What is the P-Cable system?

The P-Cable system is a unique platform for towing up to 24 short streamers from a relatively small vessel with minimal crew for high-resolution 3D seismic surveys. A schematic of a fully-deployed P-Cable array is shown below.

What is the basic seismic technology?

The P-Cable is based on Geometrics’ digital GeoEel technology. The GeoEel was developed from the ground-up as a high-resolution system, offering a Nyquist of 4 KHz. It comes in both solid and liquid-filled models.


What is the basic layout of the system?

A schematic of the P-Cable system is shown below.

What are the key components of the system?

1) CNT-2 Seismic Acquisition Controller

• Basic system control – sample interval, gain settings, etc.
• Multiple shot gathers
• Multiple common offset gathers
• Real-time noise monitor
• Real-time brute stack with semblance analysis
• AGC and playback filters for all displays
• Spectra display
• Cycle time plot
• Full survey log – keeps track of changes in acquisition parameters, where each file was written along with the date and time, alarms or error messages that have occurred during the survey, time-stamped observer comments, etc.
• Writes data to multiple hard drives simultaneously in SEG-2, SEG-Y, or SEG-D
• Fully RAID-compatible
• Operates up to four tape drives with automatic switching and dual simultaneous write-to-tape
• Full integration of incoming GPS, depth, compass, and other serial data into SEG-D extended header. GPS string also written along with shot record number to ASCII navigation file.
• Depth and compass data written along with shot record number to ASCII depth/compass reading file
• System testing – analog performance, leakage, capacitance

2) Streamer Power Supply Unit (SPSU, also “Deckbox”)
• Provides power to all in-water components
• Displays voltage, current draw, system leakage
• Accepts trigger signal and triggers system
• Contains 8 AUX channels
• Receives digital seismic data and transmits to CNT-2 controller via 100 mbs Ethernet
• Receives streamer bird signal and transmits to bird controller (birds not used in 3D deployments)

3) Signal Cable

• Transmits power to in-water components
• Transmits digital seismic data to CNT-2 Controller via 100 mbs Ethernet-over-COAX
• Up to 600m long, no repeaters required

4) Cross-cable / Junction Boxes

• Digital / power wire bundle helically wrapped around central strength member, connecting evenly spaced junction boxes
• Titanium construction
• Integrated 100 mbs Ethernet switch, digital compass, and depth sensor in each junction box
• SubConn self-bailing connector to A/D module

5) GeoEel 8-channel A/D Module

• Titanium construction
• 24-bit A/D technology
• Sample rates from 1/8 to 2 msec
• Functions also as Ethernet repeater

6a) GeoEel Solid Streamer


• Solid polyurethane design
• 44.5mm diameter
• 8 channels per section
• 1.5625, 3.125, 6.25, and 12.5m hydrophone groups
• No oil to leak or spill – very environmentally friendly
• Designed for shallow high-resolution – clustered hydrophone groups
• Extremely rugged
• Non-flammable – ships by air
• No cable-borne noise
Click for technical specifications
- OR -

6b) GeoEel Liquid Streamer


• Polyurethane tube
41 mm diameter
8 channels per section
Teledyne Geopoint hydrophones
1.5625, 3.125, 6.25 and 12.5m hydrophone groups
Filled with non-polluting silicone oil
Same electronics and front end as GeoEel Solid
Click for technical specifications

7) Tail Module and Low-profile Drogue

• Tail module includes compass and depth sensor
• Drogue provides tension to keep streamer straight

8) Paravane or Diverter

• Provides horizontal lift to cross cable
• Typically #3 Baro or equivalent

9) NavPoint TrawlerTM GPS/compass-based navigation and positioning system

• Provides positioning of array
• Displays continuously-updated array shape and position along with cross-cable depth
• Writes standard P 1/90 navigation file
• Provided along with personnel by NCS Subsea, Inc.

What is the maximum width of the array?

The maximum that has been achieved (and attempted) is 287.5m, from first to last streamer. The array consisted of 24 streamers spaced at 12.5m intervals along the cross-cable. Baro #8-2500 (2.5m x 4.75m, 2300 Kg) paravanes were used to achieve the necessary lift. The M/V Birkeland was used to tow the array.


How many streamers can there be in the P-Cable?

Assuming #3 vanes, this depends on the desired bin size and the width of the spread. Typical streamer spacing is 6.25m or 12.5m, yielding a cross-line bin size of 3.125m or 6.25m, respectively. This equates to up to 16 streamers spaced at 12.5m or 32 streamers at 6.25m. Larger arrays require more lift and hence larger paravanes.

What is the system bandwidth/sample rate?

The A/D modules can sample at up to 8 KHz. Analog bandwidth of the arrays depends on the number of hydrophones per group, which can range from 4-12. Typical is 10 Hz – 3.2 KHz. A low-end -3dB point of 5 Hz is available.

How long can the streamers be? How many channels in each?

Each hydrophone streamer section contains 8 channels. Section length depends on the group interval (1.5625m, 3.125m, 6.25m, or 12.5m), and is either 12.5m, 25m, 50m, or 100m. Sections can be connected together to build multi-section streamers. Typical systems consist of single-section, 8 channel arrays, having group intervals equal to the streamer spacing, yielding square bins. This means that, in a typical array having 6.25m streamer spacing, the streamer length is 50m.  A P-Cable multi-client geohazard survey based on 100m streamers was conducted in the Gulf of Mexico in May 2014. Some of the data from this survey can be viewed on the Example Data page.

Why are the streamers so short? Doesn’t this severely limit the depth of exploration,
and how do you deal with multiples?

Technical Note:  Mythbusting Short Streamers and Depth of Penetration

Long streamers (>~100m) cannot feasibly be towed from a cross-cable without very large paravanes. Using large paravanes requires a large vessel, and this defeats part of the purpose of the P-Cable, which is to enable high-resolution surveying from relatively small (and inexpensive) vessels.

The second half of the above question is a big one, and is best addressed by the following dialog:

Q. How can you do reflection in 2500m of water with 25m streamers?  Don’t the streamer lengths need to be about the same as the depth of interest?
A. Depth of penetration has nothing to do with the length of the streamer, and everything to do with the size of the source. The larger the source, the deeper you can see. However, the larger the source, the lower the frequencies tend to be. Also, larger sources must be towed deeper, leading to a lower source ghost notch frequency, which puts an upper limit on the useable frequency band. So for high-resolution work, the depth of penetration must be balanced against the resolution requirements. In general, an array of relatively small sources is better than a single large source in terms of frequency content.

Q. Yes, but with streamer lengths so short compared to the depth of the water, how can you get good velocity information?
A. That’s a great question, and the answer is, you can’t. If you are in 2500m of water, and you have a streamer only 25m (or 50m or 100m or even 300m) long, then over the length of the streamer, there will be very little moveout on a given reflector – the reflectors will be essentially linear. It is therefore difficult to get an accurate stacking velocity from what little measureable moveout there might be. But remember that the P-Cable is designed for shallow, high-resolution surveying. In deep water, the desired depth of investigation beneath the seabed will usually be small compared to the water depth. Since stacking velocity for any given reflector is the rms velocity of the column above the reflector, then in the case of deep water, the stacking velocity will essentially be that of water.

The data below were acquired in 800m of water with 25m streamers.  Water velocity was used to stack the data.

Q. But what about multiples?
A. If the water is shallow enough (roughly equivalent to streamer length), multiples can be dealt with in the conventional way, because velocities will be obtainable from the P-Cable data. If the water is very deep relative to the streamer length, then the multiples will usually be well below the zone of interest and will not be an issue.

Q. Yeah, but those are the ends of the spectrum. What if I’m in 500m of water, with 25m streamers, and I’m interested in the first second of data below the seabed.  The water is too deep to get velocity information from the P-Cable data, but too shallow to just use water velocities in the multiple removal.  The sea-bottom reflector will be at about 0.7 seconds, so the first multiple will be at 1.4 seconds.  That’s right in the middle of my zone of interest.  If I don’t have velocities, how can I remove these multiples?
A. In these cases, which tend to be rare, multiples can be removed quite effectively using velocity-independent multiple-suppression methods. Our P-Cable processing partner, DECO Geophysical, has developed a velocity-independent de-multipling algorithm based on a technique called “adaptive subtraction”. It works very well. You can read about it here:

Q. Ok, fine.  But you still haven’t solved the velocity problem.  I understand that I can get a good picture of structure without measured velocity information, but ultimately I am going to want to convert time to depth. Where do I get the velocity information to enable this?

A. If the water is deep enough relative to the depth of investigation such that water velocity can be used to stack the data, it follows that you can calculate a pretty good depth using the water velocity. That said, there are three possible sources of velocity, depending on the situation. In oil and gas applications, being a high-resolution tool, the P-Cable is usually used to focus on small areas of interest that have already been imaged in lower-resolution by conventional seismic methods. For instance, a common use of the P-Cable in oil and gas applications is shallow geohazard mapping – shallow gas, ice gouge, faults, submarine landslides – ahead of drilling. Another is 4D monitoring of secondary recovery operations. In either case, the velocity field is usually well-understood, having already been measured by conventional 2D or 3D surveys.

In situations where the P-Cable is the first seismic technology applied (rare in industry applications), and no velocity information is available, one option is ocean-bottom seismometers (OBS). Several of these can be deployed in the survey area to record and store data continuously. Over the course of the survey, they will record data from all of the shot points, giving plenty of near and far offset data from which to calculate velocities. Weighed against the cost of drilling, the cost of the OBS nodes is trivial.

A third alternative is to convert the P-Cable system to a long 2D system and do several 2D lines throughout the survey area. For instance, a 16 x 25m P-Cable system can quickly be converted into a 400m single-streamer (longer if you include your spares), which would allow reasonable velocity measurements to 600-800m. And like the OBS approach, the cost is small compared to that of drilling.

Q. Okay, but what you are basically saying is that, in the case of deep water, you have a point-receiver, and you are doing a normal-incidence survey.  Big deal.  They’ve been doing that forever.
A. Yes, that’s one way to look at it. But note that with long streamers, the shot/channel pairs that sample a depth point are distributed over a long distance. There is plenty of moveout from which to calculate velocities. Unfortunately, in the process of correcting that moveout, there is also significant wavelet distortion, known as “NMO stretch”. This distortion increases with increasing offset. The net effect of this distortion is a reduction in overall frequency when the data are stacked, resulting in a loss of resolving power. With the P-Cable, the channels that sample a depth point are distributed in two dimensions rather than one, minimizing the overall offset. We can therefore achieve very high fold with no measureable moveout. Wavelet distortion associated with NMO correction is eliminated, because little to no NMO exists.

In high-resolution, offset is the enemy. In fact, when industry wants to tease smaller targets out of their large-scale 3D surveys, they do something surprisingly simple -- they chop off the far offset traces prior to processing the data, in order to minimize NMO stretch and maximize overall frequency content (this only goes so far, since the airguns used in such surveys tend to be very large and low-frequency to begin with). Since the design purpose of the P-Cable is high-resolution, why record the long offsets in the first place? With a P-Cable array, we get the best of both worlds: high fold without wavelet distortion. With long streamers, you must trade one for the other.

Q. So what are you saying? That the P-Cable is going to replace long-streamer, very large-scale 3D systems? Are Ramforms on their way to obsolescence?
A. Of course not. In oil and gas exploration, long offsets will always be necessary. The P-Cable is no substitute for conventional 3D; it is not even a competitor with it. They are different technologies aimed at different objectives. Most of the time, conventional 3D, coupled with well logs and other data sources, will be sufficient for exploration and exploitation purposes. But in those instances where you need to take a closer look at the shallow sub-seabed – when you need a microscope instead of a “naked eye” – the P-Cable technology is unmatched.


What is the bollard pull of the system?

A 16-streamer array with 12.5m streamer spacings (largest recommended with #3 Baro or equivalent paravanes) has a bollard pull of about 5.5 tons at 5 knots.

What size vessel is required?

This depends on the size of the array and the configuration of the vessel. A 300m-wide array has been successfully pulled by the R.V. Helmer Hanssen (formerly R.V. Jan Mayen), operated by Tromso University. This vessel is 64m in length with a 13m beam. Smaller arrays can be pulled by smaller vessels; a survey conducted by Fugro West off California, using fourteen 50m arrays at 6.25m spacing (~80m swath), was easily accomplished using the R/V Bluefin. This vessel is 53m long with an 11.5m beam.  NCS Subsea successfully completed a program in the Gulf of Mexico with a system consisting of 18 100m streamers spaced 12.5m apart; the vessel was the M/V Bjørkhaug. Geometrics can provide a vessel assessment upon request.  A complete list of vessels that have been used in P-Cable surveys is given below.



What handling equipment is required? How many winches do you need?

Four winches are required to deploy and retrieve the P-Cable system: One for the signal cable, one for the cross-cable, and one small winch for each paravane. A storage drum for the active sections is highly recommended, but not absolutely necessary. Cranes and/or davits will be required for launching the paravanes, depending on the vessel. If the vessel has an A-frame, rear-deployment is possible in calm seas. In seas greater than 2m, a side deployment is generally preferred. A typical deck layout is shown below, although the actual layout depends on the vessel.

rvbluefindeck11 rvbluefindeck11a

How many people are required to deploy/retrieve the system? How long does it take?

Deployment time averages 45 minutes and retrieval time averages 30 minutes. Four to five people are required, depending on deck configuration and winch control method.

What is the deployment process?

Click to view an animation.

How many actual surveys have been done with the P-Cable?

Over 70 3D cubes have been acquired with the P-Cable, most of it 6.25m-bin data.

Has the P-Cable been used on any commercial projects?

Yes.  As of November 2014, six surveys have been done for commercial interests.  The first and second were done by Fugro West for Pacific Gas and Electric in December 2011 and August 2012.  The full report can be found here.  The P-Cable survey was part of a multifaceted land and marine program, and the data can be found in Chapter 3.  During the summer of 2012, WGP Group (UK) conducted four months of exploration with a 14-streamer system in the Barents Sea for Spring Energy.  In October 2013, Subsea Systems Inc. and the Scripps Institution of Oceanography completed a geohazards survey for Southern California Edison off the San Onofre Nuclear Generation Station near San Clemente, CA.  And during summer 2014, WGP teamed with TGS-NOPEC on a large multi-client shallow gas survey in the Barents Sea.  Also in summer 2014, NCS Subsea, in partnership with SPEC Partners and Geotrace, began a multi-year geohazard survey in the Gulf of Mexico, dubbed “SAFE-BAND”.

How is the P-Cable positioned?

There are two options that have been developed specifically for the P-Cable. The first is by DECO Geophysical. This method calculates the shape of the cross-cable, which tends toward a catenary, using GPS to fix the endpoints. The shape is double-checked by measuring first-break times to the near channels. Any feathering of the streamers or distortion of the cross-cable due to debris is ignored.

The second method was developed by NCS SubSea, Inc. and is based on GPS and digital compass technology. GPS fixes the endpoints of the cross-cable. A digital compass in each junction box on the cross-cable, and at the tail of each active section, allows accurate calculation of the cross cable shape and hydrophone positions throughout the array. Depth sensors in the junction boxes and streamer tails allow measurement of the third dimension. Includes real-time binning. Click here to download a data sheet. This option is provided as a service by NCS Subsea, Inc.


What is the recommended tow-depth of the system?

When using traditional de-ghosting techniques, the higher the desired frequency, the shallower the tow-depth must be. The frequency of the receiver ghost-notch, generally considered to be the upper limit of useable bandwidth, decreases with tow depth according to the following graph:
How shallow the system can be towed depends on the sea state, and the higher the required resolution, the shallower the required tow depth.  Operational efficiency can therefore be greatly impacted by the weather, especially in places like the North and Barents seas. This is not unique to the P-Cable; this applies to high-resolution work in general.

However, recent advances in broadband de-ghosting processing techniques, combined with the ultra-low noise floor of solid streamers like the GeoEel Solid, are challenging the necessity of shallow towing.  The low noise floor of solid streamers, particularly at low frequencies, means that the S/N ratio is high enough, even “in the notch”, to obtain useable data.  This is a promising development for high-resolution work to the extent that it allows deeper towing (and consequent higher operational efficiencies) without significant degradation of critical higher frequencies.

How is depth controlled? Are birds required?

Birds are not used on the P-Cable; again, the overarching philosophy behind the P-Cable is to have as little gear in the water as possible. Depth of the cross-cable is controlled by the length of the buoy rope over each tri-point. Streamers are carefully balanced to be neutrally buoyant, so they tow horizontally at the same depth as the cross-cable.

How is the depth of the array monitored?

A high-resolution depth sensor is housed in each junction box on the cross-cable and in the tail of each section. The depth is reported and recorded after each shot. Accuracy is +/- 10 cm.

What about tail buoys? Are they required?

Tail buoys are not necessary. The streamers are short, and the paravanes are visible during the day and are marked by strobes at night. We do use small, low-profile drogues on the ends of the streamers to keep them straight in the water.

What is the maximum depth of investigation?

Like any seismic survey, depth of investigation beneath the seabed is controlled mostly by gun size (see the discussion for the question above – “Why are the streamers so short?”) – not by the hydrophone array.  A more appropriate question might be, “To what depth of investigation beneath the seafloor is the P-Cable system useful?”   The answer is highly situational.  In general, depth of investigation and resolution are inversely related.  The earth acts as a low-pass filter, attenuating higher frequencies more readily than lower frequencies.  So for any given source, the frequency content of reflected energy decreases with the depth from which that energy is reflected.  Put another way, the farther the energy has to travel through the earth, the more high-frequency energy is lost to attenuation.  Given that resolution is directly related to frequency content, it follows that resolution decreases with depth.  You can stack the deck in your favor by using a high-frequency source, such as a boomer, sparker, or small air gun, but high-frequency sources are limited in energy output, putting a limit on how deep you can see with them.

So the answer to the question is that the P-Cable is useful to the depth at which the required frequencies can be obtained with the optimal source.  If your resolution requirements dictate a peak frequency of 200 Hz or more, then the depth at which the peak reflector frequency drops below 200 Hz marks the limit of the P-Cable’s usefulness.  The higher the resolution requirements, the shallower the depth of investigation.

Note that depth of investigation beneath the seafloor, which is what we have been discussing so far, is largely independent of water depth.  Since water is essentially incompressible and non-attenuating, significantly less energy and frequency content is lost in the water column.  It is therefore quite feasible to conduct shallow, high-resolution surveys in water several times the depth of investigation.  Note, however, that the deeper the water, the larger the water surface footprint of the survey needs to be to obtain full-fold and full 3D migration over the entire seafloor survey area.

To date, all P-Cable surveys that have been shared with us have featured depths of investigation of less than 1000m in 1200m of water or less, but there is nothing fundamental preventing work in deeper water. 

How are P-Cable data processed? Is any special processing required?

Fundamentally, P-Cable data are no different than any other 3D marine seismic data. The only difference is scale. Special processing software is not required. With the compass-based positioning solution, a standard P 1/90 file is provided for positioning.

Our processing partner in the development of the P-Cable is DECO Geophysical. DECO has processed most P-Cable acquired to-date, using their RadExPro processing package. They have developed a very good QC package designed specifically for the P-Cable.

It should be noted that while P-Cable data are subject to the same physics and processing flows as other 3D data, there are tricks in processing high-resolution seismic data, and it is often beneficial to bring these to bear when processing P-Cable data. In that sense, experience in processing high-resolution data is desirable.

How fast can the P-Cable system be towed?

We recommend 5 knots maximum.

How stable is the system while under tow? Is cable strumming a problem?

Measured variation in cross-cable depth during the survey is less than  +/- .15m. The helical spiral of the cable bundle about the strength member provides enough vortex shedding that, at 5 knots or less, strumming of the cross-cable is negligible.  Below is a frequency plot of depth data from all depth sensors in a 12-streamer array over a 4-day period in calm seas. The target depth was 1.2m. Eighty-five percent of the readings were within +/- one standard deviation (0.1m) of the target depth. Ninety-eight percent were within +/- two standard deviations. Total number of readings was over 250,000.


What production rate can be expected?

This depends on the width of the array, the vessel speed, and the amount of overlap between sail lines. It is best illustrated by example:  A system consisting of 16 streamers spaced at 12.5m (largest feasible with Baro #3 or equivalent paravanes), 5 knots towing speed, and two streamers of overlap would allow 30 km2 of sea surface area to be surveyed every 24 hours.  This includes turns, and little if any infill would be required due to streamer overlap.  Bin size in this example would be 6.25m x 6.25m. 

How long do the turns take?

A system consisting of 16 streamers spaced at 12.5m requires about 30 minutes to do a full turn.








This collection of frequently asked questions (FAQ) provides brief answers to many common questions about the Geometrics P-Cable system.

If you have trouble finding the information you need, please let us know.

Please click on the question to display the answer. Clicking a second time will make the answer disappear.

Frequently Asked Questions (FAQs)