Originator: U.S. Geological Survey
Publication_Date: Unpublished Material
Title: Raw MCS data
Geospatial_Data_Presentation_Form: raster digital data
Source_Citation_Abbreviation: MCS data
Source_Currentness_Reference: ground condition
Multichannel seismic-reflection data were shot using Applied Acoustics S-Boom or SIG mini sparker (ELC 1050 and 1200) sources powered by Applied Acoustics CSP-D (700 and 2400) power supplies. Both sparker and boomer sources were used to test their relative effectiveness in the delta front setting, and some lines were reoccupied with each source to provide comparative results over common sections of sea floor. Shots were recorded using a 150-m long 32-channel (50 m active section) solid-state Geometrics GeoEel streamer with 1.5625-m spaced groups connected to a Geometrics Streamer Power Supply Unit (SPSU). The seismic sources were towed from the center of the stern approximately 17 m (lines FA2017003_transit4 - FA2017003_sw8) and 22 m (lines FA2017003_sw9 - FA2017003_bonus3) aft and approximately 4 m starboard of the DGPS antenna, and the GeoEel streamer was towed approximately 3 m outboard of the port side of the vessel from a boom crane, with the center of the first and last active groups approximately 73 and 123 meters aft of the DGPS antenna, respectively. The S-Boom source was powered at 600 joules (200 J per plate) and deployed from the start of surveying up to 19:19 (UTC) 5/21/2017 (JD141), between 02:17 and 15:31 5/22/2017 (JD142), and between 20:06 5/22/2017 (JD142) and 10:08 5/23/2017 (JD143). SIG mini sparker sources were supplied between 200 and 500 joules and deployed between 15:31 and 20:04 5/22/2017 (JD142), and between 00:20 5/24/2017 (JD144) and 00:44 5/26/2017 (JD146). Geometrics CNT-1 seismic acquisition software (version 5.361) running on a Windows PC was used to control the multichannel system and digitally log traces in the Geometrics SEG-D format, and record GPS navigation coordinates to the SEG-D external headers. Data were acquired at shot rates of 0.5 and 1 s, record lengths between 400 and 800 milliseconds (ms), and a sample interval of 0.25 ms.
A SIOSEIS (version 2015.3.1) seismic processing software script (sio_segd2segy) was used to read Geometrics SEG-D formatted (SIOSEIS process SEGDDIN) traces and write them (SIOSEIS process DISKOX) to SEG-Y Rev. 1 format (IEEE floating point) for each line.
This process step and all subsequent process steps were conducted by the same person - Wayne Baldwin.
Contact_Organization: U.S. Geological Survey
Contact_Person: Wayne E. Baldwin
Address_Type: mailing and physical
Contact_Voice_Telephone: (508) 548-8700 x2226
Contact_Facsimile_Telephone: (508) 457-2310
Address: 384 Woods Hole Rd.
City: Woods Hole
OpenCPS (version 3.3.0) seismic processing software was used to perform the following series of processing flows:
1. read_segy.flow - SegyTapeRead read the traces. HeaderMath and UTMLatLong were used to convert the source lat/lon positions from seconds of arc to decimal degrees, project them to UTM Zone 16N WGS 84 meters, and write each to new header words (NRP_LAT, NRP_LON, NRP_X, and NRP_Y). DBWrite wrote the UTM positions for the first channel of each FFID to an internal OpenCPS database table. Finally, Output wrote the traces to a new file "*.sht-raw.seis" in the internal OpenCPS format.
2. geom.flow - Input read the "*.sht-raw.seis" file and sorted the traces to FFID/CHANNEL. The custom Python module ShotlineLayback (developed by Nathan Miller of USGS-WHCMSC) was used to define the source shot and streamer geometry based on measured horizontal offsets from the DGPS antenna to the source (-17.35 or -22.35) and the centers of the first (1) and last (32) channel groups (-73.43/-112.8) of the active streamer section. The algorithm interpolated a sail line from the source shot positions (NRP_X and NRP_Y), then computed layback positions for the source shot and 32 channel groups for each FFID by translating them back along the sail line by their respective measured offsets. Midpoint positions along the sail line were also computed for each shot/receiver pair, allowing the traces to be binned by common midpoints (CMPs) spaced evenly (by 0.781 m) along the sail line. Output wrote the traces to a new file "*.sht-raw_geom.seis" in which the trace header words SRC_X, SRC_Y, REC_X, REC_Y, MPT_X, MPT_Y, BIN_X, BIN_Y, and OFFSET were populated to reflect the layback Source X and Y, layback Receiver X and Y, layback Source/Receiver Midpoint X and Y, and CMP BIN X and Y positions, as well as the offset value (computed by sqrt((REC_X - SRC_X) * (REC_X - SRC_X) + (REC_Y - SRC_Y) * (REC_Y - SRC_Y))) for each trace in the resulting CMPs. *A limitation of the ShotlineLayback module is the inability to layback source shot/receiver locations that would have occurred prior to the start of the interpolated sail line (i.e. it does not project the sail line backward). This results in erroneous computation of those shot/receiver locations and their respective midpoints, making them unable to be stacked by bin location in the subsequent step. Consequently, up to 40 dead traces exist at the start of each common midpoint stack profile.*
3. brute_stack.flow - Input read the "*.sht-raw_geom.seis" file and sorted the traces to CMP/OFFSET. The Python module Butterworth applied an Obspy version (modified from Scipy version 0.17.1) of a zero-phase, four corner Butterworth bandpass filter to the traces, between 450 and 1800 Hz for data shot with the S-Boom, or between 350 and 1400 Hz for those shot with the mini sparker. NormalMoveout applied a travel time correction to each trace based on offset and a constant velocity of 1530 m/s (water column sound speed), as well as a 60 percent stretch mute with a 30 ms taper. Stack summed the traces within each CMP, computed the average amplitude for each trace sample, and wrote the computed samples to a single CMP trace, as well as averaging the Midpoint positions of the input traces and updating the header words MPT_X and MPT_Y with the resulting averaged values. Output wrote the stacked traces to a new file "*.stk-brute.seis" in the internal OpenCPS format.
4. stk2segy.flow - Input read the "*.stk-brute.seis" file and sorted the traces to CMP. UTMLatLong projected the CMP BIN_X and BIN_Y headers from UTM Zone 16N WGS84 meters to Geographic WGS84 decimal degree, latitude and longitude. HeaderMath converted the geographic coordinates from decimal degrees to seconds of arc multiplied by 100, and set the coordinate unit header to 2, both to meet the specification of the SEGY Rev. 1 format standard. Output wrote the cmp stacked traces to SEG-Y Rev. 1 format (IEEE floating point) specifying header mappings for Ensemble Number, Trace ID code), Number of Horizontally Summed Traces, Ensemble Y coordinate, Ensemble X coordinate, Coordinate scalar, and Coordinate Units.
A Seismic Unix (version 4.3) script (plot_geomet) was used to read the stacked SEG-Y files and plot the data as 8-bit greyscale Postscript files using the Seismic Unix 'psimage' algorithm. All images were created with a horizontal scale of approximately 400 traces per inch. Images were plotted within a constant 14-inch vertical window and two-way travel time durations between 200 and 800 ms. The Postscript images were then converted to PNG format using ImageMagick (version 6.9.5-4).