Tectonics in the Northwestern West Philippine Basin

INTRODUCTION

The West Philippine basin is a currently inactive marginal basin belonging to Philippine Sea plate and subducts at the Ryukyu Trench (RT) to the north (Fig.1) and at the Philippine Trench (PT) to the west.It is one of the oldest basins in West Pacific margins, with a complex formation history (Deschamps and Lallemand, 2002; Deschamps et al., 1999; Hilde and Lee, 1984).On the origin of the basin, some scholars thought that it is a trapped oceanic basin (Hilde and Lee, 1984;Mrozowski et al., 1982; Uyeda and Ben-Avraham, 1972); whereas others suggested it is a back-arc spreading basin (Deschamps and Lallemand, 2002; Hall, 2002, 1997; Okino et al., 1999; Hall et al., 1995a, b).The most referred model proposed for the spreading processes within the basin is from Hilde and Lee (1984),they analyzed magnetic anomalies and seafloor structures and indicated that the basin was formed during 58–33 Ma and was divided into two phases of spreading in an NE-SW and N-S directions, respectively, from a single spreading center.

Figure  1.  Sketch map showing topography, seismic line location and age data (Ma) of samples.Bathymetry data from Smith and Sandwell (1997) and KR03-04 cruise, red dot is dive position, and thin broken line is seismic line.

DownLoad: Full-Size Img PowerPoint

Since 1984, several geophysical surveys have been performed in the WPB, mainly over its central part.Data collected during these surveys show that the spreading history is much more complicated than that inferred in Hilde and Lee's model (Okino and Fujioka, 2003; Deschamps and Lallemand, 2002; Fujioka et al., 2000; Deschamps et al., 1999; Okino et al., 1999).Deschamps and Lallemand (2002)showed that spreading in the basin occurred from 54 Ma to 33/30Ma.The tectonics is more complicated in the western part of basin than that in eastern part, due to the presence of an active mantle plume in this region during spreading.In the northwestern WPB, the Hydrographic Department of Japan Maritime Safety Agency has recently conducted geophysical surveys including seabeam mapping, gravity and magnetic anomalies, at least three fabrics formed between two overlapping spreading centers have been identified based on these results (Yoshida et al., 2000).Deschamps and Lallemand (2002) proposed that NE-SW direction spreading, which was highly disorganized in this part of the basin, was due to the presence of an active mantle plume during spreading.The Urdenata plateau (UP) in the northern basin and the Benham plateau (BP) in the southern basin result from the interaction of such a mantle plume with the western part of the WPB spreading system (Okino and Fujioka, 2003; Deschamps and Lallemand, 2002; Okino et al., 1999).The western part of the WPB then displays rough and complicated morphology and excess magmatism.

Another important feature of the western WPB is the Okinawa-Luzon fracture zone (OLF) (Fig.1).It extends hundreds of kilometers long over its apparent part and trends perpendicular to the spreading fabric.This feature is interpreted to be a major inactive fracture zone of the WPB that was active during spreading in Eocene (Deschamps and Lallemand, 2002; Deschamps et al., 1999; Hilde and Lee, 1984).Wang et al. (1996) thought that the OLF is a right-handed fracture zone with NE trending and the main cause of the termination or displacement of the central basin ridge at its northwestern end.But they focused only on the topography and the magnetic data ofHilde and Lee (1984), and didn't discuss its fracture and development.This article mainly discusses the tectonics of the northwestern WPB and its formation, based on marine gravity, magnetic and seismic data.

DATA AND METHODS

Data

Gravity, magnetic and bathymetry are mainly from KR03-04 cruise, assisting with data from National Geophysical Data Center (NGDC) and Smith and Sandwell (1997).And the multi-channel seismic data were collected by Institute of Oceanology, Chinese Academy of Sciences during 1982–1984.Samples are mainly from R/V KR03-04 cruise.

Methods

Though the gravity data are the revised free-air gravity anomaly (g_f), the data are gained from different cruises and different times; the precisions are different from each other.Based on the data from KR03-04 cruise, closing error is made.The Bugers gravity anomaly (g_B) could be gained from free-air gravity anomaly by the following formula

g_B=g_f+0.068 7H (mGal)

where H is the water depth, H < 0, unit is meter.

And then, we can make general proceeding on Bugers gravity anomaly, such as continuation and differential.Revising the closing error is also suitable for magnetic data.The seismic data were collected by Kexue I of Chinese Academy of Sciences during1990s, and the interpretation of the data refers to Wu et al. (2007).Age dating is done by Shinjo Ryuichi, using Ar-Ar dating method.

Based on the above data, a joint inversion of the gravity, magnetic and seismic data have been done following the steps according to Yan (2004) : (1) according to the seismic profile interpretation, construct initial inversion geology model; (2) using the densities of the layers, calculate the gravity anomaly; (3) compared with observed gravity anomaly, revise the model to make the calculated gravity anomaly fit with the observed gravity anomaly.

RESULTS

As shown in Fig.2, the Bugers gravity anomaly around Oki-Daito Ridge (ODR) is high in middle and low around.Similar results can be seen in the Urdaneta plateau (UP).But Bugers anomaly in the north UP is higher than that in south.The OLF zone is disordered, high and low appearance in turn.This phenomenon is very clear on the map of 3rd trending surface Bugers gravity anomaly (Fig.3).In order to distinguish shallow and deep structures, we make downward and upward continuations of Bugers gravity anomaly (Fig.4).On Fig.4, the anomaly of OLF zone changes largely, on the 15 km upward continuation map, there is no local anomalies.It shows that most of the structures in the OLF area are shallow, only one is deep.In ODR and UP, most of the structures are deep, only few shallow.On the 2nd vertical differential map (Fig.5), we can identify many structures, mainly shallow.

Figure  2.  Continuation map showing Bugers gravity anomaly (mGal) in Okinawa-Luzon fault zone.OT.Okinawa Trough.

DownLoad: Full-Size Img PowerPoint

Figure  3.  Contour map showing 3rd trending surface residual Bugers gravity anomaly (mGal).

DownLoad: Full-Size Img PowerPoint

Figure  4.  Continuation maps showing downward and upward of Bugers gravity anomaly. (a) Downward 2km; (b) downward 5 km; (c) upward 5 km; (d) upward 15 km.

DownLoad: Full-Size Img PowerPoint

Figure  5.  Differential map showing the 2nd vertical of Bugers gravity (10⁴ m^-1·s^-2).

DownLoad: Full-Size Img PowerPoint

On reduction to pole magnetic anomaly map (Fig.6), there are five magnetic lineations from south to north, corresponding to 18–22 magnetic lineations in Hilde and Lee (1984).But in Fig.1, the rock age varies from 30 Ma to 40 Ma, which is not consistent with 18–22 magnetic lineations.The reason may be that these rocks experienced tectonic movement.In whole, the anomaly intension is small, between (-180–180) nT.Also, in OLF zone, anomaly is highly disordered.

Figure  6.  Reduction map showing pole magnetic anomaly (nT).

DownLoad: Full-Size Img PowerPoint

Seismic profile DS6-1 (Fig.7) starts at (23°8′N, 128°3.6′E), and ends at (23°23.2′N, 124°23.2′E), with 290 km length, trending NWW, and meeting Ryukyu Islands with 30°angle.It crosses four tectonic units: Philippine Sea abyssal plain, outside rise of Ryukyu Trench (that is OLF zone), Ryukyu Trench (RT), and slope of Ryukyu Islands.The slope of outer rise (Fig.7a) is very steep.Its west side is a fault, while east is a gentle slope.Across the profile (Fig.7b), the broken crust and very undulate geomorphology indicate the existence of fracture structures in the area.In the trough floor, thin sedimentary layer with the thickness of 200–400 m is filled.Under thin sediment layer, there is base rock broken zone.The thickness is about2 000–4 000 m.The oceanic crust appears under sedimentary layers on the profile (Fig.9).

Figure  7.  Seismic section images of profile DS6-1 and the interpretation (position on Fig.1). (a) CDP3300–6700 of profile DS6-1 (outside rise and RT); (b) CDP86–3486 of profile DS6-1 (Philippine basin); (c) interpretation of the seismic section.I.Deep sea sediment; II.basement; Okt.Okinawa Trough; mig.migration.

DownLoad: Full-Size Img PowerPoint

Figure  8.  Interpretation of seismic section image of profile DMII-4 (position on Fig.1).FB.fore-arc basin; RIS.Ryukyu Islands slope; G.sediment.

DownLoad: Full-Size Img PowerPoint

Figure  9.  Joint inversion result of seismic profile DS6-1.Red is basement or magma rock, yellow is sediment, deep yellow is broken rock.

DownLoad: Full-Size Img PowerPoint

Seismic profile DMII-4 (Fig.8) starts at (26°3.8′N, 126°3.0′E), and ends at (23°9.6′N, 128°0.4′E), with NW trending, crosses vertically with RT.The outer rise is composed of a series of steep ridges, about 30 km length.Some high angle normal faults trending NW form rift valleys in the outer rise.In Philippine Sea abyssal plain, an extensional fault divides it with the outer rise.The reflector of basement is very clear with a little fluctuation.It occurs at about 4.5–4.25 s in two-way travel time and has lifting to SE.Thin sediment layers with clear-layered reflectors cover over the basement and have about 200 m thickness.Because of the existence of outer ridges, some SE trending thrust fractures are well developed in the trench and inner slope, while the extensional structures occurred in the southeast of the profile (Fig.8).

In order to find out the depth of OLF zone, we make joint inversion of DS6-1 and DMII-4 profiles, according toDai (2004) and Ding et al. (2004).The results are shown in Figs.9 and 10.The magnetic data fit very well both in the seismic profiles DS6-1 (Fig.9) and DMII-4 (Fig.10).Whereas the gravity data cannot fit very well in the whole section (Fig.10), especially in the RT area, and calculated curve cannot fit well with observed curve.The OLF zone shows the magnetic curve jumps, but does not locate at the top of the topography and the highest of the gravity anomaly.We infer that the OLF zone is broken hardly with density of 2.3–2.4 g/cm³, and deep magnetic magma complex.In Philippine Sea abyssal plain, sediment layer is very thin.Under it, we inferred that there is the broken basement rock with low density of 2.35g/cm³.The model can fit well with the interpretation of the joint inversion results (Fig.9).

Figure  10.  Joint inversion result of seismic profile DMII-4.Red is magma rock, yellow is sediment, pink is broken rock, green is broken rock, brown is basement.

DownLoad: Full-Size Img PowerPoint

In profile DMII-4, outer rise consists of basalt seamount, with density of 2.7–2.8 g/cm³.For giant resistance to this seamount subduction, the rise suppressed intensive compression.It shows minimum magnetics, no change in gravity values and low density about 2.4 g/cm³.And in Philippine Sea abyssal plain, the crust structure is the same as the profile DS6-1.

DISCUSSION

NW Trending Fracture and Amagma Spreading

Based on magnetic lineations, Hilde and Lee (1984) proposed that the tectonic of this area is in NW direction and may be formed from the Central Basin Ridge.On the tomography map, gravity anomaly map and magnetic anomaly map, we also discovered the NW structures in west part of WPB and the structures parallel with the Central Basin Ridge.Obviously, in this part of WPB, it is the result of spreading from Central Basin Ridge (Lin and Li, 1999).And age of spreading is not defined.According to Hilde and Lee (1984), it happened at 60–45 Ma, magnetic lineations are 26–13.But our age data of samples derived from this area are between 40–30 Ma, this period is inactivity stage (Deschamps and Lallemand, 2002).As these samples are basalt and most are badly weathered, their formation in amagma spreading stage is not possible.Our data support that it was formed from Central Basin Ridge during 40–30 Ma.In Fig.1, we also find that the ages of north (Dive 290) and south of Urdaneta plateau are different.The age in the south is 36–39 Ma, while that in the north is 31 Ma which is same as the age around oceanic crust (Dive 291, 288).Lacking of the spreading magnetic lineations, we cannot define the data of amagma spreading stage.

Okinawa-Luzon Fracture Zone

The fracture zone appears as three faults in Fig.5 and Fig.1, and has about hundred kilometers width.In Fig.1, rock sample in the zone (Dive 122) with the31.5 Ma age also is a basalt rock not a metamorphic rock.Lacking of age of west side oceanic crust of this fracture zone, we could not identify clearly when it was formed.Wang et al. (1996) assumed it was a strike-slip fault and main cause for termination or displacement of Central Basin Ridge, and Svarichevsky and Wang (1992) believed the northwestern part of the Central Basin Ridge has been displaced along some fractured under the Ryukyu trench-arc system, presently maybe located at Miyako fracture zone (Fig.11).In inversion maps (Figs.9, 10), the OLF zone shows low density at about 2.35 g/cm³affected by broken fracture zone.This fault is not like a common strike fault, but resembles a transform fracture zone.

Figure  11.  Sketch map showing the structure of the northwest WPB.OI.Okinawa Island; RI.Ryukyu Islands.

DownLoad: Full-Size Img PowerPoint

The Urdenata Plateau

The Urdaneta plateau (UP) is not a remained arc, but possible remnant of mantle plume although its lava chemistry is similar to OIB.The Urdenata plateau is a seafloor plateau according to topography map in the WPB (Fig.1).Two views exist for the formation of its tectonics.The former geologists suggested that it origins from a remained arc (Hilde and Lee, 1984), whereas others thought it is an active mantle plume (Wang et al., 1996).In map of continuation of gravity and magnetic anomaly, the UP has high thickness of crust and no apparent plume deep structure.According to DSDP 294/5, rock age around the UP is about 19Ma by Ar-Ar dating (Ozima et al., 1983).That is to say, rocks around UP are younger than those in UP, and these in the northern UP are younger than the southern ones (Fig.1).However, the UP at least was younger than the spreading oceanic crust (during58–33 Ma, Deschamps and Lallemand, 2002).

CONCLUSIONS

Based on bathymetry, gravity, magnetic and seismic data, two groups of fracture exist in northwesternmost of WPB (Fig.11).The magnetic and topographic data provide new evidence for amagma spreading of WPB.And the deformation force may pull off the subducted Philippine Sea plate.

The OLF zone trends NNE and turns to NE off RT (126°E).It is a hardly broken zone.In northwestern corner of WPB, it appears as a right strike-slip fracture zone, with large slip distance and depth.It may be a transform fault.The presence of this fracture zone causes Ryukyu Islands and RT bend and the termination or displacement of Central Basin Ridge.