THE OCEAN/CONTINENT TRANSITION OFF WESTERN IBERIA
Crustal and Upper Mantle Structure Across the Ocean/continent Transition
The integration of all the available geophysical data and modeling results, constrained locally with data from deep ocean drilling
(DSDP Leg 47B: Groupe Galice, 1979; Sibuet and Ryan, 1979; ODP Leg 103: Boillot, Winterer, Meyer, et al., 1987; ODP Leg 149: Sawyer,
Whitmarsh, Klaus, et al., 1994), allows the definition of a reasonably well-constrained picture of the crustal and upper mantle structure across the ocean/continent transition in the three main segments of the west Iberia Margin
(Boillot et al., 1980; 1986; 1989a,b; 1992; Whitmarsh et al., 1990; Pinheiro et al., 1992;
Horsefield, 1992; Whitmarsh et al., 1993; Pinheiro, 1994; Whitmarsh and Miles, 1995). The designation ocean/continent transition is used here, instead of the term ocean/continent boundary
(OCB) often referred to in the literature, because it now seems clear that the transition between continental and oceanic crust at nonvolcanic passive margins does not generally occur at a sharp boundary but, instead, it occurs over a transition zone, in which the crustal section has characteristics intermediate between continental and oceanic crust, and beneath which serpentinized upper mantle rocks may occur.
In broad terms, the ocean/continent transition off western Iberia
(Fig. 10A-B) is characterized by an exceptionally thin crust (less than 6 km thick, typically
2–4 km thick, and, in places, virtually nonexistent), underlain by a fairly thick layer (normally
2–3 km but up to 7 km) of a material with a high P-wave velocity (7.2 
VP
7.9 km/s). These high seismic velocities are intermediate between those generally observed in the lower continental or oceanic crust (normally lower than
7.1–7.2 km/s in the absence of crustal
underplating) and those that characterize the uppermost "normal" mantle (8.0-8.1 km/s). Such material has been interpreted as serpentinized upper mantle
(Pinheiro et al., 1992; Whitmarsh et al, 1993; Pinheiro, 1994). The idea of serpentinized peridotite beneath the thinned continental crust of passive margins was initially proposed by Boillot et al. (1987; 1989a). Off Galicia Bank, serpentinized upper mantle rocks outcrop on the ocean floor at the western edge of the ocean/continent transition, and form a basement ridge that can be followed almost continuously by sampling and by seismic reflection profiles for more that 130 km
(Boillot et al., 1989a,b). Based on the interpretation of seismic reflection profiles, Beslier et al. (1993) postulated the continuation of this ridge into the southern Iberia Abyssal Plain; the proposed location of the ridge coincided with the inferred landward edge of oceanic crust formed by seafloor spreading, based on geophysical criteria
(Whitmarsh et al., 1990;1993). Drilling results from Leg 149 have shown this interpretation to be correct (Sawyer,
Whitmarsh, Klaus et al., 1994). The existence of a serpentinized peridotite ridge in an analogous location in the Tagus Abyssal Plain, although possible, has not been yet demonstrated.
The width of the area of the crust less than 6 km thick underlain by high velocity material (7.4 km/s) that characterizes the ocean/continent transition off west Iberia may reach
110–150 km in the Tagus and southern Iberia Abyssal Plains but does not exceed more than a few tens of kilometers in the Galicia segment
(Fig. 10B). The nature of this thin crust underlain by serpentinized upper mantle is not clear
but its geophysical signature (in particular the magnetic character and the seismic-velocity profile) indicates that a major change in crustal structure occurs across its width.
The western portion of the thin crust underlain by high seismicvelocity material (i.e., that which is situated to the west of the serpentinized peridotite ridge in the south Iberia Abyssal Plain and off Galicia Bank) exhibits an acoustic basement topography and character typical of oceanic crust and its seismic crustal structure is remarkably similar to those that characterize oceanic fracture zones (very thin crust with seismic velocities in the crustal section compatible with oceanic layer 2 but absence of layer 3). In particular, the basement relief and associated magnetic anomalies are generally elongated and show a fairly well-defined
north-northeast–south-southwest trend, similar to those observed on the oceanic crust further to the west. Also, in the Iberia Abyssal Plain and the Tagus Abyssal Plain, the observed magnetic anomalies can be modeled by seafloor spreading, which clearly indicates that the underlying crust is thin oceanic crust. This thin oceanic crust was formed during the first stages of oceanic accretion. Although the basement to the west of the peridotite ridge has not been drilled on the west Iberia Margin, oceanic basalts were found covering the western side of the peridotite ridge on the north-western slope of Galicia Bank
(Malod et al., 1993).
On the contrary, the transitional crust east of the peridotite ridge shows a basement topography which is often smoother but more irregular than that observed in either typical oceanic or continental domains, and it exhibits seismic velocities in the lower crust that are different from those observed in oceanic layer 3, and which are more compatible with the seismic velocities that characterize the lower continental crust further to the east (Line 1, from Whitmarsh et al., 1990, and onshore). The associated magnetic anomalies exhibit a similar trend to that observed on the oceanic crust to the west of the peridotite ridge
(Whitmarsh and Miles, 1995) but their amplitude is generally much lower, although, locally, they may reach several hundred
nT. The basement blocks tend to be more equidimensional and exhibit more variable trends that range from north-south to north-west-southeast
(Whitmarsh et al., 1990; Pinheiro et al, 1992; Pinheiro, 1994; Whitmarsh and Miles, 1995). Such a crust is interpreted as transitional, and it most probably consists largely of extremely thinned and highly intruded continental crust (similar to the Jizan area, in the Red Sea; Voggenreiter et al., 1988), within which local areas of serpentinized upper mantle rocks may occur, given the extreme degree of thinning of the crustal section
(Pinheiro, 1994; Whitmarsh and Miles, 1995). Alternative interpretations are that significant portions of this crust represent oceanic crust formed in a very slow-spreading environment or that a significant portion of the supposedly very thin crust represents serpentinized upper mantle
(Boillot et al., 1992; Whitmarsh and Miles, 1995). As shown by the drilling results from
ODP Legs 103 and 149, the seismic velocities at the top of the serpentinized peridotite ridge can be very low
(3.5–4 km/s), but increase strongly with depth, passing through values within the range normally observed in the crustal section and reaching at depth the very high values intermediate between lower crust and upper mantle. In addition, although no significant magnetic anomalies are generally associated with the peridotite ridge, locally, some prominent anomalies are associated with serpentinite breccias, as shown by drilling at Site 899 (Sawyer,
Whitmarsh, Klaus et al., 1994; Whitmarsh et al., this volume). Therefore, it is not easy to distinguish a zone of serpentinized upper mantle from the adjacent transitional or thin oceanic crust based only on geophysical criteria.
Figure 10A shows the main features of the ocean/continent transition along the west Iberia Margin, as determined from an integration of all the available geophysical data, and
Figure 10B shows typical cross sections of the crustal and upper mantle structure across the ocean/continent transition in the three main segments of the margin
(Whitmarsh et al., 1990; Horsefield, 1992; Pinheiro et al., 1992; Pinheiro, 1994).
The main features of the ocean/continent transition show similar trends in all three segments, but, northwards along the margin, the location of the ocean/continent transition is progressively offset toward the west in each consecutive segment. Such offsets could indicate hiatuses in the northward propagation of the rifting, probably caused by the intersection of the propagating tip of the rift with a major preexisting lithospheric weakness, such as a late Variscan fault. One such offset is observed in the area of the Estremadura Spur, which is located just south of the Nazaré Fault. The other major offset occurs at the southern edge of Galicia Bank, and it may also be related to an old late Variscan fault zone.
Tagus Abyssal Plain
Mauffret et al. (1989a,b) postulated the existence of a fossil Late Jurassic to Early Cretaceous spreading center in the Tagus Abyssal Plain (magnetic anomalies M-21 to M-16), abandoned prior to a ridge jump towards the Grand Banks of Canada, sometime before magnetic anomaly M-10. More recently, however, Pinheiro et al. (1992) have shown, through the integration of all the available geophysical data and magnetic modeling of selected profiles, that such an hypothesis is unlikely. These later authors have shown that the magnetic anomalies predicted by the model of Mauffret et al. (1989a,b) did not fit the observed data. Furthermore, they showed that the oldest magnetic anomaly in this area that could be modeled with seafloor spreading was anomaly M11, which gives an age for the onset of seafloor spreading in this area of 133 Ma, with a half-spreading rate of 10 mm/ yr. Recently, these results have been refined by Whitmarsh and Miles (1995; see also
Whitmarsh et al., this volume), who modeled eight magnetic profiles across the ocean/continent transition in this area, using a new magnetic chart with a slightly better data coverage than the one used by Pinheiro et al. (1992). The new results basically confirm the Pinheiro et al. (1992) interpretation and show that the most landward magnetic anomaly that can be modeled with seafloor spreading is anomaly M-10NN, which gives a timing for the onset of seafloor spreading of 132 ± 1.9 Ma (late Valanginian/early Hauterivian, according to Gradstein et al., 1994), with a half-spreading rate of 10 mm/yr.
Southern Iberia Abyssal Plain
In the southern Iberia Abyssal Plain, recent results of modeling sea-surface magnetic profiles
(Whitmarsh et al., 1990; Whitmarsh and Miles, 1995) have shown that the most landward magnetic anomaly that could be modeled with seafloor spreading was anomaly M-8, which gives an age of ca. 129.5 Ma (middle
Hauterivian) for the onset of seafloor spreading in this area, with a half-spreading rate of 9 mm/ yr. However, more detailed modeling of one deep-towed magnetic profile across the same area
(Whitmarsh and Miles, 1995) suggested that the onset of seafloor spreading in this area took place approximately at 126 Ma (beginning of anomaly M3; early
Barremian) at a half-spreading rate of 10 mm/yr.
Galicia Bank
Off Galicia Bank, no magnetic modeling has been attempted, since the J-anomaly is absent in this area and therefore the oceanic crust was formed in the Cretaceous Magnetic Quiet Interval (Ogg, 1988). As such, neither the age of the onset of seafloor spreading nor the half-spreading rate are known for this segment of the margin. Drilling results have dated the interpreted breakup unconformity at the late Aptian, near the Aptian/Albian boundary (112 ± 1.1 Ma; Boillot et al., 1989b). This coincides with the boundary between UBSs 3 and 4 (see
Fig. 7). However, in the light of the
Wilson et al. (this volume) conclusion that rifting in this area is late
Berriasian–early Valanginian, the age of the so-called breakup unconformity may significantly postdate the commencement of seafloor spreading in this area.
