A Practical Guide for Seismic Interpreters — and How SubsurfaceAI Makes Them Easy to Use

Seismic interpreters work under a familiar constraint: the subsurface never shows itself directly. We infer geology from band-limited seismic signals, and we make decisions under uncertainty—often quickly, often with incomplete well control, and often across large areas. In that environment, rock physics can feel like a specialist’s territory: full of equations, assumptions, and terms that sound far removed from day-to-day interpretation.

But rock physics was never meant to be exclusive. At its core, rock physics is simply the bridge between rock properties (mineralogy, porosity, fluids, stress) and what seismic measures (velocity, density, impedance, AVO behavior). When that bridge is made usable, interpreters gain a major advantage: they can turn elastic attributes into geological meaning with greater consistency and confidence.

Two concepts sit at the center of practical rock physics workflows:

• Rock Physics Model (RPM)
• Rock Physics Template (RPT)

They are related, but not the same. Confusing them is common—and it’s one reason rock physics sometimes feels harder than it needs to be. This article explains RPM and RPT in interpreter-friendly language, describes their relationship and differences, and shows how the SubsurfaceAI (SAI) Rock Physics module makes these tools accessible, reusable, and scalable.

The audience for this article is expected to be seismic interpreters, not rock physics specialists. The central message is simple:

The real power of rock physics modeling belongs in the hands of ordinary seismic interpreters.

1) Why rock physics matters (even if you don’t “do rock physics”)

Interpreters already use rock physics ideas, even when they don’t call them that. Every time you interpret a low impedance anomaly as “porous sand,” every time you use Vp/Vs to infer lithology, every time you look at AVO class and think “possible gas,” you’re implicitly relying on rock physics.

The problem is that implicit rock physics is often inconsistent. A bright reflection might be:

• gas effect, or
• tuning, or
• a thin hard streak, or
• a change in shale stiffness, or
• processing/inversion artifacts.

Rock physics helps you replace “pattern recognition only” with “pattern recognition plus physics.” Instead of asking only, “What does it look like?”, you also ask:

• If this is a porous sand, what should AI and Vp/Vs do as porosity increases?
• If brine changes to gas, how far should the elastic response move—and in what direction?
• If the rock frame is cemented vs unconsolidated, how does that shift the trend?
• In carbonates, how do pore types change stiffness at the same porosity?
• In unconventional reservoirs, how do mineral mix and organic content influence elastic behavior?

To answer those questions reliably and repeatedly, you need a structured way to model physics and a structured way to interpret elastic space. That’s where RPM and RPT come in.

2) What is a Rock Physics Model (RPM)?

A simple interpreter definition

A Rock Physics Model (RPM) is a set of equations and assumptions that predicts elastic properties—typically Vp, Vs, and density (or derived properties like Acoustic Impedance AI = ρVp and Vp/Vs) from rock and fluid parameters such as:

• mineral composition (e.g., quartz, calcite, dolomite, clay)
• porosity and pore geometry
• fluid type and saturation (brine, oil, gas; Sw/Sg)
• pressure/stress and compaction state
• cementation and grain contacts
• shale fraction, kerogen/TOC (for unconventional)
• fractures or anisotropy effects (when needed)

In one sentence:

RPM is the “physics engine” that converts a rock/fluid state into elastic properties.

What an RPM does (forward modeling)

RPM is forward modeling. You specify what the rock is like, and the RPM tells you what seismic-relevant properties should be.

For example, you can ask:

• “If porosity increases from 4% to 14% in a clean sandstone, what happens to AI?”
• “If the rock is the same but fluid changes from brine to gas, what happens to Vp/Vs?”
• “If the rock frame is stiffer (cemented), how much does that change the trends?”

This forward modeling is the foundation of rock physics. But interpreters usually need the reverse: they have elastic properties (from logs, inversion, attributes) and want plausible rock states. That’s where templates come in.

RPMs can be simple or advanced

RPM doesn’t automatically mean complex. In practice, RPMs fall along a spectrum:

1) Empirical / data-driven models

• Example: a regression between AI and porosity in a specific interval.
• Strength: fast, easy, often fits local data well.
• Weakness: less transferable; may break outside calibration range.

2) Physics-based models
These combine mineral mixing, frame models, and fluid substitution concepts.

• Strength: interpretable and transferable; tied to physical assumptions.
• Weakness: requires reasonable parameter choices and calibration.

A key point for interpreters is this: you do not need to memorize equations to use RPMs effectively. You need to understand what each model assumes about rock fabric and what parameters control its behavior.

3) What is a Rock Physics Template (RPT)?

A simple interpreter definition

A Rock Physics Template (RPT) is an interpretation-ready visual guide—usually a set of trend lines or grids on a crossplot—showing how elastic properties are expected to vary across realistic reservoir scenarios.

Common RPT crossplots include:

• AI vs Vp/Vs (very common in seismic inversion workflows)
• Mu-Rho vs Lambda-Rho (or related elastic attribute spaces)
• Vp vs Vs (often used in log QC and lithology separation)
• AI vs porosity or Vp/Vs vs clay volume (fast diagnostic plots)

An RPT typically contains:

• porosity trend lines (or porosity bands)
• fluid substitution trends (e.g., brine → oil → gas)
• lithology families (e.g., sand vs shale vs carbonate)
• optional bounds and uncertainty envelopes

In one sentence:

RPT is a “map” of elastic space that helps you interpret where seismic or inversion points land.

Why RPTs are so useful for interpreters

Interpreters must act on elastic information at scale. They need tools that are:

• visual,
• repeatable,
• communicable across a team,
• tied to physics but not buried in math.

RPTs do exactly that. When an interpreter sees a cluster of inverted points shift across an RPT, they can ask:

• “Does this move correspond to porosity increase, fluid change, or lithology change?”
• “Are the points landing in a physically plausible region?”
• “Do different stratigraphic intervals show distinct elastic populations (facies separation)?”

A good RPT becomes a common language between interpreters, petrophysicists, geologists, and management.

4) Relationship between RPM and RPT (and why they’re often confused)

Here is the simplest relationship:

• RPM generates elastic predictions
• RPT organizes those predictions into an interpretation template

Think of it this way:

• RPM is the recipe
• RPT is the menu that shows what dishes the recipe can produce

More formally:

1. Choose or build an RPM (physics + assumptions).
2. Define realistic ranges for parameters (porosity, saturation, mineral fractions, stress).
3. Run the RPM across those ranges to generate many scenarios.
4. Plot those scenarios in elastic crossplot space.
5. Add grids/trend families and label them for interpretation.
6. The result is the RPT.

So:

RPT is often produced from an RPM, but RPT is not the RPM.

This distinction matters because it clarifies roles:

• RPM is about how the world works (forward physics).
• RPT is about how you interpret data (practical inverse reasoning).

5) Differences between RPM and RPT (in practical terms)

Purpose

• RPM: Predict elastic properties from rock/fluid parameters.
• RPT: Interpret elastic properties in terms of rock/fluid scenarios.

Form

• RPM: A model definition (equations, assumptions, parameters).
• RPT: A template/diagram (crossplots with grids and trend families).

Typical users

• RPM: Often built and tuned by rock physicists, but can be used by interpreters with a good interface.
• RPT: Designed to be used by interpreters and multidisciplinary teams.

Scale of use

• RPM: Can be very detailed (interval-by-interval, facies-by-facies).
• RPT: Often broader and reusable (formation-, member-, play-level).

Typical failure modes

• RPM failures: wrong assumptions about rock fabric (cemented vs soft), wrong mineral mix, wrong stress regime, unrealistic fluid inputs.
• RPT failures: built from uncalibrated RPM; parameter ranges not representative; too many facies mixed together; template not linked back to well control.

6) The interpreter workflow: how RPM and RPT become usable in real projects

A practical workflow that keeps rock physics interpreter-friendly looks like this:

Step 1 — Start with well data and elastic reality

Gather what you have:

• Vp, Vs (measured or estimated), density
• lithology/facies indicators (even simple sand/shale flags help)
• porosity, saturation, clay volume, TOC (as available)

Before modeling anything, look at what the data already says:

• Where do sands and shales plot?
• How do known fluid zones separate?
• Are there clear trends vs depth or compaction?

Step 2 — Choose an RPM family that matches the rock fabric

You do not need “the perfect model” on day one. You need a model whose assumptions are broadly consistent with your reservoir type:

• Clastics: models that capture frame stiffness and fluid substitution behavior.
• Carbonates: models that acknowledge pore-type sensitivity.
• Unconventional: models that handle mineral mixing, organic effects, and (where needed) anisotropy/stress sensitivity.

Step 3 — Calibrate the RPM to your wells

Calibration is not a luxury—it’s the difference between rock physics as science and rock physics as speculation.

A calibrated model should:

• pass through (or near) the observed log cloud for the relevant facies,
• reproduce realistic behavior under fluid substitution,
• produce sensible porosity trends.

Step 4 — Build an RPT that represents the reservoir unit

Now generate the template:

• porosity lines/bands,
• fluid substitution curves,
• separate families for different facies (if needed).

When your reservoir includes distinct facies (e.g., clean sand vs shaly sand vs silt), building separate RPTs is usually better than forcing everything into one.

Step 5 — Apply the RPT to inversion and attribute interpretation

Overlay inversion/attribute points (AI, Vp/Vs, etc.) and interpret:

• are anomalies consistent with porosity vs fluid vs lithology?
• do clusters separate in ways consistent with geology?
• are there regions in elastic space that look physically impossible (QC flag)?

This is where rock physics becomes interpretation power.

7) SubsurfaceAI’s Rock Physics module: built for seismic interpreters

Many traditional rock physics tools were designed for specialists. They can be powerful but intimidating: too many inputs, too little workflow guidance, and too much manual work before interpreters can extract value.

SubsurfaceAI (SAI) takes a different approach:

Rock physics should be operational for interpreters—fast to start, easy to calibrate, and reusable across projects.

Below are the interpreter-centered features that matter most in practice.

7.1 Easy-to-start RPM workflows (without hiding the physics)

SAI supports practical RPM workflows where interpreters can begin with standard, well-understood model families and then add complexity only when needed. That prevents two common problems:

• getting stuck in model detail too early, and
• using an oversimplified generic model that doesn’t match the reservoir.

The goal is not to turn interpreters into mathematicians. The goal is to give interpreters a tool that makes physics usable.

7.2 Customize and store your own RPMs (capture team knowledge)

Every company develops local rock physics knowledge:

• “This unit behaves like a stiff-sand frame.”
• “That carbonate has pore types that weaken the rock differently.”
• “This shale interval is quartz-rich; trends are shifted.”

SAI allows users to customize RPMs and store them. This turns rock physics from a one-off exercise into a reusable asset:

• interpreters can apply the same calibrated model consistently,
• teams can standardize across interpreters and projects,
• knowledge is retained even when personnel change.

7.3 Build reservoir-specific RPTs and re-use them across projects

SAI enables users to build Rock Physicist Templates (RPTs) (often called rock physics templates in industry practice) that are tailored to specific reservoirs and then saved and reused. This is crucial because interpretation scales. Once a trusted RPT exists:

• new prospects can be screened quickly,
• inversion results can be interpreted consistently,
• anomalies can be ranked with shared criteria,
• training and onboarding become easier.

7.4 Rock physics becomes a repeatable interpretation system

When RPMs and RPTs are stored and reusable, rock physics becomes a system rather than an occasional study:

• templates become part of the standard interpretation toolkit,
• cross-disciplinary communication improves,
• uncertainty can be discussed objectively (“where are we on the template?”).

This is how rock physics stops being exclusive and becomes operational.

8) What’s coming in SAI 2026.1: more RPMs and default RPTs for major plays

SubsurfaceAI’s 2026.1 release expands both model coverage and template readiness—especially in reservoir types where rock physics is often most challenging.

8.1 More RPM options for unconventional and carbonate reservoirs

Unconventional and carbonate reservoirs often violate the intuitive rules interpreters learn in clastics:

• Unconventional reservoirs: elastic behavior depends heavily on mineral composition, organic matter, fabric anisotropy, microcracks, and stress sensitivity. Porosity alone doesn’t explain elastic variation.
• Carbonate reservoirs: pore type dominates. Two rocks can share the same porosity but have very different stiffness depending on whether porosity is intergranular, moldic, vuggy, or crack-like.

SAI 2026.1 provides more RPM options for unconventional and carbonate reservoirs, making it easier to begin with assumptions closer to reality—without having to build everything from scratch.

8.2 Default RPTs for major active basins and plays

In addition to providing tools for building your own reservoir-specific RPTs, SAI 2026.1 will include default RPTs for several major, currently active reservoirs—so interpreters can start faster and then calibrate to local conditions.

Examples include:

• Montney, Western Canada Sedimentary Basin (WCSB)
• Permian Basin, U.S.
• Eagle Ford Shale, southern Texas

Default templates are not meant to replace local calibration. They are designed to:

• accelerate early-stage interpretation,
• provide a consistent starting point for teams,
• act as a benchmark for quick QC and screening.

In most real projects, a practical workflow is:

1. start with a default RPT,
2. compare to local wells,
3. adjust ranges and assumptions,
4. store the calibrated version for future reuse.

That workflow is what makes rock physics scalable for interpreters.

9) The core idea: RPM is for understanding; RPT is for doing interpretation

A helpful way to keep roles clear:

• Use RPM to test hypotheses and understand cause-and-effect
  (“What should gas do to AI and Vp/Vs?”)
• Use RPT to interpret and communicate at scale
  (“These inverted points fall into the likely gas-sand region of elastic space.”)

If rock physics is the bridge between rocks and seismic, then:

• RPM is how you build the bridge correctly,
• RPT is how you drive across it every day.

10) Practical advice: avoiding common pitfalls (without becoming a specialist)

Separate facies before templating

If you mix clean sand, shaly sand, silt, and carbonate streaks into one RPT, you often get a confusing blob instead of meaningful trends. Build multiple templates if needed—one per facies family or stratigraphic unit.

Don’t overfit

A perfect match at one well is less valuable than a reasonable match across multiple wells. Prefer stable, interpretable models.

Treat default templates as hypotheses

Defaults are starting points. Local mineralogy, stress regime, burial history, and fluid system can shift trends.

Use RPTs for QC, not just fluid prediction

Even before interpreting fluids, RPTs help validate inversion results:

• Are points physically plausible?
• Do trends behave logically with stratigraphy?
• Are there clusters suggesting inversion artifacts?

Make rock physics collaborative

Rock physics is most valuable when it becomes a shared interpretation language. Stored RPMs and reusable RPTs help teams stay aligned and consistent.

Closing: putting rock physics where it belongs

Seismic interpretation is evolving. We rely more and more on elastic inversion, multi-attribute analysis, and play-scale analytics. In that world, interpreters who can connect elastic anomalies to plausible rock and fluid scenarios gain a significant advantage.

Rock physics does not need to remain a specialist-only discipline. It becomes practical when:

• RPMs are easy to use and calibrate,
• templates are easy to build and interpret,
• models and templates can be saved and reused,
• and workflows are designed around interpreter needs.

That philosophy is built into SubsurfaceAI’s Rock Physics module: interpreters can customize and store their own RPMs, build reservoir-specific RPTs, and reuse them across projects—turning rock physics into a repeatable interpretation capability rather than a one-off study.

And with SAI 2026.1 expanding RPM coverage for unconventional and carbonate reservoirs and adding default RPTs for key plays such as the Montney, Permian Basin, and Eagle Ford Shale, even more interpreters can start quickly, calibrate locally, and apply rock physics consistently.

Because in the end, the best place for rock physics is not only in specialist reports—it’s in the hands of the people making interpretation decisions every day:

The real power of rock physics modeling lies in the hands of ordinary seismic interpreters.