IV. 2.8.1. Interpretation of experimental field data

Some field tracer experimental data has been analysed using the PORO simulator. Some of them are described below as case studies. In all cases, the first step in the simulation was carried out under the following conditions:

• Residual oil saturation (Sw = 1 — (Sor — Swi));

• Total layer thickness watered (h = hmax);

• Reported porosity;

• Nominal water flow rates;

• 100% tracer recovery;

• No faults;

• No anisotropy;

• Dispersivity equal to 10% of the distance between wells.

Case application 1: Carmopolis oilfield (Brazil)

The pattern is illustrated in Fig. 107. The layer thickness is 10 m, the porosity 17%, the water saturation 63%, the injection flow rate 83 m3/d and the injected activity 55.5 GBq.

In the first step of the simulation of well CP-1091, using the parameters reported by the oil company, it was observed that the simulated tracer breakthrough time was double the experimental value.

Additionally, the measured cumulative tracer recovery in the well was 55.33% (extrapolated to 60%) instead of the simulated value (43.80%). It is believed that this difference is due to the fact that the streamlines were too ‘open’ because only the water flow rates in the wells were taken into account in

calculating their paths. Introducing modifications to confine the streamlines simulation became more realistic (Fig.108).

Thickness, water saturation and porosity values used in the second step of the simulation were provided by the company. The dispersivity value was in

FIG 109. Well 1 pattern (Pindori oilfield, Pakistan).

agreement with the criteria requiring that it must be equivalent to 10% of the distance between wells.

Case application 2: Pindori oilfield (Pakistan)

The pattern and the reservoir parameters are illustrated in Fig. 109. Up to day 150 after HTO injection, the only well in which tracer had been detected was well 3. On day 60, the injection conditions were modified, introducing strong perturbations in the tracer record. For that reason, only its non-perturbed portion was considered for this study.

Figure 110 shows both the experimental and the simulated tracer responses under the mentioned conditions.

The PORO simulator allowed an acceptable fit to the experimental records to be obtained by employing the reported injection water flow rate and the same water flow rates for all the producer wells. A layer thickness of only 0.35 m was used.

However, the simulation predicts a quick tracer breakthrough in well 4, which is not in agreement with actual tracer behaviour. Several reasons may be responsible for the lack of tracer detection in this well, such as different location of the fault, anisotropy and low water flow rate to the well. Additional information will be necessary for selecting the right scenario for this case.