Fossil Swimming Models: How to Judge Reliability

Updated 7 November 2026 · Standard formula, shown below

This guide explains how to critically evaluate a published model of prehistoric swimming—whether it’s a plesiosaur, ichthyosaur, mosasaur, or ancient fish. You’ll learn what to check in the model’s assumptions, data, and math, so you can separate a plausible reconstruction from a speculative one. No paleontology degree is needed, but a basic grasp of physics and biology helps.

What You’ll Need

The original paper or detailed description of the model (not just a news summary).
Access to the fossil specimen’s measurements or a clear figure with scale bars.
A calculator or spreadsheet to re-check simple arithmetic.
Basic knowledge of units (meters, seconds, newtons) and simple physics (force, speed, drag).
Patience to read through methods sections, even if they’re dense.

Step-by-Step

Identify the model’s core claim. Is it about maximum speed, cruising speed, efficiency, or maneuverability? Write down exactly what the model is trying to predict. A reliable model will state its output clearly (e.g., “estimated cruising speed of 2.5 m/s”).
Check the fossil basis. Does the model use actual measurements from a specific specimen (e.g., humerus length, tail fin span) or generic values from a related species? If it uses generic values, note that as a limitation. If it uses a single specimen, check whether that specimen is complete or reconstructed—missing bones require assumptions.
Look for explicit assumptions about body shape and density. Swimming models need body mass, surface area, and drag coefficient. Are these derived from the fossil (e.g., via volumetric estimates) or guessed? A reliable model will state the density used (usually near 1000 kg/m³ for aquatic animals) and how body volume was estimated (e.g., from skeletal reconstructions or muscle wrapping).
Examine the propulsive mechanism. How does the model generate thrust? For a tail-propelled animal, it should consider tail beat frequency and amplitude. For limb-driven swimmers (like plesiosaurs), it should consider lift-based or drag-based paddling. The model must explain the mechanics—if it just assumes a motor efficiency, that’s a red flag.
Check the hydrodynamic equations. The model should use standard drag equations (e.g., F_drag = 0.5 * ρ * v² * C_d * A). Verify that units are consistent. If the paper uses a custom coefficient, see if it’s justified. A reliable model will show the equations or at least reference them clearly.
Look for sensitivity analysis. A good model tests how changing key inputs (e.g., drag coefficient, body mass) affects the result. If the paper says “we varied C_d from 0.1 to 0.5 and speed changed by 20%,” that’s a good sign. If it gives a single number with no error bars, treat it as a rough estimate.
Compare with living analogues. Does the model’s predicted speed or efficiency match what we see in modern animals of similar size and shape? For example, a 5-meter ichthyosaur predicted to swim at 40 km/h is plausible; a 10-meter plesiosaur predicted to swim at 100 km/h is not. Use common sense and known biological limits.
Check for peer review and consensus. Has the model been published in a reputable journal? Have other researchers cited or critiqued it? A single paper with no follow-up is less reliable than one that has been tested by others. Look for subsequent studies that either support or contradict the model.
Read the limitations section. Every honest model lists its weaknesses. If the paper doesn’t mention any, be suspicious. Common limitations include missing soft tissue, unknown muscle properties, and simplified fluid dynamics. A reliable model will acknowledge these openly.
Reproduce a simple calculation. Take the paper’s own numbers and plug them into the main equation. If you get a wildly different result, there’s an error somewhere. Even a rough mental check (e.g., “speed = distance/time”) can catch obvious mistakes.

Common Mistakes

Assuming all fossils are complete. Many models rely on partial skeletons, and the missing parts are estimated. If the paper doesn’t state how missing bones were filled in, that’s a problem.
Mixing up drag coefficient units. Drag coefficient is dimensionless, but some papers use a different reference area (e.g., frontal area vs. wetted area). Check which one is used.
Overestimating muscle power. Muscle force is limited by cross-sectional area, not just mass. A model that scales muscle power directly from body mass without considering muscle attachment points is likely wrong.
Ignoring buoyancy. Aquatic animals are nearly neutrally buoyant, but not exactly. If the model ignores buoyancy, it may overestimate the energy needed to stay at depth.
Trusting a single reconstruction. If the body shape is reconstructed differently by another artist, the drag changes. A reliable model should be tested against multiple plausible body shapes.

Tips

Look for models that use computational fluid dynamics (CFD) or particle image velocimetry (PIV) on physical models—these are more rigorous than simple math.
Check whether the model’s speed estimate falls within the range of living marine mammals and reptiles of similar size. For example, a 3-meter dolphin cruises at about 2-3 m/s; a 10-meter whale at about 1-2 m/s. Prehistoric animals likely had similar constraints.
Pay attention to the tail shape. A crescent-shaped tail (like a tuna) is efficient for continuous swimming; a paddle-shaped tail is better for bursts. The model should match the tail type to the behavior.
If the paper mentions “Reynolds number,” that’s a good sign—it shows they considered flow regime (turbulent vs. laminar).
Remember that a model is a hypothesis, not a fact. Even a reliable model gives an estimate with uncertainty, not a definitive answer.

Troubleshooting

If you can’t find the original paper: Try Google Scholar or ResearchGate. If it’s paywalled, look for a preprint or a conference abstract. If you only have a news article, treat the model as unverified.
If the equations are too complex: Focus on the input values and the final result. You can often check whether the result is plausible without understanding every derivative. If you can’t follow the math, that’s a limitation of your background, not necessarily the model—but it’s also a reason to be cautious.
If the model gives a speed that seems too fast: Check if they used a burst speed (short duration) vs. sustained speed. Burst speeds can be much higher but are not sustainable. A reliable model will distinguish between the two.
If the model contradicts another published model: That’s normal. Compare the assumptions of both. Often, the difference comes down to body mass or drag coefficient. The model with more realistic inputs is usually more reliable.
If the paper doesn’t state its assumptions: That’s a red flag. Look for supplementary materials or online data. If none exist, email the authors—they may share details.

Frequently Asked Questions

How do I know if a model is “reliable” vs. “speculative”?

A reliable model is based on measured fossil dimensions, uses standard physics, and includes uncertainty. A speculative model often uses guessed values and gives a precise number without error bars.

Do I need to understand calculus to judge a model?

No. You can check the logic, the inputs, and the plausibility of the output without understanding every equation. If the paper explains its reasoning in words, that’s enough for a basic evaluation.

What if the model is based on a computer simulation I can’t run?

That’s fine. You can still evaluate the inputs and the validation. Did they test the simulation on a living animal first? If not, be cautious.

Are models of extinct animals ever “proven”?

No. Since we can’t observe the animal, we can only build increasingly plausible hypotheses. A model is reliable if it’s consistent with all known evidence and doesn’t violate physical laws.

Why do two different models give different speeds for the same animal?

Because they make different assumptions about body mass, muscle efficiency, or drag. That’s why you should always look at the assumptions, not just the final number.

Can I trust models from popular science articles?

Usually not as much as the original paper. News articles often simplify and omit caveats. Always go back to the source.

Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.