Oceanic basement roughness alongside magma-poor rifted margins: insight into initial seafloor spreading

Oceanic basement roughness alongside magma-poor rifted margins: insight into initial seafloor... Abstract The variation of oceanic basement roughness at mid-oceanic ridges is a complex trade-off between spreading rate that largely controls the thermal state of the lithosphere and its composition controlling the rheology and thus also the strength of the lithosphere. Here we estimate top basement roughness (i.e. the root-mean-square deviation of residual basement relief) over initial oceanic crust bordering the Iberia, Newfoundland, Bay of Biscay, Goban Spur, Flemish Cap, Australian and Antarctic rifted margins to provide new insights into the spreading processes at the nascent plate boundary. Although ultraslow seafloor spreading is suggested in those areas, the lack of undisputable oceanic magnetic anomalies prevents any well-constrained determination of the initial spreading rates. We compare these estimated roughness values with those determined over ultraslow-spreading crust formed at the Mid Atlantic Ridge, Southwest Indian Ridge, Arctic ridges, Mid-Cayman Spreading Center, Sheba ridge and South Pandora Ridge. The roughness values obtained at these ultraslow-spreading ridges range from 100 to >500 m and include 200–240 m roughness values which are typical of slow-spreading ridges. Roughness values larger than ∼300 m are characteristic of magma-poor sections of ultraslow-spreading ridges. The top basement roughness values determined within the inferred initial oceanic domain bordering the investigated magma-poor rifted margins are all higher than 200 m. Mean roughness values of the inferred initial oceanic domains alongside the conjugate Iberia and Newfoundland margins are greater than 300 m similarly to magma-poor sections of ultraslow-spreading ridges. The top basement in the initial oceanic crust alongside the conjugate Flemish Cap and Goban Spur margins shows roughness values and a tilted block morphology typical of slow-spreading ridges. We suggest that the roughness and the morphology of the top basement bordering the conjugate Australian and Antarctic margins indicate large tectonic extension and intermediate magma supply at either slow- or ultraslow-spreading rates. We show that estimating roughness values within transitional domains of magma-poor margins like exhumed mantle domains is less pertinent as polyphased tectonism and magmatism may have affected these domains leading to highly variable top basement roughness values that cannot be linked to a single process. Composition and structure of the oceanic crust, Atlantic Ocean, Indian Ocean, Continental margins: divergent, Mid-ocean ridge processes, Submarine tectonics and volcanism 1 INTRODUCTION Up to recently, hydrocarbon exploration has influenced our knowledge of the formation of rifted margins by collecting data mainly over sedimentary basins proximal to the continent. The acquisition of oceanward seismic data combined with deep sea drilling at magma-poor rifted margins, particularly along the conjugate Iberia and Newfoundland margins, revealed that the proximal and distal domains are very different from each other. The discovery of hyperextended continental crust and exhumed subcontinental mantle in these conjugated margins led to debated models for crustal thinning (e.g. Ranero & Perez-Gussinye 2010; Sutra & Manatschal 2012; McDermott & Reston 2015). However, acquisition of both seismic and potential field data still remains, with some exceptions, focused to the rifted margin and only few lines image the transition to unambiguous oceanic crust. As a consequence, the transition to ocean crust is still poorly constrained. Terms like initial or incipient (Hopper et al.2004), embryonic (Jagoutz et al.2007) or proto-oceanic crust (Gillard et al.2015) have been used to describe this transitional domain lacking clear geophysical characteristics of ocean crust like a well-marked reflector interpreted as the oceanic Moho on multichannel seismic (MCS) reflection data or typical velocity gradients on refraction data (Min 2009). Moreover, the lack of undisputable oceanic magnetic anomalies often did not allow the determination of initial spreading rates either because the onset of seafloor spreading fell within the Cretaceous Normal Superchron or because the ill-defined magnetic lineations may not be due to reversals of the Earth's magnetic field but to other processes (Bronner et al.2011). However, occurrence of crust thinner (<5 km) than typical oceanic crust was thought to indicate that this transitional crust was emplaced at an ultraslow-spreading ridge (Hopper et al.2004; van Avendonk et al.2006). Indeed, evidence for thin oceanic crust (<5 km) has been found at ultraslow-spreading ridges in the early 80s (Reid & Jackson 1981). This led to the suggestion that below a critical spreading rate (15–20 mm yr−1), the melt supply per increment of plate separation is dramatically decreasing (Reid & Jackson 1981). This is thought to result from a thickening of the thermal boundary layer at those ridges which in turn reduces the height of the melting regime (White et al.2001). In the last two decades exhumed mantle derived rocks have been found in wide areas of the ultraslow-spreading Southwest Indian Ridge (SWIR; Cannat et al.2006; Sauter et al.2013) and the Gakkel Ridge (Michael et al.2003) and seismic data have shown thin crust in magma-poor sections of these ridges (Muller et al.1999; Jokat et al.2003; Minshull et al.2006). However, very thick crust (8–10 km) has been recently discovered (Zhao et al.2013; Li et al.2015; Niu et al.2015) at the shallowest part of a >550 km long thick crust ridge section of the SWIR (Sauter et al.2009). Such along-axis large crustal thickness variations, while the ultraslow-spreading rate varies only slightly, are therefore challenging the current view of thin crust being associated with ultraslow spreading (Michael et al.2003; Cannat et al.2008). It is commonly inferred that seafloor roughness (i.e. the root mean square deviation of residual basement relief) increases dramatically for full spreading rates of <40 mm yr−1 (Malinverno 1991; Ehlers & Jokat 2009). Top basement roughness could thus be used to identify ultraslow-spreading lithosphere. We therefore suggest that investigating the top basement roughness of initial oceanic domains bordering magma-poor rifted margins may provide new insights in the spreading processes triggered by lithospheric breakup. In this paper, we first review seafloor roughness estimates at present day ultraslow-spreading ridges based on published data from the SWIR and from the Arctic ridges and on new roughness estimations at the Mid-Cayman Spreading Center (MCSC), the Sheba ridge and at the South Pandora Ridge (SPR) in the North Fiji Basin (Fig. 1). We also investigate the relief on flanks of the Mid Atlantic Ridge (MAR), which were created when this ridge underwent ultraslow spreading. Results from measurements of the top basement roughness along MCS profiles on inferred initial oceanic domains bordering the Iberia, Newfoundland, Bay of Biscay (BB), Goban Spur (GS), Flemish Cap, Australian and Antarctic margins are then presented in a second part (Fig. 1). These top basement roughness values are compared to those of ultraslow-spreading ridges. We discuss how estimating the roughness of the top basement may help to constrain the spreading processes at the initial oceanic ridge. We found that roughness values of the inferred initial oceanic domains alongside the conjugate Iberia and Newfoundland margins are similar to those of magma-poor sections on ultraslow-spreading ridges. In contrast, the top basement within the initial oceanic domain alongside the conjugate Flemish Cap and GS margins shows roughness values typical of slow-spreading ridges. We also show that the roughness and morphology of the top basement alongside the conjugate Australian and Antarctic margins indicate large tectonic extension and intermediate magma supply at either slow or ultraslow-spreading rates. Figure 1. View largeDownload slide Worldwide distribution of the magma-poor rifted margin systems and locations of the rifted margins and mid-oceanic ridges discussed in this paper. MAR: Mid-Atlantic Ridge; MCSC: Mid Cayman Spreading Center; SPR: South Pandora Ridge; SWIR: Southwest Indian Ridge. After Haupert et al. (2016). Background image is from Amante & Eakins (2009). Figure 1. View largeDownload slide Worldwide distribution of the magma-poor rifted margin systems and locations of the rifted margins and mid-oceanic ridges discussed in this paper. MAR: Mid-Atlantic Ridge; MCSC: Mid Cayman Spreading Center; SPR: South Pandora Ridge; SWIR: Southwest Indian Ridge. After Haupert et al. (2016). Background image is from Amante & Eakins (2009). 2 DATA AND METHOD The stochastic modelling procedure (Goff & Jordan 1988, 1989) offers a robust way to analyse the statistical properties of abyssal hill morphology. This method was applied on grids of bathymetric data that were collected using multibeam echo sounders on the flanks of the MAR (Goff 1991; Goff et al.1995; Neumann & Forsyth 1995) and the SWIR (Sauter et al.2011; Sloan et al.2012). Close to the ridge axis and far from continents, where sediments are lacking, such bathymetric grids allow a 2-D analysis of the top of the oceanic crust. On sedimented rifted margins or spreading ridges, MCS profiles are needed to investigate the top basement relief, but are spaced too far apart and give only access to 1-D analysis of the top basement. Nevertheless, calculation of the roughness of the top basement is derived the same way from 1-D or 2-D data. It is defined as the root-mean-square (RMS) height, which is the average variation of depths with respect to the regional depth trend. RMS height is therefore calculated after removing the regional trend from the data. We used a second order empirical trend of the basement instead of a theoretical subsidence curve to avoid dependence upon on a crustal age model (Ehlers & Jokat 2009). When available we use MCS data published as depth sections (Hopper et al.2004; Bullock & Minshull 2005; Shillington et al.2006; Gerlings et al.2012; Dean et al.2015; Tugend et al.2015). However, MCS data are commonly shown as two-way traveltime (TWTT) sections (e.g. Afilhado et al.2008; Welford et al.2010b, Pickup et al.1996). The seafloor and top basement reflections were thus picked to obtain the sediment thickness in second TWTT which was then depth converted and added to the seafloor bathymetry to obtain the top basement relief in meters. Sediment velocity information was provided by a compilation of wide angle and refraction seismic sections within each area of interest. Along the Iberia margin we use the TWTT-velocity relationships of (Cunha 2008) to convert MCS data in the Iberia Abyssal Plain (IAP; Pickup et al.1996) and in the Tagus Abyssal Plain (TAP; Afilhado et al.2008). On the Newfoundland margin, we built up on the work of (Shillington et al.2006; van Avendonk et al.2006) and (Hopper et al.2004) to determine the TWTT-velocity functions for the SCREECH 2 and SCREECH 1 areas that were then used to derive the depth of the top basement for the MCS data falling within these areas (Welford et al.2010b). No refraction profile has been collected on the conjugate Australian-Antarctic margins. Therefore we used compilations of velocity profiles from sonobuoys (Ball 2005; Gillard 2014) to derive TWTT-velocity functions and convert MCS data interpreted by Gillard et al. (2015, 2016a,b). 3 SEAFLOOR ROUGHNESS AT ULTRASLOW-SPREADING RIDGES We review hereafter the top basement roughness estimates at the SWIR, Gakkel, Knipovich and Sheba ridges, the MCSC, the SPR and the MAR (Table 1). We focus first on the SWIR as there, roughness values are well documented on various tectonic settings (Sauter et al.2011; Sloan et al.2012). The presence of exhumed mantle-derived rocks on >100 km long stretches of the ridge axis is one of the striking contrasts between most of the ultraslow-spreading ridges and faster spreading ridges (Dick et al.2003; Sauter et al.2004, 2013; Cannat et al.2006). However, exhumed mantle is not ubiquitous at ultraslow-spreading ridges and thick magmatic oceanic crust is also found there (Niu et al.2015). Although some fossil oceanic ridges have undergone slow to ultraslow rates while spreading was failing (e.g. in the Labrador or Somali basin) we disregarded them as a post spreading tectonic phase may have affected the top basement (Delescluse et al.2015; Sauter et al.2016). Table 1. Estimations of the top basement roughness on ultraslow-spreading mid-oceanic ridges.   Full Rate (mm yr−1)  RMS height (m)  Size and type of data set  Number of samples  References  Southwest Indian Ridge (SWIR)  15    ∼100 km long bathymetric grids    (Sauter et al.2011; Sloan et al.2012)  54–67°E (volcanic domains)  15  298 ± 23  id.  27  id.  54–61°E (volcanic domains)  15  242 ± 36  id.  12  id.  61–67°E (volcanic domains)  15  372 ± 23  id.  9  id.  61–67°E (exhumed mantle domains)  15  304 ± 20  id.  6  id.  Arctic ridges  9–15  517 ± 50  >100 km long MCS profiles  9  (Weigelt & Jokat 2001; Ehlers & Jokat 2009)  Mid-Cayman Spreading Center (MCSC)  17  371 ± 79a  >120 km long bathymetric profiles  13  GeoMapApp cruises  Western flank  17  432 ± 72a  id.  6  id.  Eastern flank  17  319 ± 34a  id.  7  id.  Sheba ridge 52–55°E  20  164 ± 59a  >140 km long MCS profiles  3  (d’Acremont et al.2010)  Western basin  20  103a  150 km long MCS profile  1  id.  Eastern basin  20  195 ± 50a  >140 km long MCS profiles  2  id.  Mid Atlantic Ridge (MAR)  18    >100 km long MCS profiles  3    Western North America basin  18  257a  >130 km long MCS profile  1  (Lizarralde et al.2004)  Western North America Basin  18  160  >100 km long MCS profile  1  (Min 1999 1999)  Eastern (Canary) basin  18  350  id.  1  id.  South Pandora Ridge (SPR)  18  158 ± 16a  >80 km long bathymetric grids  4  (Lagabrielle et al.1996)    Full Rate (mm yr−1)  RMS height (m)  Size and type of data set  Number of samples  References  Southwest Indian Ridge (SWIR)  15    ∼100 km long bathymetric grids    (Sauter et al.2011; Sloan et al.2012)  54–67°E (volcanic domains)  15  298 ± 23  id.  27  id.  54–61°E (volcanic domains)  15  242 ± 36  id.  12  id.  61–67°E (volcanic domains)  15  372 ± 23  id.  9  id.  61–67°E (exhumed mantle domains)  15  304 ± 20  id.  6  id.  Arctic ridges  9–15  517 ± 50  >100 km long MCS profiles  9  (Weigelt & Jokat 2001; Ehlers & Jokat 2009)  Mid-Cayman Spreading Center (MCSC)  17  371 ± 79a  >120 km long bathymetric profiles  13  GeoMapApp cruises  Western flank  17  432 ± 72a  id.  6  id.  Eastern flank  17  319 ± 34a  id.  7  id.  Sheba ridge 52–55°E  20  164 ± 59a  >140 km long MCS profiles  3  (d’Acremont et al.2010)  Western basin  20  103a  150 km long MCS profile  1  id.  Eastern basin  20  195 ± 50a  >140 km long MCS profiles  2  id.  Mid Atlantic Ridge (MAR)  18    >100 km long MCS profiles  3    Western North America basin  18  257a  >130 km long MCS profile  1  (Lizarralde et al.2004)  Western North America Basin  18  160  >100 km long MCS profile  1  (Min 1999 1999)  Eastern (Canary) basin  18  350  id.  1  id.  South Pandora Ridge (SPR)  18  158 ± 16a  >80 km long bathymetric grids  4  (Lagabrielle et al.1996)  aDetermined in this paper. Values with standard deviations are mean values. View Large 3.1 The Southwest Indian ridge (SWIR) The SWIR full spreading rate has been ultraslow since a dramatic decrease ∼24 myr ago, from a slow rate of 30 to 15 mm yr−1 (Patriat et al.2008). The RMS abyssal hill height mean value (298 m; Table 1), determined within polygonal areas on the flanks of the eastern part of the ultraslow-spreading SWIR, is ∼80 m greater than for slow-spreading seafloor. However, seafloor roughness also increases eastward along the SWIR axis (Sauter et al.2011). The roughness mean value (372 m) is much larger between 61 and 67°E, in the easternmost deep part of the ridge, than between 54 and 61°E (242 m; Table 1). Fig. 2 displays some bathymetric profiles to illustrate this eastward increasing roughness of the seafloor. The Mantle Bouguer Anomaly (MBA; a crustal thickness and/or mantle temperature proxy deduced from gravity data) also changes eastward, suggesting thinner crust and/or cooler mantle temperatures in the easternmost part of the SWIR (Sauter et al.2011). There, large seafloor domains of exhumed mantle derived rocks have been called ‘smooth seafloor’ because they occur in the form of broad ridges with a smooth, rounded topography with no resolvable volcanic cones or typical abyssal hills on bathymetric data (Cannat et al.2006). The mean value of RMS height for these exhumed mantle domains (304 m) is ∼70 m smaller than for the volcanic seafloor (372 m; Sloan et al.2012; Table 1 and Fig. 2). Figure 2. View largeDownload slide Across axis profiles at the Southwest Indian Ridge and the Gakkel ridge illustrating the range of roughness values (R) found at ultraslow-spreading centres. (a) Bathymetric profile crossing the SWIR ridge axis (at 56°E) showing a roughness value <200 m (calculated outside the axial valley; see Sauter et al.2011) and typical abyssal hills usually observed at slow-spreading centres; (b) bathymetric profile (at 62°E) in the easternmost magma-poor section of the ridge within a domain of exhumed mantle derived rocks (smooth seafloor); (c) bathymetric profile crossing the SWIR axis (at 64°E) within a volcanic part of the easternmost magma-poor section of the ridge; (d) top basement picked in the seismic reflection profile AWI-20 010 100 on the eastern flank of the Gakkel ridge. Fault locations are inferred from the bathymetric grids. Their depth and dip are unconstraint. Vertical exaggeration is 2×. Profiles across the SWIR are from Sauter et al. (2009) and Sauter et al. (2013). The roughness values estimated along the SWIR bathymetric profiles are slightly different from the roughness values given in Table 1 which are mean values calculated on several polygonal areas. The roughness estimation and the profile on the flank of the Gakkel Ridge are from (Ehlers & Jokat 2009). Figure 2. View largeDownload slide Across axis profiles at the Southwest Indian Ridge and the Gakkel ridge illustrating the range of roughness values (R) found at ultraslow-spreading centres. (a) Bathymetric profile crossing the SWIR ridge axis (at 56°E) showing a roughness value <200 m (calculated outside the axial valley; see Sauter et al.2011) and typical abyssal hills usually observed at slow-spreading centres; (b) bathymetric profile (at 62°E) in the easternmost magma-poor section of the ridge within a domain of exhumed mantle derived rocks (smooth seafloor); (c) bathymetric profile crossing the SWIR axis (at 64°E) within a volcanic part of the easternmost magma-poor section of the ridge; (d) top basement picked in the seismic reflection profile AWI-20 010 100 on the eastern flank of the Gakkel ridge. Fault locations are inferred from the bathymetric grids. Their depth and dip are unconstraint. Vertical exaggeration is 2×. Profiles across the SWIR are from Sauter et al. (2009) and Sauter et al. (2013). The roughness values estimated along the SWIR bathymetric profiles are slightly different from the roughness values given in Table 1 which are mean values calculated on several polygonal areas. The roughness estimation and the profile on the flank of the Gakkel Ridge are from (Ehlers & Jokat 2009). Seafloor roughness could not be estimated in the shallowest part of the SWIR (35–52°E) where the magmatic crust is the thickest (Zhao et al.2013; Niu et al.2015). There, a sudden increase of the magma supply ∼10 Ma ago is thought to have resulted in a shallow volcanic plateau at the ridge axis (Sauter et al.2009). This plateau is too narrow to estimate valid roughness values. However, we note that large topographic features (up to 60 km long and ∼900 m-high) are observed on top of this volcanic plateau as well as close to the Marion hot spot (Mendel et al.2003) suggesting still thick lithosphere in the shallowest parts of the SWIR. 3.2 The arctic ridges High roughness values (>500 m) were obtained by Weigelt & Jokat (2001) and Ehlers & Jokat (2009) along >100 km long seismic reflection profiles in the Arctic Ocean. We retain nine seismic profiles in the Amundsen, Nansen and Boreas basins on the flanks of the ultraslow-spreading Gakkel and Knipovich ridges (9–15 mm yr−1 spreading rates) and disregard one profile in the Molloy basin as it is located too close to the Spitsbergen Fracture Zone. The roughness values were estimated from profiles of seismically determined top basement by (Weigelt & Jokat 2001; Ehlers & Jokat 2009). They may not be straightforwardly compared to the statistical analysis of bathymetric grids at the SWIR. However, the mean roughness value on the ultraslow-spreading Arctic ridges is significantly larger (517 m; Table 1) than those estimated on the SWIR (Fig. 2). No correlation has been found between these values and the magmatic and amagmatic segmentation of the Arctic ridges or the highly variable thickness of the crust (2–6 km; Ehlers & Jokat 2009). 3.3 The Mid‐Cayman Spreading Center (MCSC) The deep-seated MCSC is located midway along a depression, called Cayman trough, that extends from the Belize margin to the northern edge of Jamaica. Seafloor magnetic anomalies within the Cayman trough record a ∼17 mm yr−1 ultraslow-spreading rate for the last 20 myr (Leroy et al.2000). Series of horsts and grabens are well marked about 150 km on either side of the N–S oriented spreading centre but, further away, the seafloor topography becomes smoother as the sedimentary cover thickens towards the Belize margin to the west and offshore Cuba Island to the east (Leroy et al.2000). We therefore use 13 ∼125 km long-bathymetric profiles from the WI93300 and Ka93200 cruises (from USNS Wilkes and Kane in 1973 and 1972 respectively) on each flank of the MCSC excluding a 60 km wide central part across the ridge axis. The bathymetry of the off-axis Cayman trough is highly asymmetric (Hayman et al.2011). The estimated roughness to the west of the MCSC is larger (432 m mean value) than to the east (319 m mean value; Table 1). The crustal thickness is thought to be also asymmetric, with thinner upper crust to the east than to the west (Leroy et al.1996; Hayman et al.2011). However, we choose to disregard the crustal thickness estimations on the MCSC as they are poorly constrained and debated (ten Brink et al.2002). 3.4 The Sheba ridge The 20–22 mm yr−1 spreading rate of the Sheba ridge is equal to the threshold (Fournier et al.2001; Leroy et al.2004) below which the ultraslow-spreading class of oceanic ridges is usually defined (Reid & Jackson 1981). We use the top basement relief determined in two seismic profiles from the Arabian margin to the Somalian margin between the Alula-Fartak Fracture Zone and the Socotra-Hadbeen Fracture Zone (d’Acremont et al.2010). We avoid the axial valley relief and exclude thus a central 100 km wide part to estimate the roughness of the oceanic crust. We also disregard the anomalously shallow area on the southern flank of the Sheba ridge where recent off-axis volcanism has been shown by (Leroy et al.2010a). The estimated mean roughness value is larger in the eastern basin (195 m mean value; Table 1) than in the western one (103 m), while the crust is thinner in the eastern than in the western basin (4 and 6 km average crustal thickness, respectively; Leroy et al.2010b, d’Acremont et al.2010; Watremez et al.2011). 3.5 The South Pandora ridge (SPR) The North Fiji Basin is a back-arc basin opening with a complex geometry of spreading centres trending both ∼NS and ∼EW (Garel et al.2003). We used a set of multibeam bathymetric data collected along the 500 km long SPR, which is opening with an ultraslow-spreading rate (16 mm yr−1; Lagabrielle et al.1996). We obtained a low RMS roughness mean value (158 m; Table 1). It is worth pointing out that the SPR strongly differs from other ultraslow-spreading ridges. Mantle rocks have never been recovered along any segment of the ridges of the North Fiji Basin (Garel et al.2003) contrasting with the frequent recovery of serpentinized peridotites at segment ends along the ultraslow and slow oceanic ridges (e.g. Cannat 1993). Moreover, the magma production is high, consistent with the general shallow depth of the entire back arc basin and the analysis of gravity anomalies indicating crustal thickness as large as 8 km along the SPR (Garel et al.2003). 3.6 The Mid Atlantic ridge (MAR) The spreading rate in the North Atlantic decreased by a factor of about two between M-21 (148 Ma) and M-11 (135 Ma; Sundvik et al.1984). A rough acoustic basement is associated with this period of ultraslow-spreading (∼18 mm yr−1) in the Central Atlantic west of Bermuda and in the conjugated Canary Basin (Sundvik et al.1984; Ranero & Banda 1997). Min (1999) estimated that the RMS roughness mean values only increased from 130 to 160 m in the Western (Central Atlantic) Basin while it reached ∼350 m in the eastern (Canary) Basin. Using the MCS data of Lizarralde et al. (2004) collected in the Western Basin we obtained a higher roughness value of 257 m for the ultraslow-spreading basement (Table 1). Refraction data coincident with the MCS in the western North Atlantic basin show that the crustal thickness decreased from 6.7 to 5.5 km when spreading became ultraslow (Lizarralde et al.2004). Finally, as a conclusion of this section, we note that the roughness estimations obtained at these ultraslow-spreading ridges cover a wide range of values, from ∼100 m on the Sheba Ridge to ∼580 m in the Arctic basin (Weigelt & Jokat 2001). 4 TOP BASEMENT ROUGHNESS ALONGSIDE MAGMA-POOR MARGINS Estimating the roughness of initial oceanic crust alongside rifted margins requires to define the location of the Continent Ocean Boundary (COB). Typically, ‘normal’ oceanic crust is characterized in MCS sections by a: (1) smooth top basement at a kilometric scale, commonly observed at 8 s TWTT, (2) highly reflective upper crust and an acoustically relatively transparent lower-crust, (3) ∼2 s TWTT thickness, (4) well-marked reflector at the base of the crust interpreted as the Moho, commonly observed at 10 s TWTT and (5) passive sediment infill (e.g. Klimke et al.2016). However, seismic reflection profiles beyond rifted margins often do not sample such unambiguous oceanic crust. A well-marked reflector that could be interpreted as the Moho is rarely observed, while the seismic texture within the basement is often confusing. Therefore, initial oceanic crust is usually inferred from a conjunction of arguments from seismic (refraction and/or reflection) as well as potential field data, none of these being conclusive on its own. This is the case at the Iberia–Newfoundland, GS–Flemish Cap, BB, and Australo–Antarctic margin systems that we analyse hereafter (Table 2). Although we use the location of the initial oceanic domains proposed in the literature, we were careful to verify some basic characteristics. As the COB can be associated with an important top basement topography (Unternehr et al.2010), thought to result from the transition from a deep and denser exhumed mantle domain to a shallower less dense oceanic domain (Peron-Pinvidic et al.2013), we checked that the location of such topographic features agrees with the inferred location of the COB. Moreover, domains with volcanic overprint and/or a deformed sediment cover were not assumed to be oceanic in nature as no deformation related to spreading is expected on the flanks of mid-oceanic ridges outside a narrow deformation zone close to the axis (Cannat et al.2009). Table 2. Estimations of the top basement roughness within the initial oceanic domain alongside magma-poor rifted margins. OC, Oceanic crust; EM, exhumed mantle; POC, proto-oceanic crust (see definition in main text).   Inferred crust type  RMS height (m)  Profile length (km)  Number of profiles  References  Tagus Abyssal Plain (TAP)        1    Shallow subdomain on IAM5 line  OC  456a  60  1  (Afilhado et al.2008)  Deep subdomain on IAM5 line  OC  414a  105  1  id.  Both domains on IAM5 line  OC  520a  165  1  id.  Iberia Abyssal Plain (IAP)        3    West of R4 on IAM9 line  OC  429a  80  1  (Pickup et al.1996)  East of R4 on IAM9 line  EM  368a  115  1  id.  West of R3 on Lusigal12 line  OC  566a  120  1  (Beslier 1996)  West of R3 on SO-22 line  OC  534a  160  1  (Pickup et al.1996)  Mean Value for OC  OC  510 ± 72a    3  id.  Galicia Margin (GM)            West of Peridotite Ridge on line WE1  OC  583a  85  1  (Dean et al.2015)  Northern Newfoundland Basin (NNFB)        5    East of M0 on SCREECH 2  OC  449a  75  1  (Shillington et al.2006)  West of M0 on SCREECH 2  EM  410a  120  1  id.  West of M3 on SCREECH 2  EM  268a  75  1    East of M0 on SCREECH 107/201  OC  355a  80  1  id.  East of M0 on SCREECH 302/109  OC  374a  84  1  id.  East of M0 on SCREECH E56  OC  320a  75  1  (Welford et al.2010b)  West of M0, on SCREECH E56  EM  378a  96  1    East of M0 on SCREECH E54  OC  320a  144  1  id.  Mean value for OC  OC  364 ± 53a    5  id.  Mean value for EM  EM  394 ± 23a    3  id.  Southern Margin of Flemish Cap (SMFC)        1    On line E36  OC  380a  93  1  (Welford et al.2010b)  Northern Margin of Flemish Cap (NMFC)        2    On line F87–4  OC  215a  85  1  (Welford et al.2010a)  On line F87–3  OC  322a  87  1  id.  Goban Spur Margin (GS)        1    On WAM line  OC  216  50  1  (Bullock & Minshull 2005)  On WAM  EM  128  67  1  id.  Bay of Biscay (BB)        3    On Norgasis 11–12  OC  315a  65  1  (Thinon et al.2003)  On Norgasis 14  OC  275a  64  1  id.  On IAM12  OC  328a  95  1  (Tugend et al.2014)  Great Australian Bight (GAB)        8    On GA199–10  POC  231a  56  1  (Gillard et al.2015)  On GA199–10  OC  258a  102  1  id.  On GA199–07  POC  251a  61  1  id.  On GA199–06  OC  243a  96  1  id.  On GA199–06  POC  362a  61  1  id.  On GA199–05  POC  262a  82  1  id.  On GA199–04  POC  216a  82  1  id.  On GA199–03  POC  374a  99  1  id.  On GA199–02  POC  356a  95  1  id.  On GA199–01  POC  313a  106  1  id.  On GA199–01  OC  296a  117  1  id.  Mean value for POC  POC  296 ± 63a    8  id.  Mean value for OC  OC  266 ± 27a    3  id.  Wilkes Land Basin (WLB)        6    On GA228–27  POC  333a  143  1  id.  On GA228–26  POC  336a  163  1  id.  On GA228–26  OC  208a  57  1  id.  On GA228–25  POC  270a  60  1  id.  On GA228–24  POC  298a  129  1  id.  On GA228–23  POC  256a  154  1  id.  On GA228–22  POC  313a  83  1  id.  On GA228–22  OC  269a  50  1  id.  Mean value for POC  POC  301 ± 33a    6  id.  Mean value for OC  OC  239 ± 43a    2  id.    Inferred crust type  RMS height (m)  Profile length (km)  Number of profiles  References  Tagus Abyssal Plain (TAP)        1    Shallow subdomain on IAM5 line  OC  456a  60  1  (Afilhado et al.2008)  Deep subdomain on IAM5 line  OC  414a  105  1  id.  Both domains on IAM5 line  OC  520a  165  1  id.  Iberia Abyssal Plain (IAP)        3    West of R4 on IAM9 line  OC  429a  80  1  (Pickup et al.1996)  East of R4 on IAM9 line  EM  368a  115  1  id.  West of R3 on Lusigal12 line  OC  566a  120  1  (Beslier 1996)  West of R3 on SO-22 line  OC  534a  160  1  (Pickup et al.1996)  Mean Value for OC  OC  510 ± 72a    3  id.  Galicia Margin (GM)            West of Peridotite Ridge on line WE1  OC  583a  85  1  (Dean et al.2015)  Northern Newfoundland Basin (NNFB)        5    East of M0 on SCREECH 2  OC  449a  75  1  (Shillington et al.2006)  West of M0 on SCREECH 2  EM  410a  120  1  id.  West of M3 on SCREECH 2  EM  268a  75  1    East of M0 on SCREECH 107/201  OC  355a  80  1  id.  East of M0 on SCREECH 302/109  OC  374a  84  1  id.  East of M0 on SCREECH E56  OC  320a  75  1  (Welford et al.2010b)  West of M0, on SCREECH E56  EM  378a  96  1    East of M0 on SCREECH E54  OC  320a  144  1  id.  Mean value for OC  OC  364 ± 53a    5  id.  Mean value for EM  EM  394 ± 23a    3  id.  Southern Margin of Flemish Cap (SMFC)        1    On line E36  OC  380a  93  1  (Welford et al.2010b)  Northern Margin of Flemish Cap (NMFC)        2    On line F87–4  OC  215a  85  1  (Welford et al.2010a)  On line F87–3  OC  322a  87  1  id.  Goban Spur Margin (GS)        1    On WAM line  OC  216  50  1  (Bullock & Minshull 2005)  On WAM  EM  128  67  1  id.  Bay of Biscay (BB)        3    On Norgasis 11–12  OC  315a  65  1  (Thinon et al.2003)  On Norgasis 14  OC  275a  64  1  id.  On IAM12  OC  328a  95  1  (Tugend et al.2014)  Great Australian Bight (GAB)        8    On GA199–10  POC  231a  56  1  (Gillard et al.2015)  On GA199–10  OC  258a  102  1  id.  On GA199–07  POC  251a  61  1  id.  On GA199–06  OC  243a  96  1  id.  On GA199–06  POC  362a  61  1  id.  On GA199–05  POC  262a  82  1  id.  On GA199–04  POC  216a  82  1  id.  On GA199–03  POC  374a  99  1  id.  On GA199–02  POC  356a  95  1  id.  On GA199–01  POC  313a  106  1  id.  On GA199–01  OC  296a  117  1  id.  Mean value for POC  POC  296 ± 63a    8  id.  Mean value for OC  OC  266 ± 27a    3  id.  Wilkes Land Basin (WLB)        6    On GA228–27  POC  333a  143  1  id.  On GA228–26  POC  336a  163  1  id.  On GA228–26  OC  208a  57  1  id.  On GA228–25  POC  270a  60  1  id.  On GA228–24  POC  298a  129  1  id.  On GA228–23  POC  256a  154  1  id.  On GA228–22  POC  313a  83  1  id.  On GA228–22  OC  269a  50  1  id.  Mean value for POC  POC  301 ± 33a    6  id.  Mean value for OC  OC  239 ± 43a    2  id.  aDetermined in this paper. View Large 4.1 The Iberia margin Because the J magnetic anomaly is thought to be located at the boundary between exhumed mantle and initial oceanic crust in the IAP, it has been proposed to mark the limit of the oceanic domain in the whole Iberia–Newfoundland system (Bronner et al.2011; Stanton et al.2016). However, (Nirrengarten et al.2017) recently demonstrated that the J anomaly is the result of polygenic and multiple magmatic events occurring during and after the formation of initial oceanic crust. We therefore built on the COB proposed by Nirrengarten et al. (2017) that uses a combination of geophysical and geological arguments. 4.1.1 The Tagus Abyssal Plain (TAP) In the TAP a two-layer igneous oceanic crust was recognized at the IAM5 MCS line and the wide-angle data by (Afilhado et al.2008). A sharp step of the top basement marks the limit between a westernmost shallow subdomain and a deeper easternmost subdomain (see Plate 1 of Afilhado et al.2008). MCS reflections were inferred between 10 and 11 s TWT, consistent with the Moho depth obtained from velocity models (Afilhado et al.2008). The inferred crustal thickness of the initial oceanic crust increased westward from 5 km on average in the deep subdomain to 7 km on average in the shallow one (Afilhado et al.2008). Therefore, (Nirrengarten et al.2017) took the basement step between the two subdomains as indicating the transition to an unambiguous magmatic oceanic crust. Excluding the western termination of IAM5 line at the volcanic Tore Madeira rise, we estimated the roughness values in the shallow and the deep subdomains to 456 and 414 m, respectively (Table 2). Analysing the shallow and deep subdomains together leads logically to a higher roughness value of 520 m. 4.1.2 The Iberia Abyssal Plain (IAP) Three MCS lines, IAM9, SO-22 (Pickup et al.1996) and Lusigal12 (Beslier et al.1996), reach the oceanic domain in the IAP. Constrains are given by refraction data and several drill holes drilled during ODP Legs 149 and 173 (Dean et al.2000; Whitmarsh et al.2001; Minshull et al.2014). The refraction data show that the inferred initial oceanic crust is 6–7 km thick (Minshull et al.2014). The transition between the initial oceanic crust to the west and the exhumed mantle domain to the east is marked by a series of peridotite ridges on MCS lines (Pickup et al.1996). The peridotite ridges called R3 and R4 by (Pickup et al.1996) mark the beginning of the westward shallowing of the top basement on the MCS lines. The roughness mean value for the initial oceanic crust to the west of these ridges is estimated at 510 m (Table 2). Taking anomaly J instead of the peridotite ridge R3 as the COB on the SO-22 profile results in a similar roughness value for initial oceanic crust (520 m versus 534 m using R3) although the sub-section is 50 km shorter. Because the exhumed mantle domains on the Lusigal12 and SO-22 lines were too short, we estimate the roughness over the exhumed mantle domain (at 368 m) only using the IAM9 profile. 4.1.3 The deep Galicia margin As in the IAP the transition between the initial oceanic crust and the exhumed mantle domain is marked by a series of peridotite ridges. On MCS line WE1 (Dean et al.2015) suggested a gradual transition to seafloor spreading westward of the ‘Peridotite Ridge’ (Fig. 3) where the top basement begins to shallow. There, recently published refraction data suggest a very thin oceanic crust thickening seaward (∼0.5–1.5 km; Davy et al.2016). Oceanic layer 3 is absent and the crust is thought to overlie serpentinized mantle material (Davy et al.2016). The roughness value estimated to the west of the Peridotite Ridge is 583 m (Fig. 3). Figure 3. View largeDownload slide Top basement relief in initial oceanic crust alongside magma-poor rifted margins. Roughness (R) is estimated for oceanic crust (OC) and proto-oceanic crust (POC) along MCS profiles of the (a) Galicia margin (profile WE1 from Dean et al.2015); (b) Newfoundland margin (profile Erable54 from Welford et al.2010b); (c) Northern margin of Flemish Cap (profile F87–3 from Welford et al.2010a); (d) Goban Spur margin (profile WAM from Bullock & Minshull 2005); (e) Cantabrian margin in the Bay of Biscay (profile IAM12 from Tugend et al. (2014); (f) Great Australian Bight margin (profile GA199–06 from Gillard et al.2015); (g) Wilkes Land margin (profile GA228–22 from Gillard et al.2015). Vertical exaggeration is 2×. Figure 3. View largeDownload slide Top basement relief in initial oceanic crust alongside magma-poor rifted margins. Roughness (R) is estimated for oceanic crust (OC) and proto-oceanic crust (POC) along MCS profiles of the (a) Galicia margin (profile WE1 from Dean et al.2015); (b) Newfoundland margin (profile Erable54 from Welford et al.2010b); (c) Northern margin of Flemish Cap (profile F87–3 from Welford et al.2010a); (d) Goban Spur margin (profile WAM from Bullock & Minshull 2005); (e) Cantabrian margin in the Bay of Biscay (profile IAM12 from Tugend et al. (2014); (f) Great Australian Bight margin (profile GA199–06 from Gillard et al.2015); (g) Wilkes Land margin (profile GA228–22 from Gillard et al.2015). Vertical exaggeration is 2×. 4.2 The Newfoundland margin We focus here on the Northern Newfoundland Basin (NNFB) and on the Southern margin of Flemish Cap (SMFC). Because oceanic crust is overprinted by post breakup magmatism in the Newfoundland Seamounts area, we disregard the Southern Newfoundland Basin. 4.2.1 The Northern Newfoundland Basin (NNFB) Refraction data in the NNFB show that the landward limit of a 3 km thick oceanic crust falls close to the eastern end of the J anomaly (interpreted as M0; van Avendonk et al.2006; Welford et al.2010b). Layer 3 oceanic crust is interpreted to be missing with layer 2 directly overlying serpentinized mantle (van Avendonk et al.2006). Normal oceanic crust of 6 km thickness is thought to be found at the seaward ends of the MCS lines of the SCREECH project, but a continuous Moho reflection is not observed. As along the Iberia margin, the transition between the oceanic crust and the exhumed mantle domain is marked by a series of peridotite ridges on the MCS lines. There, at ODP Site 1277, drilling into basement revealed gabbro and basalt fragments as well as serpentinized peridotites of subcontinental origin (Tucholke & Sibuet 2007). We use three MCS lines of the SCREECH project (SCREECH 2, 107/201 and 302/109; Shillington et al.2006) and E56 and E54 lines of the Erable project in the NNFB (Welford et al.2010b; Fig. 3). The mean roughness value for the oceanic crust east of M0 is estimated at 364 m while the mean roughness value for the exhumed mantle domain west of M0 is estimated at 394 m (Table 2). Moving the COB to the western end of the J anomaly (interpreted as M3) as suggested by (Shillington et al.2006) on SCREECH 2 profile results in a similar estimation for the roughness of the oceanic crust than using the eastern end of the J anomaly (458 m and 449 m, respectively), although the sub-section is 50 km shorter. Excluding the outer high drilled by the IODP Site 1277 that has been formed after the exhumation stage (Bronner et al.2011) leads to much lower roughness values for the exhumed mantle domain (268 m on SCREECH 2). 4.2.2 The Southern Margin of Flemish Cap (SMFC) At the SMFC, (Hopper et al.2004) interpreted the last ∼80 km at the seaward end of the MCS line SCREECH 1 as oceanic crust whereas a recent reinterpretation suggested a much narrower oceanic domain (<15 km; Gillard et al.2016b). Both interpretations are compatible with the seismic refraction data along this line (Funck et al.2003). Instead, we use here the Erable E36 MCS line that enters the initial oceanic domain over >90 km to the east of the J anomaly according to (Welford et al.2010b). Roughness is estimated at 380 m over this domain (Table 2). Distinguishing anomalously thin oceanic crust (with 3 km thickness) and normal oceanic crust (with 6 km thickness (Funck et al.2003; Hopper et al.2004)) along E36 line brings slightly lower roughness values of 351 and 344 m, respectively. 4.3 The Northern margin of Flemish Cap (NMFC) and the Goban Spur (GS) conjugate margin The Flame Lines F87–3 and F87–4 cross magnetic anomaly 34 at the NMFC and extend another ∼50 km seaward well into oceanic crust (Welford et al.2010a). Refraction data along profile F87–3 show a two-layer oceanic crust that is somewhat thinner (4.5–6.5 km) than typical oceanic crust (Gerlings et al.2011). Roughness values for the oceanic crust are estimated at 322 and 215 m along F87–3 and F87–4 profiles, respectively (Table 2). The Western Approaches Margin (WAM) line on the GS margin is thought to be conjugate of the F87–3 line (Welford et al.2010a; Gerlings et al.2012). The Flemish Cap and GS conjugate margins form an asymmetrical pair: the exhumed mantle domain is ∼100 km wide off GS (Bullock & Minshull 2005) while it is narrower (∼25 km) on the NMFC (Welford et al.2010a). The exhumed mantle domain off GS has a low relief basement (with a very low roughness of 128 m, Table 2) that has been interpreted as a detachment surface (Whitmarsh et al.2001). In the region of initial oceanic crust off GS, the basement is shallower and has a more subdued relief than that off Flemish Cap along F87–3 line. There, (Bullock & Minshull 2005) estimated the roughness value for a ∼6 km thick initial oceanic crust off GS at 216 m (Fig. 3). A similar low roughness value (215 m) and a normal thickness oceanic crust (inferred from gravity data; Welford et al.2010a) are found at the seaward end of the F87–4 profile (Fig. 3). 4.4 The Bay of Biscay (BB) The BB is characterized offshore by an oceanic basin where spreading began at Aptian-Albian time and stopped at Santonian time (e.g. Tugend et al.2014, and references therein). We use Norgasis 11–12 and 14 MCS lines across the Armorican Margin (Thinon et al.2003; Tugend et al.2014, 2015) and the IAM12 MCS line across the North Iberian Margin (Alvarez-Marron et al.1997; Tugend et al.2014) excluding the failed ridge axis and areas too close to the compressional accretionary prism (Fernández-Viejo et al.2012). Roughness values were estimated at 275 and 315 m for the Norgasis lines and at 328 m for the IAM12 line (Fig. 3). Refraction data along Norgasis 14 line show a 3–5 km thick ocean crust at the southern end of the line while a normal ∼6 km thick oceanic crust is found at the northern end of IAM12 line (Fernández-Viejo et al.1998; Thinon et al.2003). 4.5 The Australia-Antarctic margin system There is an ongoing debate on the location and age of the first undisputable oceanic crust within each sector of the Australian-Antarctic margin system (Tikku & Cande 1999; Tikku & Direen 2008). We focus here on the Great Australian Bight (GAB)—Wilkes Land Basin (WLB) system using Australian MCS lines from the AGSO surveys GA199 and GA228 (e.g. Sayers et al.2001). In both the GAB and the WLB, Gillard et al. (2015) used these MCS lines to distinguish a ‘proto-oceanic’ domain from a ‘steady-state’ oceanic domain. This so-called proto-oceanic domain is interpreted as composed of hybrid crust (exhumed mantle and magmatic rocks) and formed as a non-steady state, not localized seafloor spreading. The steady-state oceanic domain is formed from stable oceanic accretion. The proto-oceanic crust is characterized by a deformed base of the sedimentary coverture, the presence of volcanic additions sealing faults, and the rise of the top basement arguing for an increase of magmatism. The oceanic Moho is absent in the proto-oceanic crust but is observed locally at ∼10 s (TWT) in the steady state oceanic crust (Gillard et al.2015). However, by contrast to the Iberia–Newfoundland system, the nature of the basement is ill defined at the GAB/WLB system as neither refraction data nor samples can confirm the inferred basement type. For the GAB, the mean roughness values of the top basement on eight MCS lines in the proto-oceanic and steady-state oceanic domains are estimated at 296 and 266 m, respectively (Table 2, Fig. 3). In the basin off the Wilkes Land we used six MCS lines to estimate the mean roughness values at 301 and 239 m for the proto-oceanic and steady-state oceanic domains, respectively (Table 2, Fig. 3). Within the initial oceanic crust alongside both margins, we exclude large conic-shaped edifices that we interpreted as resulting from post-rift volcanic activity (Fig. 3). As a conclusion of this section, we note that, as for ultraslow-spreading ridges, the roughness estimations obtained over initial oceanic domain alongside magma-poor rifted margins cover a wide range of values, from ∼210 m in the Wilkes Land Basin to ∼580 m off the Galicia margin. 5 COMPARISON OF OCEANIC BASEMENT ROUGHNESS AT MID-OCEAN RIDGES AND OVER INITIAL OCEANIC DOMAIN ALONGSIDE MAGMA-POOR RIFTED MARGINS 5.1 Clues from mid-oceanic ridges 5.1.1 Top basement roughness versus spreading rate It is well known that the elastic lithosphere along slow-spreading ridges is relatively thicker than at faster spreading ridges (Chen & Morgan 1990) and is thus able to sustain greater relief through faulting and flexure (Shaw & Lin 1996). The 750 °C isotherm obtained by thermal modelling for fast to slow-spreading ridges correlates well with the spreading-rate dependence inferred from teleseismic axial earthquake depths (from 2 to 10 km depth; Phipps Morgan & Chen 1993). A recent study of the seismicity along ridge sections of the SWIR and Arctic ridges revealed that earthquakes reach maximum depths of 15 km below volcanic segments and as much as 35 km below the sea floor in exhumed mantle domains (Schlindwein & Schmid 2016). This shows that the lithosphere can be much thicker at ultraslow-spreading ridges than at slow ones. Early attempts to analyse the relationship between the roughness of oceanic seafloor and spreading rate suffered from the lack of data for ultraslow-spreading crust (Small 1994; Goff et al.1997). The new roughness values of seafloor collected in this paper supplement the global data for the ultraslow end of the spectrum. Fig. 4 shows these values compared to the power-law curve fit of (Malinverno 1991) obtained for slow to fast spreading rates. (Ehlers & Jokat 2009) noted that the roughness values obtained for the Arctic ridges were systematically above this reference curve and thus proposed an updated curve fit for slow and ultraslow-spreading crust. The mean roughness values obtained at the ultraslow-spreading SWIR, MCSC, MAR, SPR, Arctic and Sheba ridges range from 100 to >500 m and fall well above or below these two reference curves. The roughness of the flanks of the SPR and the Sheba ridge are even lower than 200–240 m, which is typical of slow-spreading ridges (e.g. at the MAR; Goff et al.1997). This scattering supports that, although the oceanic basement roughness generally increases as the spreading rate decreases from slow to ultraslow, this variation is not solely related to spreading rate (e.g. Goff et al.1997). Figure 4. View largeDownload slide Roughness versus spreading rate for ultraslow and slow-spreading ridges. Data for ultraslow-spreading ridges are given in Table 1. Those for the slow-spreading Mid Atlantic Ridge are from (Goff 1991; Goff et al.1995; Neumann & Forsyth 1995; Min 1999). The roughness value for the slow-spreading SWIR (before 24 Ma) is from Sauter et al. (2011). The continuous grey line is the power-law curve fit of (Malinverno 1991) and the dashed grey line is the curve fit of Ehlers & Jokat (2009). Figure 4. View largeDownload slide Roughness versus spreading rate for ultraslow and slow-spreading ridges. Data for ultraslow-spreading ridges are given in Table 1. Those for the slow-spreading Mid Atlantic Ridge are from (Goff 1991; Goff et al.1995; Neumann & Forsyth 1995; Min 1999). The roughness value for the slow-spreading SWIR (before 24 Ma) is from Sauter et al. (2011). The continuous grey line is the power-law curve fit of (Malinverno 1991) and the dashed grey line is the curve fit of Ehlers & Jokat (2009). 5.1.2 Top basement roughness versus magma supply A weak correlation between spreading rate and bathymetric roughness was also observed at slow-spreading centres, while roughness is relatively constant for intermediate to fast spreading ridges (Small 1994). This weak correlation has been attributed to the along-axis variable thickness of magmatic crust resulting in a more heterogeneous bulk rheology of the lithosphere at slow-spreading ridges than at faster ridges (Small 1994). Along the eastern SWIR, the west-to-east increase of the roughness toward the magma-poor easternmost section (Fig. 4) correlates well with increasing MBA values indicating progressively thinner magmatic crust and/or cooler mantle temperatures (Sauter et al.2011). Either case should produce progressive strengthening of the lithospheric mantle: reduction of the thickness of the magmatic crust relative to the rigid upper mantle will strengthen the lithosphere while the decrease in mantle temperature should produce increasing lithosphere thickness since elastic strength of the lithosphere is dependent upon thermal structure. On the Sheba ridge, roughness is also larger in the ridge section where the crust is thinner (d’Acremont et al.2010). However, no such correlation have been found for the Arctic ridges (Ehlers & Jokat 2009) and no well constrained thickness estimations are available for the MSCS. Despite few estimations of crustal thickness along ultraslow ridges, Fig. 5 shows a weak tendency for increasing roughness when magmatic crustal thickness decreases. The wide range of crustal thickness at ultraslow-spreading ridges shows that another process than enhanced cooling of the mantle when spreading rate decrease is required. A thickening of the thermal boundary layer at those ridges does reduce the height of the melting regime and thus the magma budget (White et al.2001). However, this dismisses the importance of varying mantle source composition relative to temperature (Zhou & Dick 2013). At increasingly slower spreading rates and lower extents of melting, more fertile mantle heterogeneities are sampled preferentially which may lead to greater crustal thickness variations for ultraslow-spreading ridges and thus also to greater variation in surface roughness. On the lower end of the roughness spectra, short wave heterogeneities in the back-arc mantle of the North-Fiji basin has been inferred to explain the segmentation of the SPR (Garel et al.2003). There, the low roughness may be partly explained by a smooth seafloor resulting mainly from the coalescence of numerous volcanoes (Lagabrielle et al.1996; Garel et al.2003). On the Sheba ridge, the western area appears to be affected by excess magmatism (Leroy et al.2010a) that may explain the very low roughness values there. Although error bars on crustal thickness are large, this tendency for increasing roughness when crustal thickness decreases makes sense as thinner magmatic crust, hence a greater mantle component in the lithosphere, would imply greater lithospheric strength. Roughness values larger than ∼300 m may thus be characteristic of magma-poor sections of ultraslow-spreading ridges. Figure 5. View largeDownload slide Roughness versus thickness of the magmatic crust for ultraslow-spreading ridges. Crustal thicknesses for the SWIR are from Cannat et al. (2008), those for the Sheba ridge are from d’Acremont et al. (2010), the one for the MAR is from Lizarralde et al. (2004), the one from the Arctic ridges is from Jokat & Schmidt-Aursch (2007) and Ehlers & Jokat (2009), the one for the SPR is from Garel et al. (2003). Figure 5. View largeDownload slide Roughness versus thickness of the magmatic crust for ultraslow-spreading ridges. Crustal thicknesses for the SWIR are from Cannat et al. (2008), those for the Sheba ridge are from d’Acremont et al. (2010), the one for the MAR is from Lizarralde et al. (2004), the one from the Arctic ridges is from Jokat & Schmidt-Aursch (2007) and Ehlers & Jokat (2009), the one for the SPR is from Garel et al. (2003). This tendency for higher roughness values in thinner crust areas is further complicated by the fact that the mean roughness value for exhumed mantle domains at the SWIR is slightly lower (304 m) than for adjacent magmatic crust areas (372 m; Sloan et al.2012). This is not expected as lithosphere should be thinner and weaker where magma supply is large and focused toward volcanic segment centres, and thicker and stronger where the magma supply is reduced or nearly absent (Cannat 1996). This apparent contradiction may be explained by considering the differences in both rheology and tectonic deformation style of magmatic segments and exhumed mantle domains. Within the magmatic segments, the volcanic seafloor composition is inferred to include the full complement of oceanic lithosphere components (i.e. volcanic and intrusive magmatic rocks overlying upper mantle) that tend to accommodate extension by high-angle normal faults. By contrast, the exhumed mantle domains are inferred to consist largely of serpentinized peridotites above unaltered mantle in which the formation of low-angle, long offset detachment faults is more common (Sauter et al.2013). This is supported by the distribution of the seismicity between magmatic segment and exhumed mantle domains at ultraslow-spreading ridges. At the SWIR the magmatic segments, exhibit brittle deformation throughout the lithosphere up to 8–15 km depths in the upper mantle (Schlindwein & Schmid 2016). By contrast, although earthquakes are observed as deep as 35 km depths, the upper lithosphere is entirely aseismic up to 10 km depth along the adjacent peridotite-dominated ridge sections (Schmid & Schlindwein 2016). This observation of an upper aseismic zone has been interpreted as probably reflecting substantial serpentinization within the exhumed mantle domains of ultraslow-spreading ridges (Schmid & Schlindwein 2016). The variation of oceanic basement roughness at ultraslow-spreading ridges is thus a complex trade-off between spreading rate that highly controls the thermal state of the lithosphere and its composition controlling the rheology and thus also the strength of the lithosphere. 5.2 Insight into the initial seafloor spreading The roughness values determined over inferred initial oceanic domains are all higher than 200 m (Fig. 6), a typical value for slow-spreading ridges or magmatic sections of ultraslow-spreading ridges like the SPR or the Sheba Ridge (Fig. 4). Mean roughness values of the inferred initial oceanic domains bordering the Iberia and Newfoundland margins are greater than 300 m, which is characteristic of magma-poor sections of ultraslow-spreading ridges. Figure 6. View largeDownload slide Roughness of initial oceanic basement alongside Iberia/Newfoundland, Flemish Cap/Goban Spur, Bay of Biscay, Australian Great Bight/Wilkes Land Basin margins. Open hexagons and vertical black lines correspond to the mean values and standard deviations, respectively, for each study area. EM: exhumed mantle. Roughness values and references are given in Table 2. Figure 6. View largeDownload slide Roughness of initial oceanic basement alongside Iberia/Newfoundland, Flemish Cap/Goban Spur, Bay of Biscay, Australian Great Bight/Wilkes Land Basin margins. Open hexagons and vertical black lines correspond to the mean values and standard deviations, respectively, for each study area. EM: exhumed mantle. Roughness values and references are given in Table 2. 5.2.1 The Iberian/Newfoundland margin system Roughness >500 m, as high as those found at the Arctic ridges (Ehlers & Jokat 2009), are found alongside the Iberia margin. There, oceanic domains with either thin crust or normal crustal thickness show high roughness values in the same way as along the Arctic ridges. Although significantly smaller than alongside the Iberia margin, roughness values alongside the Newfoundland margin are as large as those determined at the easternmost magma-poor SWIR. As these margins are conjugate, initial oceanic crust was emplaced at the same spreading centre and thus higher roughness values next to the Iberia margin cannot be explained by a lower spreading rate. Accretion may have been asymmetric with more detachment faulting eastward and more magmatic accretion westward as it is suggested on the MCSC (Hayman et al.2011). However, there is no obvious difference between the basement morphologies in the inferred initial oceanic domains alongside the two margins. The overall different basement topography of the shallower Newfoundland margin compared to the deeper Iberia margin has been tentatively partly attributed to an ‘inherited buoyancy’ of the continental mantle at the Newfoundland margin (Müntener & Manatschal 2006). There, high degrees of melting formed a continental mantle lithosphere that might be compositionally buoyant (Müntener & Manatschal 2006). This hypothesis is consistent with the compositional differences between the Iberia and Newfoundland continental mantle derived peridotites (Müntener & Manatschal 2006). However, although the composition of the rising asthenospheric mantle is unknown, it is likely that its density structure is different from that of the adjacent continental mantle. A more homogenous density structure is rather expected below the two flanks of the initial oceanic ridge, which thus, could not explain a different strength of the initial oceanic lithosphere on the two margins. Alternatively, an explanation may be found in the widespread magmatism, which has been found in the distal domain of the Newfoundland margin (Péron-Pinvidic et al.2010). There, by contrast to the Iberia margin, a major magmatic event, which post-dated the onset of seafloor spreading by at least 7–15 Myr (Hart & Blusztajn 2006), resulted in emplacement of sills over an area of ∼80 000 km2 above the exhumed mantle (Péron-Pinvidic et al.2010). We therefore speculate that smoother initial oceanic basement alongside the Newfoundland margin is related to the weakening of the initial oceanic lithosphere by this late magmatic event. The source of the magma is still uncertain and resolving this issue will require a combination of further geophysical studies and drilling (Péron-Pinvidic et al.2010). Anyhow, we stress that initial oceanic crust alongside both margins exhibit roughness values on the higher end of the spectrum of ultraslow-spreading ridges regardless of the thickness of this initial oceanic crust. Although basement ridges display larger relief next to the Iberia margin, these ridges are on both margins generally symmetric, either with a domal or a triangular shape, without a clear oceanward faulting pattern that would characterize an oceanic magmatic crust with a normal thickness. Landward-dipping reflectors interpreted as faults are rather found on both the exhumed mantle domain of each margin and over the oceanic crust next to it (e.g. Welford et al.2010b; Minshull et al.2014). These structures were interpreted as abandoned fault roots resulting from successive mantle unroofing episodes (Boillot & Froitzheim 2001) as new detachments repeatedly cut through the footwall of the preceding detachment (Reston & McDermott 2011), a ‘flip-flop’ mechanism that has been later shown to be active at the ultraslow-spreading SWIR (Sauter et al.2013). 5.2.2 The Flemish Cap/Goban Spur margin system By contrast to the Iberia–Newfoundland margin, the top basement in the initial oceanic crust alongside the Flemish Cap and GS conjugate margins shows a tilted block morphology typical of slow-spreading ridges (Bullock & Minshull 2005; Gerlings et al.2012). There, oceanward faults were identified in inferred oceanic domains displaying a 215 m roughness value, typical of slow-spreading ridges. A more irregular pattern of dome shaped ridges results in a rougher basement along the F87–3 line (322 m). Gerlings et al. (2012) already noted that the relief of the initial oceanic basement was rougher along the F87–3 line than on the conjugate GS line (216 m). They suggested that the initial igneous accretion was more dominant on Flemish Cap than on its conjugate. We rather argue that, although the velocity (5.8–7 km s–1) of the lower crust at the inferred initial oceanic domain of GS is ambiguous and could support either a gabbroic or serpentinized composition, the roughness of the top basement, its morphology and the ∼6 km thickness of the inferred initial crust indicate typical slow seafloor spreading. The irregular pattern of dome shaped ridges along the F87–3 line may result from asymmetric spreading with oceanic core complex development westward while volcanism dominates eastward on the GS margin. Such asymmetry is often observed locally on slow-spreading ridges, close to transform faults, where magma supply is reduced relatively to the segment centres. 5.2.3 The Bay of Biscay margin system Strong constraints are lacking to clearly determine the accretion processes that produced the initial oceanic domain in the BB. Roughness values are similar to some sections of the ultraslow-spreading SWIR, MCSC or MAR. The top basement is irregular showing either domes or symmetric ridges without tilted blocks typical of slow-spreading crust, which would indicate an ultraslow-spreading. However, compressional structures are observed in the sediment cover showing that some faults were slightly reactivated during the Iberian/European convergence (Thinon et al.2001; Tugend et al.2014). The extent of the resulting change of roughness and morphology is not yet constrained. Moreover, the crustal thickness is highly variable with overthickened crust in some places (Tugend et al.2015). Therefore, we stress that more MCS data extending well into the inferred oceanic domain are needed to resolve the initial accretion processes in the BB. 5.2.4 The Australian Great Bight/Wilkes Land Basin margin system Although there is some overlap between the ranges of roughness values, the mean values are significantly different for the proto-oceanic crust (296–300 m) and steady state oceanic crust (266–239 m) defined by (Gillard et al.2015) at the Great Australian Bight and the Wilkes Land Basin, respectively (Table 2 and Fig. 6). Reflections that could correspond to the oceanic Moho are only locally observed below steady state oceanic crust, leading to a normal thickness (∼2 s TWTT) estimation. This oceanic domain was therefore rather defined by the highly reflective top basement on the MCS data and by the absence of deformation at the base of the sediment cover (Gillard et al.2015). Roughness values of the steady state oceanic crust, only slightly larger than those typical of slow-spreading crust, are relatively low for ultraslow-spreading crust. As in the BB, the top basement is irregular without successive ridgeward facing faults typical of slow-spreading crust. By contrast, at both Australian and Antarctic margins, high structures associated with ridgeward-dipping intrabasement reflections were interpreted as oceanic core complexes produced by detachment faults (Gillard et al.2015). Numerical models show that conditions of high or very low magma supply do not favour detachment faulting and that core complexes occur with intermediate magma supply (Tucholke et al.2008). We therefore argue that both the roughness and the morphology of the top basement indicate large tectonic extension and intermediate magma supply with either slow or ultraslow-spreading rate. We discuss the roughness of the proto-oceanic crust in the next section. 5.3 Proto versus initial oceanic basement The proto-oceanic crust defined alongside the Australian and Antarctic margins is inferred to correspond to serpentinized peridotites that are intensively faulted and intruded (Gillard et al.2016a). It differs from the exhumed mantle domain by oceanward increasing volcanic additions sealing the faulted top basement and intruding or underplating this basement (Gillard et al.2016a). This type of hybrid crust with mantle derived rocks and magmatic rocks, might thus be similar to the exhumed mantle domains between initial oceanic crust and hyperextended continental crust at the Newfoundland-Iberia or Flemish Cap-GS margin systems. However, the amount of melt may be highly variable from one margin to the other or even along the margins themselves (Fig. 1). Roughness values are highly variable in both exhumed mantle domains and proto-oceanic crust ranging from low values (<130 m on GS margin (Bullock & Minshull 2005)) to very high values (>400 m on Newfoundland margin). This dispersion of the roughness values may result from variable extent of post rift volcanism, from highly intruded basement at the Australian and Antarctic margins resulting in a relatively rough seafloor, to a smoother exhumed mantle domain with little or no evidence for melting on GS margin (Bullock & Minshull 2005). However, although the Iberia margin (including Galicia bank) is part of the magma-poor class of margins, the high roughness there is not produced by volcanic edifices but by high relief peridotite ridges inferred to result from successive faulting episodes (e.g. Reston & McDermott 2011). On the Newfoundland margin relatively smooth parts made of exhumed mantle are adjacent to much rougher basement. There, the prominent peridotite ridge drilled at ODP Site 1277 is thought to result from a significant tectonic uplift accompanied by a magmatic event that may have triggered the onset of seafloor spreading (Bronner et al.2011). This illustrates the fundamental difference in nature between proto-oceanic crust or exhumed mantle domain and steady state oceanic crust. While deformation and magmatism are localized at oceanic ridges, multiple tectonic and magmatic episodes may have affected the whole proto-oceanic/exhumed mantle domains leading to highly variable top basement roughness. 6 FURTHER IMPLICATIONS The underlying issue of the determination of the spreading rate at nascent spreading centres is to understand if there, the accretion processes are somehow related to the processes of lithospheric breakup. If spreading is ultraslow alongside every magma-poor rifted margins it could be tempting to suggest that the ultraslow-spreading rate is a consequence of the breakup processes at magma-poor rifted margins. However, among the examples analysed in this paper, only the Iberia–Newfoundland margin system shows clear evidence for initial ultraslow oceanic spreading. Disputable magnetic anomalies cannot be used to estimate the spreading rate alongside these margins and corroborate this result (Barnett-Moore et al.2016; Nirrengarten et al.2017). However, a new palinspastic restoration of the southern North-Atlantic based on 3D gravity inversion, which does not take into account disputed magnetic anomalies but which integrates the rift deformation (Sutra et al.2013), allows to estimate the initial spreading rate alongside these margins (Nirrengarten et al.submitted). Ultraslow-spreading rates <17 mm yr−1 were obtained between 112 and 87Ma at the onset of accretion before reaching slow-spreading rate (>22 mm yr−1; Nirrengarten et al.submitted). This early evolution of the spreading rate between Iberia and Newfoundland is related to the motion of Iberia at that time, resulting in the propagation of seafloor spreading in the BB. When spreading stopped in the BB during Santonian time, oceanic spreading was no more distributed along three branches of a triple junction but became focused at the mid-Atlantic ridge. We therefore suggest that similarly to present day oceanic ridges, spreading at the initial oceanic ridge between Iberia and Newfoundland is controlled by global plate tectonics and is not related to the rifting processes leading to the final breakup. 7 CONCLUSIONS The investigation of the top basement roughness at ultraslow-spreading ridges and within the inferred initial oceanic domain alongside magma-poor rifted margins leads to the following conclusions. The mean roughness values obtained at ultraslow-spreading ridges range from 100 to >500 m and includes 200–240 m roughness values which are typical of slow-spreading ridges. There is a weak tendency for increasing roughness at ultraslow-spreading ridges when magmatic crustal thickness decreases. This is related to the rheology of the oceanic lithosphere; a thinner magmatic crust, hence a greater mantle component in the lithosphere, implying greater lithospheric strength. Roughness values larger than ∼300 m may be characteristic of magma-poor sections of ultraslow-spreading ridges. The top basement roughness values determined within the inferred initial oceanic domain alongside magma-poor rifted margins are all higher than 200 m, a typical value for slow-spreading ridges or magmatic sections of ultraslow-spreading ridges Mean roughness values of the inferred initial oceanic domains alongside the Iberia and Newfoundland margins are greater than 300 m which is characteristic of magma-poor sections of ultraslow-spreading ridges. The top basement in the initial oceanic crust alongside the Flemish Cap and GS margins shows roughness values and a tilted block morphology typical of slow-spreading ridges. Roughness values of the initial oceanic crust alongside the Australian-Antarctic margins are only slighter larger than those typical of slow-spreading crust. We suggest that the roughness and the morphology of the top basement along those margins indicate large tectonic extension and intermediate magma supply with either slow or ultraslow-spreading rate. Estimating roughness values within proto oceanic crust or exhumed mantle domains is less pertinent as polyphased tectonism and magmatism may have affected these domains leading to highly variable top basement roughness values that cannot be linked to a single monophase processes. Acknowledgements We acknowledge funding by ‘Actions Marges’ a TOTAL/INSU/IFP/IFREMER/BRGM program to DS and JA, and funding from Modelling of Margins Phase 3 and 4 consortium for JT. We thank Ségolène Vitoux, Elena Metz and Hugo Ketterer for their work that led to this paper. We also gratefully acknowledge the two anonymous reviewer and the Editor, Donna Blackman, for their review and helpful comments. REFERENCES Afilhado A., Matias L., Shiobara H., Hirn A., Mendes-Victor L., Shimamura H., 2008. From unthinned continent to ocean: the deep structure of the West Iberia passive continental margin at 38°N, Tectonophysics , 458, 9– 50. https://doi.org/10.1016/j.tecto.2008.03.00210.1016/j.tecto.2008.03.00210.1016/j.tecto.2008.03.002 Google Scholar CrossRef Search ADS   Alvarez-Marron J., Rubio E., Torne M., 1997. Subduction-related structures in the North Iberian Margin, J. geophys. Res. , 102, 22 497– 22 511. https://doi.org/10.1029/97JB0142510.1029/97JB0142510.1029/97JB01425 Google Scholar CrossRef Search ADS   Amante C., Eakins B.W., 2009. ETOPO1 1 Arc-minute global relief model: procedures, data sources and analysis, in NOAA Technical Memorandum NESDIS NGDC-24 , ed. NOAA, National Geophysical Data Center. Ball P.J., 2005. Break-up history and evolution of the southern passive margin of Australia, PhD thesis, Royal Holloway University, London. Barnett-Moore N., Hosseinpour M., Maus S., 2016. Assessing discrepancies between previous plate kinematic models of Mesozoic Iberia and their constraints, Tectonics , 35, 1843– 1862. https://doi.org/10.1002/2015TC00401910.1002/2015TC00401910.1002/2015TC004019 Google Scholar CrossRef Search ADS   Beslier M.O., 1996. Seismic Line LG12 in the Iberia Abyssal Plain, in Iberia Abyssal Plain , pp. 737– 739, eds Whitmarsh R.B., Sawyer D.S., Klaus A., Masson D.G., Ocean Drilling Program. Beslier M.O., Cornen G., Girardeau J., 1996. Tectono-metamorphic evolution of peridotites from the ocean/continent transition of the Iberia Abyssal Plain margin, in Proc. ODP, Sci. Results , pp. 397– 412, eds Whitmarsh R.B., Sawyer D.S., Klaus A., Masson D.G., Ocean Drilling Program. Google Scholar CrossRef Search ADS   Boillot G., Froitzheim N., 2001. Non-volcanic rifted margins, continental break-up and the onset of sea-floor spreading: some outstanding questions, Geol. Soc., London, Spec. Publ. , 187, 9– 30 Google Scholar CrossRef Search ADS   Bronner A., Sauter D., Manatschal G., Peron-Pinvidic G., Munschy M., 2011. Magmatic breakup as an explanation for magnetic anomalies at magma-poor rifted margins, Nat. Geosci. , 4, 549– 553. https://doi.org/10.1038/ngeo120110.1038/ngeo120110.1038/ngeo120110.1038/ngeo1201 Google Scholar CrossRef Search ADS   Bullock A.D., Minshull T.A., 2005. From continental extension to seafloor spreading: crustal structure of the Goban Spur rifted margin, southwest of the UK, Geophys. J. Int. , 163, 527– 546. https://doi.org/10.1111/j.1365-246X.2005.02726.x10.1111/j.1365-246X.2005.02726.x10.1111/j.1365-246X.2005.02726.x10.1111/j.1365-246X.2005.02726.x Google Scholar CrossRef Search ADS   Cannat M., 1993. Emplacement of mantle rocks in the seafloor at mid-ocean ridges, J. geophys. Res. , 98, 4163– 4172. https://doi.org/10.1029/92JB0222110.1029/92JB0222110.1029/92JB0222110.1029/92JB02221 Google Scholar CrossRef Search ADS   Cannat M., 1996. How thick is the magmatic crust at slow spreading oceanic ridges?, J. geophys. Res. , 101, 2847– 2857. https://doi.org/10.1029/95JB0311610.1029/95JB0311610.1029/95JB0311610.1029/95JB03116 Google Scholar CrossRef Search ADS   Cannat M., Sauter D., Mendel V., Ruellan E., Okino K., Escartin J., Combier V., Baala M., 2006. Modes of seafloor generation at a melt-poor ultraslow-spreading ridge, Geology , 34, 605– 608. https://doi.org/10.1130/G22486.110.1130/G22486.110.1130/G22486.110.1130/G22486.1 Google Scholar CrossRef Search ADS   Cannat M., Sauter D., Bezos A., Meyzen C., Humler E., Le Rigoleur M., 2008. Spreading rate, spreading obliquity, and melt supply at the ultraslow spreading Southwest Indian Ridge, Geochem. Geophys. Geosyst. , 9, Q04002, doi:10.1029/2007GC001676. https://doi.org/10.1029/2007GC00167610.1029/2007GC00167610.1029/2007GC00167610.1029/2007GC001676 Google Scholar CrossRef Search ADS   Cannat M., Manatschal G., Sauter D., Peron-Pinvidic G., 2009. Assessing the conditions of continental breakup at magma-poor rifted margins: What can we learn from slow spreading mid-ocean ridges?, C. R. Geosci. , 341, 406– 427 Google Scholar CrossRef Search ADS   Chen Y., Morgan W.J., 1990. Rift valley/no rift valley transition at mid-ocean ridges, J. geophys. Res. , 95, 17 571–17 581. https://doi.org/10.1029/JB095iB11p1757110.1029/JB095iB11p1757110.1029/JB095iB11p1757110.1029/JB095iB11p17571 Cunha T., 2008. Gravity Anomalies, Flexure and the Thermo-Mechanical Evolution of the West Iberia Margin and its Conjugate of Newfoundland, PhD thesis, University of Oxford, Oxford. D’Acremont E., Leroy S., Maia M., Gente P., Autin J., 2010. Volcanism, jump and propagation on the Sheba ridge, eastern Gulf of Aden: segmentation evolution and implications for oceanic accretion processes, Geophys. J. Int. , 180, 535– 551. https://doi.org/10.1111/j.1365-246X.2009.04448.x10.1111/j.1365-246X.2009.04448.x10.1111/j.1365-246X.2009.04448.x10.1111/j.1365-246X.2009.04448.x Google Scholar CrossRef Search ADS   Davy R.G. et al., 2016. Continental hyperextension, mantle exhumation and thin oceanic crust at the continent-ocean transition, West Iberia: new insights from wide-angle seismic, J. geophys. Res. , 121, 3177– 3199. Google Scholar CrossRef Search ADS   Dean S.M., Minshull T.A., Whitmarsh R.B., Louden K.E., 2000. Deep structure of the ocean-continent transition in the southern Iberia Abyssal Plain from seismic refraction profiles: The IAM-9 transect at 40°20΄N, J. geophys. Res. , 105, 5859– 5885. https://doi.org/10.1029/1999JB90030110.1029/1999JB90030110.1029/1999JB90030110.1029/1999JB900301 Google Scholar CrossRef Search ADS   Dean S.L., Sawyer D.S., Morgan J.K., 2015. Galicia Bank ocean–continent transition zone: New seismic reflection constraints, Earth planet. Sci. Lett. , 413, 197– 207. https://doi.org/10.1016/j.epsl.2014.12.04510.1016/j.epsl.2014.12.04510.1016/j.epsl.2014.12.04510.1016/j.epsl.2014.12.045 Google Scholar CrossRef Search ADS   Delescluse M., Funck T., Dehler S.A., Louden K.E., Watremez L., 2015. The oceanic crustal structure at the extinct, slow to ultra-slow Labrador Sea spreading center, J. geophys. Res. , 120, 5249– 5272. Google Scholar CrossRef Search ADS   Dick H.J.B., Lin J., Schouten H., 2003. An ultraslow-spreading class of ocean ridge, Nature , 426, 405– 412. https://doi.org/10.1038/nature0212810.1038/nature0212810.1038/nature0212810.1038/nature02128 Google Scholar CrossRef Search ADS PubMed  Ehlers B., Jokat W., 2009. Subsidence and crustal roughness of ultra-slow spreading ridges in the northern North Atlantic and the Arctic Ocean, Geophys. J. Int. , 177, 451– 462. https://doi.org/10.1111/j.1365-246X.2009.04078.x10.1111/j.1365-246X.2009.04078.x10.1111/j.1365-246X.2009.04078.x10.1111/j.1365-246X.2009.04078.x Google Scholar CrossRef Search ADS   Fernández-Viejo G., Gallart J., Pulgar J.A., Gallastegui J., Dañobeitia J.J., Córdoba D., 1998. Crustal transition between continental and oceanic domains along the North Iberian Margin from wide angle seismic and gravity data, Geophys. Res. Lett. , 25, 4249– 4252. https://doi.org/10.1029/1998GL90014910.1029/1998GL90014910.1029/1998GL90014910.1029/1998GL900149 Google Scholar CrossRef Search ADS   Fernández-Viejo G., Pulgar J.A., Gallastegui J., Quintana L., 2012. The fossil accretionary wedge of the Bay of Biscay: critical wedge analysis on depth-migrated seismic sections and geodynamical implications, J. Geol. , 120, 315– 331. https://doi.org/10.1086/66478910.1086/66478910.1086/66478910.1086/664789 Google Scholar CrossRef Search ADS   Fournier M., Patriat P., Leroy S., 2001. Reappraisal of the Arabia–India–Somalia triple junction kinematics, Earth planet. Sci. Lett. , 189, 103– 114. https://doi.org/10.1016/S0012-821X(01)00371-510.1016/S0012-821X(01)00371-510.1016/S0012-821X(01)00371-510.1016/S0012-821X(01)00371-5 Google Scholar CrossRef Search ADS   Funck T., Hopper J.R., Larsen H.C., Louden K.E., Tucholke B.E., Holbrook W.S., 2003. Crustal structure of the ocean-continent transition at Flemish Cap: seismic refraction results, J. geophys. Res. , 108( B11), 2531, doi:10.1029/2003JB002434. https://doi.org/10.1029/2003JB00243410.1029/2003JB00243410.1029/2003JB00243410.1029/2003JB002434 Google Scholar CrossRef Search ADS   Garel E., Lagabrielle Y., Pelletier B., 2003. Abrupt axial variations along the slow to ultra-slow spreading centers of the northern North Fiji Basin (SW Pacific): Evidence for short wave heterogeneities in a back-arc mantle, Mar. Geophys. Res. , 24, 245– 263. https://doi.org/10.1007/s11001-004-1060-y10.1007/s11001-004-1060-y10.1007/s11001-004-1060-y10.1007/s11001-004-1060-y Google Scholar CrossRef Search ADS   Gerlings J., Louden K.E., Jackson H.R., 2011. Crustal structure of the Flemish Cap Continental Margin (eastern Canada): an analysis of a seismic refraction profile, Geophys. J. Int. , 185, 30– 48. https://doi.org/10.1111/j.1365-246X.2011.04931.x10.1111/j.1365-246X.2011.04931.x10.1111/j.1365-246X.2011.04931.x10.1111/j.1365-246X.2011.04931.x Google Scholar CrossRef Search ADS   Gerlings J., Louden K.E., Minshull T.A., Nedimovic M.R., 2012. Flemish Cap-Goban Spur conjugate margins: New evidence of asymmetry, Geology , 40, 1107– 1110. https://doi.org/10.1130/G33263.110.1130/G33263.110.1130/G33263.110.1130/G33263.1 Google Scholar CrossRef Search ADS   Gillard M., 2014. Évolution tectono-magmatique menant à l’océanisation sur les marges passives pauvres en magma, exemple des marges Australie-Antarctique, PhD thesis, University of Strasbourg, Strasbourg. Gillard M., Autin J., Manatschal G., Sauter D., Munschy M., Schaming M., 2015. Tectonomagmatic evolution of the final stages of rifting along the deep conjugate Australian-Antarctic magma-poor rifted margins: constraints from seismic observations, Tectonics , 34, 753– 783. https://doi.org/10.1002/2015TC00385010.1002/2015TC00385010.1002/2015TC00385010.1002/2015TC003850 Google Scholar CrossRef Search ADS   Gillard M., Autin J., Manatschal G., 2016a. Fault systems at hyper-extended rifted margins and embryonic oceanic crust: Structural style, evolution and relation to magma, Mar. Pet. Geol. , 76, 51– 67. https://doi.org/10.1016/j.marpetgeo.2016.05.01310.1016/j.marpetgeo.2016.05.01310.1016/j.marpetgeo.2016.05.01310.1016/j.marpetgeo.2016.05.013 Google Scholar CrossRef Search ADS   Gillard M., Manatschal G., Autin J., 2016b. How can asymmetric detachment faults generate symmetric ocean continent transitions?, Terra Nova , 28, 27– 34. https://doi.org/10.1111/ter.1218310.1111/ter.1218310.1111/ter.1218310.1111/ter.12183 Google Scholar CrossRef Search ADS   Goff J.A., 1991. A Global and Regional Stochastic Analysis of Near-Ridge Abyssal Hill Morphology, J. geophys. Res. , 96, 21 713–21 737. https://doi.org/10.1029/91JB0227510.1029/91JB0227510.1029/91JB0227510.1029/91JB02275 Goff J.A., Jordan T.H., 1988. Stochastic modeling of seafloor morphology: inversion of sea beam data for second-order statistics, J. geophys. Res. , 93, 13 589–13 608. https://doi.org/10.1029/JB093iB11p1358910.1029/JB093iB11p1358910.1029/JB093iB11p1358910.1029/JB093iB11p13589 Goff J.A., Jordan T.H., 1989. Stochastic modeling of seafloor morphology; Resolution of topographic parameters by sea beam data, IEEE J. Ocean. Eng. , 14, 326– 337. https://doi.org/10.1109/48.3598310.1109/48.3598310.1109/48.3598310.1109/48.35983 Google Scholar CrossRef Search ADS   Goff J.A., Tucholke B.E., Lin J., Jaroslow G.E., Kleinrock M.C., 1995. Quantitative analysis of abyssal hills in the Atlantic Ocean: A correlation between inferred crustal thickness and extensional faulting, J. geophys. Res. , 100, 22 509–22 522. https://doi.org/10.1029/95JB0251010.1029/95JB0251010.1029/95JB0251010.1029/95JB02510 Goff J.A., Ma Y., Shah A., Cochran J.R., Sempéré J., 1997. Stochastic analysis of seafloor morphology on the flank of the Southeast Indian Ridge: The influence of ridge morphology on the formation of abyssal hills, J. geophys. Res. , 102, 15 521–15 534. https://doi.org/10.1029/97JB0078110.1029/97JB0078110.1029/97JB0078110.1029/97JB00781 Hart S., Blusztajn J., 2006. Age and geochemistry of the mafic sills, ODP site 1276, Newfoundland margin, Chem. Geol. , 235, 222– 237. https://doi.org/10.1016/j.chemgeo.2006.07.00110.1016/j.chemgeo.2006.07.00110.1016/j.chemgeo.2006.07.00110.1016/j.chemgeo.2006.07.001 Google Scholar CrossRef Search ADS   Haupert I., Manatschal G., Decarlis A., Unternehr P., 2016. Upper-plate magma-poor rifted margins: Stratigraphic architecture and structural evolution, Mar. Pet. Geol. , 69, 241– 261. https://doi.org/10.1016/j.marpetgeo.2015.10.02010.1016/j.marpetgeo.2015.10.02010.1016/j.marpetgeo.2015.10.02010.1016/j.marpetgeo.2015.10.020 Google Scholar CrossRef Search ADS   Hayman N.W., Grindlay N.R., Perfit M.R., Mann P., Leroy S., De Lépinay B.M., 2011. Oceanic core complex development at the ultraslow spreading Mid-Cayman Spreading Center, Geochem. Geophys. Geosyst. , 12. https://doi.org/10.1029/2010GC00324010.1029/2010GC00324010.1029/2010GC00324010.1029/2010GC003240 Hopper J.R., Funck T., Tucholke B.E., Christian Larsen H., Holbrook W.S., Louden K.E., Shillington D., Lau H., 2004. Continental breakup and the onset of ultraslow seafloor spreading off Flemish Cap on the Newfoundland rifted margin, Geology , 32, 93– 96. https://doi.org/10.1130/G19694.110.1130/G19694.110.1130/G19694.110.1130/G19694.1 Google Scholar CrossRef Search ADS   Jagoutz O., Muntener O., Manatschal G., Rubatto D., Peron-Pinvidic G., Turrin B.D., Villa I.M., 2007. The rift-to-drift transition in the North Atlantic: a stuttering start of the MORB machine?, Geology , 35, 1087– 1090. https://doi.org/10.1130/G23613A.110.1130/G23613A.110.1130/G23613A.110.1130/G23613A.1 Google Scholar CrossRef Search ADS   Jokat W., Schmidt-Aursch M.C., 2007. Geophysical characteristics of the ultraslow spreading Gakkel Ridge, Arctic Ocean, Geophys. J. Int. , 168, 983– 998. https://doi.org/10.1111/j.1365-246X.2006.03278.x10.1111/j.1365-246X.2006.03278.x10.1111/j.1365-246X.2006.03278.x10.1111/j.1365-246X.2006.03278.x Google Scholar CrossRef Search ADS   Jokat W., Ritzmann O., Schmidt-Aursch M.C., Drachev S., Gauger S., Snow J., 2003. Geophysical evidence for reduced melt production on the Arctic ultraslow Gakkel mid-ocean ridge, Nature , 423, 962– 965. https://doi.org/10.1038/nature0170610.1038/nature0170610.1038/nature0170610.1038/nature01706 Google Scholar CrossRef Search ADS PubMed  Klimke J., Franke D., Gaedicke C., Schreckenberger B., Schnabel M., Stollhofen H., Rose J., Chaheire M., 2016. How to identify oceanic crust—Evidence for a complex break-up in the Mozambique Channel, off East Africa, Tectonophysics , 693, 436– 452. https://doi.org/10.1016/j.tecto.2015.10.01210.1016/j.tecto.2015.10.01210.1016/j.tecto.2015.10.01210.1016/j.tecto.2015.10.012 Google Scholar CrossRef Search ADS   Lagabrielle Y. et al., 1996. Active oceanic spreading in the northern North Fiji Basin: Results of the NOFI cruise of R/V L’Atalante (newstarmer project), Mar. Geophys. Res. , 18, 225– 247. https://doi.org/10.1007/BF0028607910.1007/BF0028607910.1007/BF0028607910.1007/BF00286079 Google Scholar CrossRef Search ADS   Leroy S., Mercier de Lepinay B., Mauffret A., Pubellier M., 1996. Structural and tectonic evolution of the eastern Cayman Trough (Caribbean Sea) from seismic reflection data, AAPG Bull. , 80, 222– 247. Leroy S., Mauffret A., Patriat P., Mercier De Lepinay B., 2000. An alternative interpretation of the Cayman trough evolution from a reidentification of magnetic anomalies, Geophys. J. Int. , 141, 539– 557. https://doi.org/10.1046/j.1365-246x.2000.00059.x10.1046/j.1365-246x.2000.00059.x10.1046/j.1365-246x.2000.00059.x10.1046/j.1365-246x.2000.00059.x Google Scholar CrossRef Search ADS   Leroy S. et al., 2004. From rifting to spreading in the eastern Gulf of Aden: a geophysical survey of a young oceanic basin from margin to margin, Terra Nova , 16, 185– 192. https://doi.org/10.1111/j.1365-3121.2004.00550.x10.1111/j.1365-3121.2004.00550.x10.1111/j.1365-3121.2004.00550.x10.1111/j.1365-3121.2004.00550.x Google Scholar CrossRef Search ADS   Leroy S., D’acremont E., Tiberi C., Basuyau C., Autin J., Lucazeau F., Sloan H., 2010a. Recent off-axis volcanism in the eastern Gulf of Aden: Implications for plume–ridge interaction, Earth planet. Sci. Lett. , 293, 140– 153. https://doi.org/10.1016/j.epsl.2010.02.03610.1016/j.epsl.2010.02.03610.1016/j.epsl.2010.02.03610.1016/j.epsl.2010.02.036 Google Scholar CrossRef Search ADS   Leroy S. et al., 2010b. Contrasted styles of rifting in the eastern Gulf of Aden: A combined wide-angle, multichannel seismic, and heat flow survey, Geochem. Geophys. Geosyst. , 11. https://doi.org/10.1029/2009GC00296310.1029/2009GC00296310.1029/2009GC00296310.1029/2009GC002963 Li J. et al., 2015. Seismic observation of an extremely magmatic accretion at the ultraslow spreading Southwest Indian Ridge, Geophys. Res. Lett. , 42, 2656– 2663. https://doi.org/10.1002/2014GL06252110.1002/2014GL06252110.1002/2014GL06252110.1002/2014GL062521 Google Scholar CrossRef Search ADS   Lizarralde D., Gaherty J.B., Collins J.A., Hirth G., Kim S.D., 2004. Spreading-rate dependence of melt extraction at mid-ocean ridges from mantle seismic refraction data, Nature , 432, 744– 747. https://doi.org/10.1038/nature0314010.1038/nature0314010.1038/nature0314010.1038/nature03140 Google Scholar CrossRef Search ADS PubMed  Malinverno A., 1991. Inverse square-root dependence of mid-ocean-ridge flank roughness on spreading rate, Nature , 352, 58– 60. https://doi.org/10.1038/352058a010.1038/352058a010.1038/352058a010.1038/352058a0 Google Scholar CrossRef Search ADS   Mcdermott K., Reston T., 2015. To see, or not to see? Rifted margin extension, Geology , 43, 967– 970. https://doi.org/10.1130/G36982.110.1130/G36982.110.1130/G36982.110.1130/G36982.1 Google Scholar CrossRef Search ADS   Mendel V., Sauter D., Rommevaux-Jestin C., Patriat P., Lefebvre F., Parson L.M., 2003. Magmato-tectonic cyclicity at the ultra-slow spreading Southwest Indian Ridge: evidence from variations of axial volcanic ridge morphology and abyssal hills pattern, Geochem. Geophys. Geosyst. , 4, doi:9110.1029/2002GC000417. https://doi.org/10.1029/2002GC00041710.1029/2002GC00041710.1029/2002GC00041710.1029/2002GC000417 Michael P.J. et al., 2003. Magmatic and amagmatic seafloor generation at the ultraslow-spreading Gakkel ridge, Arctic Ocean, Nature , 423, 956– 961. https://doi.org/10.1038/nature0170410.1038/nature0170410.1038/nature0170410.1038/nature01704 Google Scholar CrossRef Search ADS PubMed  Minshull T.A., 1999. On the roughness of Mesozoic oceanic crust in the western North Atlantic, Geophys. J. Int. , 136, 286– 290. https://doi.org/10.1046/j.1365-246X.1999.00722.x10.1046/j.1365-246X.1999.00722.x10.1046/j.1365-246X.1999.00722.x10.1046/j.1365-246X.1999.00722.x Google Scholar CrossRef Search ADS   Minshull T.A., 2009. Geophysical characterisation of the ocean–continent transition at magma-poor rifted margins, C. R. Geosci. , 341, 382– 393 Google Scholar CrossRef Search ADS   Minshull T.A., Muller M.R., White R.S., 2006. Crustal structure of the Southwest Indian Ridge at 66°E: seismic constraints, Geophys. J. Int. , 166, 135– 147. https://doi.org/10.1111/j.1365-246X.2006.03001.x10.1111/j.1365-246X.2006.03001.x10.1111/j.1365-246X.2006.03001.x10.1111/j.1365-246X.2006.03001.x Google Scholar CrossRef Search ADS   Minshull T.A., Dean S.M., Whitmarsh R.B., 2014. The peridotite ridge province in the southern Iberia Abyssal Plain: seismic constraints revisited, J. geophys. Res. , 119, 1580– 1598. Google Scholar CrossRef Search ADS   Muller M.R., Minshull T.A., White R.S., 1999. Segmentation and melt supply at the Southwest Indian Ridge, Geology , 27, 867– 870. https://doi.org/10.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;210.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;210.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;210.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;2 Google Scholar CrossRef Search ADS   Müntener O., Manatschal G., 2006. High degrees of melt extraction recorded by spinel harzburgite of the Newfoundland margin: The role of inheritance and consequences for the evolution of the southern North Atlantic, Earth planet. Sci. Lett. , 252, 437– 452. https://doi.org/10.1016/j.epsl.2006.10.00910.1016/j.epsl.2006.10.00910.1016/j.epsl.2006.10.00910.1016/j.epsl.2006.10.009 Google Scholar CrossRef Search ADS   Neumann G.A., Forsyth D.W., 1995. High resolution statistical estimation of seafloor morphology: Oblique and orthogonal fabric on the flanks of the Mid-Atlantic Ridge, 34°–35.5°S, Mar. Geophys. Res. , 17, 221– 250. https://doi.org/10.1007/BF0120346410.1007/BF0120346410.1007/BF0120346410.1007/BF01203464 Google Scholar CrossRef Search ADS   Nirrengarten M., Manatschal G., Tugend J., Kusznir N.J., Sauter D., 2017. Nature and origin of the J-magnetic anomaly offshore Iberia-Newfoundland: implications for plate reconstructions, Terra Nova , 29, 20– 28. https://doi.org/10.1111/ter.1224010.1111/ter.1224010.1111/ter.1224010.1111/ter.12240 Google Scholar CrossRef Search ADS   Nirrengarten M., Manatschal G., Tugend J., Kusznir N.J., Sauter D., submitted. Kinematic evolution of the southern North Atlantic: implications for the formation of hyper-extended rift systems, Tectonics . Niu X. et al., 2015. Along-axis variation in crustal thickness at the ultraslow spreading Southwest Indian Ridge (50°E) from a wide-angle seismic experiment, Geochem. Geophys. Geosyst. , 16, 468– 485. https://doi.org/10.1002/2014GC00564510.1002/2014GC00564510.1002/2014GC00564510.1002/2014GC005645 Google Scholar CrossRef Search ADS   Patriat P., Sloan H., Sauter D., 2008. From slow to ultraslow: A previously undetected event at the Southwest Indian Ridge at ca. 24 Ma, Geology , 36, 207– 210. https://doi.org/10.1130/G24270A.110.1130/G24270A.110.1130/G24270A.110.1130/G24270A.1 Google Scholar CrossRef Search ADS   Péron-Pinvidic G., Shillington D.J., Tucholke B.E., 2010. Characterization of sills associated with the U reflection on the Newfoundland margin: evidence for widespread early post-rift magmatism on a magma-poor rifted margin, Geophys. J. Int. , 182, 113– 136. Peron-Pinvidic G., Manatschal G., Osmundsen P.T., 2013. Structural comparison of archetypal Atlantic rifted margins: A review of observations and concepts, Mar. Pet. Geol. , 43, 21– 47. https://doi.org/10.1016/j.marpetgeo.2013.02.00210.1016/j.marpetgeo.2013.02.00210.1016/j.marpetgeo.2013.02.00210.1016/j.marpetgeo.2013.02.002 Google Scholar CrossRef Search ADS   Phipps Morgan J., Chen Y.J., 1993. Dependence of ridge-axis morphology on magma supply and spreading rate, Nature , 364, 706– 708. https://doi.org/10.1038/364706a010.1038/364706a010.1038/364706a010.1038/364706a0 Google Scholar CrossRef Search ADS   Pickup S.L.B., Whitmarsh R.B., Fowler C.M.R., Reston T.J., 1996. Insight into the nature of the ocean-continent transition off West Iberia from a deep multichannel seismic reflection profile, Geology , 24, 1079– 1082. https://doi.org/10.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;210.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;210.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;210.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;2 Google Scholar CrossRef Search ADS   Ranero C.R., Perez-Gussinye M., 2010. Sequential faulting explains the asymmetry and extension discrepancy of conjugate margins, Nature , 468, 294– 299. https://doi.org/10.1038/nature0952010.1038/nature0952010.1038/nature0952010.1038/nature09520 Google Scholar CrossRef Search ADS PubMed  Ranero C.R., Banda E., Buhl P., 1997. The crustal structure of the Canary Basin: accretion processes at slow spreading centers, J. geophys. Res. , 102, 10185– 10201. https://doi.org/10.1029/97JB0010110.1029/97JB0010110.1029/97JB0010110.1029/97JB00101 Google Scholar CrossRef Search ADS   Reid I., Jackson H.R., 1981. Oceanic spreading rate and crustal thickness, Mar. Geophys. Res. , 5, 165– 172. Reston T.J., Mcdermott K.G., 2011. Successive detachment faults and mantle unroofing at magma-poor rifted margins, Geology , 39, 1071– 1074. https://doi.org/10.1130/G32428.110.1130/G32428.110.1130/G32428.110.1130/G32428.1 Google Scholar CrossRef Search ADS   Sauter D., Mendel V., Rommevaux-Jestin C., Parson L.M., Fujimoto H., Mével C., Cannat M., Tamaki K., 2004. Focused magmatism versus amagmatic spreading along the ultra-slow spreading southwest Indian ridge: evidence from TOBI side scan sonar imagery, Geochem. Geophys. Geosyst. , 5, doi:10.1029/2004GC000738. https://doi.org/10.1029/2004GC00073810.1029/2004GC00073810.1029/2004GC00073810.1029/2004GC000738 Sauter D., Cannat M., Meyzen C., Bezos A., Patriat P., Humler E., Debayle E., 2009. Propagation of a melting anomaly along the ultraslow Southwest Indian Ridge between 46°E and 52°20΄E: interaction with the Crozet hotspot?, Geophys. J. Int. , 179, 687– 699. https://doi.org/10.1111/j.1365-246X.2009.04308.x10.1111/j.1365-246X.2009.04308.x10.1111/j.1365-246X.2009.04308.x10.1111/j.1365-246X.2009.04308.x Google Scholar CrossRef Search ADS   Sauter D., Sloan H., Cannat M., Goff J., Patriat P., Schaming M., Roest W.R., 2011. From slow to ultra-slow: How does spreading rate affect seafloor roughness and crustal thickness?, Geology , 39, 911– 914. https://doi.org/10.1130/G32028.110.1130/G32028.110.1130/G32028.110.1130/G32028.1 Google Scholar CrossRef Search ADS   Sauter D. et al., 2013. Continuous exhumation of mantle-derived rocks at the Southwest Indian Ridge for 11 million years, Nat. Geosci. , 6, 314– 320. https://doi.org/10.1038/ngeo177110.1038/ngeo177110.1038/ngeo177110.1038/ngeo1771 Google Scholar CrossRef Search ADS   Sauter D. et al., 2016. Evidence for magma entrapment below oceanic crust from deep seismic reflections in the Western Somali Basin, Geology , 44, 407– 410. https://doi.org/10.1130/G37747.110.1130/G37747.110.1130/G37747.110.1130/G37747.1 Google Scholar CrossRef Search ADS   Sayers J., Symonds P.A., Direen N.G., Bernardel G., 2001. Nature of the continent-ocean transition on the non-volcanic rifted margin of the central Great Australian Bight, Geol. Soc., London, Spec. Publ. , 187, 51– 76. https://doi.org/10.1144/GSL.SP.2001.187.01.0410.1144/GSL.SP.2001.187.01.0410.1144/GSL.SP.2001.187.01.0410.1144/GSL.SP.2001.187.01.04 Google Scholar CrossRef Search ADS   Schlindwein V., Schmid F., 2016. Mid-ocean-ridge seismicity reveals extreme types of ocean lithosphere, Nature , 535 276– 279. https://doi.org/10.1038/nature1827710.1038/nature1827710.1038/nature1827710.1038/nature18277 Google Scholar CrossRef Search ADS PubMed  Schmid F., Schlindwein V., 2016. Microearthquake activity, lithospheric structure, and deformation modes at an amagmatic ultraslow spreading Southwest Indian Ridge segment, Geochem. Geophys. Geosyst. , 17, 2905– 2921. https://doi.org/10.1002/2016GC00627110.1002/2016GC00627110.1002/2016GC00627110.1002/2016GC006271 Google Scholar CrossRef Search ADS   Shaw W.J., Lin J., 1996. Models of ocean ridge lithospheric deformation: Dependence on crustal thickness, spreading rate, and segmentation, J. geophys. Res. , 101, 17 977– 17 993. https://doi.org/10.1029/96JB0094910.1029/96JB0094910.1029/96JB0094910.1029/96JB00949 Google Scholar CrossRef Search ADS   Shillington D.J. et al., 2006. Evidence for asymmetric nonvolcanic rifting and slow incipient oceanic accretion from seismic reflection data on the Newfoundland margin, J. geophys. Res. , 111, B09402, doi:10.1029/2005JB003981. Sloan H., Sauter D., Goff J.A., Cannat M., 2012. Abyssal hill characterization at the ultraslow spreading Southwest Indian Ridge, Geochem. Geophys. Geosyst. , 13. https://doi.org/10.1029/2011GC00385010.1029/2011GC00385010.1029/2011GC00385010.1029/2011GC003850 Small C., 1994. A global analysis of mid-ocean ridge axial topography, Geophys. J. Int. , 116, 64– 84. https://doi.org/10.1111/j.1365-246X.1994.tb02128.x10.1111/j.1365-246X.1994.tb02128.x10.1111/j.1365-246X.1994.tb02128.x10.1111/j.1365-246X.1994.tb02128.x Google Scholar CrossRef Search ADS   Stanton N., Manatschal G., Autin J., Sauter D., Maia M., Viana A., 2016. Geophysical fingerprints of hyper-extended, exhumed and embryonic oceanic domains: the example from the Iberia–Newfoundland rifted margins, Mar. Geophys. Res. , 37, 185– 205. https://doi.org/10.1007/s11001-016-9277-010.1007/s11001-016-9277-010.1007/s11001-016-9277-010.1007/s11001-016-9277-0 Google Scholar CrossRef Search ADS   Sundvik M., Larson R.L., Detrick R.S., 1984. Rough-smooth basement boundary in the western North Atlantic basin: evidence for a seafloor-spreading origin, Geology , 12, 31– 34. https://doi.org/10.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;210.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;210.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;210.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;2 Google Scholar CrossRef Search ADS   Sutra E., Manatschal G., 2012. How does the continental crust thin in a hyperextended rifted margin? Insights from the Iberia margin, Geology , 40, 139– 142. https://doi.org/10.1130/G32786.110.1130/G32786.110.1130/G32786.110.1130/G32786.1 Google Scholar CrossRef Search ADS   Sutra E., Manatschal G., Mohn G., Unternehr P., 2013. Quantification and restoration of extensional deformation along the Western Iberia and Newfoundland rifted margins, Geochem. Geophys. Geosyst. , 14, 2575– 2597. https://doi.org/10.1002/ggge.2013510.1002/ggge.2013510.1002/ggge.2013510.1002/ggge.20135 Google Scholar CrossRef Search ADS   Ten Brink U.S., Coleman D.F., Dillon W.P., 2002. The nature of the crust under Cayman Trough from gravity, Mar. Pet. Geol. , 19, 971– 987. https://doi.org/10.1016/S0264-8172(02)00132-010.1016/S0264-8172(02)00132-010.1016/S0264-8172(02)00132-010.1016/S0264-8172(02)00132-0 Google Scholar CrossRef Search ADS   Thinon I., Fidalgo-González L., Réhault J., Olivet J., 2001. Déformations pyrénéennes dans le golfe de Gascogne, C. R. Acad. Sci., Paris II , 332, 561– 568. https://doi.org/10.1016/S1251-8050(01)01576-210.1016/S1251-8050(01)01576-210.1016/S1251-8050(01)01576-210.1016/S1251-8050(01)01576-2 Thinon I., Matias L., Rehault J.P., Hirn A., Fidalgo-Gonzalez L., Avedik F., 2003. Deep structure of the Armorican Basin (Bay of Biscay): a review of Norgasis seismic reflection and refraction data, J. Geol. Soc. , 160, 99– 116. https://doi.org/10.1144/0016-764901-10310.1144/0016-764901-10310.1144/0016-764901-10310.1144/0016-764901-103 Google Scholar CrossRef Search ADS   Tikku A.A., Cande S.C., 1999. The oldest magnetic anomalies in the Australian-Antarctic Basin: Are they isochrons?, J. geophys. Res. , 104, 661– 677. https://doi.org/10.1029/1998JB90003410.1029/1998JB90003410.1029/1998JB90003410.1029/1998JB900034 Google Scholar CrossRef Search ADS   Tikku A.A., Direen N.G., 2008. Comment on “Major Australian-Antarctic Plate Reorganization at Hawaiian-Emperor Bend Time”, Science , 321, 490. https://doi.org/10.1126/science.115716310.1126/science.115716310.1126/science.115716310.1126/science.1157163 Google Scholar CrossRef Search ADS PubMed  Tucholke B.E., Sibuet J.-C., 2007. Leg 210 synthesis: tectonic, magmatic, and sedimentary evolution of the Newfoundland-Iberia rift, in Proceedings of the Ocean Drilling Program, Scientific Results , pp. 1– 56, eds Tucholke B.E., Sibuet J.-C., Klaus A., Ocean Drilling Program. Google Scholar CrossRef Search ADS   Tucholke B.E., Behn M.D., Buck W.R., Lin J., 2008. Role of melt supply in oceanic detachment faulting and formation of megamullions, Geology , 36, 455– 458. https://doi.org/10.1130/G24639A.110.1130/G24639A.110.1130/G24639A.110.1130/G24639A.1 Google Scholar CrossRef Search ADS   Tugend J., Manatschal G., Kusznir N.J., Masini E., Mohn G., Thinon I., 2014. Formation and deformation of hyperextended rift systems: insights from rift domain mapping in the Bay of Biscay-Pyrenees, Tectonics , 33, 1239– 1276. https://doi.org/10.1002/2014TC00352910.1002/2014TC00352910.1002/2014TC00352910.1002/2014TC003529 Google Scholar CrossRef Search ADS   Tugend J., Manatschal G., Kusznir N.J., Masini E., 2015. Characterizing and identifying structural domains at rifted continental margins: application to the Bay of Biscay margins and its Western Pyrenean fossil remnants, Geol. Soc., London, Spec. Publ. , 413, 171– 203. https://doi.org/10.1144/SP413.310.1144/SP413.310.1144/SP413.310.1144/SP413.3 Google Scholar CrossRef Search ADS   Unternehr P., Peron-Pinvidic G., Manatschal G., Sutra E., 2010. Hyper-extended crust in the South Atlantic: in search of a model, Pet. Geosci. , 16, 207– 215. https://doi.org/10.1144/1354-079309-90410.1144/1354-079309-90410.1144/1354-079309-90410.1144/1354-079309-904 Google Scholar CrossRef Search ADS   Van Avendonk H.J.A., Holbrook W.S., Nunes G.T., Shillington D.J., Tucholke B.E., Louden K.E., Larsen H.C., Hopper J.R., 2006. Seismic velocity structure of the rifted margin of the eastern Grand Banks of Newfoundland, Canada, J. geophys. Res. , 111, B11404, doi:10.1029/2005JB004156. https://doi.org/10.1029/2005JB00415610.1029/2005JB00415610.1029/2005JB00415610.1029/2005JB004156 Google Scholar CrossRef Search ADS   Watremez L., Leroy S., Rouzo S., D’acremont E., Unternehr P., Ebinger C., Lucazeau F., Al-Lazki A., 2011. The crustal structure of the north-eastern Gulf of Aden continental margin: insights from wide-angle seismic data, Geophys. J. Int. , 184, 575– 594. https://doi.org/10.1111/j.1365-246X.2010.04881.x10.1111/j.1365-246X.2010.04881.x10.1111/j.1365-246X.2010.04881.x10.1111/j.1365-246X.2010.04881.x Google Scholar CrossRef Search ADS   Weigelt E., Jokat W., 2001. Peculiarities of roughness and thickness of oceanic crust in the Eurasian Basin, Arctic Ocean, Geophys. J. Int. , 145, 505– 516. https://doi.org/10.1046/j.1365-246X.2001.00398.x10.1046/j.1365-246X.2001.00398.x10.1046/j.1365-246X.2001.00398.x10.1046/j.1365-246X.2001.00398.x Google Scholar CrossRef Search ADS   Welford J.K., Hall J., Sibuet J., Srivastava S.P., 2010a. Structure across the northeastern margin of Flemish Cap, offshore Newfoundland from Erable multichannel seismic reflection profiles: evidence for a transtensional rifting environment, Geophys. J. Int. , 183, 572– 586. https://doi.org/10.1111/j.1365-246X.2010.04779.x10.1111/j.1365-246X.2010.04779.x10.1111/j.1365-246X.2010.04779.x10.1111/j.1365-246X.2010.04779.x Google Scholar CrossRef Search ADS   Welford J.K., Smith J.A., Hall J., Deemer S., Srivastava S.P., Sibuet J.-C., 2010b. Structure and rifting evolution of the northern Newfoundland Basin from Erable multichannel seismic reflection profiles across the southeastern margin of Flemish Cap, Geophys. J. Int. , 180, 976– 998. https://doi.org/10.1111/j.1365-246X.2009.04477.x10.1111/j.1365-246X.2009.04477.x10.1111/j.1365-246X.2009.04477.x10.1111/j.1365-246X.2009.04477.x Google Scholar CrossRef Search ADS   White R.S., Minshull T.A., Bickle M.J., Robinson C.J., 2001. Melt generation at very slow-spreading oceanic ridges: constraints from geochemical and geophysical data, J. Pet. , 42, 1171– 1196. https://doi.org/10.1093/petrology/42.6.117110.1093/petrology/42.6.117110.1093/petrology/42.6.117110.1093/petrology/42.6.1171 Google Scholar CrossRef Search ADS   Whitmarsh R.B., Manatschal G., Minshull T.A., 2001. Evolution of magma-poor continental margins from rifting to seafloor spreading, Nature , 413, 150– 154. https://doi.org/10.1038/3509308510.1038/3509308510.1038/3509308510.1038/35093085 Google Scholar CrossRef Search ADS PubMed  Zhao M. et al., 2013. Three-dimensional seismic structure of the Dragon Flag oceanic core complex at the ultraslow spreading Southwest Indian Ridge (49°39΄E), Geochem. Geophys. Geosyst. , 14, 4544– 4563. https://doi.org/10.1002/ggge.2026410.1002/ggge.2026410.1002/ggge.2026410.1002/ggge.20264 Google Scholar CrossRef Search ADS   Zhou H., Dick H.J.B., 2013. Thin crust as evidence for depleted mantle supporting the Marion Rise, Nature , 494, 195– 200. https://doi.org/10.1038/nature1184210.1038/nature1184210.1038/nature1184210.1038/nature11842 Google Scholar CrossRef Search ADS PubMed  © The Authors 2017. Published by Oxford University Press on behalf of The Royal Astronomical Society. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Geophysical Journal International Oxford University Press

Oceanic basement roughness alongside magma-poor rifted margins: insight into initial seafloor spreading

Loading next page...
 
/lp/ou_press/oceanic-basement-roughness-alongside-magma-poor-rifted-margins-insight-09tgC0PBZm
Publisher
The Royal Astronomical Society
Copyright
© The Authors 2017. Published by Oxford University Press on behalf of The Royal Astronomical Society.
ISSN
0956-540X
eISSN
1365-246X
D.O.I.
10.1093/gji/ggx439
Publisher site
See Article on Publisher Site

Abstract

Abstract The variation of oceanic basement roughness at mid-oceanic ridges is a complex trade-off between spreading rate that largely controls the thermal state of the lithosphere and its composition controlling the rheology and thus also the strength of the lithosphere. Here we estimate top basement roughness (i.e. the root-mean-square deviation of residual basement relief) over initial oceanic crust bordering the Iberia, Newfoundland, Bay of Biscay, Goban Spur, Flemish Cap, Australian and Antarctic rifted margins to provide new insights into the spreading processes at the nascent plate boundary. Although ultraslow seafloor spreading is suggested in those areas, the lack of undisputable oceanic magnetic anomalies prevents any well-constrained determination of the initial spreading rates. We compare these estimated roughness values with those determined over ultraslow-spreading crust formed at the Mid Atlantic Ridge, Southwest Indian Ridge, Arctic ridges, Mid-Cayman Spreading Center, Sheba ridge and South Pandora Ridge. The roughness values obtained at these ultraslow-spreading ridges range from 100 to >500 m and include 200–240 m roughness values which are typical of slow-spreading ridges. Roughness values larger than ∼300 m are characteristic of magma-poor sections of ultraslow-spreading ridges. The top basement roughness values determined within the inferred initial oceanic domain bordering the investigated magma-poor rifted margins are all higher than 200 m. Mean roughness values of the inferred initial oceanic domains alongside the conjugate Iberia and Newfoundland margins are greater than 300 m similarly to magma-poor sections of ultraslow-spreading ridges. The top basement in the initial oceanic crust alongside the conjugate Flemish Cap and Goban Spur margins shows roughness values and a tilted block morphology typical of slow-spreading ridges. We suggest that the roughness and the morphology of the top basement bordering the conjugate Australian and Antarctic margins indicate large tectonic extension and intermediate magma supply at either slow- or ultraslow-spreading rates. We show that estimating roughness values within transitional domains of magma-poor margins like exhumed mantle domains is less pertinent as polyphased tectonism and magmatism may have affected these domains leading to highly variable top basement roughness values that cannot be linked to a single process. Composition and structure of the oceanic crust, Atlantic Ocean, Indian Ocean, Continental margins: divergent, Mid-ocean ridge processes, Submarine tectonics and volcanism 1 INTRODUCTION Up to recently, hydrocarbon exploration has influenced our knowledge of the formation of rifted margins by collecting data mainly over sedimentary basins proximal to the continent. The acquisition of oceanward seismic data combined with deep sea drilling at magma-poor rifted margins, particularly along the conjugate Iberia and Newfoundland margins, revealed that the proximal and distal domains are very different from each other. The discovery of hyperextended continental crust and exhumed subcontinental mantle in these conjugated margins led to debated models for crustal thinning (e.g. Ranero & Perez-Gussinye 2010; Sutra & Manatschal 2012; McDermott & Reston 2015). However, acquisition of both seismic and potential field data still remains, with some exceptions, focused to the rifted margin and only few lines image the transition to unambiguous oceanic crust. As a consequence, the transition to ocean crust is still poorly constrained. Terms like initial or incipient (Hopper et al.2004), embryonic (Jagoutz et al.2007) or proto-oceanic crust (Gillard et al.2015) have been used to describe this transitional domain lacking clear geophysical characteristics of ocean crust like a well-marked reflector interpreted as the oceanic Moho on multichannel seismic (MCS) reflection data or typical velocity gradients on refraction data (Min 2009). Moreover, the lack of undisputable oceanic magnetic anomalies often did not allow the determination of initial spreading rates either because the onset of seafloor spreading fell within the Cretaceous Normal Superchron or because the ill-defined magnetic lineations may not be due to reversals of the Earth's magnetic field but to other processes (Bronner et al.2011). However, occurrence of crust thinner (<5 km) than typical oceanic crust was thought to indicate that this transitional crust was emplaced at an ultraslow-spreading ridge (Hopper et al.2004; van Avendonk et al.2006). Indeed, evidence for thin oceanic crust (<5 km) has been found at ultraslow-spreading ridges in the early 80s (Reid & Jackson 1981). This led to the suggestion that below a critical spreading rate (15–20 mm yr−1), the melt supply per increment of plate separation is dramatically decreasing (Reid & Jackson 1981). This is thought to result from a thickening of the thermal boundary layer at those ridges which in turn reduces the height of the melting regime (White et al.2001). In the last two decades exhumed mantle derived rocks have been found in wide areas of the ultraslow-spreading Southwest Indian Ridge (SWIR; Cannat et al.2006; Sauter et al.2013) and the Gakkel Ridge (Michael et al.2003) and seismic data have shown thin crust in magma-poor sections of these ridges (Muller et al.1999; Jokat et al.2003; Minshull et al.2006). However, very thick crust (8–10 km) has been recently discovered (Zhao et al.2013; Li et al.2015; Niu et al.2015) at the shallowest part of a >550 km long thick crust ridge section of the SWIR (Sauter et al.2009). Such along-axis large crustal thickness variations, while the ultraslow-spreading rate varies only slightly, are therefore challenging the current view of thin crust being associated with ultraslow spreading (Michael et al.2003; Cannat et al.2008). It is commonly inferred that seafloor roughness (i.e. the root mean square deviation of residual basement relief) increases dramatically for full spreading rates of <40 mm yr−1 (Malinverno 1991; Ehlers & Jokat 2009). Top basement roughness could thus be used to identify ultraslow-spreading lithosphere. We therefore suggest that investigating the top basement roughness of initial oceanic domains bordering magma-poor rifted margins may provide new insights in the spreading processes triggered by lithospheric breakup. In this paper, we first review seafloor roughness estimates at present day ultraslow-spreading ridges based on published data from the SWIR and from the Arctic ridges and on new roughness estimations at the Mid-Cayman Spreading Center (MCSC), the Sheba ridge and at the South Pandora Ridge (SPR) in the North Fiji Basin (Fig. 1). We also investigate the relief on flanks of the Mid Atlantic Ridge (MAR), which were created when this ridge underwent ultraslow spreading. Results from measurements of the top basement roughness along MCS profiles on inferred initial oceanic domains bordering the Iberia, Newfoundland, Bay of Biscay (BB), Goban Spur (GS), Flemish Cap, Australian and Antarctic margins are then presented in a second part (Fig. 1). These top basement roughness values are compared to those of ultraslow-spreading ridges. We discuss how estimating the roughness of the top basement may help to constrain the spreading processes at the initial oceanic ridge. We found that roughness values of the inferred initial oceanic domains alongside the conjugate Iberia and Newfoundland margins are similar to those of magma-poor sections on ultraslow-spreading ridges. In contrast, the top basement within the initial oceanic domain alongside the conjugate Flemish Cap and GS margins shows roughness values typical of slow-spreading ridges. We also show that the roughness and morphology of the top basement alongside the conjugate Australian and Antarctic margins indicate large tectonic extension and intermediate magma supply at either slow or ultraslow-spreading rates. Figure 1. View largeDownload slide Worldwide distribution of the magma-poor rifted margin systems and locations of the rifted margins and mid-oceanic ridges discussed in this paper. MAR: Mid-Atlantic Ridge; MCSC: Mid Cayman Spreading Center; SPR: South Pandora Ridge; SWIR: Southwest Indian Ridge. After Haupert et al. (2016). Background image is from Amante & Eakins (2009). Figure 1. View largeDownload slide Worldwide distribution of the magma-poor rifted margin systems and locations of the rifted margins and mid-oceanic ridges discussed in this paper. MAR: Mid-Atlantic Ridge; MCSC: Mid Cayman Spreading Center; SPR: South Pandora Ridge; SWIR: Southwest Indian Ridge. After Haupert et al. (2016). Background image is from Amante & Eakins (2009). 2 DATA AND METHOD The stochastic modelling procedure (Goff & Jordan 1988, 1989) offers a robust way to analyse the statistical properties of abyssal hill morphology. This method was applied on grids of bathymetric data that were collected using multibeam echo sounders on the flanks of the MAR (Goff 1991; Goff et al.1995; Neumann & Forsyth 1995) and the SWIR (Sauter et al.2011; Sloan et al.2012). Close to the ridge axis and far from continents, where sediments are lacking, such bathymetric grids allow a 2-D analysis of the top of the oceanic crust. On sedimented rifted margins or spreading ridges, MCS profiles are needed to investigate the top basement relief, but are spaced too far apart and give only access to 1-D analysis of the top basement. Nevertheless, calculation of the roughness of the top basement is derived the same way from 1-D or 2-D data. It is defined as the root-mean-square (RMS) height, which is the average variation of depths with respect to the regional depth trend. RMS height is therefore calculated after removing the regional trend from the data. We used a second order empirical trend of the basement instead of a theoretical subsidence curve to avoid dependence upon on a crustal age model (Ehlers & Jokat 2009). When available we use MCS data published as depth sections (Hopper et al.2004; Bullock & Minshull 2005; Shillington et al.2006; Gerlings et al.2012; Dean et al.2015; Tugend et al.2015). However, MCS data are commonly shown as two-way traveltime (TWTT) sections (e.g. Afilhado et al.2008; Welford et al.2010b, Pickup et al.1996). The seafloor and top basement reflections were thus picked to obtain the sediment thickness in second TWTT which was then depth converted and added to the seafloor bathymetry to obtain the top basement relief in meters. Sediment velocity information was provided by a compilation of wide angle and refraction seismic sections within each area of interest. Along the Iberia margin we use the TWTT-velocity relationships of (Cunha 2008) to convert MCS data in the Iberia Abyssal Plain (IAP; Pickup et al.1996) and in the Tagus Abyssal Plain (TAP; Afilhado et al.2008). On the Newfoundland margin, we built up on the work of (Shillington et al.2006; van Avendonk et al.2006) and (Hopper et al.2004) to determine the TWTT-velocity functions for the SCREECH 2 and SCREECH 1 areas that were then used to derive the depth of the top basement for the MCS data falling within these areas (Welford et al.2010b). No refraction profile has been collected on the conjugate Australian-Antarctic margins. Therefore we used compilations of velocity profiles from sonobuoys (Ball 2005; Gillard 2014) to derive TWTT-velocity functions and convert MCS data interpreted by Gillard et al. (2015, 2016a,b). 3 SEAFLOOR ROUGHNESS AT ULTRASLOW-SPREADING RIDGES We review hereafter the top basement roughness estimates at the SWIR, Gakkel, Knipovich and Sheba ridges, the MCSC, the SPR and the MAR (Table 1). We focus first on the SWIR as there, roughness values are well documented on various tectonic settings (Sauter et al.2011; Sloan et al.2012). The presence of exhumed mantle-derived rocks on >100 km long stretches of the ridge axis is one of the striking contrasts between most of the ultraslow-spreading ridges and faster spreading ridges (Dick et al.2003; Sauter et al.2004, 2013; Cannat et al.2006). However, exhumed mantle is not ubiquitous at ultraslow-spreading ridges and thick magmatic oceanic crust is also found there (Niu et al.2015). Although some fossil oceanic ridges have undergone slow to ultraslow rates while spreading was failing (e.g. in the Labrador or Somali basin) we disregarded them as a post spreading tectonic phase may have affected the top basement (Delescluse et al.2015; Sauter et al.2016). Table 1. Estimations of the top basement roughness on ultraslow-spreading mid-oceanic ridges.   Full Rate (mm yr−1)  RMS height (m)  Size and type of data set  Number of samples  References  Southwest Indian Ridge (SWIR)  15    ∼100 km long bathymetric grids    (Sauter et al.2011; Sloan et al.2012)  54–67°E (volcanic domains)  15  298 ± 23  id.  27  id.  54–61°E (volcanic domains)  15  242 ± 36  id.  12  id.  61–67°E (volcanic domains)  15  372 ± 23  id.  9  id.  61–67°E (exhumed mantle domains)  15  304 ± 20  id.  6  id.  Arctic ridges  9–15  517 ± 50  >100 km long MCS profiles  9  (Weigelt & Jokat 2001; Ehlers & Jokat 2009)  Mid-Cayman Spreading Center (MCSC)  17  371 ± 79a  >120 km long bathymetric profiles  13  GeoMapApp cruises  Western flank  17  432 ± 72a  id.  6  id.  Eastern flank  17  319 ± 34a  id.  7  id.  Sheba ridge 52–55°E  20  164 ± 59a  >140 km long MCS profiles  3  (d’Acremont et al.2010)  Western basin  20  103a  150 km long MCS profile  1  id.  Eastern basin  20  195 ± 50a  >140 km long MCS profiles  2  id.  Mid Atlantic Ridge (MAR)  18    >100 km long MCS profiles  3    Western North America basin  18  257a  >130 km long MCS profile  1  (Lizarralde et al.2004)  Western North America Basin  18  160  >100 km long MCS profile  1  (Min 1999 1999)  Eastern (Canary) basin  18  350  id.  1  id.  South Pandora Ridge (SPR)  18  158 ± 16a  >80 km long bathymetric grids  4  (Lagabrielle et al.1996)    Full Rate (mm yr−1)  RMS height (m)  Size and type of data set  Number of samples  References  Southwest Indian Ridge (SWIR)  15    ∼100 km long bathymetric grids    (Sauter et al.2011; Sloan et al.2012)  54–67°E (volcanic domains)  15  298 ± 23  id.  27  id.  54–61°E (volcanic domains)  15  242 ± 36  id.  12  id.  61–67°E (volcanic domains)  15  372 ± 23  id.  9  id.  61–67°E (exhumed mantle domains)  15  304 ± 20  id.  6  id.  Arctic ridges  9–15  517 ± 50  >100 km long MCS profiles  9  (Weigelt & Jokat 2001; Ehlers & Jokat 2009)  Mid-Cayman Spreading Center (MCSC)  17  371 ± 79a  >120 km long bathymetric profiles  13  GeoMapApp cruises  Western flank  17  432 ± 72a  id.  6  id.  Eastern flank  17  319 ± 34a  id.  7  id.  Sheba ridge 52–55°E  20  164 ± 59a  >140 km long MCS profiles  3  (d’Acremont et al.2010)  Western basin  20  103a  150 km long MCS profile  1  id.  Eastern basin  20  195 ± 50a  >140 km long MCS profiles  2  id.  Mid Atlantic Ridge (MAR)  18    >100 km long MCS profiles  3    Western North America basin  18  257a  >130 km long MCS profile  1  (Lizarralde et al.2004)  Western North America Basin  18  160  >100 km long MCS profile  1  (Min 1999 1999)  Eastern (Canary) basin  18  350  id.  1  id.  South Pandora Ridge (SPR)  18  158 ± 16a  >80 km long bathymetric grids  4  (Lagabrielle et al.1996)  aDetermined in this paper. Values with standard deviations are mean values. View Large 3.1 The Southwest Indian ridge (SWIR) The SWIR full spreading rate has been ultraslow since a dramatic decrease ∼24 myr ago, from a slow rate of 30 to 15 mm yr−1 (Patriat et al.2008). The RMS abyssal hill height mean value (298 m; Table 1), determined within polygonal areas on the flanks of the eastern part of the ultraslow-spreading SWIR, is ∼80 m greater than for slow-spreading seafloor. However, seafloor roughness also increases eastward along the SWIR axis (Sauter et al.2011). The roughness mean value (372 m) is much larger between 61 and 67°E, in the easternmost deep part of the ridge, than between 54 and 61°E (242 m; Table 1). Fig. 2 displays some bathymetric profiles to illustrate this eastward increasing roughness of the seafloor. The Mantle Bouguer Anomaly (MBA; a crustal thickness and/or mantle temperature proxy deduced from gravity data) also changes eastward, suggesting thinner crust and/or cooler mantle temperatures in the easternmost part of the SWIR (Sauter et al.2011). There, large seafloor domains of exhumed mantle derived rocks have been called ‘smooth seafloor’ because they occur in the form of broad ridges with a smooth, rounded topography with no resolvable volcanic cones or typical abyssal hills on bathymetric data (Cannat et al.2006). The mean value of RMS height for these exhumed mantle domains (304 m) is ∼70 m smaller than for the volcanic seafloor (372 m; Sloan et al.2012; Table 1 and Fig. 2). Figure 2. View largeDownload slide Across axis profiles at the Southwest Indian Ridge and the Gakkel ridge illustrating the range of roughness values (R) found at ultraslow-spreading centres. (a) Bathymetric profile crossing the SWIR ridge axis (at 56°E) showing a roughness value <200 m (calculated outside the axial valley; see Sauter et al.2011) and typical abyssal hills usually observed at slow-spreading centres; (b) bathymetric profile (at 62°E) in the easternmost magma-poor section of the ridge within a domain of exhumed mantle derived rocks (smooth seafloor); (c) bathymetric profile crossing the SWIR axis (at 64°E) within a volcanic part of the easternmost magma-poor section of the ridge; (d) top basement picked in the seismic reflection profile AWI-20 010 100 on the eastern flank of the Gakkel ridge. Fault locations are inferred from the bathymetric grids. Their depth and dip are unconstraint. Vertical exaggeration is 2×. Profiles across the SWIR are from Sauter et al. (2009) and Sauter et al. (2013). The roughness values estimated along the SWIR bathymetric profiles are slightly different from the roughness values given in Table 1 which are mean values calculated on several polygonal areas. The roughness estimation and the profile on the flank of the Gakkel Ridge are from (Ehlers & Jokat 2009). Figure 2. View largeDownload slide Across axis profiles at the Southwest Indian Ridge and the Gakkel ridge illustrating the range of roughness values (R) found at ultraslow-spreading centres. (a) Bathymetric profile crossing the SWIR ridge axis (at 56°E) showing a roughness value <200 m (calculated outside the axial valley; see Sauter et al.2011) and typical abyssal hills usually observed at slow-spreading centres; (b) bathymetric profile (at 62°E) in the easternmost magma-poor section of the ridge within a domain of exhumed mantle derived rocks (smooth seafloor); (c) bathymetric profile crossing the SWIR axis (at 64°E) within a volcanic part of the easternmost magma-poor section of the ridge; (d) top basement picked in the seismic reflection profile AWI-20 010 100 on the eastern flank of the Gakkel ridge. Fault locations are inferred from the bathymetric grids. Their depth and dip are unconstraint. Vertical exaggeration is 2×. Profiles across the SWIR are from Sauter et al. (2009) and Sauter et al. (2013). The roughness values estimated along the SWIR bathymetric profiles are slightly different from the roughness values given in Table 1 which are mean values calculated on several polygonal areas. The roughness estimation and the profile on the flank of the Gakkel Ridge are from (Ehlers & Jokat 2009). Seafloor roughness could not be estimated in the shallowest part of the SWIR (35–52°E) where the magmatic crust is the thickest (Zhao et al.2013; Niu et al.2015). There, a sudden increase of the magma supply ∼10 Ma ago is thought to have resulted in a shallow volcanic plateau at the ridge axis (Sauter et al.2009). This plateau is too narrow to estimate valid roughness values. However, we note that large topographic features (up to 60 km long and ∼900 m-high) are observed on top of this volcanic plateau as well as close to the Marion hot spot (Mendel et al.2003) suggesting still thick lithosphere in the shallowest parts of the SWIR. 3.2 The arctic ridges High roughness values (>500 m) were obtained by Weigelt & Jokat (2001) and Ehlers & Jokat (2009) along >100 km long seismic reflection profiles in the Arctic Ocean. We retain nine seismic profiles in the Amundsen, Nansen and Boreas basins on the flanks of the ultraslow-spreading Gakkel and Knipovich ridges (9–15 mm yr−1 spreading rates) and disregard one profile in the Molloy basin as it is located too close to the Spitsbergen Fracture Zone. The roughness values were estimated from profiles of seismically determined top basement by (Weigelt & Jokat 2001; Ehlers & Jokat 2009). They may not be straightforwardly compared to the statistical analysis of bathymetric grids at the SWIR. However, the mean roughness value on the ultraslow-spreading Arctic ridges is significantly larger (517 m; Table 1) than those estimated on the SWIR (Fig. 2). No correlation has been found between these values and the magmatic and amagmatic segmentation of the Arctic ridges or the highly variable thickness of the crust (2–6 km; Ehlers & Jokat 2009). 3.3 The Mid‐Cayman Spreading Center (MCSC) The deep-seated MCSC is located midway along a depression, called Cayman trough, that extends from the Belize margin to the northern edge of Jamaica. Seafloor magnetic anomalies within the Cayman trough record a ∼17 mm yr−1 ultraslow-spreading rate for the last 20 myr (Leroy et al.2000). Series of horsts and grabens are well marked about 150 km on either side of the N–S oriented spreading centre but, further away, the seafloor topography becomes smoother as the sedimentary cover thickens towards the Belize margin to the west and offshore Cuba Island to the east (Leroy et al.2000). We therefore use 13 ∼125 km long-bathymetric profiles from the WI93300 and Ka93200 cruises (from USNS Wilkes and Kane in 1973 and 1972 respectively) on each flank of the MCSC excluding a 60 km wide central part across the ridge axis. The bathymetry of the off-axis Cayman trough is highly asymmetric (Hayman et al.2011). The estimated roughness to the west of the MCSC is larger (432 m mean value) than to the east (319 m mean value; Table 1). The crustal thickness is thought to be also asymmetric, with thinner upper crust to the east than to the west (Leroy et al.1996; Hayman et al.2011). However, we choose to disregard the crustal thickness estimations on the MCSC as they are poorly constrained and debated (ten Brink et al.2002). 3.4 The Sheba ridge The 20–22 mm yr−1 spreading rate of the Sheba ridge is equal to the threshold (Fournier et al.2001; Leroy et al.2004) below which the ultraslow-spreading class of oceanic ridges is usually defined (Reid & Jackson 1981). We use the top basement relief determined in two seismic profiles from the Arabian margin to the Somalian margin between the Alula-Fartak Fracture Zone and the Socotra-Hadbeen Fracture Zone (d’Acremont et al.2010). We avoid the axial valley relief and exclude thus a central 100 km wide part to estimate the roughness of the oceanic crust. We also disregard the anomalously shallow area on the southern flank of the Sheba ridge where recent off-axis volcanism has been shown by (Leroy et al.2010a). The estimated mean roughness value is larger in the eastern basin (195 m mean value; Table 1) than in the western one (103 m), while the crust is thinner in the eastern than in the western basin (4 and 6 km average crustal thickness, respectively; Leroy et al.2010b, d’Acremont et al.2010; Watremez et al.2011). 3.5 The South Pandora ridge (SPR) The North Fiji Basin is a back-arc basin opening with a complex geometry of spreading centres trending both ∼NS and ∼EW (Garel et al.2003). We used a set of multibeam bathymetric data collected along the 500 km long SPR, which is opening with an ultraslow-spreading rate (16 mm yr−1; Lagabrielle et al.1996). We obtained a low RMS roughness mean value (158 m; Table 1). It is worth pointing out that the SPR strongly differs from other ultraslow-spreading ridges. Mantle rocks have never been recovered along any segment of the ridges of the North Fiji Basin (Garel et al.2003) contrasting with the frequent recovery of serpentinized peridotites at segment ends along the ultraslow and slow oceanic ridges (e.g. Cannat 1993). Moreover, the magma production is high, consistent with the general shallow depth of the entire back arc basin and the analysis of gravity anomalies indicating crustal thickness as large as 8 km along the SPR (Garel et al.2003). 3.6 The Mid Atlantic ridge (MAR) The spreading rate in the North Atlantic decreased by a factor of about two between M-21 (148 Ma) and M-11 (135 Ma; Sundvik et al.1984). A rough acoustic basement is associated with this period of ultraslow-spreading (∼18 mm yr−1) in the Central Atlantic west of Bermuda and in the conjugated Canary Basin (Sundvik et al.1984; Ranero & Banda 1997). Min (1999) estimated that the RMS roughness mean values only increased from 130 to 160 m in the Western (Central Atlantic) Basin while it reached ∼350 m in the eastern (Canary) Basin. Using the MCS data of Lizarralde et al. (2004) collected in the Western Basin we obtained a higher roughness value of 257 m for the ultraslow-spreading basement (Table 1). Refraction data coincident with the MCS in the western North Atlantic basin show that the crustal thickness decreased from 6.7 to 5.5 km when spreading became ultraslow (Lizarralde et al.2004). Finally, as a conclusion of this section, we note that the roughness estimations obtained at these ultraslow-spreading ridges cover a wide range of values, from ∼100 m on the Sheba Ridge to ∼580 m in the Arctic basin (Weigelt & Jokat 2001). 4 TOP BASEMENT ROUGHNESS ALONGSIDE MAGMA-POOR MARGINS Estimating the roughness of initial oceanic crust alongside rifted margins requires to define the location of the Continent Ocean Boundary (COB). Typically, ‘normal’ oceanic crust is characterized in MCS sections by a: (1) smooth top basement at a kilometric scale, commonly observed at 8 s TWTT, (2) highly reflective upper crust and an acoustically relatively transparent lower-crust, (3) ∼2 s TWTT thickness, (4) well-marked reflector at the base of the crust interpreted as the Moho, commonly observed at 10 s TWTT and (5) passive sediment infill (e.g. Klimke et al.2016). However, seismic reflection profiles beyond rifted margins often do not sample such unambiguous oceanic crust. A well-marked reflector that could be interpreted as the Moho is rarely observed, while the seismic texture within the basement is often confusing. Therefore, initial oceanic crust is usually inferred from a conjunction of arguments from seismic (refraction and/or reflection) as well as potential field data, none of these being conclusive on its own. This is the case at the Iberia–Newfoundland, GS–Flemish Cap, BB, and Australo–Antarctic margin systems that we analyse hereafter (Table 2). Although we use the location of the initial oceanic domains proposed in the literature, we were careful to verify some basic characteristics. As the COB can be associated with an important top basement topography (Unternehr et al.2010), thought to result from the transition from a deep and denser exhumed mantle domain to a shallower less dense oceanic domain (Peron-Pinvidic et al.2013), we checked that the location of such topographic features agrees with the inferred location of the COB. Moreover, domains with volcanic overprint and/or a deformed sediment cover were not assumed to be oceanic in nature as no deformation related to spreading is expected on the flanks of mid-oceanic ridges outside a narrow deformation zone close to the axis (Cannat et al.2009). Table 2. Estimations of the top basement roughness within the initial oceanic domain alongside magma-poor rifted margins. OC, Oceanic crust; EM, exhumed mantle; POC, proto-oceanic crust (see definition in main text).   Inferred crust type  RMS height (m)  Profile length (km)  Number of profiles  References  Tagus Abyssal Plain (TAP)        1    Shallow subdomain on IAM5 line  OC  456a  60  1  (Afilhado et al.2008)  Deep subdomain on IAM5 line  OC  414a  105  1  id.  Both domains on IAM5 line  OC  520a  165  1  id.  Iberia Abyssal Plain (IAP)        3    West of R4 on IAM9 line  OC  429a  80  1  (Pickup et al.1996)  East of R4 on IAM9 line  EM  368a  115  1  id.  West of R3 on Lusigal12 line  OC  566a  120  1  (Beslier 1996)  West of R3 on SO-22 line  OC  534a  160  1  (Pickup et al.1996)  Mean Value for OC  OC  510 ± 72a    3  id.  Galicia Margin (GM)            West of Peridotite Ridge on line WE1  OC  583a  85  1  (Dean et al.2015)  Northern Newfoundland Basin (NNFB)        5    East of M0 on SCREECH 2  OC  449a  75  1  (Shillington et al.2006)  West of M0 on SCREECH 2  EM  410a  120  1  id.  West of M3 on SCREECH 2  EM  268a  75  1    East of M0 on SCREECH 107/201  OC  355a  80  1  id.  East of M0 on SCREECH 302/109  OC  374a  84  1  id.  East of M0 on SCREECH E56  OC  320a  75  1  (Welford et al.2010b)  West of M0, on SCREECH E56  EM  378a  96  1    East of M0 on SCREECH E54  OC  320a  144  1  id.  Mean value for OC  OC  364 ± 53a    5  id.  Mean value for EM  EM  394 ± 23a    3  id.  Southern Margin of Flemish Cap (SMFC)        1    On line E36  OC  380a  93  1  (Welford et al.2010b)  Northern Margin of Flemish Cap (NMFC)        2    On line F87–4  OC  215a  85  1  (Welford et al.2010a)  On line F87–3  OC  322a  87  1  id.  Goban Spur Margin (GS)        1    On WAM line  OC  216  50  1  (Bullock & Minshull 2005)  On WAM  EM  128  67  1  id.  Bay of Biscay (BB)        3    On Norgasis 11–12  OC  315a  65  1  (Thinon et al.2003)  On Norgasis 14  OC  275a  64  1  id.  On IAM12  OC  328a  95  1  (Tugend et al.2014)  Great Australian Bight (GAB)        8    On GA199–10  POC  231a  56  1  (Gillard et al.2015)  On GA199–10  OC  258a  102  1  id.  On GA199–07  POC  251a  61  1  id.  On GA199–06  OC  243a  96  1  id.  On GA199–06  POC  362a  61  1  id.  On GA199–05  POC  262a  82  1  id.  On GA199–04  POC  216a  82  1  id.  On GA199–03  POC  374a  99  1  id.  On GA199–02  POC  356a  95  1  id.  On GA199–01  POC  313a  106  1  id.  On GA199–01  OC  296a  117  1  id.  Mean value for POC  POC  296 ± 63a    8  id.  Mean value for OC  OC  266 ± 27a    3  id.  Wilkes Land Basin (WLB)        6    On GA228–27  POC  333a  143  1  id.  On GA228–26  POC  336a  163  1  id.  On GA228–26  OC  208a  57  1  id.  On GA228–25  POC  270a  60  1  id.  On GA228–24  POC  298a  129  1  id.  On GA228–23  POC  256a  154  1  id.  On GA228–22  POC  313a  83  1  id.  On GA228–22  OC  269a  50  1  id.  Mean value for POC  POC  301 ± 33a    6  id.  Mean value for OC  OC  239 ± 43a    2  id.    Inferred crust type  RMS height (m)  Profile length (km)  Number of profiles  References  Tagus Abyssal Plain (TAP)        1    Shallow subdomain on IAM5 line  OC  456a  60  1  (Afilhado et al.2008)  Deep subdomain on IAM5 line  OC  414a  105  1  id.  Both domains on IAM5 line  OC  520a  165  1  id.  Iberia Abyssal Plain (IAP)        3    West of R4 on IAM9 line  OC  429a  80  1  (Pickup et al.1996)  East of R4 on IAM9 line  EM  368a  115  1  id.  West of R3 on Lusigal12 line  OC  566a  120  1  (Beslier 1996)  West of R3 on SO-22 line  OC  534a  160  1  (Pickup et al.1996)  Mean Value for OC  OC  510 ± 72a    3  id.  Galicia Margin (GM)            West of Peridotite Ridge on line WE1  OC  583a  85  1  (Dean et al.2015)  Northern Newfoundland Basin (NNFB)        5    East of M0 on SCREECH 2  OC  449a  75  1  (Shillington et al.2006)  West of M0 on SCREECH 2  EM  410a  120  1  id.  West of M3 on SCREECH 2  EM  268a  75  1    East of M0 on SCREECH 107/201  OC  355a  80  1  id.  East of M0 on SCREECH 302/109  OC  374a  84  1  id.  East of M0 on SCREECH E56  OC  320a  75  1  (Welford et al.2010b)  West of M0, on SCREECH E56  EM  378a  96  1    East of M0 on SCREECH E54  OC  320a  144  1  id.  Mean value for OC  OC  364 ± 53a    5  id.  Mean value for EM  EM  394 ± 23a    3  id.  Southern Margin of Flemish Cap (SMFC)        1    On line E36  OC  380a  93  1  (Welford et al.2010b)  Northern Margin of Flemish Cap (NMFC)        2    On line F87–4  OC  215a  85  1  (Welford et al.2010a)  On line F87–3  OC  322a  87  1  id.  Goban Spur Margin (GS)        1    On WAM line  OC  216  50  1  (Bullock & Minshull 2005)  On WAM  EM  128  67  1  id.  Bay of Biscay (BB)        3    On Norgasis 11–12  OC  315a  65  1  (Thinon et al.2003)  On Norgasis 14  OC  275a  64  1  id.  On IAM12  OC  328a  95  1  (Tugend et al.2014)  Great Australian Bight (GAB)        8    On GA199–10  POC  231a  56  1  (Gillard et al.2015)  On GA199–10  OC  258a  102  1  id.  On GA199–07  POC  251a  61  1  id.  On GA199–06  OC  243a  96  1  id.  On GA199–06  POC  362a  61  1  id.  On GA199–05  POC  262a  82  1  id.  On GA199–04  POC  216a  82  1  id.  On GA199–03  POC  374a  99  1  id.  On GA199–02  POC  356a  95  1  id.  On GA199–01  POC  313a  106  1  id.  On GA199–01  OC  296a  117  1  id.  Mean value for POC  POC  296 ± 63a    8  id.  Mean value for OC  OC  266 ± 27a    3  id.  Wilkes Land Basin (WLB)        6    On GA228–27  POC  333a  143  1  id.  On GA228–26  POC  336a  163  1  id.  On GA228–26  OC  208a  57  1  id.  On GA228–25  POC  270a  60  1  id.  On GA228–24  POC  298a  129  1  id.  On GA228–23  POC  256a  154  1  id.  On GA228–22  POC  313a  83  1  id.  On GA228–22  OC  269a  50  1  id.  Mean value for POC  POC  301 ± 33a    6  id.  Mean value for OC  OC  239 ± 43a    2  id.  aDetermined in this paper. View Large 4.1 The Iberia margin Because the J magnetic anomaly is thought to be located at the boundary between exhumed mantle and initial oceanic crust in the IAP, it has been proposed to mark the limit of the oceanic domain in the whole Iberia–Newfoundland system (Bronner et al.2011; Stanton et al.2016). However, (Nirrengarten et al.2017) recently demonstrated that the J anomaly is the result of polygenic and multiple magmatic events occurring during and after the formation of initial oceanic crust. We therefore built on the COB proposed by Nirrengarten et al. (2017) that uses a combination of geophysical and geological arguments. 4.1.1 The Tagus Abyssal Plain (TAP) In the TAP a two-layer igneous oceanic crust was recognized at the IAM5 MCS line and the wide-angle data by (Afilhado et al.2008). A sharp step of the top basement marks the limit between a westernmost shallow subdomain and a deeper easternmost subdomain (see Plate 1 of Afilhado et al.2008). MCS reflections were inferred between 10 and 11 s TWT, consistent with the Moho depth obtained from velocity models (Afilhado et al.2008). The inferred crustal thickness of the initial oceanic crust increased westward from 5 km on average in the deep subdomain to 7 km on average in the shallow one (Afilhado et al.2008). Therefore, (Nirrengarten et al.2017) took the basement step between the two subdomains as indicating the transition to an unambiguous magmatic oceanic crust. Excluding the western termination of IAM5 line at the volcanic Tore Madeira rise, we estimated the roughness values in the shallow and the deep subdomains to 456 and 414 m, respectively (Table 2). Analysing the shallow and deep subdomains together leads logically to a higher roughness value of 520 m. 4.1.2 The Iberia Abyssal Plain (IAP) Three MCS lines, IAM9, SO-22 (Pickup et al.1996) and Lusigal12 (Beslier et al.1996), reach the oceanic domain in the IAP. Constrains are given by refraction data and several drill holes drilled during ODP Legs 149 and 173 (Dean et al.2000; Whitmarsh et al.2001; Minshull et al.2014). The refraction data show that the inferred initial oceanic crust is 6–7 km thick (Minshull et al.2014). The transition between the initial oceanic crust to the west and the exhumed mantle domain to the east is marked by a series of peridotite ridges on MCS lines (Pickup et al.1996). The peridotite ridges called R3 and R4 by (Pickup et al.1996) mark the beginning of the westward shallowing of the top basement on the MCS lines. The roughness mean value for the initial oceanic crust to the west of these ridges is estimated at 510 m (Table 2). Taking anomaly J instead of the peridotite ridge R3 as the COB on the SO-22 profile results in a similar roughness value for initial oceanic crust (520 m versus 534 m using R3) although the sub-section is 50 km shorter. Because the exhumed mantle domains on the Lusigal12 and SO-22 lines were too short, we estimate the roughness over the exhumed mantle domain (at 368 m) only using the IAM9 profile. 4.1.3 The deep Galicia margin As in the IAP the transition between the initial oceanic crust and the exhumed mantle domain is marked by a series of peridotite ridges. On MCS line WE1 (Dean et al.2015) suggested a gradual transition to seafloor spreading westward of the ‘Peridotite Ridge’ (Fig. 3) where the top basement begins to shallow. There, recently published refraction data suggest a very thin oceanic crust thickening seaward (∼0.5–1.5 km; Davy et al.2016). Oceanic layer 3 is absent and the crust is thought to overlie serpentinized mantle material (Davy et al.2016). The roughness value estimated to the west of the Peridotite Ridge is 583 m (Fig. 3). Figure 3. View largeDownload slide Top basement relief in initial oceanic crust alongside magma-poor rifted margins. Roughness (R) is estimated for oceanic crust (OC) and proto-oceanic crust (POC) along MCS profiles of the (a) Galicia margin (profile WE1 from Dean et al.2015); (b) Newfoundland margin (profile Erable54 from Welford et al.2010b); (c) Northern margin of Flemish Cap (profile F87–3 from Welford et al.2010a); (d) Goban Spur margin (profile WAM from Bullock & Minshull 2005); (e) Cantabrian margin in the Bay of Biscay (profile IAM12 from Tugend et al. (2014); (f) Great Australian Bight margin (profile GA199–06 from Gillard et al.2015); (g) Wilkes Land margin (profile GA228–22 from Gillard et al.2015). Vertical exaggeration is 2×. Figure 3. View largeDownload slide Top basement relief in initial oceanic crust alongside magma-poor rifted margins. Roughness (R) is estimated for oceanic crust (OC) and proto-oceanic crust (POC) along MCS profiles of the (a) Galicia margin (profile WE1 from Dean et al.2015); (b) Newfoundland margin (profile Erable54 from Welford et al.2010b); (c) Northern margin of Flemish Cap (profile F87–3 from Welford et al.2010a); (d) Goban Spur margin (profile WAM from Bullock & Minshull 2005); (e) Cantabrian margin in the Bay of Biscay (profile IAM12 from Tugend et al. (2014); (f) Great Australian Bight margin (profile GA199–06 from Gillard et al.2015); (g) Wilkes Land margin (profile GA228–22 from Gillard et al.2015). Vertical exaggeration is 2×. 4.2 The Newfoundland margin We focus here on the Northern Newfoundland Basin (NNFB) and on the Southern margin of Flemish Cap (SMFC). Because oceanic crust is overprinted by post breakup magmatism in the Newfoundland Seamounts area, we disregard the Southern Newfoundland Basin. 4.2.1 The Northern Newfoundland Basin (NNFB) Refraction data in the NNFB show that the landward limit of a 3 km thick oceanic crust falls close to the eastern end of the J anomaly (interpreted as M0; van Avendonk et al.2006; Welford et al.2010b). Layer 3 oceanic crust is interpreted to be missing with layer 2 directly overlying serpentinized mantle (van Avendonk et al.2006). Normal oceanic crust of 6 km thickness is thought to be found at the seaward ends of the MCS lines of the SCREECH project, but a continuous Moho reflection is not observed. As along the Iberia margin, the transition between the oceanic crust and the exhumed mantle domain is marked by a series of peridotite ridges on the MCS lines. There, at ODP Site 1277, drilling into basement revealed gabbro and basalt fragments as well as serpentinized peridotites of subcontinental origin (Tucholke & Sibuet 2007). We use three MCS lines of the SCREECH project (SCREECH 2, 107/201 and 302/109; Shillington et al.2006) and E56 and E54 lines of the Erable project in the NNFB (Welford et al.2010b; Fig. 3). The mean roughness value for the oceanic crust east of M0 is estimated at 364 m while the mean roughness value for the exhumed mantle domain west of M0 is estimated at 394 m (Table 2). Moving the COB to the western end of the J anomaly (interpreted as M3) as suggested by (Shillington et al.2006) on SCREECH 2 profile results in a similar estimation for the roughness of the oceanic crust than using the eastern end of the J anomaly (458 m and 449 m, respectively), although the sub-section is 50 km shorter. Excluding the outer high drilled by the IODP Site 1277 that has been formed after the exhumation stage (Bronner et al.2011) leads to much lower roughness values for the exhumed mantle domain (268 m on SCREECH 2). 4.2.2 The Southern Margin of Flemish Cap (SMFC) At the SMFC, (Hopper et al.2004) interpreted the last ∼80 km at the seaward end of the MCS line SCREECH 1 as oceanic crust whereas a recent reinterpretation suggested a much narrower oceanic domain (<15 km; Gillard et al.2016b). Both interpretations are compatible with the seismic refraction data along this line (Funck et al.2003). Instead, we use here the Erable E36 MCS line that enters the initial oceanic domain over >90 km to the east of the J anomaly according to (Welford et al.2010b). Roughness is estimated at 380 m over this domain (Table 2). Distinguishing anomalously thin oceanic crust (with 3 km thickness) and normal oceanic crust (with 6 km thickness (Funck et al.2003; Hopper et al.2004)) along E36 line brings slightly lower roughness values of 351 and 344 m, respectively. 4.3 The Northern margin of Flemish Cap (NMFC) and the Goban Spur (GS) conjugate margin The Flame Lines F87–3 and F87–4 cross magnetic anomaly 34 at the NMFC and extend another ∼50 km seaward well into oceanic crust (Welford et al.2010a). Refraction data along profile F87–3 show a two-layer oceanic crust that is somewhat thinner (4.5–6.5 km) than typical oceanic crust (Gerlings et al.2011). Roughness values for the oceanic crust are estimated at 322 and 215 m along F87–3 and F87–4 profiles, respectively (Table 2). The Western Approaches Margin (WAM) line on the GS margin is thought to be conjugate of the F87–3 line (Welford et al.2010a; Gerlings et al.2012). The Flemish Cap and GS conjugate margins form an asymmetrical pair: the exhumed mantle domain is ∼100 km wide off GS (Bullock & Minshull 2005) while it is narrower (∼25 km) on the NMFC (Welford et al.2010a). The exhumed mantle domain off GS has a low relief basement (with a very low roughness of 128 m, Table 2) that has been interpreted as a detachment surface (Whitmarsh et al.2001). In the region of initial oceanic crust off GS, the basement is shallower and has a more subdued relief than that off Flemish Cap along F87–3 line. There, (Bullock & Minshull 2005) estimated the roughness value for a ∼6 km thick initial oceanic crust off GS at 216 m (Fig. 3). A similar low roughness value (215 m) and a normal thickness oceanic crust (inferred from gravity data; Welford et al.2010a) are found at the seaward end of the F87–4 profile (Fig. 3). 4.4 The Bay of Biscay (BB) The BB is characterized offshore by an oceanic basin where spreading began at Aptian-Albian time and stopped at Santonian time (e.g. Tugend et al.2014, and references therein). We use Norgasis 11–12 and 14 MCS lines across the Armorican Margin (Thinon et al.2003; Tugend et al.2014, 2015) and the IAM12 MCS line across the North Iberian Margin (Alvarez-Marron et al.1997; Tugend et al.2014) excluding the failed ridge axis and areas too close to the compressional accretionary prism (Fernández-Viejo et al.2012). Roughness values were estimated at 275 and 315 m for the Norgasis lines and at 328 m for the IAM12 line (Fig. 3). Refraction data along Norgasis 14 line show a 3–5 km thick ocean crust at the southern end of the line while a normal ∼6 km thick oceanic crust is found at the northern end of IAM12 line (Fernández-Viejo et al.1998; Thinon et al.2003). 4.5 The Australia-Antarctic margin system There is an ongoing debate on the location and age of the first undisputable oceanic crust within each sector of the Australian-Antarctic margin system (Tikku & Cande 1999; Tikku & Direen 2008). We focus here on the Great Australian Bight (GAB)—Wilkes Land Basin (WLB) system using Australian MCS lines from the AGSO surveys GA199 and GA228 (e.g. Sayers et al.2001). In both the GAB and the WLB, Gillard et al. (2015) used these MCS lines to distinguish a ‘proto-oceanic’ domain from a ‘steady-state’ oceanic domain. This so-called proto-oceanic domain is interpreted as composed of hybrid crust (exhumed mantle and magmatic rocks) and formed as a non-steady state, not localized seafloor spreading. The steady-state oceanic domain is formed from stable oceanic accretion. The proto-oceanic crust is characterized by a deformed base of the sedimentary coverture, the presence of volcanic additions sealing faults, and the rise of the top basement arguing for an increase of magmatism. The oceanic Moho is absent in the proto-oceanic crust but is observed locally at ∼10 s (TWT) in the steady state oceanic crust (Gillard et al.2015). However, by contrast to the Iberia–Newfoundland system, the nature of the basement is ill defined at the GAB/WLB system as neither refraction data nor samples can confirm the inferred basement type. For the GAB, the mean roughness values of the top basement on eight MCS lines in the proto-oceanic and steady-state oceanic domains are estimated at 296 and 266 m, respectively (Table 2, Fig. 3). In the basin off the Wilkes Land we used six MCS lines to estimate the mean roughness values at 301 and 239 m for the proto-oceanic and steady-state oceanic domains, respectively (Table 2, Fig. 3). Within the initial oceanic crust alongside both margins, we exclude large conic-shaped edifices that we interpreted as resulting from post-rift volcanic activity (Fig. 3). As a conclusion of this section, we note that, as for ultraslow-spreading ridges, the roughness estimations obtained over initial oceanic domain alongside magma-poor rifted margins cover a wide range of values, from ∼210 m in the Wilkes Land Basin to ∼580 m off the Galicia margin. 5 COMPARISON OF OCEANIC BASEMENT ROUGHNESS AT MID-OCEAN RIDGES AND OVER INITIAL OCEANIC DOMAIN ALONGSIDE MAGMA-POOR RIFTED MARGINS 5.1 Clues from mid-oceanic ridges 5.1.1 Top basement roughness versus spreading rate It is well known that the elastic lithosphere along slow-spreading ridges is relatively thicker than at faster spreading ridges (Chen & Morgan 1990) and is thus able to sustain greater relief through faulting and flexure (Shaw & Lin 1996). The 750 °C isotherm obtained by thermal modelling for fast to slow-spreading ridges correlates well with the spreading-rate dependence inferred from teleseismic axial earthquake depths (from 2 to 10 km depth; Phipps Morgan & Chen 1993). A recent study of the seismicity along ridge sections of the SWIR and Arctic ridges revealed that earthquakes reach maximum depths of 15 km below volcanic segments and as much as 35 km below the sea floor in exhumed mantle domains (Schlindwein & Schmid 2016). This shows that the lithosphere can be much thicker at ultraslow-spreading ridges than at slow ones. Early attempts to analyse the relationship between the roughness of oceanic seafloor and spreading rate suffered from the lack of data for ultraslow-spreading crust (Small 1994; Goff et al.1997). The new roughness values of seafloor collected in this paper supplement the global data for the ultraslow end of the spectrum. Fig. 4 shows these values compared to the power-law curve fit of (Malinverno 1991) obtained for slow to fast spreading rates. (Ehlers & Jokat 2009) noted that the roughness values obtained for the Arctic ridges were systematically above this reference curve and thus proposed an updated curve fit for slow and ultraslow-spreading crust. The mean roughness values obtained at the ultraslow-spreading SWIR, MCSC, MAR, SPR, Arctic and Sheba ridges range from 100 to >500 m and fall well above or below these two reference curves. The roughness of the flanks of the SPR and the Sheba ridge are even lower than 200–240 m, which is typical of slow-spreading ridges (e.g. at the MAR; Goff et al.1997). This scattering supports that, although the oceanic basement roughness generally increases as the spreading rate decreases from slow to ultraslow, this variation is not solely related to spreading rate (e.g. Goff et al.1997). Figure 4. View largeDownload slide Roughness versus spreading rate for ultraslow and slow-spreading ridges. Data for ultraslow-spreading ridges are given in Table 1. Those for the slow-spreading Mid Atlantic Ridge are from (Goff 1991; Goff et al.1995; Neumann & Forsyth 1995; Min 1999). The roughness value for the slow-spreading SWIR (before 24 Ma) is from Sauter et al. (2011). The continuous grey line is the power-law curve fit of (Malinverno 1991) and the dashed grey line is the curve fit of Ehlers & Jokat (2009). Figure 4. View largeDownload slide Roughness versus spreading rate for ultraslow and slow-spreading ridges. Data for ultraslow-spreading ridges are given in Table 1. Those for the slow-spreading Mid Atlantic Ridge are from (Goff 1991; Goff et al.1995; Neumann & Forsyth 1995; Min 1999). The roughness value for the slow-spreading SWIR (before 24 Ma) is from Sauter et al. (2011). The continuous grey line is the power-law curve fit of (Malinverno 1991) and the dashed grey line is the curve fit of Ehlers & Jokat (2009). 5.1.2 Top basement roughness versus magma supply A weak correlation between spreading rate and bathymetric roughness was also observed at slow-spreading centres, while roughness is relatively constant for intermediate to fast spreading ridges (Small 1994). This weak correlation has been attributed to the along-axis variable thickness of magmatic crust resulting in a more heterogeneous bulk rheology of the lithosphere at slow-spreading ridges than at faster ridges (Small 1994). Along the eastern SWIR, the west-to-east increase of the roughness toward the magma-poor easternmost section (Fig. 4) correlates well with increasing MBA values indicating progressively thinner magmatic crust and/or cooler mantle temperatures (Sauter et al.2011). Either case should produce progressive strengthening of the lithospheric mantle: reduction of the thickness of the magmatic crust relative to the rigid upper mantle will strengthen the lithosphere while the decrease in mantle temperature should produce increasing lithosphere thickness since elastic strength of the lithosphere is dependent upon thermal structure. On the Sheba ridge, roughness is also larger in the ridge section where the crust is thinner (d’Acremont et al.2010). However, no such correlation have been found for the Arctic ridges (Ehlers & Jokat 2009) and no well constrained thickness estimations are available for the MSCS. Despite few estimations of crustal thickness along ultraslow ridges, Fig. 5 shows a weak tendency for increasing roughness when magmatic crustal thickness decreases. The wide range of crustal thickness at ultraslow-spreading ridges shows that another process than enhanced cooling of the mantle when spreading rate decrease is required. A thickening of the thermal boundary layer at those ridges does reduce the height of the melting regime and thus the magma budget (White et al.2001). However, this dismisses the importance of varying mantle source composition relative to temperature (Zhou & Dick 2013). At increasingly slower spreading rates and lower extents of melting, more fertile mantle heterogeneities are sampled preferentially which may lead to greater crustal thickness variations for ultraslow-spreading ridges and thus also to greater variation in surface roughness. On the lower end of the roughness spectra, short wave heterogeneities in the back-arc mantle of the North-Fiji basin has been inferred to explain the segmentation of the SPR (Garel et al.2003). There, the low roughness may be partly explained by a smooth seafloor resulting mainly from the coalescence of numerous volcanoes (Lagabrielle et al.1996; Garel et al.2003). On the Sheba ridge, the western area appears to be affected by excess magmatism (Leroy et al.2010a) that may explain the very low roughness values there. Although error bars on crustal thickness are large, this tendency for increasing roughness when crustal thickness decreases makes sense as thinner magmatic crust, hence a greater mantle component in the lithosphere, would imply greater lithospheric strength. Roughness values larger than ∼300 m may thus be characteristic of magma-poor sections of ultraslow-spreading ridges. Figure 5. View largeDownload slide Roughness versus thickness of the magmatic crust for ultraslow-spreading ridges. Crustal thicknesses for the SWIR are from Cannat et al. (2008), those for the Sheba ridge are from d’Acremont et al. (2010), the one for the MAR is from Lizarralde et al. (2004), the one from the Arctic ridges is from Jokat & Schmidt-Aursch (2007) and Ehlers & Jokat (2009), the one for the SPR is from Garel et al. (2003). Figure 5. View largeDownload slide Roughness versus thickness of the magmatic crust for ultraslow-spreading ridges. Crustal thicknesses for the SWIR are from Cannat et al. (2008), those for the Sheba ridge are from d’Acremont et al. (2010), the one for the MAR is from Lizarralde et al. (2004), the one from the Arctic ridges is from Jokat & Schmidt-Aursch (2007) and Ehlers & Jokat (2009), the one for the SPR is from Garel et al. (2003). This tendency for higher roughness values in thinner crust areas is further complicated by the fact that the mean roughness value for exhumed mantle domains at the SWIR is slightly lower (304 m) than for adjacent magmatic crust areas (372 m; Sloan et al.2012). This is not expected as lithosphere should be thinner and weaker where magma supply is large and focused toward volcanic segment centres, and thicker and stronger where the magma supply is reduced or nearly absent (Cannat 1996). This apparent contradiction may be explained by considering the differences in both rheology and tectonic deformation style of magmatic segments and exhumed mantle domains. Within the magmatic segments, the volcanic seafloor composition is inferred to include the full complement of oceanic lithosphere components (i.e. volcanic and intrusive magmatic rocks overlying upper mantle) that tend to accommodate extension by high-angle normal faults. By contrast, the exhumed mantle domains are inferred to consist largely of serpentinized peridotites above unaltered mantle in which the formation of low-angle, long offset detachment faults is more common (Sauter et al.2013). This is supported by the distribution of the seismicity between magmatic segment and exhumed mantle domains at ultraslow-spreading ridges. At the SWIR the magmatic segments, exhibit brittle deformation throughout the lithosphere up to 8–15 km depths in the upper mantle (Schlindwein & Schmid 2016). By contrast, although earthquakes are observed as deep as 35 km depths, the upper lithosphere is entirely aseismic up to 10 km depth along the adjacent peridotite-dominated ridge sections (Schmid & Schlindwein 2016). This observation of an upper aseismic zone has been interpreted as probably reflecting substantial serpentinization within the exhumed mantle domains of ultraslow-spreading ridges (Schmid & Schlindwein 2016). The variation of oceanic basement roughness at ultraslow-spreading ridges is thus a complex trade-off between spreading rate that highly controls the thermal state of the lithosphere and its composition controlling the rheology and thus also the strength of the lithosphere. 5.2 Insight into the initial seafloor spreading The roughness values determined over inferred initial oceanic domains are all higher than 200 m (Fig. 6), a typical value for slow-spreading ridges or magmatic sections of ultraslow-spreading ridges like the SPR or the Sheba Ridge (Fig. 4). Mean roughness values of the inferred initial oceanic domains bordering the Iberia and Newfoundland margins are greater than 300 m, which is characteristic of magma-poor sections of ultraslow-spreading ridges. Figure 6. View largeDownload slide Roughness of initial oceanic basement alongside Iberia/Newfoundland, Flemish Cap/Goban Spur, Bay of Biscay, Australian Great Bight/Wilkes Land Basin margins. Open hexagons and vertical black lines correspond to the mean values and standard deviations, respectively, for each study area. EM: exhumed mantle. Roughness values and references are given in Table 2. Figure 6. View largeDownload slide Roughness of initial oceanic basement alongside Iberia/Newfoundland, Flemish Cap/Goban Spur, Bay of Biscay, Australian Great Bight/Wilkes Land Basin margins. Open hexagons and vertical black lines correspond to the mean values and standard deviations, respectively, for each study area. EM: exhumed mantle. Roughness values and references are given in Table 2. 5.2.1 The Iberian/Newfoundland margin system Roughness >500 m, as high as those found at the Arctic ridges (Ehlers & Jokat 2009), are found alongside the Iberia margin. There, oceanic domains with either thin crust or normal crustal thickness show high roughness values in the same way as along the Arctic ridges. Although significantly smaller than alongside the Iberia margin, roughness values alongside the Newfoundland margin are as large as those determined at the easternmost magma-poor SWIR. As these margins are conjugate, initial oceanic crust was emplaced at the same spreading centre and thus higher roughness values next to the Iberia margin cannot be explained by a lower spreading rate. Accretion may have been asymmetric with more detachment faulting eastward and more magmatic accretion westward as it is suggested on the MCSC (Hayman et al.2011). However, there is no obvious difference between the basement morphologies in the inferred initial oceanic domains alongside the two margins. The overall different basement topography of the shallower Newfoundland margin compared to the deeper Iberia margin has been tentatively partly attributed to an ‘inherited buoyancy’ of the continental mantle at the Newfoundland margin (Müntener & Manatschal 2006). There, high degrees of melting formed a continental mantle lithosphere that might be compositionally buoyant (Müntener & Manatschal 2006). This hypothesis is consistent with the compositional differences between the Iberia and Newfoundland continental mantle derived peridotites (Müntener & Manatschal 2006). However, although the composition of the rising asthenospheric mantle is unknown, it is likely that its density structure is different from that of the adjacent continental mantle. A more homogenous density structure is rather expected below the two flanks of the initial oceanic ridge, which thus, could not explain a different strength of the initial oceanic lithosphere on the two margins. Alternatively, an explanation may be found in the widespread magmatism, which has been found in the distal domain of the Newfoundland margin (Péron-Pinvidic et al.2010). There, by contrast to the Iberia margin, a major magmatic event, which post-dated the onset of seafloor spreading by at least 7–15 Myr (Hart & Blusztajn 2006), resulted in emplacement of sills over an area of ∼80 000 km2 above the exhumed mantle (Péron-Pinvidic et al.2010). We therefore speculate that smoother initial oceanic basement alongside the Newfoundland margin is related to the weakening of the initial oceanic lithosphere by this late magmatic event. The source of the magma is still uncertain and resolving this issue will require a combination of further geophysical studies and drilling (Péron-Pinvidic et al.2010). Anyhow, we stress that initial oceanic crust alongside both margins exhibit roughness values on the higher end of the spectrum of ultraslow-spreading ridges regardless of the thickness of this initial oceanic crust. Although basement ridges display larger relief next to the Iberia margin, these ridges are on both margins generally symmetric, either with a domal or a triangular shape, without a clear oceanward faulting pattern that would characterize an oceanic magmatic crust with a normal thickness. Landward-dipping reflectors interpreted as faults are rather found on both the exhumed mantle domain of each margin and over the oceanic crust next to it (e.g. Welford et al.2010b; Minshull et al.2014). These structures were interpreted as abandoned fault roots resulting from successive mantle unroofing episodes (Boillot & Froitzheim 2001) as new detachments repeatedly cut through the footwall of the preceding detachment (Reston & McDermott 2011), a ‘flip-flop’ mechanism that has been later shown to be active at the ultraslow-spreading SWIR (Sauter et al.2013). 5.2.2 The Flemish Cap/Goban Spur margin system By contrast to the Iberia–Newfoundland margin, the top basement in the initial oceanic crust alongside the Flemish Cap and GS conjugate margins shows a tilted block morphology typical of slow-spreading ridges (Bullock & Minshull 2005; Gerlings et al.2012). There, oceanward faults were identified in inferred oceanic domains displaying a 215 m roughness value, typical of slow-spreading ridges. A more irregular pattern of dome shaped ridges results in a rougher basement along the F87–3 line (322 m). Gerlings et al. (2012) already noted that the relief of the initial oceanic basement was rougher along the F87–3 line than on the conjugate GS line (216 m). They suggested that the initial igneous accretion was more dominant on Flemish Cap than on its conjugate. We rather argue that, although the velocity (5.8–7 km s–1) of the lower crust at the inferred initial oceanic domain of GS is ambiguous and could support either a gabbroic or serpentinized composition, the roughness of the top basement, its morphology and the ∼6 km thickness of the inferred initial crust indicate typical slow seafloor spreading. The irregular pattern of dome shaped ridges along the F87–3 line may result from asymmetric spreading with oceanic core complex development westward while volcanism dominates eastward on the GS margin. Such asymmetry is often observed locally on slow-spreading ridges, close to transform faults, where magma supply is reduced relatively to the segment centres. 5.2.3 The Bay of Biscay margin system Strong constraints are lacking to clearly determine the accretion processes that produced the initial oceanic domain in the BB. Roughness values are similar to some sections of the ultraslow-spreading SWIR, MCSC or MAR. The top basement is irregular showing either domes or symmetric ridges without tilted blocks typical of slow-spreading crust, which would indicate an ultraslow-spreading. However, compressional structures are observed in the sediment cover showing that some faults were slightly reactivated during the Iberian/European convergence (Thinon et al.2001; Tugend et al.2014). The extent of the resulting change of roughness and morphology is not yet constrained. Moreover, the crustal thickness is highly variable with overthickened crust in some places (Tugend et al.2015). Therefore, we stress that more MCS data extending well into the inferred oceanic domain are needed to resolve the initial accretion processes in the BB. 5.2.4 The Australian Great Bight/Wilkes Land Basin margin system Although there is some overlap between the ranges of roughness values, the mean values are significantly different for the proto-oceanic crust (296–300 m) and steady state oceanic crust (266–239 m) defined by (Gillard et al.2015) at the Great Australian Bight and the Wilkes Land Basin, respectively (Table 2 and Fig. 6). Reflections that could correspond to the oceanic Moho are only locally observed below steady state oceanic crust, leading to a normal thickness (∼2 s TWTT) estimation. This oceanic domain was therefore rather defined by the highly reflective top basement on the MCS data and by the absence of deformation at the base of the sediment cover (Gillard et al.2015). Roughness values of the steady state oceanic crust, only slightly larger than those typical of slow-spreading crust, are relatively low for ultraslow-spreading crust. As in the BB, the top basement is irregular without successive ridgeward facing faults typical of slow-spreading crust. By contrast, at both Australian and Antarctic margins, high structures associated with ridgeward-dipping intrabasement reflections were interpreted as oceanic core complexes produced by detachment faults (Gillard et al.2015). Numerical models show that conditions of high or very low magma supply do not favour detachment faulting and that core complexes occur with intermediate magma supply (Tucholke et al.2008). We therefore argue that both the roughness and the morphology of the top basement indicate large tectonic extension and intermediate magma supply with either slow or ultraslow-spreading rate. We discuss the roughness of the proto-oceanic crust in the next section. 5.3 Proto versus initial oceanic basement The proto-oceanic crust defined alongside the Australian and Antarctic margins is inferred to correspond to serpentinized peridotites that are intensively faulted and intruded (Gillard et al.2016a). It differs from the exhumed mantle domain by oceanward increasing volcanic additions sealing the faulted top basement and intruding or underplating this basement (Gillard et al.2016a). This type of hybrid crust with mantle derived rocks and magmatic rocks, might thus be similar to the exhumed mantle domains between initial oceanic crust and hyperextended continental crust at the Newfoundland-Iberia or Flemish Cap-GS margin systems. However, the amount of melt may be highly variable from one margin to the other or even along the margins themselves (Fig. 1). Roughness values are highly variable in both exhumed mantle domains and proto-oceanic crust ranging from low values (<130 m on GS margin (Bullock & Minshull 2005)) to very high values (>400 m on Newfoundland margin). This dispersion of the roughness values may result from variable extent of post rift volcanism, from highly intruded basement at the Australian and Antarctic margins resulting in a relatively rough seafloor, to a smoother exhumed mantle domain with little or no evidence for melting on GS margin (Bullock & Minshull 2005). However, although the Iberia margin (including Galicia bank) is part of the magma-poor class of margins, the high roughness there is not produced by volcanic edifices but by high relief peridotite ridges inferred to result from successive faulting episodes (e.g. Reston & McDermott 2011). On the Newfoundland margin relatively smooth parts made of exhumed mantle are adjacent to much rougher basement. There, the prominent peridotite ridge drilled at ODP Site 1277 is thought to result from a significant tectonic uplift accompanied by a magmatic event that may have triggered the onset of seafloor spreading (Bronner et al.2011). This illustrates the fundamental difference in nature between proto-oceanic crust or exhumed mantle domain and steady state oceanic crust. While deformation and magmatism are localized at oceanic ridges, multiple tectonic and magmatic episodes may have affected the whole proto-oceanic/exhumed mantle domains leading to highly variable top basement roughness. 6 FURTHER IMPLICATIONS The underlying issue of the determination of the spreading rate at nascent spreading centres is to understand if there, the accretion processes are somehow related to the processes of lithospheric breakup. If spreading is ultraslow alongside every magma-poor rifted margins it could be tempting to suggest that the ultraslow-spreading rate is a consequence of the breakup processes at magma-poor rifted margins. However, among the examples analysed in this paper, only the Iberia–Newfoundland margin system shows clear evidence for initial ultraslow oceanic spreading. Disputable magnetic anomalies cannot be used to estimate the spreading rate alongside these margins and corroborate this result (Barnett-Moore et al.2016; Nirrengarten et al.2017). However, a new palinspastic restoration of the southern North-Atlantic based on 3D gravity inversion, which does not take into account disputed magnetic anomalies but which integrates the rift deformation (Sutra et al.2013), allows to estimate the initial spreading rate alongside these margins (Nirrengarten et al.submitted). Ultraslow-spreading rates <17 mm yr−1 were obtained between 112 and 87Ma at the onset of accretion before reaching slow-spreading rate (>22 mm yr−1; Nirrengarten et al.submitted). This early evolution of the spreading rate between Iberia and Newfoundland is related to the motion of Iberia at that time, resulting in the propagation of seafloor spreading in the BB. When spreading stopped in the BB during Santonian time, oceanic spreading was no more distributed along three branches of a triple junction but became focused at the mid-Atlantic ridge. We therefore suggest that similarly to present day oceanic ridges, spreading at the initial oceanic ridge between Iberia and Newfoundland is controlled by global plate tectonics and is not related to the rifting processes leading to the final breakup. 7 CONCLUSIONS The investigation of the top basement roughness at ultraslow-spreading ridges and within the inferred initial oceanic domain alongside magma-poor rifted margins leads to the following conclusions. The mean roughness values obtained at ultraslow-spreading ridges range from 100 to >500 m and includes 200–240 m roughness values which are typical of slow-spreading ridges. There is a weak tendency for increasing roughness at ultraslow-spreading ridges when magmatic crustal thickness decreases. This is related to the rheology of the oceanic lithosphere; a thinner magmatic crust, hence a greater mantle component in the lithosphere, implying greater lithospheric strength. Roughness values larger than ∼300 m may be characteristic of magma-poor sections of ultraslow-spreading ridges. The top basement roughness values determined within the inferred initial oceanic domain alongside magma-poor rifted margins are all higher than 200 m, a typical value for slow-spreading ridges or magmatic sections of ultraslow-spreading ridges Mean roughness values of the inferred initial oceanic domains alongside the Iberia and Newfoundland margins are greater than 300 m which is characteristic of magma-poor sections of ultraslow-spreading ridges. The top basement in the initial oceanic crust alongside the Flemish Cap and GS margins shows roughness values and a tilted block morphology typical of slow-spreading ridges. Roughness values of the initial oceanic crust alongside the Australian-Antarctic margins are only slighter larger than those typical of slow-spreading crust. We suggest that the roughness and the morphology of the top basement along those margins indicate large tectonic extension and intermediate magma supply with either slow or ultraslow-spreading rate. Estimating roughness values within proto oceanic crust or exhumed mantle domains is less pertinent as polyphased tectonism and magmatism may have affected these domains leading to highly variable top basement roughness values that cannot be linked to a single monophase processes. Acknowledgements We acknowledge funding by ‘Actions Marges’ a TOTAL/INSU/IFP/IFREMER/BRGM program to DS and JA, and funding from Modelling of Margins Phase 3 and 4 consortium for JT. We thank Ségolène Vitoux, Elena Metz and Hugo Ketterer for their work that led to this paper. We also gratefully acknowledge the two anonymous reviewer and the Editor, Donna Blackman, for their review and helpful comments. REFERENCES Afilhado A., Matias L., Shiobara H., Hirn A., Mendes-Victor L., Shimamura H., 2008. From unthinned continent to ocean: the deep structure of the West Iberia passive continental margin at 38°N, Tectonophysics , 458, 9– 50. https://doi.org/10.1016/j.tecto.2008.03.00210.1016/j.tecto.2008.03.00210.1016/j.tecto.2008.03.002 Google Scholar CrossRef Search ADS   Alvarez-Marron J., Rubio E., Torne M., 1997. Subduction-related structures in the North Iberian Margin, J. geophys. Res. , 102, 22 497– 22 511. https://doi.org/10.1029/97JB0142510.1029/97JB0142510.1029/97JB01425 Google Scholar CrossRef Search ADS   Amante C., Eakins B.W., 2009. ETOPO1 1 Arc-minute global relief model: procedures, data sources and analysis, in NOAA Technical Memorandum NESDIS NGDC-24 , ed. NOAA, National Geophysical Data Center. Ball P.J., 2005. Break-up history and evolution of the southern passive margin of Australia, PhD thesis, Royal Holloway University, London. Barnett-Moore N., Hosseinpour M., Maus S., 2016. Assessing discrepancies between previous plate kinematic models of Mesozoic Iberia and their constraints, Tectonics , 35, 1843– 1862. https://doi.org/10.1002/2015TC00401910.1002/2015TC00401910.1002/2015TC004019 Google Scholar CrossRef Search ADS   Beslier M.O., 1996. Seismic Line LG12 in the Iberia Abyssal Plain, in Iberia Abyssal Plain , pp. 737– 739, eds Whitmarsh R.B., Sawyer D.S., Klaus A., Masson D.G., Ocean Drilling Program. Beslier M.O., Cornen G., Girardeau J., 1996. Tectono-metamorphic evolution of peridotites from the ocean/continent transition of the Iberia Abyssal Plain margin, in Proc. ODP, Sci. Results , pp. 397– 412, eds Whitmarsh R.B., Sawyer D.S., Klaus A., Masson D.G., Ocean Drilling Program. Google Scholar CrossRef Search ADS   Boillot G., Froitzheim N., 2001. Non-volcanic rifted margins, continental break-up and the onset of sea-floor spreading: some outstanding questions, Geol. Soc., London, Spec. Publ. , 187, 9– 30 Google Scholar CrossRef Search ADS   Bronner A., Sauter D., Manatschal G., Peron-Pinvidic G., Munschy M., 2011. Magmatic breakup as an explanation for magnetic anomalies at magma-poor rifted margins, Nat. Geosci. , 4, 549– 553. https://doi.org/10.1038/ngeo120110.1038/ngeo120110.1038/ngeo120110.1038/ngeo1201 Google Scholar CrossRef Search ADS   Bullock A.D., Minshull T.A., 2005. From continental extension to seafloor spreading: crustal structure of the Goban Spur rifted margin, southwest of the UK, Geophys. J. Int. , 163, 527– 546. https://doi.org/10.1111/j.1365-246X.2005.02726.x10.1111/j.1365-246X.2005.02726.x10.1111/j.1365-246X.2005.02726.x10.1111/j.1365-246X.2005.02726.x Google Scholar CrossRef Search ADS   Cannat M., 1993. Emplacement of mantle rocks in the seafloor at mid-ocean ridges, J. geophys. Res. , 98, 4163– 4172. https://doi.org/10.1029/92JB0222110.1029/92JB0222110.1029/92JB0222110.1029/92JB02221 Google Scholar CrossRef Search ADS   Cannat M., 1996. How thick is the magmatic crust at slow spreading oceanic ridges?, J. geophys. Res. , 101, 2847– 2857. https://doi.org/10.1029/95JB0311610.1029/95JB0311610.1029/95JB0311610.1029/95JB03116 Google Scholar CrossRef Search ADS   Cannat M., Sauter D., Mendel V., Ruellan E., Okino K., Escartin J., Combier V., Baala M., 2006. Modes of seafloor generation at a melt-poor ultraslow-spreading ridge, Geology , 34, 605– 608. https://doi.org/10.1130/G22486.110.1130/G22486.110.1130/G22486.110.1130/G22486.1 Google Scholar CrossRef Search ADS   Cannat M., Sauter D., Bezos A., Meyzen C., Humler E., Le Rigoleur M., 2008. Spreading rate, spreading obliquity, and melt supply at the ultraslow spreading Southwest Indian Ridge, Geochem. Geophys. Geosyst. , 9, Q04002, doi:10.1029/2007GC001676. https://doi.org/10.1029/2007GC00167610.1029/2007GC00167610.1029/2007GC00167610.1029/2007GC001676 Google Scholar CrossRef Search ADS   Cannat M., Manatschal G., Sauter D., Peron-Pinvidic G., 2009. Assessing the conditions of continental breakup at magma-poor rifted margins: What can we learn from slow spreading mid-ocean ridges?, C. R. Geosci. , 341, 406– 427 Google Scholar CrossRef Search ADS   Chen Y., Morgan W.J., 1990. Rift valley/no rift valley transition at mid-ocean ridges, J. geophys. Res. , 95, 17 571–17 581. https://doi.org/10.1029/JB095iB11p1757110.1029/JB095iB11p1757110.1029/JB095iB11p1757110.1029/JB095iB11p17571 Cunha T., 2008. Gravity Anomalies, Flexure and the Thermo-Mechanical Evolution of the West Iberia Margin and its Conjugate of Newfoundland, PhD thesis, University of Oxford, Oxford. D’Acremont E., Leroy S., Maia M., Gente P., Autin J., 2010. Volcanism, jump and propagation on the Sheba ridge, eastern Gulf of Aden: segmentation evolution and implications for oceanic accretion processes, Geophys. J. Int. , 180, 535– 551. https://doi.org/10.1111/j.1365-246X.2009.04448.x10.1111/j.1365-246X.2009.04448.x10.1111/j.1365-246X.2009.04448.x10.1111/j.1365-246X.2009.04448.x Google Scholar CrossRef Search ADS   Davy R.G. et al., 2016. Continental hyperextension, mantle exhumation and thin oceanic crust at the continent-ocean transition, West Iberia: new insights from wide-angle seismic, J. geophys. Res. , 121, 3177– 3199. Google Scholar CrossRef Search ADS   Dean S.M., Minshull T.A., Whitmarsh R.B., Louden K.E., 2000. Deep structure of the ocean-continent transition in the southern Iberia Abyssal Plain from seismic refraction profiles: The IAM-9 transect at 40°20΄N, J. geophys. Res. , 105, 5859– 5885. https://doi.org/10.1029/1999JB90030110.1029/1999JB90030110.1029/1999JB90030110.1029/1999JB900301 Google Scholar CrossRef Search ADS   Dean S.L., Sawyer D.S., Morgan J.K., 2015. Galicia Bank ocean–continent transition zone: New seismic reflection constraints, Earth planet. Sci. Lett. , 413, 197– 207. https://doi.org/10.1016/j.epsl.2014.12.04510.1016/j.epsl.2014.12.04510.1016/j.epsl.2014.12.04510.1016/j.epsl.2014.12.045 Google Scholar CrossRef Search ADS   Delescluse M., Funck T., Dehler S.A., Louden K.E., Watremez L., 2015. The oceanic crustal structure at the extinct, slow to ultra-slow Labrador Sea spreading center, J. geophys. Res. , 120, 5249– 5272. Google Scholar CrossRef Search ADS   Dick H.J.B., Lin J., Schouten H., 2003. An ultraslow-spreading class of ocean ridge, Nature , 426, 405– 412. https://doi.org/10.1038/nature0212810.1038/nature0212810.1038/nature0212810.1038/nature02128 Google Scholar CrossRef Search ADS PubMed  Ehlers B., Jokat W., 2009. Subsidence and crustal roughness of ultra-slow spreading ridges in the northern North Atlantic and the Arctic Ocean, Geophys. J. Int. , 177, 451– 462. https://doi.org/10.1111/j.1365-246X.2009.04078.x10.1111/j.1365-246X.2009.04078.x10.1111/j.1365-246X.2009.04078.x10.1111/j.1365-246X.2009.04078.x Google Scholar CrossRef Search ADS   Fernández-Viejo G., Gallart J., Pulgar J.A., Gallastegui J., Dañobeitia J.J., Córdoba D., 1998. Crustal transition between continental and oceanic domains along the North Iberian Margin from wide angle seismic and gravity data, Geophys. Res. Lett. , 25, 4249– 4252. https://doi.org/10.1029/1998GL90014910.1029/1998GL90014910.1029/1998GL90014910.1029/1998GL900149 Google Scholar CrossRef Search ADS   Fernández-Viejo G., Pulgar J.A., Gallastegui J., Quintana L., 2012. The fossil accretionary wedge of the Bay of Biscay: critical wedge analysis on depth-migrated seismic sections and geodynamical implications, J. Geol. , 120, 315– 331. https://doi.org/10.1086/66478910.1086/66478910.1086/66478910.1086/664789 Google Scholar CrossRef Search ADS   Fournier M., Patriat P., Leroy S., 2001. Reappraisal of the Arabia–India–Somalia triple junction kinematics, Earth planet. Sci. Lett. , 189, 103– 114. https://doi.org/10.1016/S0012-821X(01)00371-510.1016/S0012-821X(01)00371-510.1016/S0012-821X(01)00371-510.1016/S0012-821X(01)00371-5 Google Scholar CrossRef Search ADS   Funck T., Hopper J.R., Larsen H.C., Louden K.E., Tucholke B.E., Holbrook W.S., 2003. Crustal structure of the ocean-continent transition at Flemish Cap: seismic refraction results, J. geophys. Res. , 108( B11), 2531, doi:10.1029/2003JB002434. https://doi.org/10.1029/2003JB00243410.1029/2003JB00243410.1029/2003JB00243410.1029/2003JB002434 Google Scholar CrossRef Search ADS   Garel E., Lagabrielle Y., Pelletier B., 2003. Abrupt axial variations along the slow to ultra-slow spreading centers of the northern North Fiji Basin (SW Pacific): Evidence for short wave heterogeneities in a back-arc mantle, Mar. Geophys. Res. , 24, 245– 263. https://doi.org/10.1007/s11001-004-1060-y10.1007/s11001-004-1060-y10.1007/s11001-004-1060-y10.1007/s11001-004-1060-y Google Scholar CrossRef Search ADS   Gerlings J., Louden K.E., Jackson H.R., 2011. Crustal structure of the Flemish Cap Continental Margin (eastern Canada): an analysis of a seismic refraction profile, Geophys. J. Int. , 185, 30– 48. https://doi.org/10.1111/j.1365-246X.2011.04931.x10.1111/j.1365-246X.2011.04931.x10.1111/j.1365-246X.2011.04931.x10.1111/j.1365-246X.2011.04931.x Google Scholar CrossRef Search ADS   Gerlings J., Louden K.E., Minshull T.A., Nedimovic M.R., 2012. Flemish Cap-Goban Spur conjugate margins: New evidence of asymmetry, Geology , 40, 1107– 1110. https://doi.org/10.1130/G33263.110.1130/G33263.110.1130/G33263.110.1130/G33263.1 Google Scholar CrossRef Search ADS   Gillard M., 2014. Évolution tectono-magmatique menant à l’océanisation sur les marges passives pauvres en magma, exemple des marges Australie-Antarctique, PhD thesis, University of Strasbourg, Strasbourg. Gillard M., Autin J., Manatschal G., Sauter D., Munschy M., Schaming M., 2015. Tectonomagmatic evolution of the final stages of rifting along the deep conjugate Australian-Antarctic magma-poor rifted margins: constraints from seismic observations, Tectonics , 34, 753– 783. https://doi.org/10.1002/2015TC00385010.1002/2015TC00385010.1002/2015TC00385010.1002/2015TC003850 Google Scholar CrossRef Search ADS   Gillard M., Autin J., Manatschal G., 2016a. Fault systems at hyper-extended rifted margins and embryonic oceanic crust: Structural style, evolution and relation to magma, Mar. Pet. Geol. , 76, 51– 67. https://doi.org/10.1016/j.marpetgeo.2016.05.01310.1016/j.marpetgeo.2016.05.01310.1016/j.marpetgeo.2016.05.01310.1016/j.marpetgeo.2016.05.013 Google Scholar CrossRef Search ADS   Gillard M., Manatschal G., Autin J., 2016b. How can asymmetric detachment faults generate symmetric ocean continent transitions?, Terra Nova , 28, 27– 34. https://doi.org/10.1111/ter.1218310.1111/ter.1218310.1111/ter.1218310.1111/ter.12183 Google Scholar CrossRef Search ADS   Goff J.A., 1991. A Global and Regional Stochastic Analysis of Near-Ridge Abyssal Hill Morphology, J. geophys. Res. , 96, 21 713–21 737. https://doi.org/10.1029/91JB0227510.1029/91JB0227510.1029/91JB0227510.1029/91JB02275 Goff J.A., Jordan T.H., 1988. Stochastic modeling of seafloor morphology: inversion of sea beam data for second-order statistics, J. geophys. Res. , 93, 13 589–13 608. https://doi.org/10.1029/JB093iB11p1358910.1029/JB093iB11p1358910.1029/JB093iB11p1358910.1029/JB093iB11p13589 Goff J.A., Jordan T.H., 1989. Stochastic modeling of seafloor morphology; Resolution of topographic parameters by sea beam data, IEEE J. Ocean. Eng. , 14, 326– 337. https://doi.org/10.1109/48.3598310.1109/48.3598310.1109/48.3598310.1109/48.35983 Google Scholar CrossRef Search ADS   Goff J.A., Tucholke B.E., Lin J., Jaroslow G.E., Kleinrock M.C., 1995. Quantitative analysis of abyssal hills in the Atlantic Ocean: A correlation between inferred crustal thickness and extensional faulting, J. geophys. Res. , 100, 22 509–22 522. https://doi.org/10.1029/95JB0251010.1029/95JB0251010.1029/95JB0251010.1029/95JB02510 Goff J.A., Ma Y., Shah A., Cochran J.R., Sempéré J., 1997. Stochastic analysis of seafloor morphology on the flank of the Southeast Indian Ridge: The influence of ridge morphology on the formation of abyssal hills, J. geophys. Res. , 102, 15 521–15 534. https://doi.org/10.1029/97JB0078110.1029/97JB0078110.1029/97JB0078110.1029/97JB00781 Hart S., Blusztajn J., 2006. Age and geochemistry of the mafic sills, ODP site 1276, Newfoundland margin, Chem. Geol. , 235, 222– 237. https://doi.org/10.1016/j.chemgeo.2006.07.00110.1016/j.chemgeo.2006.07.00110.1016/j.chemgeo.2006.07.00110.1016/j.chemgeo.2006.07.001 Google Scholar CrossRef Search ADS   Haupert I., Manatschal G., Decarlis A., Unternehr P., 2016. Upper-plate magma-poor rifted margins: Stratigraphic architecture and structural evolution, Mar. Pet. Geol. , 69, 241– 261. https://doi.org/10.1016/j.marpetgeo.2015.10.02010.1016/j.marpetgeo.2015.10.02010.1016/j.marpetgeo.2015.10.02010.1016/j.marpetgeo.2015.10.020 Google Scholar CrossRef Search ADS   Hayman N.W., Grindlay N.R., Perfit M.R., Mann P., Leroy S., De Lépinay B.M., 2011. Oceanic core complex development at the ultraslow spreading Mid-Cayman Spreading Center, Geochem. Geophys. Geosyst. , 12. https://doi.org/10.1029/2010GC00324010.1029/2010GC00324010.1029/2010GC00324010.1029/2010GC003240 Hopper J.R., Funck T., Tucholke B.E., Christian Larsen H., Holbrook W.S., Louden K.E., Shillington D., Lau H., 2004. Continental breakup and the onset of ultraslow seafloor spreading off Flemish Cap on the Newfoundland rifted margin, Geology , 32, 93– 96. https://doi.org/10.1130/G19694.110.1130/G19694.110.1130/G19694.110.1130/G19694.1 Google Scholar CrossRef Search ADS   Jagoutz O., Muntener O., Manatschal G., Rubatto D., Peron-Pinvidic G., Turrin B.D., Villa I.M., 2007. The rift-to-drift transition in the North Atlantic: a stuttering start of the MORB machine?, Geology , 35, 1087– 1090. https://doi.org/10.1130/G23613A.110.1130/G23613A.110.1130/G23613A.110.1130/G23613A.1 Google Scholar CrossRef Search ADS   Jokat W., Schmidt-Aursch M.C., 2007. Geophysical characteristics of the ultraslow spreading Gakkel Ridge, Arctic Ocean, Geophys. J. Int. , 168, 983– 998. https://doi.org/10.1111/j.1365-246X.2006.03278.x10.1111/j.1365-246X.2006.03278.x10.1111/j.1365-246X.2006.03278.x10.1111/j.1365-246X.2006.03278.x Google Scholar CrossRef Search ADS   Jokat W., Ritzmann O., Schmidt-Aursch M.C., Drachev S., Gauger S., Snow J., 2003. Geophysical evidence for reduced melt production on the Arctic ultraslow Gakkel mid-ocean ridge, Nature , 423, 962– 965. https://doi.org/10.1038/nature0170610.1038/nature0170610.1038/nature0170610.1038/nature01706 Google Scholar CrossRef Search ADS PubMed  Klimke J., Franke D., Gaedicke C., Schreckenberger B., Schnabel M., Stollhofen H., Rose J., Chaheire M., 2016. How to identify oceanic crust—Evidence for a complex break-up in the Mozambique Channel, off East Africa, Tectonophysics , 693, 436– 452. https://doi.org/10.1016/j.tecto.2015.10.01210.1016/j.tecto.2015.10.01210.1016/j.tecto.2015.10.01210.1016/j.tecto.2015.10.012 Google Scholar CrossRef Search ADS   Lagabrielle Y. et al., 1996. Active oceanic spreading in the northern North Fiji Basin: Results of the NOFI cruise of R/V L’Atalante (newstarmer project), Mar. Geophys. Res. , 18, 225– 247. https://doi.org/10.1007/BF0028607910.1007/BF0028607910.1007/BF0028607910.1007/BF00286079 Google Scholar CrossRef Search ADS   Leroy S., Mercier de Lepinay B., Mauffret A., Pubellier M., 1996. Structural and tectonic evolution of the eastern Cayman Trough (Caribbean Sea) from seismic reflection data, AAPG Bull. , 80, 222– 247. Leroy S., Mauffret A., Patriat P., Mercier De Lepinay B., 2000. An alternative interpretation of the Cayman trough evolution from a reidentification of magnetic anomalies, Geophys. J. Int. , 141, 539– 557. https://doi.org/10.1046/j.1365-246x.2000.00059.x10.1046/j.1365-246x.2000.00059.x10.1046/j.1365-246x.2000.00059.x10.1046/j.1365-246x.2000.00059.x Google Scholar CrossRef Search ADS   Leroy S. et al., 2004. From rifting to spreading in the eastern Gulf of Aden: a geophysical survey of a young oceanic basin from margin to margin, Terra Nova , 16, 185– 192. https://doi.org/10.1111/j.1365-3121.2004.00550.x10.1111/j.1365-3121.2004.00550.x10.1111/j.1365-3121.2004.00550.x10.1111/j.1365-3121.2004.00550.x Google Scholar CrossRef Search ADS   Leroy S., D’acremont E., Tiberi C., Basuyau C., Autin J., Lucazeau F., Sloan H., 2010a. Recent off-axis volcanism in the eastern Gulf of Aden: Implications for plume–ridge interaction, Earth planet. Sci. Lett. , 293, 140– 153. https://doi.org/10.1016/j.epsl.2010.02.03610.1016/j.epsl.2010.02.03610.1016/j.epsl.2010.02.03610.1016/j.epsl.2010.02.036 Google Scholar CrossRef Search ADS   Leroy S. et al., 2010b. Contrasted styles of rifting in the eastern Gulf of Aden: A combined wide-angle, multichannel seismic, and heat flow survey, Geochem. Geophys. Geosyst. , 11. https://doi.org/10.1029/2009GC00296310.1029/2009GC00296310.1029/2009GC00296310.1029/2009GC002963 Li J. et al., 2015. Seismic observation of an extremely magmatic accretion at the ultraslow spreading Southwest Indian Ridge, Geophys. Res. Lett. , 42, 2656– 2663. https://doi.org/10.1002/2014GL06252110.1002/2014GL06252110.1002/2014GL06252110.1002/2014GL062521 Google Scholar CrossRef Search ADS   Lizarralde D., Gaherty J.B., Collins J.A., Hirth G., Kim S.D., 2004. Spreading-rate dependence of melt extraction at mid-ocean ridges from mantle seismic refraction data, Nature , 432, 744– 747. https://doi.org/10.1038/nature0314010.1038/nature0314010.1038/nature0314010.1038/nature03140 Google Scholar CrossRef Search ADS PubMed  Malinverno A., 1991. Inverse square-root dependence of mid-ocean-ridge flank roughness on spreading rate, Nature , 352, 58– 60. https://doi.org/10.1038/352058a010.1038/352058a010.1038/352058a010.1038/352058a0 Google Scholar CrossRef Search ADS   Mcdermott K., Reston T., 2015. To see, or not to see? Rifted margin extension, Geology , 43, 967– 970. https://doi.org/10.1130/G36982.110.1130/G36982.110.1130/G36982.110.1130/G36982.1 Google Scholar CrossRef Search ADS   Mendel V., Sauter D., Rommevaux-Jestin C., Patriat P., Lefebvre F., Parson L.M., 2003. Magmato-tectonic cyclicity at the ultra-slow spreading Southwest Indian Ridge: evidence from variations of axial volcanic ridge morphology and abyssal hills pattern, Geochem. Geophys. Geosyst. , 4, doi:9110.1029/2002GC000417. https://doi.org/10.1029/2002GC00041710.1029/2002GC00041710.1029/2002GC00041710.1029/2002GC000417 Michael P.J. et al., 2003. Magmatic and amagmatic seafloor generation at the ultraslow-spreading Gakkel ridge, Arctic Ocean, Nature , 423, 956– 961. https://doi.org/10.1038/nature0170410.1038/nature0170410.1038/nature0170410.1038/nature01704 Google Scholar CrossRef Search ADS PubMed  Minshull T.A., 1999. On the roughness of Mesozoic oceanic crust in the western North Atlantic, Geophys. J. Int. , 136, 286– 290. https://doi.org/10.1046/j.1365-246X.1999.00722.x10.1046/j.1365-246X.1999.00722.x10.1046/j.1365-246X.1999.00722.x10.1046/j.1365-246X.1999.00722.x Google Scholar CrossRef Search ADS   Minshull T.A., 2009. Geophysical characterisation of the ocean–continent transition at magma-poor rifted margins, C. R. Geosci. , 341, 382– 393 Google Scholar CrossRef Search ADS   Minshull T.A., Muller M.R., White R.S., 2006. Crustal structure of the Southwest Indian Ridge at 66°E: seismic constraints, Geophys. J. Int. , 166, 135– 147. https://doi.org/10.1111/j.1365-246X.2006.03001.x10.1111/j.1365-246X.2006.03001.x10.1111/j.1365-246X.2006.03001.x10.1111/j.1365-246X.2006.03001.x Google Scholar CrossRef Search ADS   Minshull T.A., Dean S.M., Whitmarsh R.B., 2014. The peridotite ridge province in the southern Iberia Abyssal Plain: seismic constraints revisited, J. geophys. Res. , 119, 1580– 1598. Google Scholar CrossRef Search ADS   Muller M.R., Minshull T.A., White R.S., 1999. Segmentation and melt supply at the Southwest Indian Ridge, Geology , 27, 867– 870. https://doi.org/10.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;210.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;210.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;210.1130/0091-7613(1999)027%3c0867:SAMSAT%3e2.3.CO;2 Google Scholar CrossRef Search ADS   Müntener O., Manatschal G., 2006. High degrees of melt extraction recorded by spinel harzburgite of the Newfoundland margin: The role of inheritance and consequences for the evolution of the southern North Atlantic, Earth planet. Sci. Lett. , 252, 437– 452. https://doi.org/10.1016/j.epsl.2006.10.00910.1016/j.epsl.2006.10.00910.1016/j.epsl.2006.10.00910.1016/j.epsl.2006.10.009 Google Scholar CrossRef Search ADS   Neumann G.A., Forsyth D.W., 1995. High resolution statistical estimation of seafloor morphology: Oblique and orthogonal fabric on the flanks of the Mid-Atlantic Ridge, 34°–35.5°S, Mar. Geophys. Res. , 17, 221– 250. https://doi.org/10.1007/BF0120346410.1007/BF0120346410.1007/BF0120346410.1007/BF01203464 Google Scholar CrossRef Search ADS   Nirrengarten M., Manatschal G., Tugend J., Kusznir N.J., Sauter D., 2017. Nature and origin of the J-magnetic anomaly offshore Iberia-Newfoundland: implications for plate reconstructions, Terra Nova , 29, 20– 28. https://doi.org/10.1111/ter.1224010.1111/ter.1224010.1111/ter.1224010.1111/ter.12240 Google Scholar CrossRef Search ADS   Nirrengarten M., Manatschal G., Tugend J., Kusznir N.J., Sauter D., submitted. Kinematic evolution of the southern North Atlantic: implications for the formation of hyper-extended rift systems, Tectonics . Niu X. et al., 2015. Along-axis variation in crustal thickness at the ultraslow spreading Southwest Indian Ridge (50°E) from a wide-angle seismic experiment, Geochem. Geophys. Geosyst. , 16, 468– 485. https://doi.org/10.1002/2014GC00564510.1002/2014GC00564510.1002/2014GC00564510.1002/2014GC005645 Google Scholar CrossRef Search ADS   Patriat P., Sloan H., Sauter D., 2008. From slow to ultraslow: A previously undetected event at the Southwest Indian Ridge at ca. 24 Ma, Geology , 36, 207– 210. https://doi.org/10.1130/G24270A.110.1130/G24270A.110.1130/G24270A.110.1130/G24270A.1 Google Scholar CrossRef Search ADS   Péron-Pinvidic G., Shillington D.J., Tucholke B.E., 2010. Characterization of sills associated with the U reflection on the Newfoundland margin: evidence for widespread early post-rift magmatism on a magma-poor rifted margin, Geophys. J. Int. , 182, 113– 136. Peron-Pinvidic G., Manatschal G., Osmundsen P.T., 2013. Structural comparison of archetypal Atlantic rifted margins: A review of observations and concepts, Mar. Pet. Geol. , 43, 21– 47. https://doi.org/10.1016/j.marpetgeo.2013.02.00210.1016/j.marpetgeo.2013.02.00210.1016/j.marpetgeo.2013.02.00210.1016/j.marpetgeo.2013.02.002 Google Scholar CrossRef Search ADS   Phipps Morgan J., Chen Y.J., 1993. Dependence of ridge-axis morphology on magma supply and spreading rate, Nature , 364, 706– 708. https://doi.org/10.1038/364706a010.1038/364706a010.1038/364706a010.1038/364706a0 Google Scholar CrossRef Search ADS   Pickup S.L.B., Whitmarsh R.B., Fowler C.M.R., Reston T.J., 1996. Insight into the nature of the ocean-continent transition off West Iberia from a deep multichannel seismic reflection profile, Geology , 24, 1079– 1082. https://doi.org/10.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;210.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;210.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;210.1130/0091-7613(1996)024%3c1079:IITNOT%3e2.3.CO;2 Google Scholar CrossRef Search ADS   Ranero C.R., Perez-Gussinye M., 2010. Sequential faulting explains the asymmetry and extension discrepancy of conjugate margins, Nature , 468, 294– 299. https://doi.org/10.1038/nature0952010.1038/nature0952010.1038/nature0952010.1038/nature09520 Google Scholar CrossRef Search ADS PubMed  Ranero C.R., Banda E., Buhl P., 1997. The crustal structure of the Canary Basin: accretion processes at slow spreading centers, J. geophys. Res. , 102, 10185– 10201. https://doi.org/10.1029/97JB0010110.1029/97JB0010110.1029/97JB0010110.1029/97JB00101 Google Scholar CrossRef Search ADS   Reid I., Jackson H.R., 1981. Oceanic spreading rate and crustal thickness, Mar. Geophys. Res. , 5, 165– 172. Reston T.J., Mcdermott K.G., 2011. Successive detachment faults and mantle unroofing at magma-poor rifted margins, Geology , 39, 1071– 1074. https://doi.org/10.1130/G32428.110.1130/G32428.110.1130/G32428.110.1130/G32428.1 Google Scholar CrossRef Search ADS   Sauter D., Mendel V., Rommevaux-Jestin C., Parson L.M., Fujimoto H., Mével C., Cannat M., Tamaki K., 2004. Focused magmatism versus amagmatic spreading along the ultra-slow spreading southwest Indian ridge: evidence from TOBI side scan sonar imagery, Geochem. Geophys. Geosyst. , 5, doi:10.1029/2004GC000738. https://doi.org/10.1029/2004GC00073810.1029/2004GC00073810.1029/2004GC00073810.1029/2004GC000738 Sauter D., Cannat M., Meyzen C., Bezos A., Patriat P., Humler E., Debayle E., 2009. Propagation of a melting anomaly along the ultraslow Southwest Indian Ridge between 46°E and 52°20΄E: interaction with the Crozet hotspot?, Geophys. J. Int. , 179, 687– 699. https://doi.org/10.1111/j.1365-246X.2009.04308.x10.1111/j.1365-246X.2009.04308.x10.1111/j.1365-246X.2009.04308.x10.1111/j.1365-246X.2009.04308.x Google Scholar CrossRef Search ADS   Sauter D., Sloan H., Cannat M., Goff J., Patriat P., Schaming M., Roest W.R., 2011. From slow to ultra-slow: How does spreading rate affect seafloor roughness and crustal thickness?, Geology , 39, 911– 914. https://doi.org/10.1130/G32028.110.1130/G32028.110.1130/G32028.110.1130/G32028.1 Google Scholar CrossRef Search ADS   Sauter D. et al., 2013. Continuous exhumation of mantle-derived rocks at the Southwest Indian Ridge for 11 million years, Nat. Geosci. , 6, 314– 320. https://doi.org/10.1038/ngeo177110.1038/ngeo177110.1038/ngeo177110.1038/ngeo1771 Google Scholar CrossRef Search ADS   Sauter D. et al., 2016. Evidence for magma entrapment below oceanic crust from deep seismic reflections in the Western Somali Basin, Geology , 44, 407– 410. https://doi.org/10.1130/G37747.110.1130/G37747.110.1130/G37747.110.1130/G37747.1 Google Scholar CrossRef Search ADS   Sayers J., Symonds P.A., Direen N.G., Bernardel G., 2001. Nature of the continent-ocean transition on the non-volcanic rifted margin of the central Great Australian Bight, Geol. Soc., London, Spec. Publ. , 187, 51– 76. https://doi.org/10.1144/GSL.SP.2001.187.01.0410.1144/GSL.SP.2001.187.01.0410.1144/GSL.SP.2001.187.01.0410.1144/GSL.SP.2001.187.01.04 Google Scholar CrossRef Search ADS   Schlindwein V., Schmid F., 2016. Mid-ocean-ridge seismicity reveals extreme types of ocean lithosphere, Nature , 535 276– 279. https://doi.org/10.1038/nature1827710.1038/nature1827710.1038/nature1827710.1038/nature18277 Google Scholar CrossRef Search ADS PubMed  Schmid F., Schlindwein V., 2016. Microearthquake activity, lithospheric structure, and deformation modes at an amagmatic ultraslow spreading Southwest Indian Ridge segment, Geochem. Geophys. Geosyst. , 17, 2905– 2921. https://doi.org/10.1002/2016GC00627110.1002/2016GC00627110.1002/2016GC00627110.1002/2016GC006271 Google Scholar CrossRef Search ADS   Shaw W.J., Lin J., 1996. Models of ocean ridge lithospheric deformation: Dependence on crustal thickness, spreading rate, and segmentation, J. geophys. Res. , 101, 17 977– 17 993. https://doi.org/10.1029/96JB0094910.1029/96JB0094910.1029/96JB0094910.1029/96JB00949 Google Scholar CrossRef Search ADS   Shillington D.J. et al., 2006. Evidence for asymmetric nonvolcanic rifting and slow incipient oceanic accretion from seismic reflection data on the Newfoundland margin, J. geophys. Res. , 111, B09402, doi:10.1029/2005JB003981. Sloan H., Sauter D., Goff J.A., Cannat M., 2012. Abyssal hill characterization at the ultraslow spreading Southwest Indian Ridge, Geochem. Geophys. Geosyst. , 13. https://doi.org/10.1029/2011GC00385010.1029/2011GC00385010.1029/2011GC00385010.1029/2011GC003850 Small C., 1994. A global analysis of mid-ocean ridge axial topography, Geophys. J. Int. , 116, 64– 84. https://doi.org/10.1111/j.1365-246X.1994.tb02128.x10.1111/j.1365-246X.1994.tb02128.x10.1111/j.1365-246X.1994.tb02128.x10.1111/j.1365-246X.1994.tb02128.x Google Scholar CrossRef Search ADS   Stanton N., Manatschal G., Autin J., Sauter D., Maia M., Viana A., 2016. Geophysical fingerprints of hyper-extended, exhumed and embryonic oceanic domains: the example from the Iberia–Newfoundland rifted margins, Mar. Geophys. Res. , 37, 185– 205. https://doi.org/10.1007/s11001-016-9277-010.1007/s11001-016-9277-010.1007/s11001-016-9277-010.1007/s11001-016-9277-0 Google Scholar CrossRef Search ADS   Sundvik M., Larson R.L., Detrick R.S., 1984. Rough-smooth basement boundary in the western North Atlantic basin: evidence for a seafloor-spreading origin, Geology , 12, 31– 34. https://doi.org/10.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;210.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;210.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;210.1130/0091-7613(1984)12%3c31:RBBITW%3e2.0.CO;2 Google Scholar CrossRef Search ADS   Sutra E., Manatschal G., 2012. How does the continental crust thin in a hyperextended rifted margin? Insights from the Iberia margin, Geology , 40, 139– 142. https://doi.org/10.1130/G32786.110.1130/G32786.110.1130/G32786.110.1130/G32786.1 Google Scholar CrossRef Search ADS   Sutra E., Manatschal G., Mohn G., Unternehr P., 2013. Quantification and restoration of extensional deformation along the Western Iberia and Newfoundland rifted margins, Geochem. Geophys. Geosyst. , 14, 2575– 2597. https://doi.org/10.1002/ggge.2013510.1002/ggge.2013510.1002/ggge.2013510.1002/ggge.20135 Google Scholar CrossRef Search ADS   Ten Brink U.S., Coleman D.F., Dillon W.P., 2002. The nature of the crust under Cayman Trough from gravity, Mar. Pet. Geol. , 19, 971– 987. https://doi.org/10.1016/S0264-8172(02)00132-010.1016/S0264-8172(02)00132-010.1016/S0264-8172(02)00132-010.1016/S0264-8172(02)00132-0 Google Scholar CrossRef Search ADS   Thinon I., Fidalgo-González L., Réhault J., Olivet J., 2001. Déformations pyrénéennes dans le golfe de Gascogne, C. R. Acad. Sci., Paris II , 332, 561– 568. https://doi.org/10.1016/S1251-8050(01)01576-210.1016/S1251-8050(01)01576-210.1016/S1251-8050(01)01576-210.1016/S1251-8050(01)01576-2 Thinon I., Matias L., Rehault J.P., Hirn A., Fidalgo-Gonzalez L., Avedik F., 2003. Deep structure of the Armorican Basin (Bay of Biscay): a review of Norgasis seismic reflection and refraction data, J. Geol. Soc. , 160, 99– 116. https://doi.org/10.1144/0016-764901-10310.1144/0016-764901-10310.1144/0016-764901-10310.1144/0016-764901-103 Google Scholar CrossRef Search ADS   Tikku A.A., Cande S.C., 1999. The oldest magnetic anomalies in the Australian-Antarctic Basin: Are they isochrons?, J. geophys. Res. , 104, 661– 677. https://doi.org/10.1029/1998JB90003410.1029/1998JB90003410.1029/1998JB90003410.1029/1998JB900034 Google Scholar CrossRef Search ADS   Tikku A.A., Direen N.G., 2008. Comment on “Major Australian-Antarctic Plate Reorganization at Hawaiian-Emperor Bend Time”, Science , 321, 490. https://doi.org/10.1126/science.115716310.1126/science.115716310.1126/science.115716310.1126/science.1157163 Google Scholar CrossRef Search ADS PubMed  Tucholke B.E., Sibuet J.-C., 2007. Leg 210 synthesis: tectonic, magmatic, and sedimentary evolution of the Newfoundland-Iberia rift, in Proceedings of the Ocean Drilling Program, Scientific Results , pp. 1– 56, eds Tucholke B.E., Sibuet J.-C., Klaus A., Ocean Drilling Program. Google Scholar CrossRef Search ADS   Tucholke B.E., Behn M.D., Buck W.R., Lin J., 2008. Role of melt supply in oceanic detachment faulting and formation of megamullions, Geology , 36, 455– 458. https://doi.org/10.1130/G24639A.110.1130/G24639A.110.1130/G24639A.110.1130/G24639A.1 Google Scholar CrossRef Search ADS   Tugend J., Manatschal G., Kusznir N.J., Masini E., Mohn G., Thinon I., 2014. Formation and deformation of hyperextended rift systems: insights from rift domain mapping in the Bay of Biscay-Pyrenees, Tectonics , 33, 1239– 1276. https://doi.org/10.1002/2014TC00352910.1002/2014TC00352910.1002/2014TC00352910.1002/2014TC003529 Google Scholar CrossRef Search ADS   Tugend J., Manatschal G., Kusznir N.J., Masini E., 2015. Characterizing and identifying structural domains at rifted continental margins: application to the Bay of Biscay margins and its Western Pyrenean fossil remnants, Geol. Soc., London, Spec. Publ. , 413, 171– 203. https://doi.org/10.1144/SP413.310.1144/SP413.310.1144/SP413.310.1144/SP413.3 Google Scholar CrossRef Search ADS   Unternehr P., Peron-Pinvidic G., Manatschal G., Sutra E., 2010. Hyper-extended crust in the South Atlantic: in search of a model, Pet. Geosci. , 16, 207– 215. https://doi.org/10.1144/1354-079309-90410.1144/1354-079309-90410.1144/1354-079309-90410.1144/1354-079309-904 Google Scholar CrossRef Search ADS   Van Avendonk H.J.A., Holbrook W.S., Nunes G.T., Shillington D.J., Tucholke B.E., Louden K.E., Larsen H.C., Hopper J.R., 2006. Seismic velocity structure of the rifted margin of the eastern Grand Banks of Newfoundland, Canada, J. geophys. Res. , 111, B11404, doi:10.1029/2005JB004156. https://doi.org/10.1029/2005JB00415610.1029/2005JB00415610.1029/2005JB00415610.1029/2005JB004156 Google Scholar CrossRef Search ADS   Watremez L., Leroy S., Rouzo S., D’acremont E., Unternehr P., Ebinger C., Lucazeau F., Al-Lazki A., 2011. The crustal structure of the north-eastern Gulf of Aden continental margin: insights from wide-angle seismic data, Geophys. J. Int. , 184, 575– 594. https://doi.org/10.1111/j.1365-246X.2010.04881.x10.1111/j.1365-246X.2010.04881.x10.1111/j.1365-246X.2010.04881.x10.1111/j.1365-246X.2010.04881.x Google Scholar CrossRef Search ADS   Weigelt E., Jokat W., 2001. Peculiarities of roughness and thickness of oceanic crust in the Eurasian Basin, Arctic Ocean, Geophys. J. Int. , 145, 505– 516. https://doi.org/10.1046/j.1365-246X.2001.00398.x10.1046/j.1365-246X.2001.00398.x10.1046/j.1365-246X.2001.00398.x10.1046/j.1365-246X.2001.00398.x Google Scholar CrossRef Search ADS   Welford J.K., Hall J., Sibuet J., Srivastava S.P., 2010a. Structure across the northeastern margin of Flemish Cap, offshore Newfoundland from Erable multichannel seismic reflection profiles: evidence for a transtensional rifting environment, Geophys. J. Int. , 183, 572– 586. https://doi.org/10.1111/j.1365-246X.2010.04779.x10.1111/j.1365-246X.2010.04779.x10.1111/j.1365-246X.2010.04779.x10.1111/j.1365-246X.2010.04779.x Google Scholar CrossRef Search ADS   Welford J.K., Smith J.A., Hall J., Deemer S., Srivastava S.P., Sibuet J.-C., 2010b. Structure and rifting evolution of the northern Newfoundland Basin from Erable multichannel seismic reflection profiles across the southeastern margin of Flemish Cap, Geophys. J. Int. , 180, 976– 998. https://doi.org/10.1111/j.1365-246X.2009.04477.x10.1111/j.1365-246X.2009.04477.x10.1111/j.1365-246X.2009.04477.x10.1111/j.1365-246X.2009.04477.x Google Scholar CrossRef Search ADS   White R.S., Minshull T.A., Bickle M.J., Robinson C.J., 2001. Melt generation at very slow-spreading oceanic ridges: constraints from geochemical and geophysical data, J. Pet. , 42, 1171– 1196. https://doi.org/10.1093/petrology/42.6.117110.1093/petrology/42.6.117110.1093/petrology/42.6.117110.1093/petrology/42.6.1171 Google Scholar CrossRef Search ADS   Whitmarsh R.B., Manatschal G., Minshull T.A., 2001. Evolution of magma-poor continental margins from rifting to seafloor spreading, Nature , 413, 150– 154. https://doi.org/10.1038/3509308510.1038/3509308510.1038/3509308510.1038/35093085 Google Scholar CrossRef Search ADS PubMed  Zhao M. et al., 2013. Three-dimensional seismic structure of the Dragon Flag oceanic core complex at the ultraslow spreading Southwest Indian Ridge (49°39΄E), Geochem. Geophys. Geosyst. , 14, 4544– 4563. https://doi.org/10.1002/ggge.2026410.1002/ggge.2026410.1002/ggge.2026410.1002/ggge.20264 Google Scholar CrossRef Search ADS   Zhou H., Dick H.J.B., 2013. Thin crust as evidence for depleted mantle supporting the Marion Rise, Nature , 494, 195– 200. https://doi.org/10.1038/nature1184210.1038/nature1184210.1038/nature1184210.1038/nature11842 Google Scholar CrossRef Search ADS PubMed  © The Authors 2017. Published by Oxford University Press on behalf of The Royal Astronomical Society.

Journal

Geophysical Journal InternationalOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial