In conventional reservoirs, the correlation of reservoir quality (RQ) to production is a standard practice. RQ is defined as the product of various rock properties, including saturation (S), porosity (???), thickness (h) and, in some cases, permeability (k). Plotting RQ against normalized well production gives an indication of rock type and hence production trends.

To use this methodology in unconventional reservoirs, it is necessary to modify the relationship to account for the complexity and heterogeneous nature of shale rock. Modification of the method involves the identification of the most representative combination of rock properties used to calculate reservoir quality, which are then compared to multivariable, normalized cumulative production at different time intervals. Applying this relation in the Eagle Ford shale resulted in the observation of two trends, implying that different production trends were in play and should be further investigated.

This paper presents a detailed well-by-well study conducted in the Eagle Ford shale to define the rock characteristics that caused the presence of two production trends. History matches using numerical modeling and rate transient analysis were performed to verify whether two production trends were present. When available, borehole images can be used to validate and extrapolate this localized production analysis to a basin-level understanding.

Methodology Background

The objective of this study was to determine a correlation between RQ parameters as obtained from well logs and production performance in the Eagle Ford shale. This led to the identification of different rock types, implying different production trends. This correlation technique has been used in the past for carbonates, sandstones (Pickett and Artus, 1970) and muddy sands (Aguilera, 1995) and is known as " petrophysics-to-production" methodology. The original methodology correlates recoverable production from decline curves with an RQ index (a function of So, ???h and k) as depicted in the left plot in Figure 1a, which shows three different production trends. Pickett (1970) identified that each trend corresponds to a different rock type with a different production performance where the gentler trend is due to a non-fractured rock type (k/???)2. For this study, the original approach has been adapted as depicted in the right plot in Figure 1b by modifying the production term as cumulative production normalized to lateral length and the RQ index defined using a petrophysical model.

URTeC 1619181

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