AASPI · University of Oklahoma
Grains and pore space set how stiff a rock is. The fluid in the pores changes that stiffness, but not in every direction equally. Stiffness and density together set the velocities. Two rocks meeting at a boundary set a reflection. And once the sound arrives at an angle rather than straight down, that reflection stops being a single number and becomes a curve.
These modules build that chain one step at a time, with every panel computed from the controls on screen. The later ones work the chain backwards, which is harder and does not have one answer. Nothing to install, and nothing leaves your machine.
Start with the resolution modules. This set assumes What Can You REALLY See in Seismic?, where impedance, the reflection coefficient, the wavelet and tuning are built from scratch. Module 00 here restates those results and picks up where they stop.
A rock resists being squeezed and resists being sheared, and those are separate properties. A pore fluid changes the first and leaves the second almost alone, which is what makes a fluid visible to seismic data at all.
A wave arriving straight down only compresses the rock. One arriving at an angle also shears it, so the reflection changes across a recording — and changes differently depending on what is in the pores.
Going from a rock to an amplitude is exact. Going the other way is not, because several rocks arrive at the same number. The last modules measure how many, and how much the second measurement removes.
Ten modules, in the order they build on each other, from the grains outward. Each card says what it assumes.
Work through the resolution modules first. Then 00, 01, 02 and 03, and stop there. That is a complete and useful course on its own.
Start at 04 and go straight through. Module 00 is still worth ten minutes if only to see which simplifications the rest of the set comes back for, and 01 to 03 are quick if the rock physics is already familiar.
07, 08 and 09 are the ones with something new in them: how many rocks share an amplitude, the size of the error bars, and a count of how much the gradient adds.
A recap of what the resolution modules established, the assumption every one of them shared, and the observation this set begins from: a reflection coefficient is not one number.
Grains and pore space, where bulk density comes from, the two separate ways a rock resists deformation, and why a change in one of them moves Vp and leaves Vs exactly where it was.
Brine, oil and gas as elastic materials, Gassmann's relation putting one into the frame, and why the first few percent of gas does almost all of the work.
A reflection needs two rocks. Velocity and density collapse into impedance, impedance into a contrast, and the shear velocity does not enter at all.
Where offset comes from, why it is not the same thing as the incidence angle, what moveout costs to remove, and what stacking throws away.
The ray stops arriving straight down, the shear velocities become visible, and one measurement becomes a curve.
Two numbers instead of a curve, the crossplot they live in, the wet trend that makes an anomaly anomalous, and why Classes I to IV are regions somebody drew.
Run the chain backwards. Fix a stacked amplitude and search for every rock that could have produced it: porosity against saturation, the shale nobody measured, and tuning.
Real gathers carry noise and a mute. How large the error bars are, why the gradient takes almost all the damage, and whether the rocks can still be told apart.
Every loss added up and measured, a count of what the second measurement removes, four rules of thumb with computed counterexamples, and a statement the numbers will support.
They are built for students meeting seismic interpretation for the first time, and for anyone who arrived in an interpretation role from an adjacent discipline and is expected to be productive in weeks. Beyond the resolution modules named above, they assume first-year geology and nothing else.
The title is a promise about the later modules rather than the early ones. "Actually" means that each rule of thumb gets tested until it breaks, and that where it breaks is measured rather than asserted. The early modules build the forward calculation without qualifying it, because running that calculation confidently is what the later ones take apart.
Every panel is computed from the parameters on screen. There are no stored images and no curves drawn to look plausible, which means the tool can be wrong — and during construction it repeatedly was. A drawing cannot disagree with theory; a calculation can, and the disagreements are where the corrections came from. Every module carries a Method tab listing what has been left out and where the implementation departs from production software, and every number quoted in the exercises is read off the running page rather than estimated.
No slider, no click and no trace you generate is transmitted anywhere, and the modules make no network requests at all. The state of every control is written into the address bar, so a configuration can be handed out as a link. The one thing recorded is that a page was opened, with no cookie and nothing about you.
A further companion set on geometric attributes covers dip, coherence and curvature, and is independent of this one and of the resolution modules.