Prado bodies. The Serra do Lima complex contains a number of subtly different rock types, some of which resemble those present in the other bodies. Insufficient information was accumulated on this body, to which only passing mention is made in the rest of the article. Within the country rocks the dominant, approximately NE structural trend is defined by a foliation whose direction corresponds to that of a number of shear zones. The most important of these is the Portalegre shear zone (Fig. 2), which is at least 200 km long and 1 to 2 km wide (Hackspacher and Oliveira, 1984; Hackspacher and Legrand, 1989). The latest transcurrent movement on this fault was dextral, and retrograde mineralogical assemblages are developed in rocks cut by it.

THE BRASILIANO GRANITOIDS Geology and Petrography In this section, only the basic features of the petrography will be described. More detailed discussions of petrogenetical aspects have been presented (Galindo, 1993) and are being prepared. Modal analyses are given in Table 3, and the results are plotted in the QAP diagram (Fig. 3). Field relationships and textural aspects clearly show that the Umarizal granitoid is the latest intrusion. It has almost no preferred mineral orientations associated with the magmatic stage, it was not deformed in the solid state, and retrograde secondary minerals are rarely seen. It occupies slightly more than 300 km 2 in the central and southwestern parts of the area, and also occurs as small bodies, including dikes, within the Tour~o granite and Jucurutu

Evolution o f Brasiliano-age granitoid types in a shear-zone environment, NE Brazil

83

Table 3: Modal Analyses of Granitoids of the Umarizal-Carafibas Region

UMARIZAL No.

1

QU1XABA

2

3

4

5

6

7

8

9

Quartz

8.6

10.7

17.0

27.9

24.2

2.4

7.4

8.7

28.0

Plagioclase

31.9

31.9

16.0

24.7

15.7

43.6

52.7

23.7

21.6

K-Feldspar

47.7

48.1

57.6

31.9

51.4

20.4

15.8

41.4

31.4

Biotite

019

0.0

1.0

9.6

1.4

21.9

8.2

17.6

5.5

Amphibole

4.4

4.4

7.1

5.1

7.3

3.8

15.0

8.2

11.8

Clinopyroxene

3.7

1.7

0.8

0.0

0.0

3.4

0.0

0.0

0.0

Orthopyroxene

0.0

2.9

0.0

0.0

0.0

3.1

0.0

0.0

0.0

Fayalite

2.1

0.0

0.2

0.0

0.0

0.0

0.0

0.0

0.0

Titanite

0.0

0.0

0.0

0.0

tr

tr

0.3

tr

0.4

Opaques

0.6

0.3

0.2

0.3

tr

1.3

0.3

0.3

0.2

Allanite

0.0

0.0

0.1

0.4

0.0

0.0

tr

0.1

1.1

Epidote

0.0

0.0

0.0

0.0

tr

0.0

0.0

0.0

0.0

Others

0.1

tr

tr

tr

tr

tr

tr

tr

tr

TOURAO

CARAI]BAS

10 (3)

11 (8)

12 (10)

13 (2)

Quartz

24.5

26.5

30.0

28.1

30.7

31.3

Plagioclase

29.4

~.9

31.9

40.0

28.0

K-Feldspar

21.5

38.1

31.4

27.4

Biotite

17.0

6.6

5.4

Amphibole

4.8

tr

0.0

Clinopyroxene

0.0

0.0

Orthopyroxene

0.0

Fayalite Titanite

No.

14 (12)

PRADO

15 (3)

16

17

18 (5)

19 (3)

4.7

16.6

27.6

34.1

31.9

53.0

42.3

35.2

31.8

26.2

32.7

2.5

6.4

29.4

27.2

3.5

11.3

3.9

27.9

23.5

6.8

6.4

0.0

2.1

0.0

9.3

5.3

~

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.3

0.3

0.2

0.0

1.1

tr

1.8

4.8

0.5

0.1

Opaques

0.3

0.2

0.3

0.2

0.2

0.1

0.8

0.3

0.1

0.1

Allanite

0.4

0.1

0.3

tr

0.2

0.1

tr

0.2

0.2

0.1

Epidote

0.7

0.2

0.2

0.5

0.2

0.1

tr

0.5

0.1

0.1

Others

0.1

0.1

0.1

0.5

tr

tr

~

tr

tr

Table 3. Nos. 1-9 and 16, 17; selected representative samples; others, averages of several modal analyses: nos. of samples used - sample 10, n=3; 11, n=8; 12, n=10; 13, n=2; 14, n=12; 15, n=3; 18, n=5; 19, n=3. Sample types: I=FCAQMzN; 2=COAQMzN; 3=BAQS with CF; 4=BAmzG (Lagoa); 5=BAsG (Aggo); 6=OCABMzD (Umarf); 7=BADMzD; 8=ABQMzN; 9=BAMzG; 10=PABmzG; l l=PBMzG; 12=BmzG (Timbafiba); 13=BMG; 14=PABmzG; 15=BMG; 16=ABQD (diorite facies); 17=TABGD; 18=PBmzG; 19=BMG'. Symbols, Minerals: F=Fayalite; C=Clinopyroxene; O= Orthopyroxene; A=Amphibole; B=Biotite; Rocks and qualifying names: P=porphyritic; Q=quartz; MzN=Monzonite; S=Syenite; mzG=monzogranite; sG;Syenogranite; MzD=monzodiorite; MG=microgranite; D=Diorite; GD=granodiorite.

4

A. GALINDO, R. DALL'AGNOL, I. MCREATH, et aL

Q

l(

A

rl

&

rl

+

/

II

P Fig. 3. QAP diagram after Streckeisen (1976) for samples reported in Table 3. Symbols: see Fig. 4. gneisses. The northern and northwestern contacts with country rocks are hidden by Mesozoic sediments of the small Gangorra basin. Part of the eastern contact with the Tour~o granite is formed by indented faults (Fig. 2). Three petrographic facies are present in this body. The pink to dark green or grey Umarizal facies is best exposed around and south of the town of Umarizal, while the pink porphyritic, sometimes Rapakivi-like A ~ o facies occurs around Aqao farm, and occupies the northeastern tip of the body. Contacts between the facies are usually covered, and no contact between the two facies is shown on the map. Small bodies near the town of Patu form the Lagoa facies which is petrographically more closely related to the Umarizal and A ~ o facies than to any other granite type present in the region. The area of the Umarizal pluton corresponds to a Bouguer high relative to the surrounding gravity field (Lins, 1987), which could indicate the presence at depth of denser rocks than those exposed at the surface. Magmatic breccias associated with the Umarizal facies are composed of angular to tabular xenoliths, mainly of Jucurutu gneisses but including rarer examples of Caic6 Group gneisses. The xenoliths show localized preferred orientations, but consistent orientation at outcrop scale is absent. Larger bodies of Jucurutu gneisses totally surrounded by granite, such as the occurrences along the main road westwards from the town of Umarizal, ha~e garnet porphyroblasts formed by contact metamorphism. Further south of the main exposure of this body, small bodies of Umarizal granite penetrate the Jucurutu gneisses. Altogether, it appears that the present erosion level exposes the top of the intrusion. The Umarizal facies rocks are quartz monzonites and quartz syenites with fayalite or ferrohypersthene, hedenbergite, ferro-edenitic hornblende and subsidiary biotite.

The anhydrous ferromagnesian minerals are usually surrounded with hornblende-quartz symplectitic coronas. Larger masses of hornblende-quartz symplectites are probably products of total transformation of hedenbergite, while biotite-quartz symplectites are probably products of crystallization of late-stage liquids. Allanite, magnetite, ilmenite, zircon and apatite form the accessory mineral assemblage. Although the syenogranites of the A~ao facies have neither magmatic nor tectonic foliations, they locally have Rapakivi-like K-feldspar phenocrysts. Ferro-edenitic hornblende and biotite occur only as isolated crystals without either anhydrous nuclei or symplectites with quartz. Zircon and apatite are never as abundant as in the Umarizal facies, allanite was observed in only one thin section, ilmenite is rare and magnetite is absent. Sphene occurs as mantles to ilmenite crystals and as isolated crystals. The monzogranites of the Lagoa facies have hornblende-quartz symplectites which are possibly pseudomorphs after Caclinopyroxene, but are otherwise similar to the A ~ o rocks. The older Tour~o granite occupies about 350 km 2 in the south-central part of the region (Fig. 2). It intrudes rocks of the Jucurutu Formation and the Caic6 Group. Dikes of the Tourao granite cut the Carafibas granite. Contact relationships with the Serra do Lima complex to the southeast are not well established. Two main porphyritic facies are present in the Tourao granite. In the Tour~o facies, feldspar phenocrysts are larger and the rock compositions are generally mor e mafic than those of the Timbafiba facies. The latter facies has localized occurrences of sub-parallel hypermicaceous schlieren. Both facies have a NE-trending preferred orientation of biotite flakes whose direction corresponds to that of a penetrative foliation present in the country rocks. Magmatic flow structures such as en-echelon flow stacking

Evolution of Brasiliano-age granitoid types in a shear-zone environment, NE Brazil of K-feldspar phenocrysts are present, but very little evidence for solid-state deformation of crystal grains is observed. Inconclusive intrusion sequence relationships suggest that the two facies suffered penecontemporaneous intrusion. Late stage hydrothermal transformations occurred in these rocks, and the secondary assemblage includes fluorite. A third facies is represented by minor, fine to medium grained aplite or leucogranite dikes which cut both of the main facies. The limits of the ca. 60 km 2 almond-shaped Prado body in the northwestern part of the region are defined by shear zones (Fig. 2), which generated well-foliated mylonites in the rocks. In these mylonites, retrograde mineral assemblages are developed. The granitoids are predominantly porphyritic, but equigranular, medium to finegrained types are also present. Fine to medium-grained diorites are present in two situations. In the southwestern part of the body, massive rocks are intruded by veins and irregularly shaped networks of porphyritic granite. The diorites contain a few small feldspar phenocrysts, some of which are K-feldspar. In contrast, in the northeastern part of the body the diorites are present as swarms of oriented, roughly ellipsoid bodies contained in porphyritic granitoid. Contacts between the two are often sharp and regularly curved, though lobate contacts could signify that the rheological conditions of the diorite masses and/or the granitoid crystal suspension were heterogeneous, while sometimes diffuse contacts suggest that some limited interaction between intermediate and felsic materials occurred. Tonal differences between enclaves are not great, suggesting that more extensive chemical interactions between mafic and felsic components did not occur. K-feldspar phenocrysts are both larger and more abundant in the diorite enclaves than in the massive diorite. This is probably the result of mechanical injection of crystals formed in the granite suspension, and introduced into the plastic diorite masses during kneading together of the two different materials (Cantagrel et aI., 1984; Vernon 1986). The main interaction between the two materials was limited to mingling (Reid et al., 1983). The Prado diorites contain about 40 vol.% ferromagnesian minerals, mostly biotite with subordinate amphibole. The amphibole frequently occurs in symplectites with quartz, which is probably a transformation product of clinopyroxene. Sphene and opaque Fe-Ti oxide minerals are the principal accessory minerals, accompanied by traces of zircon and apatite. The ca. 250 km 2 Carafibas granite occupies the northeastern quadrant of the region, and mainly intrudes Caic6 Group orthogneisses. The principal facies is coarse to very c o a r s e - g r a i n e d and p o r p h y r i t i c , while the second (unnamed) facies is composed of leucogranites occupying the northwestern part of the body. Leucogranite also forms dikes which cut rocks of the porphyritic facies, and composes a satellite stock south of Carafibas town. Mafic rock occurrences are limited to hypermicaceous clots and schlieren.

85

This is the oldest and most deformed Brasiliano-age pluton presently identified in the region. It is cut by a number of shear zones, and has a strong NE-SW oriented penetrative deformation most strongly developed in or close to the shear zones. In low strain domains, it is still possible to identify crystal stacking and other flow orientations of feldspar phenocrystals. In higher strain domains, conjugated fracturing and rotation of feldspar phenocrystals accompanied by recrystallisation in pressure shadows is commonly observed. In the axial parts of the shear zones, mylonitic or ultramylonitic textures (Spry, 1969; Sibson 1977) are developed. Some retrograde minerals are present in the mylonites. Although the Caratibas, Prado and Tour~o granitoids are partly of different ages, the felsic rocks are petrographically similar. The dominant porphyritic biotite + hastingsitic hornblende monzogranites have a characteristic epidote- and magnetite-bearing accessory mineral assemblage. Allanite is frequently rimmed by epidote, which also occurs as isolated crystals. When in contact with or included in biotite, epidote crystals have well-developed crystal faces, but these are more irregular where epidote is in contact with felsic minerals. Sphene, zircon, apatite and opaque minerals (essentially magnetite) complete the accessory assemblage. In some samples of the Carafibas pluton, the abundance of sphene is sufficiently great for it to be considered a varietal mineral. The textural features of epidote are a good indication of its magmatic origin. The types found here correspond to Sial's (1990) types 2 (rims around allanite) and 3 (subhedral), found in other Brasiliano-age granitoids of NE Brazil, and also w i t h t y p e s d e s c r i b e d by Zen and Hammarstrom (1984; their Fig. 2b). The Quixaba body occupies the NNE part of the region and is the least well-exposed of the bodies. It forms an irregularly elongated body of ca. 100 km 2 whose northern contact with basement rocks is by shear, while faulting has brought Cretaceous sediments of the Potiguar basin into direct contact with the granitoid. Most of the other contacts are unexposed, and those shown on the map are largely photo-interpreted. Coarse grained quartz monzonites predominate, but a small roughly circular mafic body occurs in the central southwestern part in an area of low relief and very poor outcrop. The contact relationships between the mafic Umarf facies and the rest of the body are largely unknown. A strong gravity high is geographically associated with this mafic body (Lins, 1987), suggesting that the mass of mafic rocks present is quite substantial (ca. 10 km3). NE to EW trending shear zones are common in the main Quixaba quartz monzonite facies, but associated mylonitization is discontinuous, has very variable intensity, and apparently does not strongly affect the Umarf facies. On the other hand, Morais Neto (1987) considered that the deformation patterns of the Quixaba and Caratibas bodies are similar.

86

A. GALINDO, R. DALL'AGNOL, I. MCREATH, et al.

1,5

A TiOt

z.o

B

~~

o°o

1.0

2.0

0.5

i.O

0.0

62

I

I

65

70

c

MOO

,

~o '6~ 65

SiO=

7'0 '

I 7o ~u-'-s

°.°~z

75

NozO

r/~p6

1.O

J 75

3 , 5 ~

D

4.0

\ 0.5

*_e,f f "

"/

.'x

3.0-

¢

1

GO

65

62

E

-

n la

\A

70

SiO z

a

~. I

-75

A]

I

2.¢ 62

I

65

i

70 o,u":"z

Log loZr

75

LOOlosr

3.C

2.5 ¸

A

.x * X ~X

*

X

I

215

-

2.5

. / 2.0

20

LOgloRb

25

2.0

I

I

zo

I

[

I

Log|oRb

Fig. 4. a-d. Harker diagrams for rocks from the Umarizal-Carafibas region. Insets show analyses for mafic/intermediate rocks: open triangles, Prado diorites (Pd), open lozanges, Quixaba rocks (Qd), hachured area, trend for granitoids. Main graph shows granitoids: CTPc, Caratibas (crosses), Tour~o (asterisk) and Prado (empty triangle); U, Umarizal and related rocks (empty squares). (e) LogloZr vs. LOgl0Rb and (f) Log]0Sr vs. LogloRb. In Figures 4-9, all analysed samples are plotted using the same symbols.

Evolution of Brasiliano-age granitoid types in a shear-zone environment, NE Brazil

O O~|*'(:

' ¢'*

'

* r .-0.69

I '

(~)

°F o.o,.,

| II-O I~-II O.O', ,r. , .

t~

]

(

I

f°7I

I -,.or '-°° o-o

87

'

.'x,I

•

n,9~ r •-0.93

50

60

70

80

50

I 60

70

80

Fig. 5. SiO2 vs. log (Ca)/(Na20 + K20) diagram for determination of Peacock index, a=alkaline; a-c=alkali-calcic; c-a=calc-alkaline; c=calcic; n=number of samples; r=correlation coefficient. For Umarizal, the best fit line with n=13 is for Umarizal + Lagoa + Ag~o facies, while n=10 corresponds to Umarizal + Lagoa. The Quixaba facies is composed of quartz monzonites and monzogranites with biotite and hastingsitic hornblende as mafic minerals and sphene, allanite, ilmenite, zircon and apatite as accessory phases. The more mafic Umarf facies is composed of monzodiorites which contain Fe-rich diopside and ferrohypersthene in addition to biotite and amphibole. Allanite is absent, ilmenite predominates over rare magnetite and pyrite is present in the accessory assemblage. The two-pyroxene assemblage found in the Umarf facies is not common in the more mafic rocks of Brasiliano-age granitoids of the Borborema Province. Other examples are found in some plutons in the Serid6 region further to the east (Leterrier et aL, 1990; McReath, unpublished data).

(ferro)hypersthene and associated hornblende-biotite, magnetite- and ilmenite-bearing granitoids of the Umarizal body, and syenogranites and monzogranites of the Aqgo and Lagoa facies; (3) Two-pyroxene, ilmenite-bearing monzonitic association of the Quixaba body; (4) Diorites of the Prado body.

GEOCHEMISTRY

Linear regression of data points in Brown's (1982) diagram (Fig. 5) shows that the Umarizal rocks are alkaline or alkali-calcic, the Tour~o and Prado rocks are alkali-calcic, while the Carafibas rocks are possibly calc-alkaline in the classic Peacock definition. Although this procedure is not very satisfactory due to the absence in most bodies of intermediate or mafic rocks which control the calculated index, the diagnosis that true calc-alkaline rocks are not common in the region is amply confirmed using other classification diagrams, such as the R1-R2 (La Roche et al., 1980; Fig. 6) and Q-P and Q-B-F (Debon and Le Fort, 1983; Fig. 7) diagrams.

Details of the analytical methods are given in the Appendix, and analytical results are reported in Table 4. Most of the samples included here correspond to those analyzed modally (Table 3). The field and petrographical information indicated the following separation of the granitoid facies and associations: (1) Hornblende-biotite and biotite granitoids containing magmatic epidote and magnetite in the Carafibas and Tourao bodies, and form part of a bimodal dioritegranitoid association in the Prado body; (2) Fayalite or SAES 8 / 1 ~

, Most of the geochemical data obtained confirms these sub-divisions. Harker diagrams (Fig. 4a-d) show that compositional gaps separate the mafic-intermediate rocks of the Quixaba and Prado bodies from the other rock types present, and also demonstrate that these two mafic associations are chemically different. For the more felsic rocks, the similarity of the Caratibas, Tourao and Prado granitic rocks, and the different nature of the Umarizal rocks, are well shown by the groupings of data points. This conclusion is reinforced by the Zr-Rb and Sr-Rb plots (Fig. 4e-f).

88

A. GALINDO, R. DALL'AGNOL, I. MCREATH, et al. &

A

R2

i ioo-

Q 254

A

,,~ 20(

8

+

+~

700

i 500

I000

1500

o~

15(

50

2000

- 200

2500 Rf

Fig. 6. R1-R2 diagram. La Roche et al., 1980.

-

150

-I00

-50

I SO

0

P

I00

Fig. 7. Q-P diagram. Debon and Lefort, 1983.

1.6 A

80

-I~

CALC-ALKALINE AND STRONGLY PERALUMI NOUS

1.5 CI uJ

75

1.4

o o

~70

9

÷

v

+ 13

%

F~ ¢ , - ju

~

xA = ~ X • ~ A ALKALINE

x

65 + 1.2

50

6O

I.I

55 50 - 0.5

I 0

I 0.5

I I

I 1.5

I 2

I 2.5

Log(K20/MgO)

I t 0

I 2

I 4

I 6

I 8

I00( MgO+FeOt+Ti02

I I0 )

sir

Fig. 8. Diagram for separating calc-alkaline and alkaline types. Rogers and Greenberg, 1981.

Fig. 9. Sylvester's 1989 diagram for separating granitic types.

The Quixaba rocks are sub-alkaline, mafic monzonites o f S A L K D type, the Prado diorites are transitional between sub-alkaline and calc-alkaline, the Umarizal rocks are saturated alkaline (ALKOS), while the Carafibas, Tourgo and Prado granitoids are subalkaline (Fig. 5). Further reinforcement to this idea is given by other diagrams (Figs. 8 and 9).

Although the Quixaba and Prado mafic-intermediate rocks conform to shoshonitic trends on some diagrams (e.g., Fig. 7), some of their chemical features, such as the higher TiO2, Nb and Y and much higher Zr contents are more typically alkaline.

Plutonic rocks of shoshonitic affinity (Pagel and Leterrier, 1980; T h o m p s o n and Fowler, 1986) have been encountered in Northeast Brazil (Silva Filho et al., 1987; Sial and Ferreira, 1988; Guimar~es and Silva Filho, 1990) including the Serid6 region (Leterrier et al., 1990).

In the Pearce et al. (1984) Rb vs Nb + Y diagram (Fig. 10), most of the data points for Umarizal and all points for Quixaba fall within the within-plate field. Most of the rest occupy the collision granite field. No points lie well within the volcanic arc granite field. Most of the rocks of the Umarizal-Carafibas region are somewhat richer in Rb than those of the Serid6 region.

1 12

Evolution of Brasiliano-age granitoid types in a shear-zone environment, NE Brazil

/ SYN G

**

~

A

aa 10C

Rb

/

/

/

/

/

/

/

/

~ /

~

'~

]A "4" D A

A-I4-

4-

WPG

VAG

/

LO

10

Nb+Y

1OO

Fig. 10. Pearce et al. (1984) Rb vs. Nb + Y diagram. Discontinuous line limits field of compositions of Brasiliano-age granotoids from the Serid6 region, after Leterrier et al. (1990). GEOCHRONOLOGY Summaries of analytical methods are given in the Appendix, and results are given in Table 5. In the following, "age" is the calculated Rb-Sr whole rock isochron age, while Ri = (87Sr/86Sr) initial. All errors quoted are at the lcy level. Samples from the Caratibas body including one sample of microgranite yield an age of 631 _+ 23 Ma with Ri = 0.70860 + 106. The MSWD is 2.44 (Fig. lla). The inclusion of two samples (UCG 254 and UCG 268) from the Prado body, which may be related to the Carafibas body, yields an age of 654 + 24 Ma with Ri = 0.70751 + 99 and MSWD = 3.1 (Fig. llb). Although both age and MSWD values are slightly higher in the second case, ages and Ri values are identical within statistical error. Since it is more likely that only the Carafibas samples are cogenetic, the ca. 630 Ma age is preferred. Even so, the dispersion of points reflected by the high MSWD values could have either a primary magmatic origin, or a secondary cause related to mobilization of elements during shearing. The age of the Tour~o granite is 592 + 10 Ma, with Ri -0.70105 + 34 (Fig. 1 lc). Data points are rather dispersed (MSWD = 2.72) and exclusion of the obviously discrepant point (UCG 78) results in a better fit (MSWD = 0.87) with no significant changes in age (600 + 7 Ma) or Ri (0.70999 + 22). Considering that no geological or petrographic justi-

89

fication was found for this manipulation, it is accepted that the petrogenesis of the body may have been more complex than that required for a satisfactory isochron treatment, but data are at present insufficient to test other possibilities. Five samples of the Umarizal facies and one of the Lagoa facies define an age of 545 + 7 Ma with Ri = 0.71208 + 21 (Fig. 1 ld). The fit is good (MSWD = 0.45). The inclusion or exclusion of the Lagoa rock (UCG 11) would make no signifcant difference to these results, since the point lies exactly on the isochron. Although rocks of the Aq~o and Lagoa facies may not be cogenetic with those of the Umarizal facies, the inclusion of three points (UCG 45, UCG 48 and UCG 57) for A~fio rocks results in no significant changes of age (545 + 19 Ma) or Ri (0.71199 _+42), but the dispersion (MWD = 4.99) becomes much larger than analytical error. An attempt to date the Quixaba body was frustrated by the very smal ! range of Rb/Sr ratios of the rocks, and an exceptionally large dispersion of their 87Sr/86Sr ratios. The latter fact clearly indicates that the petrogenesis of these rocks was not a simple process, at least as far as the Rb - Sr system was concerned. There is satisfactory correspondence between the sequence of events recorded by the geological and geochronological data. Ri values increase with decreasing age. DISCUSSION Granite Typology a n d E v o l u t i o n

Brasiliano granitoid genesis in the Serid6 region of the Borborema Province occurred over a total time span from ca. 840 Ma to ca. 480 Ma, though the majority of ages, obtained mostly by the Rb-Sr whole r o c k isochron method, lie in the range ca. 650-550 Ma (Table 2). Regional scale (Jardim de Sfi et al., 1987; Leterrier et al., 1990) and more local (Hackspacher et al., 1987) studies demonstrate that petrographically similar rock types were emplaced in the Serid6 and adjacent regions at different times during the Brasiliano cycle. There is, however, a tendency for more mafic rock types (generically called diorites from now on) to occur with greater frequency in the earliest bodies. The "diorites" have a wide range of chemical affinities from tholeiitic to shoshonitic (Leterrier et al., 1990), the latter types considered as products derived mainly from enriched upper mantle whose presence has also been suggested by Sial and Ferreira (1990). To what extent some of the "diorites" owe their potassic nature to mechanical incorporation of K-feldspar phenocrysts from surrounding granite crystal suspensions (or other types of chemical interchange) is a question which still needs evaluation. The granitoids of the northern half of the Borborema province include calc-alkaline types (Sial and Ferreira, 1990), but K-alkali-calcic types are more abundant. Evolved compositions include leucogranites, some of which are biotite + amphibole- bearing types (Jardim de Sfi et al., 1987), while others are metaluminous but contain traces of muscovite and garnet (McMurry et al., 1987).

90

A. GALINDO, R. DALL'AGNOL, I. MCREATH, et al. .77

.77

.76

26

CARAUBAS • + PRA D

25 ~1 .74

~" m

.73

/

.72

,:631

(/)

_+Z3Mo

Ri : 0 . 7 0 8 6 0

.71

_+ 1 0 6

'

2:o

'

i o

'

,.o'

'

MO Ri : 0 . 7 0 7 5 1 t" 99 MSWD--3 . 1 (n=6)

"

t:654

/ ,71

MSWD= 2 . 4 4 ( n = 4 )

~Jo

r31 72

l

/ .70

~

.741

qD

Io ~

O

51o

'

I

6.o

I

?0

1.IO

n

I 2.0

e7 R b / e ( k 3 r

n

I 3.0

+ 24

u

4.no

,

a 5.0

n

6.nO

87Rb / 86Sr

.77

,00,-

"t u , , , , , , , ,

._.w

.75 @

,%

r,=

.73

Q

•710, O

,

1.0

,

2.0

,

,

.s,

3.0

,

4.0

5.0

6.0

O7Rb / 8 6 S r

0.0

0.5

,

,

,

1.0 87 R b /

,

1.5

2.0

2.5

3.0

3.5

86Sr

Fig. 11. Rb-Sr whole rock isochron diagrams. See text for discussion and Table 5 for analytical data. Rocks of saturated syenitic or transitional alkaline compositions are also known (Jardim de S~i et al., 1987; Leterrier et al., 1990). The only other known fayalite-bearing granite occurs as a facies of the Meruoca granite, situated at the extreme northwestern limit of the province (Fig. lb). Intrusive activity in this part of the Borborema Province has been dated by Rb-Sr whole-rock isochron studies in the range ca. 550 Ma (Sial et al., 1981) to ca. 480 Ma (Tavares et al., 1991). The granitoids of the Umarizal-Carat~bas region repeat in a geographically small area many of the features found in granitoids of the Serid6 region. The oldest Carafibas and Kalkali-calcic Prado and Quixaba (in part) bodies suffered solid state deformation by the Portalegre shear zone, which may have also controlled the intrusion of the Prado and part of the Quixaba bodies. The Tourfio body, further from the main loci of shear but also geochronologically younger, was intruded under the effect of waning deformation, but retains K-alkali-calcic character. Biotite leucogranites accompany these rocks, and were usually intruded later than the main masses. Finally, the more frankly alkaline Umarizal body, which was emplaced into rigid crust about 90 Ma after the Carafibas body, has some of the geochemical features of anorogenic, continental within-plate granites (Pearce et al., 1984; Whalen et al. 1987), but is better considered as posttectonic in relation to movement on the Portalegre shear zone. Comparisons with West A f r i c a

The transition from calk-alkaline orogenic rocks to alkaline post-tectonic or anorogenic rocks registered in the Umarizal-Carafibas region is thus similar to that described from other parts of the world (Bonin, 1990). In western

Africa, Pan-African plutonic associations range from calkalkaline in the Adrar des Iforas (Boullier et al. 1986) to potassic varieties including syenites. The overall time span of intrusion of granitoids in the Pan-African belt of NigerNigeria and neighboring countries is from ca. 660 to ca. 550 Ma, but it is possible that the cycle was complex, with docking episodes and accompanying granite genesis at different times during the overall time interval (Black and Li6geios, 1991). Fayalite-bearing syenites or monzonites (bauchites) occur in central-northern Nigeria (Oyawoye, 1965; Dada, 1989), while potassic syenites, including two-pyroxenebearing varieties, are found in southwestern Nigeria (Rahaman et al., 1991). Although long range correlations are hazardous, in pre-drift reconstructions (e.g., Caby, 1989), the granitoids of southwestern Nigeria (Rahaman et al., 1991) would have occurred approximately NNE of the UmarizalCarafibas region (using the present day Brazilian latitudelongitude grid as reference). The intervening region includes a number of regional faults or shear zones. U-Pb zircon ages of the potassic syenites (ca. 610-620 Ma: Rahaman et al., 1991) and of the bauchites (ca. 640 Ma: Dada and Respaut, 1989) are older than the Rb-Sr age of the Umarizal body. Part of this age difference might be attributed to different thermal behaviors of the U-Pb zircon and Rb-Sr whole rock systems, which frequently result in the Rb-Sr whole rock isochron age being younger than the UPb zircon age. On the other hand, the potassic syenites of southwestern Nigeria are syntectonic with relation to local shear zones and their U-Pb zircon ages are close to the ca. 630 Ma Rb-Sr age of the Carafbas granitoid. The bauchites are also syn-tectonic rocks (Dada and Respaut, 1989), in contrast to the post-tectonic Umarizal body.

Evolution of Brasiliano-age granitoid types in a shear-zone environment, NE Brazil

9

Table 4: Chemical Analysis of Whole Rocks from the Umarizal-Carafbas Area UMARIZAL No.

QUIXABA

1

2

3

4

5

6

7

8

SiO2

63.62

66.27

66.02

67.84

69.77

53.07

58.97

59.59

TiO2

0.50

0.61

0.46

0.68

0.34

1.81

0.83

0.80

A1203

15.86

15.66

14.25

13.86

13.75

16.61

16.73

17.53

Fe203

1.33

1.57

1.54

1.96

1.36

2.98

1.99

2.68

FeO

3.82

3.74

2.97

3.39

2.23

7.05

5.42

4.60

MnO

0.11

0.12

0.11

0.07

0.07

0.14

0.12

0.11

MgO

0.19

0.40

0.27

0.72

n.d.

2.45

0.81

0.83

CaO

2.08

2.30

1.94

2.15

1.18

5.00

4.16

3.29

Na20

3.97

4.10

3.85

3.12

3.40

4.04

4.19

4.08

KzO

6.39

5.56

5.79

5.50

5.44

3.90

4.25

5.48

P205

0.16

0.16

0.14

0.26

0.19

1.08

0.44

0.32

LOI

0.21

0.41

0.62

0.57

0.88

0.27

Total

98.24

100.46

97.64

99.91

98.35

98.70

98.79

100.19

Rb

105 125

99 242

162 236

162 201

155 156

88

105

125

Sr

614

407

395

Y

27

33

36

52

46

39

40

33

Zr

962

665

678

569

470

Nb

29

26

27

29

29

437 48

>1000 27

>1000 31

TOURAO

CARAUBAS

PRADO

10

11

12

13

14

15

16

18

SiO2

63.95

70.82

72.27

74.56

67.15

73.54

53.19

68.65

YiO 2

0.98

0.38

0.24

0.10

0.69

0.16

1.97

0.53

A1203

14.79

14.61

13.91

12.82

14.60

13.46

15.71

14.38

No.

Fe203

2.17

0.08

1.03

0.38

1.29

0.90

3.81

1.34

FeO

4.68

2.69

1.01

1.01

3.03

0.72

5.69

2.02

0.02 0.22

0.11 4.12

0.05

0.98

6.16

1.92

MnO

0.12

0.05

0.02

0.02

0.06

MgO CaO

1.13

0.51

0.15

n.d.

3.24

1.55

1.18

0.88

0.88 2.41

Na20

3.86

3.66

3.27

2.83

3.25

3.04

3.30

3.45

K20

4.44

5.60

5.41

5.67

5.32

5.62

2.97

5.09

P205

0.31

0.11

0.13

0.10

0,28

0.28

0.86

0.26

0.70 99.64 385

0.98 98.87 83

0.33

160 10 149

917 27 362

319 20

22

26

---

0.95

0.65

0.29

Total Rb

99.67 163

100.19 376

99.57 325

99.04 264

99.25 272

Sr

368

338

208

110

Y Zr

48 438

360 349

22 228

11 148

325 37

Nb

31

33

21

9

LOI

339 24

0.68

98.76 228

333 23

Table 4. Numbers correspond to those of Table 2. Where mean modal analyses are reported in Table 2, the corresponding chemical analysis in this table is for a representative sample. No equivalent of nos. 9, 17, and 19 (Table 2) were chemically analysed.

92

A. GALINDO, R. DALL'AGNOL, I. MCREATH, et al.

Table 5: Whole Rock Rb/Sr Data WHOLE ROCK Rb/Sr DATA SAMPLE NO.

Rb ppm

Sr ppm

87Rb/ 86Sr

1G

87Sr/ 86Sr

UMARIZAL UCG334

99.2

241.9

1.188

0.010

0 . 7 2 1 2 9 0.00008

UCG44

108.9

183.8

1.718

0.015

0 . 7 2 5 6 4 0.00017

UCG54

121.1

107.5

3.268

0.028

0 . 7 3 7 5 6 0.00010

UCG02

105.3

125.2

2.440

0.021

0 . 7 3 0 9 5 0.00011

UCG18

106.9

122.9

2.524

0.022

0 . 7 3 7 5 6 0.00016

162.4

201.1

2.341

0.044

0 . 7 3 0 2 0 0.00007

UCG45

154.7

178.6

2.513

0.023

0 . 7 3 0 8 4 0.00006

UCG48

154.8

155.7

2.883

0.056

0 . 7 3 3 7 8 0.00008

UCG57

116.5

104.7

3.229

0.028

0 . 7 3 7 8 9 0.00008

UCG124B

182.2

368.9

1.434

0.014

0 . 7 2 2 3 1 0.00007

UCG183A

202.8

270.9

2.170

0.023

0 . 7 2 8 5 7 0.00017

UCG222

209.4

228.4

2.659

0.027

0 . 7 3 2 3 9 0.00005

UCG78

320.4

219.7

4.324

0.054

0 . 7 4 4 7 1 0.00004

UCG9

306.7

162.9

5.474

0.066

0 . 7 5 7 0 8 0.00005

UCG10B

246.0

122.8

5.826

0.070

0 . 7 6 0 0 4 0.00012

Lagoa UGC11 Aqfio

Tou~o

CARAIJBAS UCG135

273.1

318.2

2.488

0.029

0 . 7 3 0 9 2 0.00012

UCG79

261.8

248.6

3.056

0.036

0 . 7 3 0 6 8 0.00007

UCG82

261.3

238.6

3.179

0.037

0.73780

UCG301

332.8

161.4

6.000

0.077

0 . 7 6 2 3 1 0.00009

UCG254

278.1

378.8

2.128

0.025

0 . 7 2 6 8 4 0.00017

UCG268

281.5

260.4

3.137

0.036

0 . 7 3 7 2 6 0.00014

0.00007

PRADO

Charnockitic rocks with two-pyroxene assemblages form significant masses among Pan-African granitoids in Nigeria (Dada et al., 1989), a situation which contrasts with the rare preservation of this assemblage in Brasiliano rocks of Northeast Brazil. The Nigerian examples are, however, accompanied by granitoids and hybrid rocks with hydrated mineral assemblages, similar to the Brazilian case.

Mineralogy and Geochemistry The presence of primary epidote in Brasiliano granitoids of the Borborema Province was noted by Almeida et al. (1967) and m a n y cases have subsequently been

described by Sial (1990) and Sial and Ferreira (1990). The latter authors used the Johnson and Rutherford (1989) A1in-hornblende geobarometer to infer amphibole crystallization pressures in the range 7.3-2.9 Kb, most values corr e s p o n d i n g to the "high" p r e s s u r e s w h i c h m a y be necessary for igneous epidote crystallization from felsic magmas (Zen and Hammarstrom, 1984). On the other hand, many samples studied by Sial (1990) and Sial and Ferreira (1990) are from bodies which intrude supracrustal sequences in low metamorphic grade. The magmas must therefore have risen through perhaps 10-15 km of crust after crystallization of early mineral phases but before total solidification. This fact challenges either the validity

Evolution of Brasiliano-age granitoid types in a shear-zone environment, NE Brazil of the current concepts about epidote stability in granitic magmas and the Al-in-hornblende geobarometer, or some current notions about possible physical mechanisms of granitic magma or crystal suspension ascent, such as slow diapiric rise. In the Umarizal-Caratibas region, the apparent conflict is less intense. The geological evidence suggests that all granites were intruded into the mesozone. Textural evidence from the Umarizal-Carat~bas granitoids demonstrate that rising oxygen fugacity may also exercise a control on the appearance of epidote rather than allanite. The former, occurring as cores to epidote in a few situations, contain essentially reduced iron, while epidote contains oxidized iron (Deer etal,. 1966). The epidote-bearing granites contain magnetite as the only Fe-Ti oxide mineral. Although ferrosilite is stable at higher pressures than fayalite, a increase of the En content of iron-rich orthopyroxene reduces the lower pressure stability limit (Bohlen and Boettcher, 1981). Thus, the ferrohypersthene-bearing rocks of the Umarizal facies did not necessarily commence their crystallization under the pressure conditions of deep continental crustal levels. The mutual incompatibility of fayalite and ferrohypersthene in the Umarizal granitoid suggests that the two rock types followed different evolutionary path. Apart from constant-sum effects, the straight line or slightly curved correlations observed in Harker diagrams (Fig. 5a-5d) could have a number of origins. No petrographic signs of inefficient restite unmixing (Chappell et al., 1987) were encountered. The absence of wide tonal variations in the dioritic enclaves of the Prado body, and the wide compositional gaps between the Prado and Umarf/Quixaba marie to intermediate rocks on the one hand, and the remaining granitoids on the other suggest that magma mixing, if it occurred, was limited to interactions in which either the marie/intermediate pole or the felsic pole was the dominant component. The possible roles of partial melting or fractional crystallization are being investigated.

93

post-tectonic bodies have geochemical signatures of within-plate granites. The mafic rocks have some shoshonitic geochemical features, but also present some more distinctly alkaline characteristics. Acknowledgements--This presentation is based on part of Galindo's doctoral thesis at the UFPA, which was supported by a scholarship and other financing from FINEP (PADCT) and CAPES (Brazil). Galindo wishes to thank the staff at CRPG (Nancy), especially J. Leterrier, for their support during a short visit to the laboratories, and the personnel of the chemical and geochronological laboratories of CRPG and UFPA for their careful analytical work. McReath thanks the DG/UFRN and the CGC/UFPA for the opportunity to participate in this work. We also acknowledge Emanuel E Jardim de S~i and Jaziel M. S~i (UFRN) for discussions on this manuscript.

APPENDIX 40 samples were analysed for major, minor and trace elements at the Centre de Recherches Prtrographiques et Grochimiques, Vandoeuvre, France, using sample attack and ICP measurement as described by Govindaraju et al. (1976). A additional 24 samples were analysed at the laboratories of the Centro de Geociancias of the UFPA for major, minor and trace elements using XRF procedures for Si, A1, Ti, Fe total, Ca, K, P, Rb, St, Nb, Y and Zr; and AAS procedures for Na, Mg and Mn. FeO and LOI were determined for all 64 samples at the UFPA. Analytical performance was monitored by duplicate analyses, and comparative performances of the two laboratories were monitored by repeat analysis of a small number of samples at the two laboratories. In general, the correspondences were satisfactory. A number of detailed mineral identifications were done on a CAMECA Camebax microprobe at the Universit6 des Sciences de Nancy II. After pre-selection of samples using XRF determination of Rb and Sr contents, samples for Rb-Sr dating were analysed by the standard ion exchange separation and mass spectrometric procedures adopted by the Laborat6rio de Geologia Isot6pica, UFPA (see Gastal et al., 1987). Element contents were determined by isotope dilution. Procedural blanks for Rb and Sr did not exceed 5 ng. Correction for mass discrimination assumed 86Sr/88Sr = 0.1194. Isochron regressions were carried out using York's (1966, 1969) method, incorporating modifications suggested by Williamson (1968) Errors are quoted at the 1~ level, and ~ = 1.42 x 10"11 (Steiger and Jager, 1977).

REFERENCES

CONCLUSIONS The granitoids of the Umarizal-Carafbas region were intruded over a ca. 90 Ma time interval during the Brasiliano orogenic cycle. They range from syn-tectonic at about 630 Ma to post-tectonic at about 545 Ma, in relation to activity along a major shear zone. The bodies were intruded at a similar crustal level. Syn- and late-tectonic bodies include an important magmatic epidote- and magnetite-bearing granite type. The main masses of these granitoids are accompanied by later intrusions of biotiteleucogranitoids. Chemical compositions of the felsic rocks evolve from sub-alkaline (K-alkali-calcic) in the earliest bodies to more frankly alkaline in the latest body. Maficintermediate rocks accompany two of the earlier bodies. These are likely to have originated in enriched upper mantle. All granitoids of the Umarizal-Caratibas region have (87Sr/86Sr)i ratios which suggest strong contributions from old crustal sources. Geochemical signatures of the felsic components of the earlier intrusions are mostly comparable with those of modern collision granites. The latest

Almeida, EEM., Leonardos Jr., O.H., Valenca, J., 1967. Review on granitic rocks of northeast South America. Symposium of lUGS/ UNESCO, Recife, Brazil. 41 p. Almeida, EEM., Hasui, Y., Bfito Neves, B.B., Fuck, R.A., 1977. Provfncias estruturais brasileiras. VII Simp6sio de Geologia do Nordeste, Campina Grande, Brazil. Anais, 363-391. Arafijo, T.C.M., 1985. Estudo geol6gico-geofi'sico da regi~o de Umarizal (RN): Area III. Unpublished Graduation Report, Departamento de Geologia, Universidade Federal do Rio Grande do Norte, Brazil. 93 p. Black, R. and Lirgeois, J-P., 1991. Pan-African plutonism of the Damagaram inlier, Niger Republic. Journal of African Earth Sciences 13, 471-482. Bohlen, S.R. and Boettcher, A.L., 1981. Experimental investigations and geological applications of orthopyroxene geobarometry. American Mineralogist 66, 951-964. Bonin, B., 1990. From orogenic to anorogenic settings: evolution of granitoid suites after a major orogenesis. Geological Journal 25, 261-270. Brown, G.C., 1982. Calc-alkaline intrusive rocks: their diversity, evolution and relation to volcanic arcs. In: Andesites (edited by R.S. Thorpe), John Wiley and Sons, England.

94

A. GALINDO, R. DALL'AGNOL, I. MCREATH, et al.

Caby, R,, 1989. Precambrian terraues of Benin-Nigeria and northeast Brazil and the Late Proterozoic south Atlantic fit. Geological Society of America Special Paper 230, 145-158.

Jardim de S~i, E.E, Legrand, J-M. and McReath, I., 1981b. Estratigrafia de rochas granRicas na regiSo do Serid6 (RN/PB), com base em crittrios estruturais. Revista Brasileira de Geoci~ncias 11, 50-57.

Campos. M., Braga, P.S., Mello, A.A., Souza, E.M., Silva, F.A.F., Franca, J.B., 1979. Projeto Rio Jaguaribe. Relatdrio Final DNPM/CPRM, Brasflia. 149 p.

Jardim de S~i,E.E, Legrand, J-M, Sena de S~l, C., Lima, P.C., Medeiros Neto, EA., Rego, J.M. and Souza, L.C., 1981a. Esbo~o da geologia prt-cambriana do extremo oeste potiguar. X Simp6sio de Geologia do Nordeste, Recife. Atas, 315-324.

Cantagrel, J.M., Didier, J., Gourgaud, A., 1984. Magma mixing: Origin of intermediate rocks and enclaves from volcanism to plutonism. Physics of the Earth and Planetary Interiors 35, 63-76. Chappel, B.W., White, A.R. and Wyborn, D., 1987. The importance of residual source material (restite) in granite petrogenesis. Journal of Petrology28, 1111-1138. Curioso, R.L., 1987. Estudo geoltgico/geoffsico da regiSo de Umarizal (RN) - drea 1. Unpublished Graduation Report, Departamento de Geologia, UFRN, Natal. 90 p. Dada, S.S. and Respaut, J.R, 1989. La monzonite/t fayalite de Bauehi (bauchite), nouveau tem6in d'un magmatisme syn-tectonique PanAfricain an nord du Nigtria. Comptes Rendus Acaddmie des Sciences de Paris, t. 309, serie II, 88%892. Dada, S.S., Lancelot, J.R. and Briqueu, L., 1989. Age and origin of the annular charnockitic complex at Toro, Northern Nigeria: U-Pb and Rb-Sr evidence. Journal of African Earth Sciences 9, 227-234. Dantas, J.R.A., 1974. Carta geol6gica do Brasil ao miliondsimo: Folha Jaguaribe-Fortaleza (SB 24). DNPM, Brasflia. 95 p. Debon, E and Le Fort, E, 1983. A cationic classification of common pintonic rocks and their magmatie associations: principles, method, applications. Bulletim Mindralogique 111,493-510. Deer, W.A., Howie, R.A. and Znssman, J., 1966. An introduction to the rock-forming minerals. Longman, London. 528 p. DNPM 1984. Mapa geol6gico do Brasil e da drea oce8nica adjacente. Escala 1:2.500.000. Brasflia.

Jardim de S~, E.F., Kawashita, K., Macedo, M.H.de F., S~i J.M., 1986. Supracrustais monoclclicas do extremo Oeste do Rio Grande do Norte. XII Simpdsio de Geologia do Nordeste, Jo~o Pessoa. Atas, 6272. Jardim de S~i, E.F., Macedo, M.H.F., Legrand, J-M., McReath, I., Galindo, A.C. and S~i, J.M., 1987. Proterozoic granitoids in a polycyclic setting: the Serid6 region, N.E. Brazil. International Symposium on Granites and Associated Mineralizations, Salvador. Extended Abstracts, 103-110. Johnson, M.C. and Rutherford, M.J., 1989. Experimental calibration of the aluminium-in-hornblende geobarometer with application to Long Valley caldera (California) volcanic rocks. Geology 17, 837-841. La Roche, H. de, Leterrier, J., Granclaude, P. and Marchal, M., 1980. A classification of volcanic and plutonic rocks using R1-R2 diagrams and major element analyses: its relationship with current nomenclature. Chemical Geology 29, 183-210. Legrand, J-M., Deutsch, S. and Souza, C., 1991. Data~go U/Pb e granitogSnese do maci~o de Acari, RN. XIV Simpdsio de Geologia do Nordeste, Recife. Boletim 12 (Resumos), 172-177. Leterrier, J., Jardim de Sg, E.E, Macedo, M.H.E and Amaro, V.E., 1990. Magmatic and geodynamic signature of the Brasiliano cycle plutonism in the Serid6 belt, NE Brazil. XXXVI Congresso BrasiIeiro de Geologia, Natal. Anais v. 4, 1640-1655.

Galindo, A.C. 1993. Petrologia dos granitdides brasilianos da regiSo de Cara~tbas e Umarizal (Oeste do Rio Grande do Norte). Unpublished Doctoral Thesis, UFPA. 386 p.

Lins, F.A.P.L., 1987. Geof(sica aplicada ao estudo do arcaboufo tectOnico de bacias sedimentares entre as Bacias Potiguar e Rio do Peixe. Unpublished Masters Thesis, Instituto de Geoci~ncias, Universidade Federal de Pernambuco. 74 p.

Gastal, M.C.P., Macambira, M.J.B., Medeiros, H. and DalrAgnol, R. 1987. Idades e geoquimica isot6pica Rb-Sr do Granito Musa e Granodiorito Rio Maria, Amaz6nia Central Geochimica Brasiliensis 1, 247-259.

Macedo, M.H.F., Jardim de S~i, E.F., Sato, K. and Kawashita, K., 1990. Dados isot6picos preliminares de Nd e Sr na regi~o do Serid6 (RN/ Pb), e suas implicag0es tect6nicas. XXXVI Congresso Brasileiro de Geologia, Natal. Resumos, 305.

Gorayeh, RS. de S.Tavares Jr., S.S. and Lafon, J.M. 1991. Novos dados geocronol6gicos na regi~o de Forquilha - Santa Quittria, NW do Cear~t. XIV Simpdsio fe Geologia do Nordeste, Recife. Boletim 12 (Resumos), 260-263.

McMurry, J., Long, L.E. and Sial, A.N., 1987a. Two-stage evolution of a texturally homogeneous granite: REE and isotopic evidence from the Dona In~s pluton, Northeast Brazil. International Symposium on Granites and Associated Mineralizations, Salvador. Extended Abstracts, 155-156.

Govindaraju, K., Mevelle, G. and Chouard, C., 1976. Automated optical emission spectrochemieal bulk analysis of silicate rocks with microwave plasma excitation. Analytical Chemistry 48, 1325-1331. Guimar~es, I de E and da Silva Filho, A.E, 1990. Magmatismo intrnsivo shoshonftico na faixa Pajeti-Paraiba: o complexo Born Jardim. 36 Congresso Brasileiro de Geologia, Natal. Anais v. 4, 1739-1757.

McMurry, J., Long, L.E. and Sial, A.N., 1987b. Petrology and isotope systematics of magma/crystal mushes: some porphyritic granitoids of northeastern Brazil. International Symposium on Granite and AssociatedMineralizations, Salvador. Extended Abstracts, 173-176.

Hackspacher RC. and Oliveira, D.C., 1984. A evolu~o estrutural das zonas de cisalhamento Portalegre e Santa M6nica (RN) XXXIII Congresso Brasileiro de Geologia, Rio de Janeiro. Anais v. 4, 1679-1692.

Morals Neto, J.M., 1987. Mapeamento geoldgico da zona de cisalhamento Portalegre (ZCP) e encaixantes, numa drea entre Cara~bas, Apod[ e Umarizal (RN). Unpublished Graduation Report, Departamento de Geologia, UFRN, Natal. 224 p.

Hackspacher, P.C. and S~i, J.M., 1984. Critdrio litoestratigr~ifico para a difereneia~o do Embasamento Caic6 do Grupo Serid6, RN, Brazil. XI Simpdsio de Geologia do Nordeste, Natal. Atas, 263-277.

Oyawoye, M.O., 1965. Bauchite: a new variety in the quartz monzonite series. Nature 205, 689.

Hackspacher, P.C. and Legrand, J-M., 1984. Microstructural and metamorphic evolution of the Portalegre shear zone, northern Brazil. Revista Brasileira de Geoci~ncias 19, 63-75. Hackspacher, RC., Macambira, M., McReath, I. and Scheller, T., 1987. Tectonic-magmatic evolution of the Taiptt-Cardoso polydiapiric granitoid bodies, Rio Grande do Norte, Brazil. International Symposium on Granites and Associated Mineralizations, Salvador. Extended Abstracts, 91-95. Jardim de S~I, E.E, 1984. A evolu~o proteroz6ica da Provfncia Borborema. XI Simpdsio de Geologia do Nordeste, Natal. Atas, 297-316.

Pagel, M. and Leterrier, J., 1980. The subalkaline potassic magmatism of the Ballons Massif (southern Vosges, France): shoshonitic affinity. Lithos 13, 1-10. Pearce, J.A., Harris, N.B.W. and Tindle, A.G., 1984. Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. Journal of Petrology 25, 956-982. Rahaman, M.A., Tubosun, I.A. and Lancelot, J.R., 1991. U-Pb geochronology of potassic syenites from Southwestern Nigeria and the timing of deformational events during the Pan-African Orogeny. Journal of African Earth Sciences 13, 387-395.

E v o l u t i o n o f B r a s i l i a n o - a g e granitoid types in a shear-zone e n v i r o n m e n t , N E Brazil Reid, J.B., Jr., Evans, O.C. and Fates, D.G., 1983. Magma mixing in granitic rocks of the Sierra Nevada, California. Earth and Planetary Science Letters 66, 243-261. Sgt, J.M., Bezerra, F.H.R., Macedo, M.H.E and Pereira, R., 1988. Middle Proterozoic supracrustals and Brasiliano orogeny in the southeast Cear~ state; a monocyclic evolution. VII Congresso Latino-Americano de Geologia, Bel6m. Anais 1, 35-48. Sial, A.N., 1987. Granitic rocks of northeast Brazil. International Symposium on Granites and Associated mineralizations, Salvador. Extended Abstracts: 61-69. Sial, A.N., 1990. Epidote-bearing ealc-alkatic granitoids in Northeast Brazil. Revista Brasileira de Geoci~ncias 20, 88-100. Sial, A.N. and Ferreira, V.P., 1988. Brasiliano age peralkaline plutonic rocks of the central structural domain, northeast Brazil. Rendiconti Societd Italiana di Mineralogia e Petrologia 43, 307-341. Sial, A.N. and Ferreira, V.P., 1990. Granitoids in northeast Brazil: oxygen and sulfur isotope compositions and depths of emplacement. Journal of South American Earth Sciences 3, 103-112. Sial, A.N., Figueiredo, M.C.H. and Long, L.E., 1981. Rare-earth element geochemistry of the Meruoca and Mocambo plutons, Cear~i, northeast Brazil. Chemical Geology 31, 271-283. Sibson, R.H., 1977. Fault rocks and fault mechanisms. Journal of the Geological Society of London 133, 191-213. Silva Filho, A.E da, Thompson, R.N. and Leat, P.T., 1987. Petrology of the Terra Nova pluton, Brazil and associated ultrapotassic dykes. Revista Brasileira de Geoci~ncias 17, 481-487. Spry, A., 1969. Metamorphic Textures. Pergamon, Oxford. 350 p.

95

Tavares JR., S.S., Lafon, J-M. and Gorayeb, ES. de S., 1991a. O granito Serra da Barriga, noroeste do Cear4: caracterfsticas petrol6gicas, geoqu/micas e geocronologia Rb-Sr. X W Simp6sio de Geologia do Nordeste, Recife. Bol. 12 (resumos), 178-181. Tavares JR., S.S., Gorayeb, RS. de S. and Lafon, J.M., 1991b. Petrografia e geoeronologia Rb-Sr do feixe de diques da borda oeste do granito Meruoca (CE). 36 Congresso Brasileiro de Geologia, Natal. Resumos, 337. Thompson, R.N. and Fowler, M.B., 1986. Subduction-related shoshonitic and ultrapotassic magmatism: a study of Siluro-Ordovician syenites from the Scottish Caledonides. Contributions to Mineralogy and Petrology 94, 507-522. Torquato, J.R. and Cordani, U.G., 1981. Brazil-Africa geological links. Earth Science Review 17, 155-176. Vernon, R.H., 1986. K-feldspar megacrysts in granites-phenocrysts, not porphyroblasts. Earth Science Review 23, 1-63. Whalen, J.B., Currie, K.L. and Chappel, B.W., 1987. A-type granites: geochemical characteristics, discrimination and petrogenesis. Contributions to Mineralogy and Petrology 95, 407-419. Williamson, J.H., 1968. Least-squares fitting of a straight line. Canadian Journal of Physics 46, 1845-1847. York, D., 1966. Least-squares fitting of a straight line. Canadian Journal of Physics 44, 1079-1086.

Steiger, R.H. and Jager, E., 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters 36, 359-362.

York, D., 1969. Least-squares fitting of a straight line with correlated errors. Earth and Planetary Science Letters 5, 320-324.

Streekeisen, A., 1976. To each plutonic rock its proper name. Earth Science Review 12, 1-33.

Zen, E-A., and Hammarstrom, J.M., 1984. Magmatic epidote and its petrologic significance. Geo',.ggy 12, 515- 518.