Top Banner
Signal Estimation Technology Inc. Maher S. Maklad Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++
15

Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Mar 26, 2015

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Maher S. MakladMaher S. Maklad

Optimal Resolution of Noisy Seismic Data

Resolve++

Page 2: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

• Your team is under constant pressure to extract the most information from corporate assets as accurately and swiftly as possible.

Resolution Optimization: Motivation

• This information provides the foundation on which your business makes decisions.

• These decisions are based on a perception of reality. The result of these decisions depends on the accuracy of the perception.

• How to use seismic attributes to enable more informed decisions for the identification, reduction and management of risk while maximizing reward? One answer is to investigate both standard and alternative interpretation workflows available to determine ways of validating and/or improving upon “current practices”.

Page 3: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

•Produce a geologically faithful higher resolution seismic image.

Resolution Optimization: Goals & Rewards

The Rewards:

• More accurate and informative seismic attributes.

Our Goal:

• Improved horizon picking.• More accurate maps.• Easier detection of important geologic features (faults, wedges,

etc.).

Page 4: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Anatomy of Seismic Data

= Consists of several components:

SEISMIC

Seismic(t) = Wavelet(t) * Reflectivity(t) + noise(t) Convolutional Model

Seismic attribute analysis uses information extracted from the seismic data or its constituents.

Seismic Response

Tim

e

Energy Source

Wavelet

*

EarthReflectivity

Reflectivity

+

Noise

Noise

Page 5: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Earth Filter

=

Seismic Response

EarthReflectivity

Noise

+

Tim

e

Noise AttenuationObservations: Signal-to-Noise Ratio (SNR) is often not stressed.

*

Consequences: The two amplitude maps below show the impact of removing noise on interpretation.

GCWS_top Amplitude map

Before After

Page 6: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Resolution Optimization…..procedure

Seismic Data

Estimate signal power spectrum

Decompose into a waveletand a reflectivity component

Estimate wavelet spectrum

Estimate SNRDesign desired wavelet

Operator = Desired wavelet / estimated input wavelet

Strategy: Divide and conquer!Address the additive noise problem (Get the signal).•Address the convolutional noise problem (Get the wavelet).

Contrast with conventional techniques where operators are designed in the data domain – signal plus noise!

Page 7: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Resolution Optimization…..objectives

EnergySource

=

Seismic response

EarthReflectivity

Noise

+

Tim

e *

The objectives are:

• Estimate the wavelet in the presence of noise.

• Shape the wavelet according to SNR.

• Improve resolution while controlling noise.

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

0 100 200 300 400 500Frequency (Hz)

Mag

nit

ud

e (d

B)

• Preserve the colour of the reflectivity.

Well log generated Reflectivity Spectrum

Page 8: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Resolution Optimization ….resultsBefore After

Page 9: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

0 100 200 300 400 500

Frequency (Hz)

Ma

gn

itu

de

(d

B)

Before

Resolve has made improvements in the following areas:

Resolution Optimization ….validationPeak Frequency

• Increased peak frequency of the data from ~ 140 Hz to > 250 Hz.

After

• Made the spectrum of the data follow the spectrum of the log generated reflectivity more closely providing confidence in the spectral gains, and the enhanced stratigraphic and structural interpretation that the Resolve data will enable.

• Increased the bandwidth of the data from ~ 200 Hz to ~ 300 Hz.

Bandwidth

Before

After

Well log generated Reflectivity Spectrum

Page 10: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Brute Stack Processor’s Final Stack Resolve on Brute Stack

Resolve++ on Brute Stack versus Conventional Processing

Page 11: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Original Processed Volume

Spectrally Shaped Volume

Input to Resolve++™

Page 12: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Original Processed Volume

Spectrally Shaped Volume

Output of Resolve++™

Page 13: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Scanned Data

Scanned DataData from scanned seismic section

Page 14: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

Scanned data after Resolve++

Page 15: Signal Estimation Technology Inc. Maher S. Maklad Optimal Resolution of Noisy Seismic Data Resolve++

Signal Estimation Technology Inc.

• Resolve++ increases the resolution of seismic data without amplification of noise (i.e. constrained by SNR).

• Enables the extraction of better quality attributes for reservoir characterization.

• This results in a better spectral representation of earth reflectivity.

• More accurate interpretation of horizons and faults.

• A viable alternative to reprocessing old data.

Conclusions