Processed seismic data of Cruise SO98 GIGICS 1994

Main target of the project GIGICS (Cooperative German-Indonesian Geoscientific Investigations in the Celebes Sea) is the investigation of the internal crustal structure and the plate tectonic evolution of the Celebes Sea and its active continental margins off Mindanao and Northern Sulawesi. These investigations were carried out during the cruise SO98 of RV SONNE by the Federal Institute for Geosciences and Natural Resources (BGR), Hannover; the German Research Centre for Geosciences (GFZ), Potsdam; the GEOMAR, Kiel; the Institute of Oceanography (IfM), Hamburg; the Mines and Geoscience Bureau, Manila; the Agency for the Assessment and Application of Technology, Jakarta, and the Institute of Oceanography, Wormley. The cruise SO98 consisted of three legs of two weeks duration and one leg of four weeks duration. The total amount of data acquired during the cruise were: - 3,300 km of multichannel reflection seismics, - over 6,800 km of gravimetric and magnetic data and approximately 10.000 km of swath bathymetric and sediment echosounder data, - 3 wideangle-/refractionseismic profiles, each of 120 - 150 km length, - geological, geochemical sampling and oceanographical measurements at a total of 37 stations. During the cruise SO98 a widespaced but regular grid of magnetic and gravimetric profiles were acquired in the eastern part of the Celebes Sea from which up to then reliable data were very sparse. WEISSEL (1980) recognized in the western Celebes Sea WSW-ENE striking magnetic lineations, which he interpreted as chrons 18 - 20 (39 - 43 Ma according to the timescale of HARLAND et al. (1990)). The data from cruise SO98 show that there is no continuation of these anomalies to the east. In the eastern part the magnetic field of the Celebes Sea is less clear and much more disturbed. Nevertheless, E-W-striking anomalies are recognizable. Because amplitudes of local magnetic anomalies are higher than the lineations, the correlation of these lineations with the magnetic reversal scale is still somewhat ambiguous. The gravity map compiled from the measured gravimetric data shows elongated positive anomalies in the eastern part of the Celebes Sea. Exceptions occur at the deep sea trenches off North Sulawesi (North Sulawesi Trench) and Mindanao (Cotabatu Trench) and at the Sulu Archipelago where strong negative gravity anomalies were found. A remarkable NW-striking gravity high of up to 60 mgal was found in the central eastern part of the Celebes Sea. Gravimetric modelling suggests that this high can be correlated with the gravimetric effect of the Molucca Sea Plate subducting from the east under the Sangihe Arc. The reflection seismic data from the northern part of the Celebes Sea show indications for a juvenile subduction of oceanic Celebes Sea crust under the Sulu Archipelago. The oceanic crust bends down towards the Sulu Arc with angles between 2° and 5° and the sedimentary sequence above is deformed indicating a compressional stress regime. With the exception of two linear arranged seamount-like basement highs the Celebes Sea is dominated by two different oceanic crustal types showing distinct differences in the topography. The first one is showing a very similar reflection seismic pattern as it is found for oceanic crust of the Atlantic (HINZ et al., 1994). This type is characterized by a small-scale block-faulted relief of the top basement and a low reflectivity in lower crustal levels typically related as to be accreted at slow to intermediate spreading ridges. This type is found in the western, northern and southern part of the investigated area. In the eastern and especially in the southeastern part the igneous crust shows a very different image. The reflection of the top of the basement is less distinct and of lower frequency. The relief is very much smoother than in the previous type. This reflection seismic image indicates a volcanic/magmatic overprinting of the oceanic crust in this part of the Celebes Sea. Another target of cruise SO98 was the area of the active continental margin off North Sulawesi and its accretionary complex. The internal structure of the accretionary complex should be investigated to decide whether this active margin is also of the 'splinter-type' or not. During former geophysical cruises with RV SONNE oceanic crustal splinters were discovered in the accretionary wedges of the Sulu Sea and off Costa Rica (e.g. HINZ et al., 1991). From our reflection seismic measurements this active continental margin is morphologically subdivided into three units and consists of two accretionary complexes of different internal structural style: the lower and middle continental slope is underlain by an intensively thrusted, sedimentary accretionary wedge. This wedge was most probably formed during the last 5 Ma. Landward of this wedge an older and seismically very complex accretionary unit is present which is overlain at its landward termination by a sedimentary fore-arc basin. Within this older accretionary complex, units with a strong, low frequency reflection pattern were found which are interpreted to represent crustal splinters of igneous oceanic or ophiolitic nature. This interpretation is supported by our gravity and magnetic data. The magnetic profiles show an increase of the magnetic field towards the north arm of Sulawesi across the continental margin. This increase of the magnetic field suggests an increase of magnetized material within the older accretionary wedge towards the northern arm of Sulawesi where ophiolites are emplaced. During the interpretation of the reflection seismic data of the project GIGICS BSR's (bottom simulating reflectors) were discovered for the first time along the active continental margin of North-Sulawesi. BSR's are the seismic expression of a velocity decrease at the bottom of a gas hydrate zone. The distribution and depth of the BSR's correlates with the geochemical and geothermal results. Radiometric age dating and geochemical analyses from pillow basalts of a seamount from the southeastern Celebes Sea indicate hot-spot activity in this part of the Celebes Sea during or shortly after the formation of the oceanic crust approximately at 43 Ma ago. Three NW-striking ridges or seamount-chains in the northeastern Celebes Sea were mapped and investigated in detail. They are thought to represent a wrench fault system extending through the northeastern Celebes Sea. At the flank of one of these ridges a strongly alterated plagioclase-olivine basalt sample was dredged which was overlain by non-fossiliferous clay stone. A similar lithostratigraphic sequence was drilled during ODP leg 124 (RANGIN et al., 1990). The geochemical composition of these basalts is different from typical MORB. The existence of a large crustal splinter within the accretionary wedge off southwestern Mindanao obviously is responsible for a high thermal conductivity which in turn could have enhanced heat flow (108.1 mW/m2) and methanogenesis (405 ppb). The heat flow of 103.0 mW/m2 at the deformation front of the Mindanao wedge and the high methane concentration of 5.555 ppb suggests tectonically induced fluid transport within the wedge. High methane concentrations between 8.044 and 49.006 ppb at the lower slope off Sulawesi and in the North Sulawesi Trench are accompanied by high heat flow values of up to 100.5 mW/m2. Heat flow is significantly lower upslope (31.3 mW/m2). This general heat flow distribution pattern is seen over a large portion of the accretionary wedge. The elevated heat flow values and high methane concentrations near the deformation front most likely result from heat transport by fluids squeezed out from vertically and laterally compacting sediments. The reduced heat flow towards the coast is compatible either with a cooling effect of slow subduction of the oceanic crust, or stacking of cool slabs of compacted sediments. A subduction of oceanic crust with a heat flow around 60 mW/m2 over a period of more than 3 million years would have produced the low heat flow values of the upper slope if the wedge consists of claystone with a low thermal conductivity (1.2 - 1.7 W/mK). Even in the low-heat flow area isolated fluid venting is possible. Lateral variations in the heat flow pattern (e.g. broadening of the anomalies in the west) may be due to different thermal regimes within the subducted crust.