Theoretical Details

Use of iMineralysis™ to optimize petrophysical multimineral analysis


In areas with mixed, complex, or simply uncommon lithologies, petrophysicists often use multimineral analysis to estimate volume fractions of individual fluid and mineral constituents of the rock after making a few simple assumptions about the relation between the log responses and the volume fractions. In this process, the log data is “inverted” at every depth for the volumes of the constituents that form the rock. 


A crucial assumption behind the multimineral analysis is that the response of any log selected for the analysis can be described as a linear combination of the individual log responses of each constituent weighted by the fraction of that constituent. Another important assumption is that the response of different logging tools when used to measure a “pure” constituent is also a known constant sometimes called “coefficient" (or end-point). This means, for instance, that the response of the density tool for pure quartz is always in the close vicinity of 2.65 g/cc. However, the response of other constituents to other tools may not always be known.


If the multimineral analysis is performed understanding the assumptions of the process and correctly choosing the right constituents and their properties, the result is a very good approximation to the measured well logs with good estimates of the volume fractions of the different constituents. However, when the assumptions behind the method are violated, fractions estimated from the analysis may be incorrect and further analyses (such as rock physics modeling, for instance) may also yield erroneous results.


In the white paper below, we review the method of multimineral analysis, its advantages and limitations, and propose solutions for some of these limitations that are implemented into the iMineralysis™ software. In particular, we discuss a solution to the problem of estimating the most appropriate “coefficient” that describes the response of a given constituent to a particular logging tool.  We also explain ways to improve the mutimineral analysis by incorporating other types of data and information not commonly used in conventional petrophysical analyses.