The cementation factor in carbonates varies considerably with rock properties and that effects the calculation of water saturation from resistivity logs significantly. Using constant values of cementation factor (m) may lead to inaccurate saturation profiles thus uncertainty in hydrocarbon volumes. Several empirical models covering wide range of carbonates have been developed and published relating cementation factor to porosity. However, there exists no published studies examining the reliability of using such models for carbonate reservoirs in Libya. In fact, using constant (average) cementation factor is a common exercise in the evaluation of carbonate reservoirs in Libya.

In the present paper, 30 representative core plugs cut from two different wells in the Garian reservoir, a low porosity carbonate reservoir in Libya were used for the experimental work. These plugs were selected based on a comprehensive rock typing workflow that covered the entire reservoir column cored and encountered at these two wells. Selected plugs underwent resistivity measurements at reservoir conditions. Cementation factor was then estimated and modeled as functioon of porosity and finally compared with four famous correlations.

The study concludes that the model presented by Masoud et. al in 2008 fits the experimental data very well and is reliable for use in the Garian reservoir. It was also observed that vuggs and fractures have opposite effects on the cementation factor where vuggs increase cementation facor value while fractures decrease it.


The cementation exponent, m, is considered one of the important parameters in the calculation of hydrocarbon/water saturations from electrical wireline logs. Archie in 1946 [1] offered an experimental relationship that relates the resistivity of a rock to its porosity, from which the cementation factor (m) can be estimated. In carbonate reservoirs, where the heterogeneity is typically high, the variation of this factor is significant. This makes the calculation of water saturation with an average value of m highly uncertain. [2]

This content is only available via PDF.
You can access this article if you purchase or spend a download.