Incorrect initialization of saturations

[olwijn]

In FlowNet, the saturation tables (e.g. SWOF) extend to Sw values of 1, consistent with the requirement stated in the OPM Flow manual. As a result, however, the equilibration produces initial water saturations of 1 below the OWC. According to the Eclipse manual, the final Sw value in the SWOF table is interpreted as 1- Sor. When the table extends to 1, Sor would then be interpreted as 0, leading to an underestimation of the oil saturation below the contact. This appears to be wat is happening in OPM Flow. The impact in FlowNet simulations with OPM Flow was illustrated in issue #416 . It is not explicitly stated in the OPM Flow manual how Sor is determined by OPM Flow. If the table extends to an Sw value of 1-Sor, a more consistent result (initial oil and water saturations and volumes) is obtained. The requirement stated in the OPM Flow manual for a final Sw value of 1 may therefore not be correct. In that case, we need to change how we write the SWOF and similar saturation tables.