Corner-Point Cell Geometry

Eight Corners, Four Pillars

Chapter 3 built the whole corner-point grid; here we zoom into a single cell to see how it is defined. A corner-point cell is a hexahedron specified by eight corner points, two on each of four near-vertical pillars. The pillars are fixed lines, often placed along faults; the cell is whatever the eight corners enclose. Tilt and shear the cell in the widget to see how those corners define a general six-faced solid.

Why the Corners Move

The power of the scheme is that each corner slides independently along its pillar. A box cell is rigid, always a rectangular block aligned to the axes. A corner-point cell is free: its top and base can dip, its sides can shear, and adjacent cells share corners so the grid stays watertight. That freedom is exactly what lets the grid follow a fault plane (corners on either side of a pillar at different depths) and steep dip (top and base parallel to the dipping layers) without the distortion a box grid suffers.

The Price of Freedom

This flexibility is not free. A cell that is too sheared or twisted becomes non-orthogonal, and the flow solver, which assumes flow crosses cell faces roughly perpendicular, loses accuracy. So corner-point geometry buys the ability to honor the structure, but it hands the modeler a responsibility: keep the cells as well-shaped as the geology allows. The rest of this chapter is about spending that freedom well, on resolution, pinchouts, and flow.