Abstract: The intense deformation zone in the central Indian Ocean, south of Indian continent is one of the most complex regions in terms of its structure and geodynamics. The deformation zone has been studied and debated in 1990s for its genesis. It was argued that deformation is mainly confined to sedimentary and oceanic crustal layers, while the large wave length geoidal anomalies, on which the deformation region lies, called for deeper sources. The inter connection between deeper and the shallower sources is found missing. The current study focuses on the complexities of this region by analyzing OBS (ocean bottom seismometer) data. The data acquired by five OBS systems along a 300 km long south-north profile in the CIOB (central Indian Ocean basin) have been modeled and the crustal and sub-crustal structure has been determined using 2-D tomographic inversion. Four subsurface layers are identified representing the sediment column, upper crustal layer, lower crustal layer and a sub-crustal layer (upper mantle layer). A considerable variation in thickness as well as velocity at all interfaces from sedimentary column to upper mantle is observed which indicates that the tectonic forces have affected the entire crust and sub-crustal configuration. The sediments are characterized by higher velocities (2.1 km/s) due to the increased confining pressure. Modeling results indicated that the velocity in upper crust is in the range of 5.7-6.2 km/s and the velocity of the lower crust varies from 7.0-7.6 km/s. The velocity of the sub-crustal layer is in the range of 7.8-8.4 km/s. This high-velocity layer is interpreted as magmatic under-plating with strong lateral variations. The base of the 7.0 km/s layer at 12-15 km depth is interpreted as the Moho.
　　Key words: OBS (ocean bottom seismometer), CIOB (central Indian Ocean basin), tomographic inversion.
　　1. Introduction
　　Investigation of crustal structure [1] is the key to understand the correlation between seismogenic structure and seismicity and to evaluate the earthquake hazard for nearby areas. Direct evidence of active deformation in the eastern Indian Oceanic place is available [2]. Many regions of the world are difficult to explore using conventional reflection profiling because a strongly reflected horizon lying high in stratigraphic column effectively masks the underlying structure [3]. A number of observations reveal large periodic undulations within the oceanic and continental lithospheres. The presence of the several moderate magnitude earthquakes, east west striking folds, thrust faults in the sediments, abnormally high heat flow, east west striking undulations of high wavelength (200-300 km) in topography and gravity, unusual structures and geophysical features make the CIOB (central Indian Ocean basin) the locus of anomalous deformation and unique among all the oceanic basins [4-8]. Mid plate deformation of CIOB and the evidences for long-term diffuse deformation of lithosphere of central Indian Ocean, and intra plate deformations are well documented [9-12]. The deformation in the central Indian Ocean interpreted as a diffuse boundary between the two tectonic plates [13-19]. Evidence from magnetic and seismic reflection data suggests abandoned Paleocene spreading center in the northeastern Indian ocean [20]. Information on periodic deformation of oceanic crust in the central Indian Ocean is made available by Neprochnov et al.[21].
　　The first investigations of the intraplate deformations in CIOB were performed by Soviet scientists of P. P. Shirchov Institute of Oceanology in 1976 on R/V “Vityaz” cruise 58 [22] and continued in 1980 on R/V “Dimitrg Mendeleev” cruise 25 and in 1981 on R/V “Academician Kurchatov” cruise 32 [23] further in 1989 by R/V “Professor Shtokman” cruise 22. According to previous seismic refraction data, the structure of the earth’s crust in the CIOB is typical three-layered [24, 25]. The crustal structure of the CIOB exhibits strong heterogeneity both in velocity and crustal thickness. Heat flow characteristics are considered to derive the velocity models imaged by OBS (ocean bottom seismometer) surveys and analyze these models with geodynamic model, which accounts for the tectonic activity in this region.
　　
　　Fig. 1 Location map of the study area, with all the deployed OBSs on different locations.
　　2. Geodynamic Setting
　　The study area is located between 2.5°N to 2.0°S latitude and 81°E to 82°E longitude as illustrated in Fig. 1. The oceanic crust in this part of CIOB was evolved by intraplate seismicity, spreading systems of India-Antractica and Wharton ridge [26-28]. Intraplate compression has probably caused widespread deformation within the CIOB during late Miocene time(7.7-8.2 Ma), as inferred by DSDP (deep sea drilling project) and piston core evidences [5, 6, 11, 29, 30]. The faulting has significantly modified the regular long-wavelength basement topography to produce a“Sharp-Crest-Broad-Trough” profile [30].
　　The study area is marked by the rectangle and it lies in a region of 2°N to 2°S latitude and 81°E to 82°E longitude. The location of the deployed OBSs is marked on the profiles. The heat flow values in mWm^-2(after Geller et al. [31]) are marked on the profile. These characteristic indicates the uniqueness of the basin.
　　The structural boundaries of CIOB in the north is marked by continental margin of India and Sri Lanka, in the west it is marked by the Chagos-Laccadive Ridge, in the east it is marked by Ninety-east Ridge, in the south it is marked by the Southeast Indian Ridge. The area is covered with 1.2 km to 2.5 km thick sediments and is associated with seismicity, excess geothermal energy. The mid-plate deformation of sediments and basement in the central Indian Basin were first reported by Ewing et al. [32], who suggested that their existence was the result of increasing compressional stress in the interior of the Indo-Australian plate due to the collision between the Indian subcontinent and Asia. The beginning of these deformations is believed to be connected with the Miocene Himalayan orogenic phase of this collision [5].
　　The collision of the continents of Asia and India has been proposed in the literature as the cause of the intraplate deformations of sediments and basement in the northern CIB [5, 31-34]. However, the mechanism that generates these deformations has not been sufficiently studied; it has only been suggested that they may have resulted from N-S horizontal compressive stress caused by the continental collision. We have attempted to develop this reasoning and propose a possible mechanism for the originating of the intraplate deformations.
　　High heat-flow is another geophysical parameter, which points to significant intraplate tectonic activity in this part of Indian Ocean. Localized heat-flow of more than 50 W·m^-2 above regional background, typical for the oceanic lithosphere of late Cretaceous-early Tertiary age has been measured in the northern central Indian basin [5, 35-39]. It is characterized by non-linear gradients [35] and suggested to have been generated by hydrothermal circulation along fault planes [36]. Heat flow data provide significant information for understanding the evolution of active continental margins and the occurrence of natural hazards, because the up-dip and down-dip limits of rupture during great subduction thrust earthquakes are suggested to be controlled by the thermal structure of the down going slab [40-42]. The CIOB through heat flow is a system of surface currents flowing from the Pacific Ocean to Indian Ocean through the Indonesian seas. It is the only flow between ocean basins at low latitudes and consequently plays an important role in the meridional transport of heat in the climate system.
　　Widespread tectonic deformation of sediments and basement is consistent with a near-equatorial E-W trending seismic zone in the central Indian Ocean [13, 23, 43]. The present day intraplate high seismicity in this region is considered as unique among oceanic basins of the world [4, 9, 43-49]. Tele-seismic focal mechanisms indicate maximum N-S compressive stress.
　　3. Seismic Data Acquisition and Processing
　　OBS measurements were conducted under the project B-2.3 “Crustal Structure of the Indian Ocean”by R/V Professor Shtoman, Cruise 22 in the first half of 1989. The study area covers about 300 km of oceanic crust of intraplate deformation with anomalous properties of geophysical fields & crustal structure in CIOB and is unique among the world oceans. A total of 18 OBSs were deployed on various profiles, out of which the Profile-4, 5, 6 & 7 are selected for the study of the lithospheric structures of CIOB. Among all the selected profiles, data of five OBSs are available. OBS data processing included relocation of the instrument position by analyzing the direct arrivals, application of frequency filtering and other processing steps such as coherency mixing which made the interpretation of OBS data less ambiguous. The seismic energy was sufficient to trace signals on the record sections. The data were preprocessed with a minimum phase spectral deconvolution (whitening), a Butterworth band-pass filtering (4 Hz and 15 Hz for cutoff frequencies); an AGC (automatic gain control) was applied and finally subjected to the spectral matrix filtering. Spectral whitening efficiently removed ringing effects caused by the narrow frequency band of the source signature at the expense of signal-to noise ratio that was improved by the subsequent band pass filtering and spectral matrix filtering. The latter consists in projecting the signal on the main eigenvectors of the frequency domain cross-correlation matrices in order to separate the coherent signal from uncorrelated noise. Numerous phases can be distinguished, among which refracted and reflected waves are recognized. The strong lateral variation between arrivals recorded along the profile illustrates the complexity of the medium. In this study, we only refer to P wave arrivals.
　　4. Materials and Methods
　　A method of seismic travel time inversion (ZS 92) for simultaneous determination of 2-D velocity and interface structure is applied to map the lithospheric configuration of CIOB. The advantage of inversion, as opposed to trial-and-error forward modeling, is that it provides estimates of model parameter resolution, uncertainty and non-uniqueness, and an assurance that the data have been fit according to a specified norm. In addition, the time required to interpret data is significantly reduced. The inversion scheme is iterative and is based on a model parameterization and a method of ray tracing suited to the forward step of an inverse approach. The number and position of velocity and boundary nodes can be adapted to the shot-receiver geometry and subsurface ray coverage, and to the complexity of the near-surface. The model parameterization also allows ancillary amplitude information to be used to constrain model features not adequately resolved by the travel time data alone. The method of ray tracing uses an efficient numerical solution of the ray tracing equations, an automatic determination of take-off angles, and a simulation of smooth layer boundaries that yields more stable inversion results. The partial derivatives of travel time with respect to velocity and the depth of boundary nodes are calculated analytically during ray tracing and a damped least-squares technique is used to determine the updated parameter values, both velocities and boundary depths simultaneously. The stopping criteria and optimum number of velocity and boundary nodes are based on the trade-off between RMS (root mean square) travel time residual and parameter resolution, as well as the ability to trace rays to all observations. Strategies for modelling seismic refraction/wide-angle reflection travel times to obtain 2-D velocity and interface structure are presented along with methods for assessing the reliability of the results. Emphasis is placed on using inverse methods, but a discussion of pre-modelling considerations such as arrival picking and classification, data uncertainties and fitting, travel time reciprocity, crooked line geometry, and the selection of a starting model is also applicable to trial-and-error forward modelling. The most important advantages of an inverse method are the ability to derive simpler models for the appropriate level of fit to the data, and the ability to assess the final model in terms of resolution, parameter bounds and non-uniqueness. Given the unique characteristics of each dataset and the local earth structure, there is no single approach to modelling wide-angle data that is best. The modeling strategies according to the model parameterization, inclusion of prior information, complexity of the earth structure, characteristics of the data, and utilization of coincident seismic reflection data should be considered before any attempt. There are two natural end member inversion styles: (1) a regular, fine-grid parameterization when seeking a minimum-structure model, and (2) an irregular grid, minimum-parameter model when considering certain forms of prior information. The former style represents the “pure” tomography approach. The latter style is closer to automated forward modelling, and can be applied best with a parameter-selective algorithm, that is one that allows any subset of model parameters to be selected for inversion. Data from regions with relatively strong lateral heterogeneity in the near-surface are best treated using layer stripping, whereas data from regions that are generally complex throughout are best treated using whole-model inversion that is determining all model parameters simultaneously after careful construction of a starting model that allows the appropriate rays to be traced to all pick locations. The lateral spacing of model nodes will depend on the type of inversion and whether detailed prior information, such as from reflection data, is included, but a general guideline is the shot spacing(receiver spacing for typical marine data), except perhaps in the upper layers where about half this may be necessary. Travel times picked from pre-stack, unmigrated or migrated coincident reflection data can be (1) used to develop the starting model, (2) inverted simultaneously with the wide-angle data, or (3) inverted after modelling the wide-angle data to constrain interfaces that “float” within the velocity model. Model assessment establishes the reliability of the final model. Presenting model statistics, travel time fits, ray diagrams, and resolution kernels is useful, but can only indirectly address this issue. Direct model assessment techniques that derive alternate models that satisfactorily fit both the real data. Prior information are the best means for establishing the absolute bounds on model parameters [50]. The same procedure is applied here to examine and validate the lithospheric structure of CIOB. The quality of the model which is derived by the analysis of OBS travel time data depends upon the picking of reflectors because picking of first arrival and other lateral arrivals are used to develop the preliminary model. A preliminary model of the study area is assumed as shown in Fig. 2. This model is used to identify first and later arrivals by overlying the predicted time on the data and then using the enlarged dataset to develop the model further in bootstrap fashion. The next important step is to assign the uncertainties to the arrival picks to avoid over- or under-fitting the data and to allow the appropriate up and down-weighting of the contribution to the solution from relatively certain and uncertain picks respectively. An automated scheme is possible whereby an empirical relationship between signal-to-noise ratio and uncertainty is applied to each lick [51].
　　The uncertainties are estimated for each picking in the OBS dataset and found under the limit. The arrival picks were checked before modelling other locations depending upon the source receiver geometry. If the sources are located near the sea surface and the receivers are on sea floor, then a correction is required[52], which is incorporated in the algorithm given by Zelt et al. [50]. The important step before inversion is the construction of a starting model. Depending upon the study area location, previous and nearby experiments and geological/tectonic information different starting models were prepared and the effect of ignoring or down weighting certain pieces is tested. The best starting model with minimum error is prepared by trial and error method, which is shown in Fig. 3. One dimensional velocity depth sections for all the OBSs have been calculated and superimposed on the preliminary model to get the better correlation (Fig. 3). With a starting model that allows utilization of the data of all phases, an inversion can be stable and converge rapidly [53].
　　
　　Fig. 2 Preliminary constructed crustal model for the travel time modeling. All the five OBSs are marked on the profile.
　　
　　Fig. 3 Starting crustal model constructed by trial and error method for the travel time modeling.
　　A modeling technique [50, 54, 55] was chosen for OBS data, and resulting lithospheric modal for the study area (all the available OBS) is shown in Fig. 4 which demarcates the ray tracing and time distance plot for all the five OBSs deployed along a north-south profile in the CIOB. By this method, it is possible to find the simplest model that fits all at the data without over or under-fitting. And therefore it could avoid over or under interpretation or under utilization of data. The starting model contains relatively simple features from prior information e.g. vertical velocity gradient or discontinuities from amplitude modeling. The forward modeling is based on the trapezoidal methods, in which theoretical rays and their corresponding travel times are calculated from two dimensional laterally heterogeneous, isotropic models and subsequently compared to the acquired OBS data. For this iterative approach a velocity depth model, like the one displayed in Fig. 2, is constructed until the calculation travel times adequately fit the observed data as shown in Fig. 4. The quality of the achieved velocity model mostly depends on the qualitative estimate at phase identification uncertainty. First arrivals at the OBS data in near offset range (< 25), including refracted wave in the sedimentary section at the outer high and refracted wave through the oceanic crust, could be accurately identified with uncertainties at less then ±40 ms. At large offsets, accuracy declined to about ±80 ms due to a lower signal to noise ratio. Events from the plate boundary are clear and continue upto 30-40 km on different stations beneath the outer high of dip line with decreasing amplitude and clearing further landward at grater depth. Their travel times could be picked with errors varying from ±40 ms to ±80 ms. P_mP reflection from the Moho were clearly recorded by station deployed on the outer high, though the picking accuracy only lies between ±80 ms -±130 ms due to their late arrivals which partially interfere with earlier events and noise. OBS data from profile-4 have a very complex travel time pattern due to the rapid change in water depth and the variable composition at the frontal accretionary prism outer high. The velocity tie points were later used to determine the velocity field. The velocity model of the profile was developed after modeling identified arrivals in the OBS data and incorporating prominent events in the reflection data by applying a top to bottom approach. Thus further constraints on the velocity structure were achieved by the two dimensional tomographic inversion, which represents a good quality lithospheric model of COIB with less ambiguity.
　　
　 Fig. 4 Ray tracing and travel time plot for all the five OBSs. (a) Rays reflected on the major discontinuities and turning rays are observed. The shallowest reflection actually corresponds to the undulated upper crustal layer reflector in direct continuity with the top of the backstop; (b) Superposition of travel time curves corresponding to the above illustrated ray paths. The travel time pattern curves show the best fitting in forward model.
　　5. Interpretation
　　Study area is 300 km long and extends from north to south to the CIOB (Fig. 1). Five instruments recorded data on the ocean floor, which allows some insight into the lithospheric structure. Over the entire cross section the sediments cover varies from 6.56 km to 9.36 km in thickness. The sediments covered by 2-3 km thick layer, and crystalline thickness of about 7.8 km on an average were observed. The upper oceanic crust was modeled with a higher velocity gradient (layer 2 and 3) then the lower crust (layer 4), which displays a velocity increase from 6.7 km/s to 8.0 km/s (on an average) for all the OBS. Integrated interpretation is done for all the OBS data independently and is discussed in the following section.
　　5.1 OBS-1
　　OBS-1 was deployed at 36.6 km from the starting point on study area and the sea floor depth was calculated 4.8 km. Recorded data was processed using known processing steps. An integrated interpretation was done and total three crustal layers of thicknesses 2.4 km, 2.3 km and 2.8 km have been identified respectively and are shown in Fig. 5.
　　
　　Fig. 5 Time distance plot for OBS1. The inset features of the deepest reflection (P_x) and refraction (P_y) are observed on the data. The subscripts x and y correspond to the velocity of the reflected and refracted waves.
　　Two dimensional tomography was performed by using ZS 92 code and it is found that two upper crustal layers (layer 2 and 3) at 7.2 km and 9.5 km depth have higher velocity gradient than the lower crustal layer(layer 4) at 12.3 km depth. The velocities of all the crustal layers were calculated 2.1 km/s to 2.4 km/s(layer 2), 5.8 km/s to 6.2 km/s (layer 3), 6.9 km/s to 6.9 km/s (layer 4) respectively. The depth to Moho was found 12.31 km with velocity of 7.65 km/s. One layer of thickness 4.3 km and of 8.9 km/s velocity was also modelled in upper Mantle.
　　5.2 OBS-2
　　OBS-2 was deployed on 4.68 km deep ocean bottom at about 35 km apart from OBS-1 on study area and the data was recorded carefully. Total three crustal layers of thicknesses 2.9 km, 1.9 km and 1.3 km respectively were observed and are shown in Fig. 6. After applying ZS 92 2D tomography approach it is observed that upper crustal layers (layer 2 and 3) at 7.6 km and 9.6 km depth have higher velocity gradient than the lower crustal layer (layer 4) at 11.9 km depth. The velocities of all the crustal layers were calculated as 2.5 km/s to 4.3 km/s (layer 2), 5.7 km/s to 6.1 km/s (layer 3), 7.0 km/s to 7.6 km/s (layer 4) respectively. The depth to Moho was observed 11.9 km with a high velocity of 7.8 km/s. Pn is recorded by OBS-2 at ocean bottom. Mantle velocity is calculated 8.9 km/s at 14.7 km depth.
　　5.3 OBS-3
　　OBS-3 was deployed at about 35 km apart from OBS-2 in the study area, the sea floor depth was calculated 4.7 km and the data was recorded. Three prominent crustal layers were identified with varying thicknesses of 3.4 km, 2.0 km and 4.2 km respectively and illustrated in Fig. 7. The velocity depth function of the incoming igneous crust at this point shows a very high velocity gradient from 2.1 km/s to 5.1 km/s (layer 2), 5.9 km/s to 6.2 km/s (layer 3), and 7.1 km/s to 7.4 km/s (layer 4). P_n is recorded by OBS-3 on the ocean basin in a well manner and Mantle velocity was calculated 8.9 km/s.
　　
　　Fig. 6 Time distance plot for OBS2. The inset features of the deepest reflection (P_x) and refraction (P_y) are observed on the data.
　　
　　Fig. 7 Time distance plot for OBS3. The inset features of the deepest reflection (P_x) and refraction (P_y) are observed on the data.
　　5.4 OBS-4
　　OBS-4 was deployed about 65 km apart from OBS-3 at 4.63 km sea floor depth. Because of reverberations the data appeared noisy. Some frequency filters and coherency mixing was applied to remove these reverberations. Total three crustal layers of thicknesses 2.4 km, 1.9 km and 3.9 km were observed respectively. After 2D tomography the velocity depth function at this point shows that two upper crustal layers (layer 2 and 3) at 7.0 km and 8.9 km depth have higher velocity gradient than the lower crustal layer (layer 4). The velocities of all the layers were calculated 2.1 km/s to 3.9 km/s (layer 2), 5.6 km/s to 6.0 km/s (layer 3), 6.9 km/s to 7.5 km/s (layer 4) respectively. The Moho reflection was recorded only in the northern side. The depth to Moho was calculated 12.9 km with velocity of 7.7 km/s. The velocity of upper Mantle was calculated as 8.7 km/s.
　　5.5 OBS-5
　　OBS-5 was deployed about 80 km apart from OBS-4 at 4.7 km depth of sea floor. As the case of OBS-4, due to reverberations the data appeared to be noisy and the same frequency filters and coherency mixing was applied to get the better results. Total three crustal layers of thicknesses 1.7 km, 2.8 km and 4.7 km respectively were calculated. After the tomographic approach the velocity depth function at this point shows that two upper crustal layers (layer 2 and 3) at 6.32 km and 9.10 km depth have higher velocity gradient than the lower crustal layer (layer 4). The velocities of all the layers were calculated 2.2 to 4.1 km/s (layer 2), 6.2 km/s to 6.3 km/s (layer 3), 7.1 km/s to 7.5 km/s (layer 4) respectively. As Moho reflection was recorded only northern side well due to the reverberations, the depth to Moho was calculated 13.8 km with velocity of 7.8 km/s. The upper mantle layer was dissolved well and the thickness of that layer was calculated as 1.3 km. The velocity of upper Mantle was calculated as 8.6 km/s.
　　5.6 Lithospheric Structure and Tectonics
　　The retrieved interfaces are well resolved in the entire region of the model (Fig. 8a) and the ray convergence is dense. The two-dimensional travel time inversion method has been applied on OBS data (Fig. 8b). The resolution is assessed with a local exploration of the objective function around the final model. The modeling/inversion of wide-angle travel times usually involves some aspects that are quite subjective like: (i) identifying and including later phases; (ii) assigning specific layers to arrivals; (iii) incorporating pre-conceived structure not specifically required by the data; (iv) model parameterization; (v) inclusion of prior information; (vi) complexity of the earth structure, and(vii) characteristics of the data. These steps are applied to maximize model constraint and minimize model non-uniqueness. For datasets with Moho reflections, the tomographic velocity model is used to invert the PmP times for a minimum-structure Moho.
　　
　　Fig. 8 Ray tracing and time distance plot derived by two dimensional travel time inversion for all the five OBSs: (a) Rays reflected on the major discontinuities and turning rays are observed. The shallowest reflection actually corresponds to the undulated upper crustal layer reflector in direct continuity with the top of the backstop; (b) Superposition of travel time curves corresponding to the above illustrated ray paths. The travel time pattern curves show the minimum error in inverted model.
　　In this way, crustal velocity and Moho models are obtained (Fig. 9) that requires the least amount of subjective input. The model structure that is obtained by the wide-angle data with a high degree of certainty is differentiated from structure that is merely consistent with the data. The tomographic models are not intended to supersede the preferred models, since the latter model is typically better resolved and more interpretable. This form of tomographic assessment is intended to lend credibility of model features common to the tomographic and preferred models. The thickness of the crust is increasing towards north and clearly resolved by the inversion (Fig. 9). A large lithospheric single layer folding is also prominent and demarcated on the final crustal model.
　　A wide angle seismic survey with OBS in CIOB provides data that enables the identification and mapping of the depth of crystalline basement, Moho and upper Mantle by ray tracing and inversion technique. Considering the two dimensional tomographic inversion of OBS data acquired in central Indian Ocean basin, crustal sections were constructed along the 300 km long north south profile (Fig. 9). Four layers namely the sediment layer (layer 1), the upper crustal (layer 2), the lower crustal (layer3) and a sub-crustal layer (layer 4) have been delineated by modeling the OBS data. Over the entire cross section the sediments cover varies from 2.3 km to 3.38 km in thickness. The thick sediment layer under OBS2 and OBS3 surprisingly veneers to about 1.65 km thickness under OBS5 in the north. The thicken sediment layer under OBS2 and OBS3 is underlain by thin upper crustal layer. This thickens out again towards north producing single layer folding in the vicinity of OBS3. The second crustal layer exhibits similar pattern as of layer 2, however no folding is seen in this layer. The second crustal rapidly thickens very much towards north. Under this crustal configuration the Moho undulates from around 12 km in the south to about 14 km in the north. The Moho is shallow under OBS2 giving rise to thinned crust under OBS2 and OBS3. The Moho is underlain by a sub-crustal layer, whose thickness varies appreciably from south to north. This layer is very thin towards the north. The seismic velocity varies appreciable in the sedimentary column((2.1-2.47 (upper surface) to 2.35-5.12 (lower surface)) km/s with very high gradient under OBS3, where the lower interface is at greater depth. By virtue of lower surface becoming deeper, the sediments are more compacted resulting in higher value of velocity. In the upper crustal layer, interface velocities of the order of((5.62-6.13 (upper surface) to 6.21-6.34 (lower surface)) km/s and in the lower crustal layer((6.87-7.07 (upper surface) to 6.97-7.61 (lower surface)) km/s are obtained. The sub-crustal layer has the velocity in the range 7.6-7.82 (upper surface) to 7.76-8.25 (lower surface) km/s. Very high gradients are encountered in the sedimentary column in comparison to all other layers.
　　
　　Fig. 9 Final smoothed crustal configuration model obtained by the 2D travel time inversion technique. The locations of all the deployed ocean bottom seismometers are clearly marked. Undulations in the entire crustal layers are clearly resolved and thinning (dashed black color) and thickening (solid black color) of the crust is noticeable. A large-scale single layer folding in the upper crustal layer is demarcated. The Moho boundary and the upper mantle layer is determined on the entire profile.
　　6. Results and Discussion
　　In the present investigation the wide angle seismic data collected along 300 km N-S profile (Fig. 1) was modeled to delineate shallow as well as deeper lithospheric configuration in the deformation region. the obtained lithospheric configuration shows interesting characteristics under the region of intense deformation area. From the inversion of OBS data we are able to resolve four layers including the top sedimentary layer. The compressional velocity in the sedimentary column ranges from 2.1-2.47 (upper surface) to 2.35-5.12 (lower surface) km/s. There is appreciable increase in velocity at the lower surface due to changes in the thickness of this layer. The increase in velocity with in the sediment layer agrees well to the results obtained in this region [23, 56] and the higher velocity values may be ascribed to the presence of pelagic at the bottom of sediments [57]. The mean velocity value for this layer varies from 3.1 km/s to 3.6 km/s, except for the first OBS, where the velocity value for this layer is significantly smaller.
　　Results indicate that the tectonic framework of NCIOB (northern central Indian Ocean basin) is controlled by tectonic element of two genetic types. The old structural pattern seems to have been affected by the largest near-meridional transform faults formed during lithospheric constructions in late cretaceous. Fracturing of the oceanic crust developes as a result of sea floor spreading and tend to get align in two directions. It is likely that besides the observed near-meridional transform fault there may exist smaller near latitudinal cracks and initial zone of weakness in the crust parallel to the paleo rift valley. The intermittent structures observed in many compressional zones are the result of large-scale lithospheric folding occurring during different phasesat the premature stages, resolved folding propagates downwards as compressive stresses build up. When the whole competent layer is at the yield state, folding starts to grow rapidly with a stable prevailing wavelength. Although preexisting zones of weakness are certainly present in nature, it is not necessary to introduce them into the models to trigger development of folding. These effects propose that the heterogeneous lithosphere preserve the considerable horizontal strength and can efficiently behave as a strong layered media. The results can be further used to resolve the apparent contradictions between insights from intraplate stress fields and the plate dynamics.
　　6.1 Oceanic Crust
　　Velocity interface structures and velocity-depth function of the incoming igneous crust in the study area show a high velocity gradient in the upper crust. The integrated interpretation of OBS data reveals a relatively detailed vertical and lateral velocity distribution observed from refracted diving waves. The most characteristic aspect of the shallow sediments is much higher sea floor velocity in the northern part of the profile. The variation in the velocities indicates lateral lithology change.
　　6.2 Upper Crystalline Crust
　　The top of the crystalline basement consists of two layers of varying thicknesses and velocities. These sediments represent the low- and high-velocity layers, which disrupt the wave field, and make the data difficult to interpret and model. In most part of the upper crystalline basement the velocities varies from 2 km/s to 6.3 km/s and high velocity gradient is observed. A large scale lithospheric single layer folding is observed as demarcated in Fig. 9.
　　6.3 Lower Crystalline Crust
　　The lower crystalline basement consists of single layer of varying thickness from 1.3 km to 4.6 km. The thickness of this lower crustal layer increases from south to north, which may be caused by increasing sediment load. The velocity varies from 6 km/s to 7.5 km/s in most part of the lower crust. This high velocity in the lower crust has been interpreted due to magmatic under plate body. In this model as mantle goes down passively beneath the continental lithosphere, huge amount of basaltic melt are generated by adiabatic decompression melting of the hot asthenospheric mantel. This basaltic melt migrates rapidly towards upward unit; it is partially extruded as basalt flows and partially intruded into or beneath the crust. This causes increase in seismic velocity of the igneous rocks emplaced in the crust from 6.8 km/s for normal temperature to 7.6 km/s or higher. The thickness of the lower crustal high velocity body varies considerably within the area from 5.9-10.0 km. The strong lateral variation in the thickness and velocity of the layer might be caused by variation in magma distribution process.
　　6.4 The Moho and the Upper Mantle
　　The base of the lower crustal layer (layer 4) is interpreted as the Moho (top of the Mantle). The upper mantle velocity is estimated to be 8.9 in the southern and northern most part of the study area. In the middle part of the study area the mantle reflections are also clearly resolved. The deepest interfaces may be interpreted as upper mantle shear-zones which are inferred also on the Lofoten Margin [58, 59]. This is the deepest layer, which is clearly resolved by the OBS data.
　　A very high heat flow, which is observed in the study area, shows typical physical characteristics of deeper parts of the central Indian Ocean basin. The observed basement trend in the CIO at 83°E and 0°N suggests that it has been formed due to the change in the physical characteristics and partially because of the sediment load. The distributed oceanic basement and overlying deformed sedimentary layers may indicate the presence of the earliest deformation in the central Indian Ocean basin. The present topography (2°N to 2°S and 80°E to 84°E) of the younger unconformities present in the CIOB has gradually developed due to the interrupted deformation. There is some correlation existing between the annual cycle of heat flow into the CIOB via meridional heat transport and the change in energy associated with an increase in temperature. It may conclude that the oceanic heat transport (by any mechanism) regulates the temperature structure in central Indian Ocean basin. An extensive nature of heat transport is observed on the basis of the available heat flow values. The central Indian Ocean proves the area of highest oceanic neotectonic activity manifested in unusual deformation and extremely high seismicity and heat flow. On the contrary, large N-S fracture zones in the central Indian Ocean are oriented perpendicularly to speculate India-Australia diffuse plate boundary. There is no evidence of E-W deep faults over the basins, which would penetrate through entire oceanic lithosphere. On the basis of the above-mentioned facts it can be predicted that the upper brittle crustal sub layer is dissected into individual geo blocks, which are capable of horizontal displacements over underlying astheno layer relative to the mantle part of the lithosphere. If any deformation will affect the entire crustal and uppermost mantle structure only in the plate interior and rest lower lithosphere is unbroken, this deformation seems to be intraplate one. More typical continental diffuse plate boundaries imply mostly existence of deepest faults. The OBS data revealed the development of the inelastic folding within a continental lithosphere, depending on the strength and thickness of the lower crust. OBS data acquisition presents a promising perspective for much improved deep crustal studies, both qualitatively and quantitatively. From a data processing aspect, 2-D travel time inversion approaches (ZS 92) could be investigated for quantifying the model properties.
　　In general, it can be argued that the intraplate deformation appears to be associated with the tectonic activity. But, the deep structure of crust and upper mantle, which was determined by limited seismic refraction studies, is still unclear beneath the broad scale basement deformation. On the basis of the result of the OBS modelling it is interpreted that under reasonable tectonic stresses, folds can develop the brittle parts of a lithosphere. The unusual inelastic folding can be considered as the mode of deformation.
　　7. Conclusions
　　In this work the compressional waves observed on the vertical component of five OBS data in the CIOB have been modelled using two dimensional ray tracing and tomographic inversion. These results incorporate the new findings including unresolved upper mantel layer characteristics. The OBS data provide a reliable estimate of the velocities of the sediments from the sea floor to the upper mantle layer. The shallow and intermediate depth sediments are characterized by a vertical increase in velocity due to the increased confining pressure. There is also considerable lateral variation in the velocities within the sedimentary layer at all levels and most of this variation can probably be attributed to varying depth of burial.
　　The analysis of the acquired OBS data have provided models of the entire crust in an area where other seismic techniques have been hampered with relatively large uncertainties, as far as the imaging of the sub-sedimentary part of the crust is concerned. The velocity variation in the crust indicates the lateral change in lithology. The measured velocity of the upper and middle crystalline crust increases with depth from 5.7 km/s to 6.8 km/s, which suggests that the crust in the CIOB is of continental origin. In most parts of the study area the velocity of the lower crust is very high, 7.0-7.6 km/s. This high velocity is interpreted as indicating the presence of a magmatic under plated body with varying thickness. The strong lateral variations in the thickness of the high-velocity layer might be caused by variations in the pre-breakup structure and/or spatial variations in the magma distribution process. The base of the lower crustal layer is defined by an interface at about 10 km depth, which is interpreted as the Moho boundary. The depth to Moho is increasing from south to north (12.2-13.8 km) due to high compressional forces which are increasing with the sediment load.
　　Acknowledgments
　　The author would like to thank Director, National Geophysical Research Institute (NGRI), Hyderabad, for his permission to publish this article. A detailed discussion and useful suggestion of Dr. V.K. Gahalaut and Prof I.V.R. Murthy helped to improve the manuscript. N.K. Thakur acknowledges the financial support by CSIR, New Delhi under Emeritus Scientist scheme.
　　References
　　[1] M.F. Coffin, O. Eldholm, Large igneous provinces: Crustal structure, dimensions and external consequences, Rev. Geophys. 32 (1994) 1-36.
　　[2] C. Deplus, M. Diament, H. Hebert, G. Bertrand, S. Dominquez, J. Dubois, et al., Direct evidence of active deformation in the eastern Indian Oceanic plate, Geology 26 (1998) 131-134.
　　[3] C.M. Jarchow, R.D. Catchings, W.J. Lutter, Large-explosive source, wide-recording aperture, seismic profiling on the Columbia Plateau, Washington, Geophysics 59 (1994) 259-271.
　　[4] S. Stein, E.A. Okal, Seismicity and tectonics of the Ninety-east Ridge area: Evidence for internal deformation of the Indian plate, J. Geophys. Res. 83 (1978) 2233-2245.
　　[5] J.K. Weissel, R.N. Anderson, C.A. Geller, Deformation of the Indo Australian plate, Nature 287 (1980) 284-291.
　　[6] J.R. Curray, T. Munasinghe, Timing of intraplate deformation, northeastern Indian Ocean, Earth and Planet Sci. Lett. 94 (1989) 71-77.
　　[7] J.R. Curray, G.G. Shor, R.W. Raitt, M. Henry, Seismic refraction and reflection studies of crustal structure of the eastern sunda and western Banda Arcs, J. Geophys. Res. 82 (1977) 2479-2489.
　　[8] O.V. Levchenko, L.R. Merklin, The M31 cruise seismic reflection studies in intraplate deformation of the central Indian Ocean basin, Geol. Soc. India Memoir 39 (1998) 95-102.
　　[9] K.S. Krishna, D. Gopala Rao, M.V. Ramana, V. Subrahmanyam, K.V.L.N.S. Sarma, A.I. Pilipenko, et al., Tectonic model for the evolution of oceanic crust in the northeastern Indian Ocean from the late Cretaceous to the early Tertiary, J. Geophys. Res. 100 (1995) 20011-20024.
　　[10] Yu. P. Evsykov, C. Subrahmanyam, K.S.R. Murthy, S.N. Rai, Sea floor geomorphology: Bathymetric investigations in the S22 cruise area, Geology & Geophysical Sciences 39 (1998) 29-37.
　　[11] J.M. Bull, J. Martinod, Ph. Davy, Buckling of the oceanic lithosphere from geophysical data and experiments, Tectonics 11 (1992) 537-548.
　　[12] D.A. Wiens, C. DeMets, R.G. Gordon, S. Stein, D. Argus, J.F. Engeln, et al., A diffuse plate boundary model for Indian Ocean tectonics, Geophys. Res. Lett. 12 (1985) 429-432.
　　[13] R.G. Gordon, C. DeMets, D.F. Argus, Kinematic constraints on distributed Lithospheric deformation in the equatorial Indian Ocean from present motion between the Australian and Indian plates, Tectonics 9 (1990) 409-422.
　　[14] R.G. Gordon, K.E. Petronotics, G.D. Acton, B. Johnson, S. Vasas, Apparent polar wander of the hotspots: The view from the Pacific, Ann. Geophy. 16 (1998) C56.
　　[15] J.Y. Royer, T. Chang, Evidence for relative motions between Indian and Australian plates during the last 20 Myr from the plate tectonic reconstructions: Implications for the deformation of the Indo-Australian plate, J. Geophys. Res. 96 (1991) 11779-11802.
　　[16] C. DeMets, R.G. Gordon, P. Vogt, Location of the Africa-Australia-India triple junction and motion between the Australian and Indian plates: Result from an aeromagnetic investigation of the central Indian and Carlsberg ridges, Geophys. J. Int. 119 (1994) 893-930.
　　[17] J.Y. Royer, R.G. Gordon, The motion and boundary between the Capricorn and Australian plates, Science 277(1997) 1268-1274.
　　[18] J.Y. Royer, R.G. Gordon, C. DeMets, P.R. Vogt, New limits on the motion between India and Australia since chron 5 (11Ma) and implications for lithospheric deformation in the equatorial Indian Ocean, Geophys. J. Int. 129 (1997) 41-74.
　　[19] K.S. Krishna, D. Gopala Rao, Abandoned Paleocene spreading center in the northeastern Indian Ocean: Evidence from magnetic and seismic reflection data, Mar. Geol. 162 (2000) 215-224.
　　[20] K.S. Krishna, M.V. Ramana, D. Gopala Rao, K.S.R. Murthy, M.M. Malleswara Rao, V. Subrahmanyam, et al., Periodic deformation of oceanic crust in the central Indian Ocean, J. Geophys. Res. 103 (1998) 17859-17875.
　　[21] Yu.P. Neprochnov, L.R. Merklin, V.E. Milanovsky, The thickness and structure of sedimentary cover based on seismic reflection data, Geophysics of the Oceanic 1 (1979) 206-242.
　　[22] O.V. Levchenko, L.R. Merklin, Yu.P. Neprochnov, Folded structures in the central trough of the Indian Ocean, Geotectonics 19 (1985) 10-15.
　　[23] Yu.P. Neprochnov, O.V. Levchenko, L.R. Merklin, V.V. Sedov, The structure and tectonics of the intraplate deformation area in the Indian Ocean, Tectonophysics 156(1988) 89-106.
　　[24] Yu.P. Neprochnov, D. Gopala Rao, C. Subrahmanyam, K.S.R. Murthy, Intraplate deformations in the central Indian Ocean basin, Geol. Soc. India. Memoir 39 (1998) 71-86.
　　[25] R.L. Larson, G.B. Carpenter, J.B. Diebold, A geophysical study of the Wharton basin near the investigator fracture zone, J. Geophys. Res. 83 (1978) 773-782.
　　[26] R.L. Larson, J.C. Mutter. J.B. Diebold, G.B. Carpenter, Cuvier basin: A product of ocean crust formation by early Cretaceous rifting off western Australia, Earth and Planet Sci. Lett. 45 (1979) 105-114.
　　[27] B.R. Naini, A marine geophysical survey (site 211 DSDP) in the Wharton basin, Indian Ocean, Init Rep. of DSDP, V XXII, National Science Foundation, Washington, 1974, pp. 835-841.
　　[28] D. Gopala Rao, K.S. Krishna, D. Sar, Crustal evolution and sedimentation history of the bay of Bengal since the cretaceous, J. Geophys. Res. 102 (1997) 17747-17768.
　　[29] K.S. Krishna, J.M. Bull, R.A. Scrutton, Evidence for multiphase folding of the central Indian Ocean lithosphere, Geology 29 (2001) 715-718.
　　[30] S.L. Eittreim, J. Ewing, Mid-plate tectonics in Indian Ocean, J. Geophys. Res. 77 (1972) 6413-6421.
　　[31] C.A. Geller, J.K. Weissel, R.N. Anderson, Heat transfer and intraplate deformation in the central Indian Ocean, J. Geophys. Res. 88 (1983) 1018-1032.
　　[32] M. Ewing, S. Eitttreim, M. Truchan, J.I. Ewing, Sediment distribution in the Indian Ocean, Deep-sea Res. 16 (1969) 231-248.
　　[33] T.J.G. Francis, R.W. Raitt, Seismic refraction measurements in the southern Indian Ocean, J. Geophys. Res. 72 (1967) 3015-3041.
　　[34] T.F. Gaskell, M.N. Hill, J.C. Swallow, Seismic measurements made by H.M.S. challenger in the Atlantic, Pacific and Indian Oceans and in the Mediterranean Sea, Phil. Trans. R. Soc. (A) 251 (1958) 23-28.
　　[35] J.R. Cochran, D.A.V. Stow, Proceedings of the Ocean Drilling Program, College Station, Texas, USA, 1989, p. 388.
　　[36] C.A. Stein, J.K. Weissel, Constraints on the central Indian basin thermal structure from heat flow, seismicity and bathymetry, Tectnophysics 176 (1990) 315-332.
　　[37] C.A. Stein, M.A. Hobart, D.H. Abbott, Has the Wharton Basin’s heat flow been perturbed by the formation of a diffuse plate boundary in the Indian ocean?, Geophy. Res. Let. 15 (1988) 455-458.
　　[38] C.A. Stein, S. Cloetingh, R. Wortel, Seasat-derived gravity constraints on stress and deformation in the northeastern Indian Ocean, Geophys. Res. Lett. 16 (1989) 823-826.
　　[39] B.W. Tichellar, L.J. Ruff, Depth of seismic coupling along subduction zones, J. Geophys. Res. 98 (1993) 2017-2037.
　　[40] R.D. Hyndman, K. Wang, Thermal constraints on the zone of major thrust earthquake failure: The Cascadia subduction zone, J. Geophys. Res. 98 (1993) 2039-2060.
　　[41] R.D. Hyndman, M. Yamano, D.A. Oleskevich, The seismogenic zone of subduction thrust faults, The Island Arcs 6 (1997) 244-260.
　　[42] O.V. Levchenko, Tectonic aspects of intraplate seismicity in the northeastern Indian Ocean, Tectonophysics 170(1989) 125-139.
　　[43] C.W. Stover, Seismicity of the Indian Ocean, J. Geophys. Res. 71 (10) (1966) 2575-2581.
　　[44] L.R. Sykes, Seismicity of the Indian Ocean and a possible nascent island arc between Ceylon and Australia, J. Geophys. Res. 75 (26) (1970) 5041-5055.
　　[45] E.A. Bergman, S.C. Solomon, Oceanic intraplate earthquakes implications for local and regional intraplate stress, J. Geophys. Res. 85 (1980) 5389-5410.
　　[46] E.A. Bergman, S.C. Solomon, Earthquake source mechanisms from body waveform inversion and intraplate tectonics in the northern Indian Ocean, Phys. Earth Planet. Inter. 40 (1985) 1-23.
　　[47] J.M. Bull, Structural style of intra-plate deformation, central Indian Ocean basin: Evidence for the role of fracture zones, Tectonophysics 184 (1990) 213-228.
　　[48] D.D. Singh, Strain deformation in the northern Indian Ocean, Mar. Geol. 79 (1988) 105-118.
　　[49] C.A. Zelt, 3-D velocity Structure from simultaneous travel time inversion of in-line seismic data along intersecting profiles, Geophy. J. Int. 118 (1994) 795-801.
　　[50] C.A. Zelt, D.A. Forsyth, Modeling wide-angle seismic data for crustal structure: Southeastern Greenville province, J. Geophys. Res. 99 (1994) 11687-11704.
　　[51] G.M. Purdy, The correction for the travel time effects of sea floor topography in the interpretation of marine seismic data, J. Geophys. Res. 87 (1982) 8389-8396.
　　[52] C.A. Zelt, Seismic velocity structure of the central Chilean margin near the subducting Juan Fernandez ridge: Effective inversion of travel time data across complex, laterally varying structure, in: 7th International Symposium, Asilomar, California, 1996.
　　[53] C.A. Zelt, R.B. Smith, Seismic travel time inversion for 2D crustal velocity structure, Geophys. J. Int. 108 (1992) 16-34.
　　[54] C.A. Zelt, Modeling strategies and model assessment for wide-angle seismic traveltime data, Geophy. J. Int. 139(1999) 183-204.
　　[55] G.T. Leger, K.E. Louden, D.M. Forsyth, J.K. Weissel, Crustal structure and faulting in the intraplate deformation region, central Indian Ocean basin, Trans. Am. Geophys. Union. 68 (1987) 1484.
　　[56] J.M. Bull, R.A. Scrutton, Fault reactivation in the central Indian Ocean and the rheology of oceanic lithosphere, Nature 344 (1990) 855-858.
　　[57] R. Mjelde, M.A. Sellevoll, H. Shimamura, T. Iwasaki, T. Kanazawa, A crustal study off Lofoten, N. Norway, by use of 3-component ocean bottom seismographs, Tectonophysics 212 (1992) 269-288.
　　[58] R. Mjelde, M.A. Sellevoll, H. Shimamura, T. Iwasaki, T. Kanazawa, Crustal structure beneath Lofoten, N. Norway, from vertical incidence and wide angle data, Geophys. J. Int. 114 (1993) 116-126.
　　[59] R. Mjelde, M.A. Sellevoll, Possible shallow crustal anisotropy off Loften, Norway, inferred from 3C ocean bottom seimographs, Geophys. J. Int. 115 (1993) 159-167.