Peak marine sparker amplitude data from calibrated source and receive hydrophones collected in April 2021 offshore Santa Cruz, California (USGS field activity 2021-619-FA), and pressure vs. offset plots

Aboard the source vessel (R/V Parke Snavely; RVPS), sparker source (either the SIG ELP790 or the Applied Acoustics Delta Sparker) was towed on a 20-meter cable from the stern of the source vessel (R/V Parke Snavely; RVPS) along seven planned survey tracklines in water depths ranging from 25 to 600 meters. The SIG ELP790 sparker was powered at 700 joules at all water depths. The Applied Acoustics Delta Sparker was powered at 1000 joules at the 200-meter depth site and powered at 2.4 kilojoules at the 600-meter depth site. A Reson TC4034 hydrophone and an RBR depth logger were towed on a second cable and positioned in the water just below the sparker to measure the near-field outgoing source signal and depth. Continuous waveform data (TDMS format) from the near-field hydrophone were acquired using SignalExpress software sampling at 192 kHz. A second, receiver vessel (R/V San Lorenzo; RVSL), maintained a nearly stationary position at the midpoint of each trackline and recorded the signal on another hydrophone hanging in the water at mid-vessel. A unique code name was given to each navigation line where PS=the source vessel R/V Parke Snavely; L=line followed by 2- or 3-numerals representing the water depth (the water depth may be followed by an A denoting a planned replicate transect line); S=SIG ELP790 700-joule minisparker source, D10=1.2 kilojoule Delta sparker source, or D24=2.4 kilojoule Delta sparker source; subsequent passes are followed with a, b, c, d, and a numeral as needed. The source and receiver vessels’ navigation data (SrcNav, RecNav) and hydrophone signal data (SrcSegy, RecSegy) are provided elsewhere in these data release.

Navigational data from both the source and receiver vessels were recorded in raw GGA format. The UTC timestamp, latitude and longitude were extracted using a series of Python notebooks. The layback position of the sparker source was estimated to be 20 meters from the stern of the source vessel based on the length of the cable. The python tool PyProj.Geod WGS84 was used to merge the data frames of the source layback latitude and longitude positions (lbsrc_lat and lbsrc_lon, respectively) and to calculate and assign the offset distance, in meters, between the sparker source and the hydrophone hanging below the receiver vessel for each UTC time sample (utc_dt). To orient the offsets from separate passes relative to each other, a signed offset (signoffet) value was assigned in which the source offset value is positive when the source vessel (RVPS) was northwest of the receiver vessel (RVSL) and negative when the source vessel is southeast of the receiver vessel.

SignalExpress software was used to output TDMS float files from the Reson TC4034 hydrophone on the source vessel (RVPS). The python package npTDMSb (https://nptdms.readthedocs.io/en/stable/apireference.html#module-nptdms) was used to read the recorded hydrophone float voltage and write to an ObsPy trace (https://docs.obspy.org). Aboard the receiver vessel (RVSL), SpectraPlus .wav files were output in 16-bit format from the Cetacean Research model CR3 receiver hydrophones. ObsPy and NumPy (https://numpy.org) python packages were used to unscale the 16-bit voltage readings to float format ObsPy traces. For each pass at each depth site, the timing of the electromagnetic spike as the power supply (aboard the source vessel) discharges power to the sparker source relative to the nearest offset shot time was used to shift the continuous waveform start time. For both source and receiver ObsPy traces, a 70-7000Hz bandpass filter was applied. ObsPy traces were sliced from the continuous waveform using the shot timing (record lengths for RVPS SEG-Y are 0.2 seconds and the RVSL SEG-Y are 2 seconds with offset dependent start times written to the Delay Recording time DELRT header in header bytes 109-110), resampled to 16 kHz, and standard SEG-Y headers were written (https://seg.org/Portals/0/SEG/News%20and%20Resources/Technical%20Standards/seg_y_rev1.pdf). The source and receiver vessel navigation, shot times, and RBR hydrophone depths (when recorded) were written to SEG-Y headers.

SEG-Y data files were imported into Shearwater Reveal seismic software (https://www.shearwatergeo.com/4/reveal-software) where data were viewed for quality control. The peak absolute value amplitude per seismic trace was selected (csv attribute lav) and the time of the sample point was recorded in milliseconds (csv attribute lav_ms). Using equations 3 and 5 from NUWC-NPT Technical Report 12, 203 24 March 2016 (Crocker and Fratantonio, 2016), Sound Pressure Level (SPL), field pressure, and Sound Level (SL) were calculated from the largest absolute value trace amplitudes as well as the calibrated hydrophone sensitivities (CR3101 = -208.2; CR3102 = -208.6; TC4034-3 = -225.4). These values were mean filtered over 51-traces to represent the overall trend of Sound Pressure Level decay as a function of offset distance from the sparker sound source. These data were exported in csv format and merged with the source and receiver navigation, shot times, and RBR hydrophone depths (when recorded).

CSV data were plotted in PNG format per depth site in a final python notebook using tools SciPy (https://scipy.org), NumPy (https://numpy.org) and Matplotlib (https://matplotlib.org). Receiver sound pressure level (csv attribute spllav) was plotted as a function of offset (csv attribute signoffset) and source sound level (sllav) also plotted as a function of offset (signoffset) are plotted in the upper portion of each scatter plot contained in the zip file ‘2021-619-FA_peakamplitude_scatter_plots.zip’. The mean source calculated sound level (referenced to 1-meter distance) was calculated for all passes and source types for each depth site and is noted in the legend as well as the published sound level for each sparker model. A lower scatter plot in the same PNG image displays the receiver sound pressure level measurements with the spherical divergence reference calculated using the mean measured sound level as a function of offset. These same data, 51-trace mean filtered, are plotted in the PNG images contained in the zipfile ‘2021-619-FA_peakamplitude_MeanFilterCurve_plots.zip’ using the mean filtered csv attributes mfspllav and mfsllav, mean-filtered sound pressure level and sound level, respectively.