Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Conversion
Felix Obere
This presentation includes forward-looking statements. Actual future conditions (including economic conditions, energy demand, and energy supply) could differ materially due to changes in technology, the
1
development of new supply sources, political events, demographic changes, and other factors discussed herein (and in Item 1A of ExxonMobil’s latest report on Form 10-K or information set forth under "factors
affecting future results" on the "investors" page of our website at www.exxonmobil.com). This material is not to be reproduced without the permission of Exxon Mobil Corporation.
Objectives
• Understand importance of time-depth conversion in the
Exploration process and the importance of interpretation in depth
vs. time
Vp/s (x,y,z)
Seismic Rock Property/
Reservoir
Survey
Planning Velocity Prediction
is the Key
Logs , Cores
VSPs , Checkshots
Depth Definitions
Z Depth
Z Layer or interval thickness
Zw Well depth
Zs Seismic depth
Zn Normalization depth
Zmid Midpoint depth (Zn+Zn+1)/2
Velocity Definitions
V Velocity
Vi Interval velocity (Vint) _
Va Average velocity (Vavg, VBAR,V )
Vrms Root mean square (RMS) Velocity
Vnmo Normalized moveout velocity (stacking velocity)
Vnorm Depth normalized interval velocity
Ve Instantaneous velocity (Vinst)
Vw Well velocity
Vs Seismic velocity
Velocity Options for Depth Conversion
• Well Velocities
– Sonics
– Checkshot / VSP
– Pseudo Velocity – from well tops and seismic times
• Seismic Velocities
– Stacking velocities – Vave & Vint from Dix equation
– Tomography – travel time inversion using horizons &
stacking velocities
What Type of Velocity Data Should You Use?
Choosing A Time-Depth Conversion Method
The choice of method(s) for a particular area depends
on several factors:
• Exploration
– Verify structure
– Basin Modeling
Complexity &
• Development Data Availability
– Establish volumetrics/economics Increases
Available
• Production Time Decreases
– Well planning
– Geologic modeling
Time Interpretation Pitfalls
Scenario 1: Lithologies with anomalously high velocities can produce structural pull-up in the
time domain. Proper depth conversion will remove structures.
V = 5000 m/s
Limestone
Target Target
Sand Sand
Shale Shale
Target
Target
Target
Target
Target
flat water
waterbottom dipping rugose water bottom
flat bottom dippingwater
waterbottom
bottom rugose water bottom
Time
Depth Time Time
Depth Depth
Simple
Not so simple
Time
Section
(sec.)
Depth
Sectio
n
(ft.)
The interpreter is most knowledgeable in geologic changes which may impact the depth
conversion.
Would you drill ?
Time
Sectio
n
(sec.)
Depth
Section
(ft.)
Exercise 2
QUIZ
Is the channel fill in Figures A and B of higher or lower
velocity than the surrounding material ?
A B
Water Water
Seismic time
Time
Methods to consider:
Time vs Vavg
Time-slice
V0+kz
Vint Depth
Time-to-Depth Conversion Methods
Single/Constant Function Example
Data Analysis
Check Shot Data
Time vs Depth
Time
Time
Depth
Depth 10
Single/Constant Function Error
Time
Depth
Polynomial for depth conversion
•Fitting a polynomial to several checkshots will result in an average velocity
function.
•This is best for depth conversion of maps over field
•An average from many wells might have larger errors at a proposed well location
•Plot two way time on X-axis and depth (ft TVDSS) on Y-axis
•Fit 2nd and 3rd order curves to the data. In rare cases in the JV, a 4th
order might be required for depths in over-pressured intervals
•Set the number of decimal places in the formula to 10
•Carry out error analysis
Time-to-Depth Conversion (Polynomial)
Time-to-Depth Conversion (Polynomial)
20
QC and Error Analysis
•Estimate depths for 2nd and/or 3rd
order polynomials based on measured
two-way times from the checkshot
CI: 5 ms CI: 20 ft
0 2500
m
•Seismic Velocity QC
Location Map
• Seismic migration Vrms is obtained during Velocity Model AOI
seismic data processing
• Vint and Vavg velocities calculated from
Vrms using dix equation
• Good seismic velocity trend with no
significant outliers N
Seismic-to-Well Velocity Calibration
Seismic Vavg Velocities along wells Ratio of Well to Seismic Vavg Velocities
Ratio
TWT(ms)
N
Seismic-to-Well Velocity Calibration
Upscaled Well-to-Seismic Vavg Ratio Petrophysical modeling of the ratio
• Upscaled the ratios (along well path) into the model (50m by
50m by 4ms grid size)
• Used a defined 1D trend away from influence of well control
• Used Kriging interpolation algorithm to obtain a 3D ratio
volume
N
Average Velocity Model
• 3D Average velocity model = 3D Seismic Vavg * 3D Final Well-to-seismic Vavg ratio
TWT(ms) vs Well-Seismic Vavg Ratio (Initial) TWT(ms) vs Well-Seismic Vavg Ratio (Final)
Advantages Disadvantages
· Handles vertical and lateral velocity changes in · Takes more effort, time, and computer resources
complex geology · Each method has its strengths and pitfalls - when
· Utilizes numerous sources of velocity data in one we ignore the assumptions of the method error occurs
depth conversion i.e. well data, isopachs, velocity · Detailed seismic interpretation needs to be complete to
functions, seismic velocities, sonic velocities, analyze velocity data and determine functions. A shallow
interval velocities etc. velocity anomaly may need horizons picked that are not
· Optimizes method used for each layer. needed for prospect economic evaluation.
· Requires more velocity data for each seismic horizon
so that a function or relationship can be defined
Layer Cake Method
Depth Migration v. Depth Conversion
• Depth migration = an imaging issue used to properly focus &
position events laterally. PSDM data are not depth
converted!
• What was the business objective when the model was created?
• Deviated wells
•Positioning uncertainty
•Time Slice Interpolation
- 50
- 35
+200
Note: A reliable depth conversion will tie existing wells and predict depths
accurately away from well control.
Well Calibration
Step 1: Calculate the mistie between the well marker (t, d, velocity)
and its map equivalent