Fine Structure of an Oceanic Crustal Section near the East Pacific Rise by D. v. Helmberger and G. Engen

In this study we model synthetically a complete seismic profile of roughly 20m.y.-old crust located to the west of the East Pacific Rise, 3.37S, 114.13W. The results indicate a rather strong velocity gradient below the sediments with little evidence of layering in the upper crust and a slightly dipping oceanic layer. The crust-to-mantle transition zone appears sharp providing a relatively good wave guide for multiple Moho reflections which are modeled synthetically to further test the usefulness of a layered earth model in explaining entire seismograms. The mantle head waves decay abruptly near 50 km which can be explained by the onset of a low-velocity zone in the upper mantle at about a depth of 12 km below the ocean surface. INTRODUCTION Recent studies suggest that the crust contains a substantial low-velocity zone (LVZ) near the crest of the East Pacific Rise which disappears as the crust ages [see Orcutt et al. (1976)]. There is also evidence that the oceanic crust thickens with age, Shor et al. (1970) and others. Woo]lard (1975) suggests this growth occurs at the base of the crust and is essentially transformed mantle material. The seismic parameters describing this basal layer are not well established, with Sutton et al. (1971) arguing for the existence of a high-velocity zone (HVZ) while Lewis and Snydsman (1976) suggest the development of a LVZ increasing with age. Most of these conclusions are based on prominent second arrivals which are apparent in many seismic profiles. To clarify the effects of basal HVZ and LVZ, we took the Raitt (1963) standard oceanic mode] modifying it so as to preserve the first-arrival times. The model parameters are given in Table 1 with the synthetics shown in Figures I, 2, and 3. These synthetics were generated by the application of the generalized ray theory (GRT) as discussed by Helmberger (1968) and used in previous oceanic modeling attempts [see Helmberger (1977) and Helmberger and Morris (1969)]. For simplicity, we included only those rays which penetrate the 6.7-km/sec layer describing the lower crustal and upper mantle response. The upper crustal response at these ranges is usually found to be quite simple in most observed profiles and can be modeled by a smooth velocity depth function as discussed in Helmberger (1977). Thus we can visualize a simple pulse arriving at nearly constant velocity on the heavy line indicated in the figures. In the step responses of Figure I we can see the reflection from the top of the 7.5km/sec layer followed by the reflection from the top of the mantle. The first four multiple reflections in the 7.5-1ayer were included in these calculations, although only two multiples were necessary for convergence to the same synthetics obtained for model A when the 7.5-1ayer is allowed to become very thin. The synthetics on the right are appropriate for a 20-pound charge fired with the standard shooting setup [see Helmberger (1968)]. The synthetics in the middle column were generated by compressing the time scale of the transfer function by two which allows the apparent velocities of 7.5 and 8.1 to be more easily recognized. The apparent velocities between 6.7 and the mantle velocity of 8.2 are observed 369 3 7 0 D . V . H E L M B E R G E R A N D G. E N G E N on numerous air gun profiles at small ranges, 15 to 30 km, as reported by Maynard (1970) and Sutton et al. (1971). The short-period nature of these experiments makes it much easier to observe such details as can be seen in Figure 1. Unfortunately, air gun data is essentially processed mechanically and is usually not available to the synthetic modeling technique. The LVZ case is displayed in Figure 2 where the most distinguishing feature is T A B L E 1 MODEL PARAMETERS Thickness P Velocity S Velocity Density (km) (km/sec) (kin/see) (gr/cm :~) A. (Rait t Oceanic Model) 3.00 1.50 0.0 1.00 0.45 2.00 0.50 1.10 1.75 5.00 3.50 2.50 4.70 6.70 3.80 2.60 8.10 4.60 3.40 B. (HVZ Transi t ion) 4.50 1.50 0.0 1.00 0.45 2.00 0.50 1.10 1.75 5.00 3.50 2.50 4.10 6.70 3.80 2.60 1.00 7.50 4.26 3.06 8.10 4.60 3.40 C. (LVZ Transi t ion) 4.50 1.50 0.0 1.00 0.45 2.00 0.50 1.10 1.75 5.00 3.50 2.50 3.37 6.70 3.80 2.60 1.00 6.00 3.68 2.56 8.10 4.60 3.40 D. (Down-Dip) 4.50 1.50 0.0 1.00 0.45 2.00 0.50 1.10 1.75 5.00 3.50 2.50 4.70 6.70 3.80 2.60 8.38 4.72 3.52 E. (Up-Dip) 4.50 1.50 0.0 1.00 0.45 2.00 0.50 1.10 1.75 5.00 3.50 2.50 4.70 6.70 3.80 2.60 7.86 4.47 3.26 the lateness of the Moho reflection. This feature is evident in many of the excellent profiles presented by Lewis and McClain (1977) and Lewis and Snydsman (1976). A small amount of dipping structure can also produce this effect. In Figure 3 we present results from models D and E where the oceanic layer has a 1.5 ° dip. The synthetics for model A are superimposed for comparison. For these small angles the transmission and reflection coefficients are essentially the same as in the flat case with the only significant change taking place in the travel times [see Hong and Helmberger (1977) for a general description of this problem]. If we examine the STRUCTURE OF OCEANIC CRUSTAL SECTION NEAR THE EAST PACIFIC RISE 371 standard synthetics it is clear that the exact arrival time of the later pulses is difficult if not impossible to pick. In reality these seismograms would be complicated more by the upper crustal arrival at 6.7 and wave shapes and relative amplitudes become important in deciding which of the above cases suits a given profile. Also, the delay times and apparent velocities could be direction dependent and the need for reversed profiles becomes obvious. ( T A / 6 . 7 ) , sec 6~ Z2 78 66 Z2 z8 6~ 7.2 7.8 FIG. 1. Responses for the model B (HVZ transition) given in Table 1, plotted on a reduced traveltime section. The solid lines at times of 6.68 sec correspond to the crustal arrival times with the crossover occurring near 38 km. ( 7 5 / 6 . 7 ) , sec 6.6 7.2 7.8 6.6 7 2 7 8 6.6 7.2 7.8