ABSTRACT

At deposition much of the petrophysical character of reservoir sandstones is defined by the grain size distribution. It would be desirable to model reservoir permeability using grain size alone, but diagenesis alters the configuration of porosity. Any effort to further understand this relationship must account for both the initial configuration of 'effective' porosity as well as any modifications accompanying diagenesis.

At deposition sands tend to self-assemble into a mixture of close-packed grain clusters separated by packing flaws. These flaws propagate throughout the sediment forming a network of expanded porosity 'circuits' that are responsible for virtually all of the fluid flow. A significant body of research indicates that grain size defines the flaw size at deposition, while cementation serves to progressively reduce flaw size (and permeability) as diagenesis proceeds.

The basic understanding of the relationship between petrology and permeability has led to two different methods for estimating permeability. The first uses flaw size, lithology, and diagenetic state to determine permeability directly from thin section. It has an advantage over other techniques in that it is applicable to fractured and deformed samples such as sidewall cores, where normal laboratory techniques can be unreliable.

The second method was designed for reservoir modeling. The extreme variability of permeability has led to the use of Monte Carlo and other simulation techniques to model reservoir permeability. Both median grain size and total porosity are less variable than permeability and can be modeled throughout a reservoir. Using the second method, permeabilities can be estimated for each block. At best the model is superior to simulation. At worst, it provides a better starting point for Monte Carlo techniques.