I like to believe that depending on the specific needs and goals, each reservoir needs a customized characterization workflow. In the case presented in this post, the goal was to understand the reasons behind the high water production in 3 recent drilled wells:
Well 1: Decent initial hydrocarbons production followed by lots of water.
Well 2: Lots of water
Well 3: Low rate – mainly water, some hydrocarbons.
In order to accomplish the goal, a step-by-step petrophysical characterization is presented including the following aspects: Lithology, Porosity, Water saturation, Permeability, Fractures and secondary porosity, and capillary pressure.
There are other aspects from different disciplines that could be integrated to this post, but I wanted to keep the emphasis on applied petrophysics.
Actions taken as a consequence of the findings described in this post added significant economic value to the project.
Step 1. Lithology Calibration (Mineralogy)
The multimineral model implemented contains the minerals observed in the drilling cuttings and described by the well site geologist. XRD or FTIR data used when available.
Based on the selected minerals and input parameters, theoretical logs responses are calculated and compared with the actual logs. The probabilistic model is calibrated until getting a satisfactory match.
Step 2. Porosity Calculation
The calibrated probabilistic model generates a porosity based on the multimineral matrix and fluids properties. This porosity must be consistent with all the logs including Density, Neutron, Sonic, and PEF.
Thin sections were analyzed and integrated to the interpretation helping to understand the porosity architecture, rock texture and composition.
When core or MSCT plugs available, the log porosity was calibrated to the core porosity as well as the matrix density.
Step 3. Water Saturation Estimation
Once the lithology and the porosity were properly calibrated, the water saturation was calculated.
Due to the uncertainty and the specific characteristics of the field, Induction and Laterolog tools were run in several wells. The final conclusion was that Laterolog has better performance in this environment.
Rw was initially defined based on samples from neighbour wells (DSTs and Catalogs). These values were subsequently verified with produced formation water resistivity.
In the absence of core measurements, m and n parameters were defined as 2.2 and 2.0 respectively by correlation with similar reservoirs according with available analog catalogs (these values are believed to be conservative).
N-D crossover, Oil shows (drilling cuttings), Gas shows (mud logging) and all the information suggesting the presence of hydrocarbons was integrated and put in context with the results.
Step 4. Permeability Calculation
In this analysis it was observed that the Micrologs was consistently a good qualitative permeability indicator (always checking that logs were not affected by washouts and borehole rugosity).
In addition, mercury injection capillary pressure (Pc), Porosity (Phi), and Permeability (K) were measured in cores and sidewall cores. This information was used to develop a permeability model based on Pore throat sizes and porosity. Results were more realistic, having a wide range of permeability for the same porosity value and the correlation between estimated and core permeability was very good.
Step 5. Fractures and Secondary Porosity
Image Logs (FMI) from 3 different wells were processed and the results (fractures, fracture density, secondary-vuggy porosity) integrated with the petrophysical analysis.
A vuggy porosity model using Neutron, Density and Sonic data was developed and calibrated with the vuggy porosity calculation from FMI.
Step 6. Capillary Pressure – Water Saturation Height
16 Capillary pressure curves were measured in sidewall plugs. Based on this Pc curves, K, Phi, and the Pore Throat Size, the Free Water Level (FWL) was estimated using Leverett J functions matching the Sw calculated from logs.
After integrating the available data and performing the described petrophysical analysis in all the wells, the potential of the perforated intervals seemed to be good and significant water production was not expected.
… What could be happening with these wells?
Typical Water Production Problems
In the picture below are presented some of the typical water production problems that we believed could be affecting the wells presented in this post.
Well 1: Decent initial hydrocarbons production followed by lots of water
Perforated interval in Well 1 has favorable petrophysical properties (good porosity, low clay content, low water saturation), evidenced by strong mud logging gas shows, cuttings oil staining, Neutron-Density crossover and high resistivity.
In the picture to the right is presented a profile of water saturation vs. Depth. Points are coloured by the intensity of the mud logging gas shows. It can be seen that perforated interval has low water saturation and strong was shows. It is also evident that water saturation is increasing with depth, as expected.
This well was expected to produce mainly hydrocarbons, and the actual high water was a surprise.
For some specialists this might be obvious (but is not always the case, even more when multidisciplinary teams/information is not integrated): after checking the Cement Bond Log for this well, it was concluded that there was a high probability that the produced water was flowing behind the casing (see picture below).
After a remedial cementing job was performed, this well produced hydrocarbons with negligible water cut.
Well 2: Lots of water.
Well 2 had a similar situation than Well 1, petrophysical properties were favorable and no water production was expected. It also had a poor cement bond, and after a remedial job was performed, the production water cut was very low.
Well 3: Low rate – mainly water, some hydrocarbons.
Situation in Well 3 was different than in the other 2 wells: cement bond was good, but the calculated water saturation was higher and Microlog response suggested lower permeability. Interval still had mud logging gas shows and drilling cuttings presented strong hydrocarbon smell and fluorescence.
The conclusion was that this well was more likely to be completed in a transition zone. The second option (less probable) was that the completed interval has fractures in contact with wet zones. No remedial action was taken in this well.
It is not only the rock and fluid properties, do not forget to check wellbore mechanical properties.
It was critical to have a robust and reliable petrophysical interpretation before deciding to spend resources in expensive remedial cementing jobs.