The Carbonate Porosity Paradox
Learning objectives
- State the carbonate problem: in a rock where pore shape varies, velocity no longer maps to porosity
- Anchor the Xu-Payne numbers on calcite: a moldic pore (aspect ratio 0.8) at 15 percent porosity gives Vp 5.813 km/s while an interparticle pore (0.15) at 8 percent gives 5.645
- Read the paradox: the higher-porosity moldic rock is faster than the lower-porosity interparticle rock, and a crack pore (0.02) at 15 percent collapses to 2.716
- Draw the remedy: porosity-from-velocity in carbonates fails unless the pore type is identified first, by thin section, borehole image, or a Vp/Vs-plus-DEM inversion
Where Velocity Stops Meaning Porosity
In a clean sand the story is simple enough to invert: more porosity, lower velocity, and a calibrated transform reads one from the other. Carbonates break that contract. The reason is pore shape. A limestone can hold the same total porosity as rounded, stiff moldic vugs, as flat crack-like microfractures, or as ordinary interparticle space, and those three geometries carry wildly different stiffness. A round pore barely weakens the frame; a flat crack devastates it. So in a carbonate the velocity is set as much by the shape of the pores as by their volume, and a single number can no longer stand for porosity.
The Xu-Payne approach makes this quantitative. Model the carbonate as a calcite matrix with pores of a chosen aspect ratio, use the differential effective medium theory of Part 6 to add those pores, then saturate with brine through Gassmann. The aspect ratio, the ratio of a pore's short dimension to its long one, is the whole story: 0.8 for a nearly round moldic pore, 0.15 for interparticle space, 0.02 for a thin crack.
The Paradox in Three Numbers
Run the model and the contract breaks in the open. A moldic rock, aspect ratio 0.8, at 15 percent porosity gives a P-wave velocity of km/s. An interparticle rock, aspect ratio 0.15, at only 8 percent porosity gives 5.645. The rock with nearly twice the porosity is the faster of the two. Any transform that reads porosity from velocity would rank them backwards, calling the tight interparticle rock the better reservoir. And a crack-dominated rock, aspect ratio 0.02, at the same 15 percent porosity as the moldic case collapses to 2.716 km/s, less than half the moldic velocity at identical porosity. Three rocks, one paradox: velocity in carbonates encodes pore geometry first and porosity second.
The physical reading is the aspect-ratio ladder of Part 6 seen in a real rock. Round moldic pores are stiff inclusions: they remove volume without cutting many load paths, so the frame stays fast and a high-porosity moldic limestone can outrun a low-porosity one. Flat cracks are the opposite, compliant inclusions that sever load paths and crush the velocity for very little pore volume. Interparticle pores sit between. The same porosity distributed into different shapes produces velocities that span kilometres per second.
Identify the Pore, Then Read the Porosity
The remedy is not to give up on velocity but to add the missing variable. Porosity-from-velocity is only valid once the pore type is known, so the workflow becomes two-stage: identify the dominant pore geometry first, then apply the transform calibrated for that geometry. Thin sections and borehole image logs read the pore type directly. Where those are absent, the elastic data can be pushed harder: because moldic and crack pores affect and differently, a joint -and-DEM inversion can estimate an effective aspect ratio and separate stiff-pore from crack-dominated rock. Either way the discipline is the same, name the pore shape before trusting the porosity. The next section returns to clean sands but keeps the theme of two causes wearing one signature: a rock can be slow because it is overpressured or because it holds gas, and only the velocity ratio tells them apart.
References
- Anselmetti, F. S., & Eberli, G. P. (1993). Controls on sonic velocity in carbonates. Pure and Applied Geophysics, 141(2-4), 287-323.
- Xu, S., & Payne, M. A. (2009). Modeling elastic properties in carbonate rocks. The Leading Edge, 28(1), 66-74.