Infomar SBP Data User Guide
Infomar SBP Data User Guide
Infomar SBP Data User Guide
Version History
2.1 Celtic Explorer (CE) and Celtic Voyager (CV) data overview ..................................5
3 RV Keary ................................................................................................................. 8
SBP data is primarily available for surveys acquired by the RV Celtic Voyager and RV Celtic Explorer
(operated by MI) and the RV Keary (operated by GSI). The systems used are survey dependent and
include Pinger, Chirp and Sparker systems. The MI vessels routinely collect SBP data on each of their
surveys, however this is not the same for the GSI, which collect SBP on an ad hoc basis. If SBP data
has been collected it is not necessarily available for every survey line.
2.1 Celtic Explorer (CE) and Celtic Voyager (CV) data overview
The majority of CE and CV data has been captured in ‘.COD’ (CODA). Conversion of the CODA to
SEG-Y data to date has been completed using CODA file Utilities (Format Converters – CODA to SEG-
Y Converter – Settings are: Generic SEG-Y/Meters/.sgy Filters: All Types/Channels/Triggers).
The majority of CV data was collected using a Pinger (SES 5000 2x2 array) until late 2019 when it was
replaced by a Knudsen Chirp system. The CE has used a Pinger (SES 5000 4x4 array) until 2015 when
it was replaced with a iXblue Echo3000 chirp system.
For surveys in UTM, you may also need to divide/multiply whatever you extract by the nav data
scalar (specified in the trace header also - bytes 71-72) - normally it is set to divide by 100.
For the small minority of surveys collected in WGS84 (arc seconds). To view this correctly in WGS84
- EPSG 4326 (as decimal degrees); the values need to be divided by 3600 (converting from arc
seconds to degrees) – the result will then need to be divided/multiplied by the scalar value stored in
bytes 71-72.
By default, the data stored in CODA files are very raw, with no signal enhancement. Routine signal
enhancement would be applied during CODA playback – which was not possible in every seismic
package. Data in these files generally benefit from a frequency bandpass filter (around 3.5kHz for
older data); applying a linear gain from the seabed are also needed.
Figure 2 - Profile before a bandpass filter centred on the 3.5kHz transmit frequency (no gain adjustments) is
applied.
To convert the raw .jsf data to SEG-Y a ‘how to’ document and .jsf to SEG-Y converter can be made
available upon request.
By default, the data stored in JSF files is very raw, with no signal enhancement. Data in these files
generally benefits from a high-pass filter. Applying a linear gain from the seabed is also required.
Data acquired by the CHIRP system installed on the RV Keary is stored as a unipolar ‘envelope’ signal
(rather than as a bipolar response signal, as common). Some frequency filter designs will generate a
lot of ‘ringing’/artifacts when applied to data of this nature.
If the frequency filters available do not provide good results, the below approach may work:
1) Design a wavelet of appropriate frequency content / sample rate / phase characteristics.
2) Convolve this with the unipolar seismic data – generating a bipolar output.
3) Frequency filter as needed.
4) Continue with bipolar data or use an envelope function to convert back to unipolar.
5) Attention needs to be given that the convolution and wavelet phase don’t shift the time
of signal.
Figure 5 – An example of RV Keary data profile showing source/receiver ghost and impact of heave on data.
.