Scribd
Upload
8 views14 pages

2011 M5 DepthErrorCorrection New

The document outlines the objectives and methods for depth error analysis and correction, including the handling of residual depth errors and the use of variograms in geostatistics. It discusses the importance of grid spacing in reducing residual errors and the setup of depth correction models using velocity models. Additionally, it emphasizes the significance of reference levels in depth conversion processes and provides recommendations for handling seismic reference datum.

Uploaded by

sandeep s
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
8 views14 pages

2011 M5 DepthErrorCorrection New

The document outlines the objectives and methods for depth error analysis and correction, including the handling of residual depth errors and the use of variograms in geostatistics. It discusses the importance of grid spacing in reducing residual errors and the setup of depth correction models using velocity models. Additionally, it emphasizes the significance of reference levels in depth conversion processes and provides recommendations for handling seismic reference datum.

Uploaded by

sandeep s
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 14

Depth error analysis and correction

Objectives

Depth error analysis


Depth error correction
Variogram principles
Depth error residual handling
Depth correction model
Reference levels
Residual depth error
After the velocity correction (typically based on well tops) a residual depth error is
sometimes observed

Error before velocity correction Residual error after velocity correction

Known reasons for residual error:


1. Clustering of wells with strongly deviating well tops
2. Method ‘Same as above’. The corrected velocity field is used for two (or more)
velocity intervals
Correction of residual error
In many cases the residual error can be considerably reduced by re-gridding the input TWT
surfaces using a smaller grid spacing (Exception: depth errors resulting from the velocity
function ‘Same as above’ option)

X/Y- increment: X/Y- increment:


100 feet 25 feet

X / Y increment in the Make velocity model process:


1. Controls the grid spacing of the output velocity
surfaces
2. Generally has no significant impact on the residual
depth error
Geostatistics – variogram concept

Variogram:
A quantitative description of the variation in a property as a function of separation
distance between data points
Based on the principle that two points close together are more likely to have
similar values than points far from each other

Two main aspects of a variogram:


1. How similar are two values right next to each other?
2. How far apart are two points before they bear no relation to each other?
Geostatistics – variogram parameters

Variance
Variance: A measure of how different
members of a collection are from Sill
each other
Lag distance: Separation distance

Nugget
between points
Sill: Variance at the point where the Range Separation
summary plot flattens out to random distance
1 2 3 4 5
similarity (lag)

Range: Correlation distance; distance


beyond which data points no longer
exhibit any statistical similarity
Nugget: Degree of dissimilarity at
zero distance
Note: The separation distance is in project units
Basic statistics – variogram calculation

Calculation process and Lag distance


setup
Defines max
1.A Search radius and distance pairs of
Number of lags must be data could be apart,
defined=> decides the Lag when compared
increment and consequently (within each lag)
the Lag distance
2.All pairs of points in each Search radius
Lag (bin) will be compared
Defines how far away
3.For each lag, the average data will be collected
variation is calculated
Basic statistics – experimental variogram
Variogram plot
1.The average
variance per lag is
plotted in a
Separation distance
vs. Semivariance
crossplot
2.These points make
up the Experimental
variogram (green
points)
Handling of residual depth errors
Typically the residual depth error is of random character. Consequently its sample
variogram shows only a small range (if any)

Sample variogram of residual


Residual depth error depth error

To interpolate the depth error, a gridding algorithm should be applied that allows to
define a range of influence (kriging, moving average)
Setup of a depth correction model
Velocity model

Depth surfaces from a previously run velocity model can be used as input for the
correction model to address residual depth errors

Note: The domain conversion needs to be done in two steps: first the velocity model and then the depth correction
model is applied to the data.
Depth correction model - gridding algorithm

Min. curvature Cos. expansion Convergent Moving average: Quadruple

The gridding algorithm for interpolating the


residual correction surfaces V0 can be
selected in the Correction and output tab

Only Moving average algorithm allows to


control the range of influence of the data
points. However, it does not honor the well
Adjust addresses the residual error
tops precisely when using low point weights.
resulting from the interpolation method

If Use influence radius is selected, all surfaces will be gridded by the influence radius
method. Otherwise, all surfaces will be gridded using the interpolation method
Reference level handling
The Seismic Reference Datum (SRD) is Petrel assumes MSL (Mean Sea Level)
defined in the Templates pane under the as depth reference level. Exception: well
Datums folder. Petrel assumes that ALL time depth axes MD, TVD
objects (incl. well data) refer to the SRD
The depth surfaces provided by the
Make velocity model process refer to
MSL, independently of the chosen SRD

Recommendation: keep the Datum default values!


Default Reference level handling: Example
SRD is 200 feet above MSL V=f(z) Elevation
After depth conversion of
surfaces, subtract 200 feet 0 200 SRD
from the depth surfaces to
make sure that the depth
axis starts at MSL 0 MSL

1200 1000
Depth

Note: Handling well time- and depth-datums when importing checkshots and other well data: this is equally
important as handling the velocity model reference correctly. Correct well TDR is paramount when doing linear
velocity extractions or corrections of velocity models to well tops
EXERCISE
Depth error analysis and correction

You might also like