Preliminary Woodlark 3-D Seismic Tomography Results

woodlark basin

The following figures show preliminary tomography results from the Woodlark "big shot" experiment. Results are based on data recorded at all 14 working OBSs and OBHs. Note that 2 of the instruments still require drift corrections, although these corrections will likely be quite small over the duration of the "big shot" experiment, and thus should not significantly affect the images shown here. A total of 31421 first arrival data from all 14 shot lines were used in the inversions. Forward modeling was performed on a cubic 0.5 km grid. Regularized inversion was performed on a grid with a vertical node spacing of 0.5 km, and lateral node spacing of 2 km. Data are fit to within a normalized chi-squared of 1.

Figure 1 shows bathymetry in study area, location of OBSs (squares) and OBHs (triangles) and shiptrack lines (numbered dotted lines). Contour interval is 0.2 km. Thick black line in the southeast denote spreading center axes. Black regions are land.

Figure 2 shows horizontal slices of estimated lateral resolution of the velocity model. Regions within the white contour have better than 5 km resolution; white regions have worse than 20 km resolution or are unsampled.

Figure 3 shows horizontal slices through the velocity model at depths of z=3-11 km. OBS/H locations are white dots; ship track lines are heavy lines. Contour interval is 0.5 km/s. White regions are unconstrained (no ray coverage).

Figure 4 is a vertical slice through the velocity model along east-west shiptrack line 1. Contour interval is 0.5 km/s. Color scale at right is velocity in km/s. White regions are unconstrained (no ray coverage). Dotted line is bathymetry. Moresby Seamount is the topographic high at x=50 km.

Figure 5 is a vertical slice through the velocity model along north-south shiptrack line 5. Contour interval is 0.5 km/s. Color scale at right is velocity in km/s. White regions are unconstrained (no ray coverage). Dotted line is bathymetry.

Figure 6 shows the seismicity located within the 3-D velocity model using a grid search method. There are a total of 694 events (black and white dots), 343 of which have error ellipses smaller than 12 km in the X-Y plane (black dots). Note that much of the seismicity is west of the OBS/H array. Panel at left shows projection of all events onto the Y-Z plane. Depth control for most events is poor.

Preliminary results from co-location of 1999 seismiicity using arrivals from both the OBS array and land station array.

In addition the following table lists the relocated northing, easting, depth, latitude and longitude of the 14 working OBS and OBHs. Relocations were based on water wave arrival times.