Good comment but there are actually three high recovery areas - Ain Dar, Shedgum, and North Uthmaniyah.
Best I remember production was reversed engineered to 0 .8, 1.2, and 2.5 respectively.
North Uthmaniyah has the oil strip down the middle with water encroaching both sides.
Ghawar has a 1,300' oil column (peak structure above the oil water contact) . That provided the driving force that filled it with oil. As a result, the connate or interstitial water saturation was probably driven down to about 11% over much of the Arab D structure. Good **** the news just keeps getting better and better you know.
The Buckley Leverett displacement model is the ohms law of waterflooding. What is great about the low connate water saturation is it makes a huge mobile oil window, Aramco says residual oil is 21%, so 68% of that pore volume is recoverable by water ( One(1)- connate water saturation(.11)- residual oil saturation(.21). I have reservoir engineering waterflooding experience in 14 US states over 25 years. I always assume the best you can do is 40% of a pore volume. Figure up your porosity * thickness * area for your oil zone and multiiply by 40%, that's your ultimate recovery.
Now with the Buckley Leverett technique one builds something called a fractional flow curve. It is a graph of the water cut versus water saturation in the reservoir. It is generally an S shaped curve. Oil viscosity also plays a role in the shape of this curve. Ghawar's oil viscosity in situ is less than that of water. Good **** the news just keeps getting better you know.
One constructs a tangent to the curve from connate water saturation on the x axis to 100% water cut on the Y axis and that tells you everything you need to know. The flatter (more horizontal) this curve the better the waterflood. Ghawar is as flat as they come. There is not much oil left after the water breaksthrough, the waterflood is extremely efficient.
Those are the physics of the process. The Scot Euan Mearns probably did as good a job as any of taking these physics to a volumetric estimate and production schedule on the Oil Drum a few years back.
Assuming water drive reservoiirs natural or induced, I will state again that any oil man in his right mind would rather have 10 million barrels per day at a 90% water cut (Russia) than 10 million barrels a day at a 35% water cut (SA). Especially at these prices.