Petrophysics is the study of the physical and chemical properties of rocks and their included fluids, if any. Petrophysical data can be obtained from well logs and from laboratory data. Petrophysical data is used both qualitatively and quantitatively. Both forms are discussed in appropriate sections of this Handbook.
Although "petrophysics" was used by G. E. Archie in the 1940's, the word has only become popular in the last 20 years. The terms "log interpretation" or "log analysis" are widely used in the literature (inaccurately) to mean the same thing.
Petrophysics is a more inclusive term, encompassing core analysis,
sample descriptions, X-ray diffraction, petrography, scanning
electron microscopy, and other forms
of detailed laboratory data, in addition to well log data. lab data
can also be considered as a "well log", because the depth of each
sample is usually known.
Petrophysicists offer services in the areas of well logging supervision, log analysis and interpretation, computer analysis of logs, seismic modeling, synthetic seismograms, and reconciliations of log data with geological, geophysical and exploration prospects, field studies and simulations, reserves estimates, and submissions to regulatory agencies.
These services are essential functions in
modern oil and gas companies and cannot be accomplished without
input from trained petrophysicists. The financial health and long-term
success of a company depends on the central role of petrophysicist
in all aspects of the company’s exploration and development
Thus the art and science of "petrophysical analysis" was born. A petrophysicist is the person who does the analysis of the data. Log analysis and log analyst are terms still widely used as synonyms.
"Integrated Petrophysics" is the most recent buzzword, but is a little redundant, as the combined use of well logs and lab data has always been a fact of life for anyone trying to calibrate well log analysis results to "ground truth".
Data analysis, or data reduction, gives us "answers" that can be ambiguous at best or dead-wrong at worst. To consider "answers" as the final result of petrophysics is not adequate; the answers have to make sense when compared to all known facts and probably need calibration to those other facts. This step may be iterative, so petrophysical analysis is usually not a one-pass process. . The next step is to interpret the answers, and hopefully, gain an understanding of the answers. There is a huge gulf between getting answers and understanding the answers. Most of this Handbook deals with getting answers. However, numerous case histories are provided to help start the understanding process. Trial and error, experience, and a great memory will continue the process.
Calgary Herald cartoon circa 1978, showing the public's impression of reservoir evaluation. That's me in the parka, looking into the borehole, while supervising a logging crew for PanArctic Oils Ltd on Melville Island in the Canadian High Arctic.
By integrating the data from other geoscience disciplines, we can do more than mere log analysis. We can actually define the rocks and fluids AND calibrate our work.
State of the art well log analysis involves the intelligent use of a multiplicity of log curves and lab data in complex computer programs which evaluate many unknowns at once. Calibration of results usually requires statistical analysis, correlating laboratory and log analysis parameters. Sophisticated analysis of this type requires highly trained technical staff for programming, data entry, and evaluation. Software is often deterministic, using a fixed or user-defined sets of mathematical equations to derive answers from raw data and parameters supplied by the analyst. Other programs use probabilistic, statistical, or neural network methods, in addition to some deterministic code, to obtain answers. No known software can do the interpretation and understanding phase of the job - that is up to YOU.
To reduce dependence on the expertise
of the programmer and user, a number of companies are studying
the use of artificial intelligence (expert systems) to guide less
experienced analysts through the analysis procedures. However,
after more than 25 years of research, none have become commercial
Page Views ---- Since 01 Jan 2015
Copyright 1978 - 2018 E. R. Crain, P.Eng. All Rights Reserved