 |
|
|
||||||||||||
R/V Langseth Seismic Capabilities and cruise planning information
First, please review the guidance found here: http://www.ldeo.columbia.edu/res/fac/oma/langseth/Chief_Scientist_instructions.html
Acquisition - Solid State Hydrophone Cables Langseth can tow up to four seismic hydrophone cables. The cable itself is Solid State [not oil-filled] manufactured by Thales [now supported by Sercel] in 150-meter sections. Each section has 12 hydrophone groups, which are, therefore, 12.5 meters long. Enough cable is being purchased to allow up to four 6-km streamers to be deployed at the same time. If a smaller number of cables are to be deployed, greater lengths are possible. The manufacturer recommends a 10-km maximum length, due to strength limitations, and this refers to new cables. Lamont policy is to restrict maximum length to 8 km. It is claimed that solid state cables are more robust and quieter than oil-filled cables. Experience so far, and these test results bear out the "quieter" claim:
Separation between cables towed by R/V LANGSETH can vary between 50 and 200 meters. With dual sound sources [deployed between 25 and 100 meters apart, respectively] this results in CDP line spacing between 12.5 and 50 meters. CDP line spacing should be chosed according to the desired frequency content and maximum cross-dips to be imaged. The Langseth seismic source array comprises four identical 10-airgun strings. Four 2D or other single-source work, between one and four strings can be deployed, depending on the needs of the survey. During twin-source multi-streamer 3D surveys, the sources are split, each "side" with one or two strings. Source and receiver navigation are determined and logged using the 3D version of the Concept Systems "Spectra" software. Positioning inputs include:
Concept's "Sprint" software and personnel are available for post processing of the real time acquisition positioning data, and a new package, "Reflex," which does 3D CDP binning analysis, has been purchased. Data can be recorded on 3590 tape drives but are principally stored on disk in real time. Langseth is equipped to tow four linear subarrays of airguns, each up to 16 meters long. Each string looks like this:
For 2D surveys, between one and four strings can be towed as a single array, depending on penetration requirements. For 3D surveys, two arrays are formed, each with either one or two strings. The 4-string 2D array design is shown here in schematic plan view:
The modelled signature for this array shows excellent source characteristics:
Modelled statistics for this array show that while its source characteristics are superior to the Ewing 20-airgun array, the overall energy and sound levels are similar:
The four identical source strings can be subdivided for dual source 3D acquisition. Either one or two strings can be used for each "side," depending on source level requirements. In addition, one or two strings can be used as alternative, lower-power 2D sources, if desired.
Cruise Planning - 2D and 3D MCS surveys Some guidelines for planning and allocation of time for deployment, maintenance and recovery of MCS equipment during surveys is given here: Deployment of four streamers aturally takes longer than it does for one. When working in a new area where reballasting is required, at least two days should be allowed for deployment. When this is not required, the process may be completed in a 24-hour period. The rule of thumb for the radius and time spent for turns is: Diameter = streamer length + spread:
With four 6-km streamers with a 600-meter spread, the path along a 180 degree turn is about 11 km, and the turn itself will therefore take about 1 1/4 hours. Time must also be allowed for the 1/2 streamer length runout required to build full CDP fold on the completed line. In addition, rule-of-thumb for achieving a straight streamer when beginning a line is that this takes between 1.0 and 1.5 x cable length.. Nominal ship speed will be about the same as it is for Ewing 2D MCS: 4.5 knots, or roughly 8.5 km/hour, so the minimum time for a line change will be 3 - 4 hours, and 5 - 6 hours will be typical. Efficiency is gained by minimizing turns, which means shooting a rectangular survey with lines running in the longer direction. To maximize data quality, however, the shooting direction is usually chosen to be in the dip direction, as inline CDP spacing [6.25 meters] is usually much finer than the crossline spacing [12.5 - 50 meters] and these two objectives may be contradictory. Another important consideration is that turning in shallow nearshore areas [often rife with fishing craft] is dangerous, which may lead to compromises when acquiring 3D data along continental margins. The Leading Edge article on shooting direction Despite the many improvements in exploration industry 3D equipment, there are still enough differences between data shot in two opposite directions that a "seam" is visible in the processed data. For this reason, industry data are usually shot in "racetrack" fashion, which exploits the large minimum turning radius illustrated above and minimizes the number of "seams:"
It also makes sense to place the seams outside the principal area of interest. Assuming that this is in the center of the survey area, The diagram below presents a schematic method for doing this:
Trackline spacing Minimum distance between sail lines [the tracks actually followed by the acquisition vessel) is dictated by the number of streamers and the distance between streamers:
Langseth will be able to pull four streamers. Max streamer spacing will be 200 meters, so the maximum distance between 3D tracklines will be (4 x 200)/2 = 400 meters. For a standard streamer spacing of 100 meters, trackline spacing will be 200 meters, and for high resolution work with a spacing of 50 meters, the tracklines must be 100 meters apart. Migration Aperture The main purpose of 3D MCS surveys is to provide data for 3D migration. Migration is a process that gathers energy scattered from a single point and gathers it back to provide an image of that point in its true and proper spatial location. To capture the needed extent of scattered energy requires collection of data in a fringe, or aperture outside the area which is to be fully migrated. To calculate the size of this aperture [which may have a large effect on the time required to survey an area] three aspects are usually considered - dip, fresnel zones and diffractions.
This illustration shows both the migration aperture due to dipping structure (a little over 1 km) and the half CDP spread "fold taper" required to obtain a complete CDP imaging a point on a dipping reflector. Added to the dip-dependent and taper apertures is a fresnel zone aperture. A simple way to visualize the fresnel zone is to consider straight rays in a medium with an average velocity:
The maximum fresnel zone aperture, X, is dependent on water depth (VT/2) and the lowest frequency to be imaged. X is the horizontal distance from normal where destructive interference is complete. Here we plot the fresnel aperture against reflector two-way-time for a representative suite of average velocities and a fairly low minimum frequency:
When faults, or other discontinuous structures are to be imaged, the fresnel zone aperture may not be large enough to capture a sufficient amount of energy diffracted from those discontinuities. The fresnel zone typically includes energy out to 15 degrees from vertical, but an aperture that captures thirty degrees is a more desirable window.
As with the fresnel aperture graph above, this calculation assumes straight rays - usually the worst case answer when rays are bending. Note that diffraction moveout must be included in the recording window, wich may affect record lengths and shot cycle times. Note also that diffraction moveout is dependent on two-way-time but not velocity. For the same set of average velocities as before, we see that the diffraction aperture is very dependent on velocity, and nearly always exceeds the fresnel aperture:
Altogether, these and similar considerations dictate the overall survey area required to adequately migrate the desired zone:
As this schematic shows, the area covered during a survey may be considerably larger than that enclosing the final, well-migrated target. This figure indicates an efficiency of 45% , but this is an ideal figure. Added time is required for transit in and out of the work area, deploying and recovering the equipment, down time due to weather and other problems, and infill. Actual efficiency may typically lie between 20% and 30%. Data qualilty is enhanced and CDP binning simplified when shotpoints are regularly spaced. The spacing is usually some multiple of the desired in-line CDP spacing. In the ancient past, before the near world-wide availability of differential GPS [DGPS] this was achieved by triggering the source with a constant time interval, while trying to control the ship speed to produce the desired physical interval. In deep water acquisition, this practise could result in the unfortunate circumstance that reverberations from earlier shots would appear at the same times for many sequential records, and thus be "stacked in" producing spurious horizons [see McBride, et al, 1994, Geophysics 59, 1160-1165.] To prevent this, Lamont -built source controllers introduced a randomizing factor, which was employed between 1975 and the early 2000's, when world-wide DGPS became available. When shooting by distance, it is usually claimed that the inherent jitter in the necessary prediction process introduces sufficient randomization. To test this, we examine a portion of line EW0207-3, shot parallel to the Juan de Fuca Ridge:
The average time difference between successive shot intervals is about 90 msec, and the overall pattern is non-stationary. To see how effective the shoot-by-distance prediction is, we plot distance between shots for the same sequence:
The intention was to trigger the seismic source every 37.5 meters, which was accomplished, within typical standards. To gain an appreciation of how good this shot spacing is, it is useful to re-plot these data at the same spatial scale as a "typical" Langseth source array and hydrophone group:
langseth source array strings will be 15 meters long, individual hydrophone array groups are 12.5 meters in length. These dimensions impose smearing effects on CDP resolution which are an order of magnitude greater than that resulting from variations in shot spacing. ... to be continued... last updated 13 January, 2009 |
