Lab 05: Sorting, Velocity Analysis and Stacking

At this particular stage, we are ready to prepare the data for revealing a true image of the subsurface where it involves compressing the seismic data to infer a first approximation for such image which includes:

1.      Sorting the shot gathered data into common mid-point gathers (CMP).

2.      Picking appropriate stacking velocities and applying accordingly normal-move-out (NMO) corrections.

3.      Stacking all CMPs and then concatenate all stacked CMP traces together.

Surface seismic reflection surveys are commonly been acquired using the common mid-point (CMP) method. There are also other names for this method for example common reflection point CRP) and common depth point CDP). Traces which reflected from the same mid-point form a CMP gather, while the number of traces in a gather is called the fold of that gather.

As the seismic data processing is performed in the CMP-offset mode, generally we need a way to sort the traces between these modes. In term of MATLAB function, ssort.m is used for sorting those traces accordingly. Later, one will obtain the fold or number of traces per CMP based on another MATLAB function, extracting_cmp_fold_num.m as shown in Figure 1 below.

Figure 1: The fold (number of traces per CMP) versus the CMP numbers

From the Figure 1, one will see the data is started at CMP 203 with one single trace then increases gradually until it reaches the maximum fold of 18 at CMP number 235. Followed by that, it then ends at CMP 266 with once trace. Figure 2 below showing some examples of the CMP gather taken for viewing the different number of folds at different CMP gathers..

Figure 2: CMP gathers 208 (left), 234 (middle) and 250 (right) of CMP-sorted seismic data

For velocity analysis, it is primarily aim to determine the seismic velocities of layers in the subsurface. At this stage most of the working involves with stacking velocities which will greatly used for the NMO correction of the seismic data. In term of methodology, one can obtain or pick stacking velocities by simply running a velocity analysis on a given CMP gather, where the aim is to flatten out the hyperbolas on the CMP gathers and the most common stacking velocities used are the velocity spectrum and the constant velocity stack techniques. The examples of velocity picking are as shown in Figure 3 below.

Figure 3: Velocity analysis for CMP gathers 220, 240 and 250.

After velocity analysis, the NMO correction is what we need to applied next. However, as per referred to the manual, there is a technical error toward the NMO correction function in the MATLAB. It is believed to have a technical error towards the looping syntax of the NMO correction function where the looping does not terminate after the 11^th iteration of the stacking function done in the CMP part. Figure 4 below explained and illustrates the main problem been faced throughout the completion of the NMO correction function.

Figure 4: Error in the NMO correction function