Seismic mis-tie

Seismic mis-tie software

Vintages rarely agree at the crossings. g-Space measures the time and phase difference at every intersection by cross-correlation, lets you correct one crossing by dragging the section, and then balances the whole 2D network — or ties two 3D volumes to a reference — with a solver of your choosing.

Download free trial See the workspace Watch Demos

Time & phase

Both, per line, per crossing

Five solvers

Equal split through polarity

2D and 3D

Line networks and volume ties

In view

Mis-tie workspace

A mis-tie is a measurement before it is a correction. Where two lines cross, the same reflector should arrive at the same time with the same phase — and when it does not, the difference is a number you can compute rather than a discrepancy you argue about. g-Space builds that number by cross-correlating the two sections at every intersection, holds it in a table with the shift and rotation charged to each line, and only then asks how you want it distributed. Correct one crossing by hand, or solve the whole network at once; either way the horizons and faults you go on to pick sit on data that agrees with itself.

Working the intersections

The crossing is the unit of work.

The mis-tie workspace opens from the bar as a set: a location map, the cross-points table, four section views of the selected crossing, and for volumes a shifts table and a 3D cross-correlation view.

Every crossing on the map

The location map draws the line network with the 2D intersections picked out in yellow and 3D volumes in their own colour, so a survey's worth of crossings is one view rather than a list to work through blind.

Both sections, side by side

Section views show the crossing lines together at the intersection, so the reflector's arrival time, its phase and its continuity across the join are judged on the data rather than on a number alone.

Drag to shift

Hold the left button and drag a section up or down to apply a shift to that part of it. The value is written straight into the cross-points table; a right-click on the section clears it again.

Phase, not just time

A phase rotation in degrees is entered per line against the crossing, so a vintage that differs in wavelet phase rather than in timing is corrected for what is actually wrong with it.

One row per crossing

The cross-points table holds the intersection's coordinates, the measured misfit, the names of both lines, the shift and the phase rotation charged to each of them, and a checked flag — ten columns, one row per crossing.

Mark it checked

Double-click the box to sign an intersection off; it turns green on the map. Checked crossings can then be excluded from the automatic calculation so reviewed work is not silently overwritten.

Misfits shown as a map

Turn on misfit points and every intersection is coloured by its mis-tie value, in time or in phase — which turns a table of residuals into a spatial pattern you can read for its cause.

Take the table with you

Copy the whole table with its headers, or just the cells you selected, straight into a spreadsheet — or save it as CSV, or export it as tab-separated ASCII to be reviewed and loaded back later.

Automatic correction

Five ways to spread the difference.

Once the misfits are measured, the question is who pays for them. The solve method decides how each intersection's mis-tie is charged to the two lines that cross — run it on the crossing in front of you, or on the entire dataset in one go.

Even split

Equal distribution

Splits each mis-tie value evenly between the two intersecting lines. The simplest reading of the problem, and the right one when neither line has a claim to be the more reliable.

Global

Least squares

Minimises the sum of the squared mis-tie corrections across the network, so the shifts settle into the arrangement that leaves the least total residual rather than the one you happened to work through first.

Constrained

Least squares, constant time

The same minimisation with the time shift held constant along each line — a bulk correction per line instead of a value that drifts along it, which is what a static acquisition or datum difference actually looks like.

Constrained

Constant time and phase

Adds the phase shifts into the same constrained solution, so timing and wavelet phase are resolved together across the network rather than in two passes.

Polarity

Polarity estimation

Estimates and corrects polarity differences between intersecting lines — the case where a vintage is not late, it is inverted, and no amount of time shifting will make it agree.

Scope

One crossing, or all of them

Run the calculation on the current intersection while you tune the parameters, then run it across every intersection: the cross-points table fills with the computed values in one pass.

Workflow

Six steps from a mismatched survey to one that ties.

Open the mis-tie views

One button on the mis-tie bar creates the workspace: location map, cross-points table and the section views of the selected crossing, arranged the way the job needs them.

Set the comparison window

Choose a fixed time gate, or centre the window on a horizon so it follows the structure. A horizon used this way needs a map first — the calculation stops and asks for one otherwise.

Try one intersection

Run the calculation on a single crossing, with the maximum shift, the trace-averaging distance and the correlation criteria set, and look at the two sections to see whether the answer is believable.

Solve the network

Pick the solve method, decide whether to skip crossings already checked and whether to down-weight lines along the structural dip, then run it across every intersection at once.

Review before applying

Leave the shifts as a map display first: colour the misfit points by time or by phase, look for a pattern, correct anything odd by hand in the table or on the section, and sign crossings off as checked.

Apply, and keep the record

Apply the shifts to the seismic data, then export the cross-points table so the correction can be quality-checked, archived, or loaded straight back into the project later.

Settings & 3D volumes

Everything the calculation asks, and the volume case.

The parameters live on the mis-tie bar and stay with the project. The same bar carries the volume-to-volume half of the tool, for surveys that were acquired or processed apart and now have to sit in one interpretation.

What the calculation asks for

The parameters behind an automatic run

Window

A fixed time gate with its own start and end, or a gate centred on a horizon with a chosen width — the horizon must already have a map, or the run stops and says so.

Max shift

The largest correction the calculation is allowed to propose, so a bad correlation cannot pull a line somewhere it clearly does not belong.

Trace averaging

A distance over which traces are averaged before comparison, and a threshold deciding which of them are included in that average.

Dip lines

Shifts on lines running along the structural slope can be reduced by naming their azimuth — such lines often migrate better and are the more trustworthy of the pair.

Criteria threshold

The level of cross-correlation the automatic calculation works to, so the balancing stops once the network is tied to the accuracy you asked for.

Group priority

Where a project holds several processing groups, their priority for correction is set explicitly rather than left to the order they were loaded in.

Tying 3D volumes

The volume-to-volume half of the same tool

Intersections

g-Space calculates where the volumes overlap and creates the intersection objects; the location map shows each volume in its own colour.

The measurement

Pick an intersection point and the mis-tie and the cross-correlation are computed there, in the same fixed or horizon-centred window as the 2D case.

Auto fit

Nominate the volume everything else is corrected to, and solve for the time residual, the phase residual, or both together.

Volumes shift table

The calculated shifts land in a table of their own, where a value can also be typed in by hand when you know better than the correlation does.

Applied to the data

Once calculated the shifts are applied to the seismic volume, and the corrected result is dropped into the active view for an immediate look.

Why it matters

Different acquisitions and different processing flows leave volumes that disagree; tying them is what makes multi-survey and reprocessing work interpretable as one dataset.

Views and tools in this group

The parts of g-Space this topic is built from

Mis tie seismic bar Cross points table Volumes shift table Location map Intersection 3D cross-correlation Section views Misfit points Seismic group priority Import / export cross-points

More g-Space capabilities

A tied survey is where interpretation starts.

Horizons that cross a line network, maps built from them and any depth conversion downstream all assume the data agrees with itself first. These are the other topics in the g-Space workflow.

g-Space overview Horizon picking Fault picking Well tie Geological interpretation Velocity modeling Attribute analysis & AVO Petrophysics Geological modeling Reservoir characterization

FAQ

Questions, answered.

What does the mis-tie workspace show?

One click on the mis-tie bar opens a workspace built for the job. A location map draws the survey with the 2D intersections marked in yellow and 3D volumes in their own colours. The cross-points table lists every 2D intersection with its measured misfit and the time shift and phase rotation held against each line. Section views show the crossing lines beside each other so the two can be compared directly, and for volumes there is a shifts table and a 3D cross-correlation view.

How are mis-ties calculated automatically?

By cross-correlation inside a window you define — either a fixed time gate, or a gate centred on a horizon so the comparison follows the structure rather than a flat time. A horizon used this way must already have a map, and the calculation stops and says so if it does not. You then set the maximum shift allowed, the distance over which traces are averaged for the comparison and the threshold for including them, and a criteria threshold on the correlation level. The calculation runs on the current intersection or across every intersection in the survey at once.

How are the shifts distributed across a line network?

By a solver you choose. Equal distribution splits each mis-tie evenly between the two lines that cross. Least squares minimises the sum of squared corrections over the whole network, and two variants constrain it further — one holding the time shift constant along each line, one adding phase to that solution. Polarity estimation looks for and corrects polarity differences between intersecting lines. Intersections you have already checked can be excluded, shifts on lines running along the structural dip can be given less weight by azimuth, and the result can be applied to the data or left as a map display only.

Can 3D volumes be tied to each other?

Yes — that is a separate part of the same tool, for surveys acquired or processed differently. g-Space calculates where the volumes intersect, computes the mis-tie and the cross-correlation at a chosen intersection point in the same fixed or horizon-centred window, then auto-fits the volumes against a reference volume you nominate, solving for time, for phase, or for both. The results land in a volumes shift table where they can also be typed in by hand, and are applied to the volume.

Get started

Tie your own line network and see what is left.

Load the lines, compute the misfits at every crossing and colour the map by what comes back — take g-Space for a trial run, or talk to Geomage about a demo on your own vintages.

Download free trial See pricing