Unconventional reservoirs have numerous challenges, low porosity and permeability, complex pore geometries and uncertainties associated with interpretation of dynamic properties after fracture stimulation. Often questions are asked: Is this really a poor-quality rock (non-reservoir)? Has the rock been properly fracture stimulated? Can we use any log in such a predictive way that will have a good correlation with where is the flow coming from? In this paper, the authors will focus on how data integration played a key role in linking the results from production logs (PLT) of target zones that were perforated and how the use of pre & post fracture stimulation logs helped determine fracture height from a log based perspective.
Petrophysical analysis identified the zones of interest with the NMR log providing initial insights of movable and bound fluid volumes. Two different types of cased-hole logs were used to address fracture height:
Cross dipole sonic (anisotropy) and
Pulsed neutron logs (sigma mode and inelastic capture mode).
Both logs were run before and after fracture stimulation operations, so a "like for like" comparison could be made to reduce uncertainties associated with comparing logs from different service companies. A third quality control check was performed using temperature logs, however due to elapsed time between operations (pre & post fracture logs) the temperature difference was negligible.
Fracture stimulation operations were performed in 6 different intervals and the use of non-radioactive traceable proppant in all stages helped better understand fracture height with both cased-hole logs exhibiting good correlation. It also confirmed that all stages were successfully fracture stimulated. Overall, the log derived interpretation of fracture height did not correlate with the net pressure match derived from the post-frac analysis, where both cased-hole logs were characterised by lower than expected fracture height. Temperature logs indicated a better correlation but was used in a more qualitative manner.
Integration of static petrophysical properties with dynamic data correlated well and combined with the pre & post fracture logs were critical in identifying better quality prospective reservoirs. All stages were successfully fracture stimulated with the post fracture logs clearing identifying the traceable proppant. Some uncertainty exists in the true height of the fracture and limitations of each method and will be deliberated further in the paper.