serie NOVA TERRA nº 48

6. PROVENANCE OF THE HP-HT UPPER ALLOCHTHON

123

10), pointing to a crustal recycling due to the initial

development of the Cadomian arc system (‘proto

–

arc’ stage; Fig. 8). The

590

–

490 Ma Banded

Gneisses zircon is very abundant and shows a maxi-

mum peak at

522

–

512 Ma, and other abundant

peaks at

575, 560 and 545 Ma (Fig. 4). The Lu

–

isotopic pattern (Fig. 7a,b) shows that this Banded

Gneisses zircon is arranged vertically, with

(

)

val-

ues up to

9.5 (zircon almost directly derived from a

DM source), and

(

)

values down to 38. These

patterns can be explained by the intrusion of juvenile

magmas that triggered mixing processes with an

Eburnean and Archean crust (and with a small pro-

portion of reworked early Cadomian crustal mate-

rial), consistent with a peripheral arc activity at the

Northern WAC.

According to the data presented here, the MDA

for the siliciclastic series of the HP

–

HT upper units is

521 Ma. Taking into account crystallization ages

of intrusive igneous rocks in this formation (

506

–

484 Ma, Albert

et al.

, 2013), the protolith of the

eclogite facies paragneisses of the Cabo Ortegal Com-

plex was a Middle Cambrian siliciclastic sedimentary

series.

As discussed by Cawood

et al.

(2012), convergent

margin basins have a high proportion of detrital zir-

con with ages close to the age of the sediment, where

back-arc basins have an increasing input of older

detritus from the adjoining mainland. The Banded

Gneisses U

–

Pb age distribution pattern is a strongly

bimodal one, where the majority of the Palaeozoic

–

Neoproterozoic fraction is concentrated near the age

of sedimentation (Fig. 4). These observations suggest

that the protoliths of the sedimentary rocks involved

in the Banded Gneiss formation may have been

deposited in a back-arc type basin, where the vol-

canic arc system was very active, shedding its juvenile

materials into the basin at the same time as the

adjacent WAC supplied the Eburnean and Archean

detritus.

Comparison between the provenance of the IP and the

–

HT upper units of the Cabo Ortegal Complex

When comparisons are made with the unit overlap-

ping the HP

–

HT Banded Gneisses (the IP Cari

Gneisses, Albert

et al.

, 2015) big similarities are

observed. Their MDAs are very similar (Cari

Gneisses MDA

510 Ma,

Banded Gneisses

MDA

521 Ma). As no big differences in their

youngest U

–

Pb ages are apparent, it seems reasonable

to conclude that both units are temporally related.

–

Nd isotopic experiments also reveal the simi-

larity between the gneissic units. The

(

)

values

for the Banded Gneisses (from 6.1 to 9.8, Fig. 9)

are in the range of those for the Cari

no Gneisses

( 7.7 to 10.2, Albert

et al.

, 2015), except for sam-

ple GCH-17 (

(

)

2.5, this study). This higher

(

)

value is possibly due to the high Palaeozoic

–

Neoproterozoic juvenile input in the sample. Of the

zircon cores analysed with Lu/Hf in this sample, 64

are Palaeozoic

–

Neoproterozoic (

65/101) of which

62% are juvenile (

40/65,

(

)

0).

A Kolmogorov

–

Smirnoff test to evaluate the statis-

tical similarities between age distributions of the sam-

ples of the unit studied in the present contribution

failed. This does not necessarily imply that the sam-

ples do not come from the same detrital sources, but

seem to resemble the heterogeneity of the Banded

Gneisses in contrast to the homogeneity exhibited by

the Cari

no Gneisses (see Fig. 6).

The main difference between their U

–

Pb age den-

sity distributions is that the Banded Gneiss formation

has an abundant

590

–

540 Ma Ediacaran popula-

tion (Fig. 4) and the Cari

no Gneiss formation has

not. The Cari

no Gneisses were tentatively related to

the late or post-activity of the Cadomian Orogen

(and in a broader sense to the activity of a Pan-Afri-

can orogen). The abundance of this Ediacaran popu-

lation in the Banded Gneisses favours this

interpretation. If the magmatic arc, revealed by the

isotopic information of the Upper Allochthon detrital

units, is the Cadomian magmatic arc (as described by

Linnemann

et al.

, 2014) the magmatic activity

inferred by the Banded Gneisses lasted at least until

510 Ma in the NW Iberian section of the Gond-

wana margin, instead of

540 Ma (as described in

the Bohemian Massif, Linnemann

et al.

, 2014). This

could imply a diachronous scissor-like continental arc

generation, due to oblique vector of subduction, that

propagated westwards (cf. Linnemann

et al.

, 2008).

But more likely, it could imply that the

540

–

510 Ma arc activity was not registered in the Bohe-

mian Massif sector of the Gondwana margin,

because magmatic activity can strongly vary between

segments of the same arc system.

Cari

no Gneiss and Banded Gneiss formations rep-

resent the same section of the Gondwanan margin.

Both have the same detrital sources (despite the rela-

tive higher abundance of the Banded Gneisses Edi-

acaran population, see discussion above). Their

bimodal detrital populations suggest that both forma-

tions had the same geological setting, i.e. sedimenta-

tion in a back-arc type basin. Their MDAs and

(

)

values are very similar (Fig. 9). Banded Gneiss

formation was profusely intruded by acid and basic

magmas (protolith ages of

506

–

484 Ma, see geo-

logical context) and Cari

no Gneisses were not. All

these observations lead one to suppose that they were

probably deposited in the same back-arc basin at

521

–

506 Ma. The first sediments deposited were

those of the Banded Gneiss formation (due to its

higher isotopic heterogeneity and higher presence of

intruded igneous rocks), formed by the mixture of

the old components from the WAC (Eburnean and

Archean detritus) and abundant arc related

590

–

520 Ma sediments. The Cari

no Gneiss protoliths

deposited afterwards, filling the same basin with

2015 John Wiley & Sons Ltd

PROVENANCE OF THE HP

–

HT VARISCAN TERRANE

973