Skip to content

Flattening

A dimensionless axial-compression measure for an ellipse or spheroid, ordinarily the semiaxis difference divided by the semimajor axis, with explicitly convertible alternative normalizations.

Version
v3 · 2026-09-19 · History
Domain-specific #
1851
Domain group
Formal Sciences
Origin domain
Mathematics
Subdomain
Geometry → Mathematics
Aliases
First flattening, Ellipticity, Oblateness parameter

Core Idea

Flattening quantifies how far an ellipse or spheroid departs from circular or spherical shape through axial compression. For semimajor axis \(a\) and semiminor axis \(b\), with \(a\ge b>0\), the ordinary or first flattening is

\[ f=\frac{a-b}{a}=1-\frac ba. \]

It is dimensionless, equals zero for a circle or sphere, and increases as the minor axis shrinks relative to the major axis. Geodetic reference ellipsoids are often specified by \(a\) and reciprocal flattening \(1/f\).[1]

Alternative conventions include second flattening \(f'=(a-b)/b\) and third flattening \(n=(a-b)/(a+b)\). Names vary, so the formula or symbol must accompany any value.[2]

The recognition invariant is ordered semiaxes + normalized axial difference + declared denominator convention + dimensionless departure from equality.

How would you explain it like I'm…

How Squished Is It?

Think of a round ball, and then imagine squishing it a little from the top so it gets shorter. Flattening is a number that says how squished it is. A perfect ball gets zero, and the more you squish it, the bigger the number gets.

How Squashed Is It?

An oval or a squished ball has a long half-width, called a, and a short half-width, called b. Flattening measures how far the shape is from perfectly round using the formula f = (a − b) / a. A circle or a perfect sphere has flattening 0, because a and b are equal. The shorter b gets compared to a, the bigger the flattening. It has no units, so it works the same whether you measure in centimeters or kilometers.

Ellipse Flattening Ratio

Flattening measures how far an ellipse or spheroid departs from a circle or sphere because it's compressed along one axis. With semimajor axis a and semiminor axis b (a ≥ b > 0), the usual (first) flattening is f = (a − b)/a = 1 − b/a. It's dimensionless, equals 0 for a circle or sphere, and increases as b shrinks relative to a. Map makers often describe reference ellipsoids for the Earth by a and the reciprocal flattening 1/f. There are other versions with different denominators—second flattening (a − b)/b and third flattening (a − b)/(a + b)—so you should always state which formula you're using.

 

Flattening is a dimensionless measure of how much an ellipse or spheroid is compressed along one axis relative to a circle or sphere. For semiaxes a ≥ b > 0, the first (ordinary) flattening is f = (a − b)/a = 1 − b/a, which is zero exactly when a = b and increases as the minor axis shrinks relative to the major one. In geodesy, reference ellipsoids are commonly specified by the semimajor axis a together with the reciprocal flattening 1/f. Alternative conventions normalize the same axial difference by other quantities: second flattening f′ = (a − b)/b and third flattening n = (a − b)/(a + b). Since naming varies across sources, any reported value should come with its formula or symbol. The essential ingredients are ordered semiaxes, a normalized axial difference, a declared denominator, and a dimensionless measure of departure from equality.

Structural Signature

  • An ellipse or spheroid of revolution.
  • Semimajor axis \(a\) and semiminor axis \(b\).
  • An ordering convention \(a\ge b>0\).
  • Axial difference \(a-b\).
  • Normalization by \(a\), \(b\), or \(a+b\).
  • A dimensionless shape parameter.
  • Zero at the circular or spherical limit.
  • Monotone increase under greater oblate compression with fixed \(a\).
  • Algebraic conversion to axis ratio and eccentricity.
  • Reciprocal flattening convention in geodesy.
  • A declared sign convention if prolate cases are permitted.

What It Is Not

Flattening is not a length, force, or literal deformation history; geometrically identical ellipsoids have the same value at every scale. It is not eccentricity, although first flattening satisfies \(e^2=2f-f^2\) for an ellipse. It is not automatically the physical dynamical ellipticity used in rotational theory, which can involve moments of inertia rather than geometric semiaxes.

The term “flattening” in data processing, neural networks, or ontology maintenance is unrelated unless it denotes this axial shape parameter.

Scope of Application

Flattening parameterizes terrestrial and planetary reference ellipsoids, map-projection formulae, geodesics, orbit and figure calculations, and comparisons of rotating bodies. Because Earth's flattening is small, series in \(f\) or third flattening \(n\) are efficient in high-accuracy geodesy.[3]

The simple two-axis definition applies directly to spheroids. A fully triaxial ellipsoid needs more than one axial ratio or a clearly selected pair of axes.

Clarity

Report the formula, axes, units for axes, and whether the value is \(f\), \(1/f\), \(f'\), or \(n\). State whether the body is oblate or prolate and whether dimensions are fitted geometric parameters or physical equilibrium predictions. Do not call two values inconsistent until their conventions have been converted.

Manages Complexity

One scalar captures the leading departure from spherical symmetry and allows a large family of geodetic formulae to be organized as perturbation series. Conversion identities translate among standards without reconstructing the entire ellipse. Reciprocal flattening also makes small terrestrial departures easier to tabulate.

Abstract Reasoning

  1. Identify the relevant figure and its principal semiaxes.
  2. Order or label the axes consistently.
  3. Choose the flattening convention and write it explicitly.
  4. Check dimensional cancellation and the spherical limit.
  5. Convert other reported parameters before comparison.
  6. Propagate uncertainty from axis estimates, including covariance if material.
  7. Test whether a spheroidal reduction is adequate for a triaxial body.
  8. Use expansion formulae only within their stated error range.

Knowledge Transfer

The portable pattern is normalize a directional deficit by a reference extent to obtain a scale-free departure from isotropy. It transfers to aspect ratios, anisotropy indices, strain measures, and normalized geometric residuals. The proposed immediate parent is Measurement.

Examples

Sphere. If \(a=b\), then \(f=0\).

Reference ellipsoid. The Geodetic Reference System 1980 specifies a semimajor axis and reciprocal flattening, from which the polar semiaxis and eccentricity follow.[4]

Convention mismatch. For \(a=10\) and \(b=9\), first flattening is \(0.1\), second flattening is \(1/9\), and third flattening is \(1/19\); all describe the same ellipse.

Structural Tensions

  • Simple scalar summary versus full three-dimensional shape.
  • Direct flattening versus reciprocal reporting.
  • Geometric figure versus dynamical mass distribution.
  • Oblate convention versus prolate generalization.
  • Small-parameter efficiency versus approximation error.
  • Standard symbols versus inconsistent variant names.

Structural–Framed Character

Normalization, difference, ratio, scale invariance, and limiting equality are structural. Ellipses, spheroids, semiaxes, eccentricity, and reference ellipsoids provide the constitutive geometric and geodetic frame.

Structural Core vs. Domain Accent

The portable core is a normalized directional deficit. The domain accent is the exact semiaxis formula and its convertible first, second, third, and reciprocal conventions for ellipses and spheroids.

This entry is a kind of Measurement.

Measurement is the broader abstraction this entry instantiates. Ratio, Normalization, Shape, Scale Invariance, Approximation, Coordinate System, and Uncertainty are related.

Relationships to Other Abstractions

Local relationship map for FlatteningParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.FlatteningDOMAINPrime abstraction: Measurement — is a kind ofMeasurementPRIME

Current abstraction Flattening Domain-specific

Parents (1) — outgoing typed relations

  • Flattening is a kind of Measurement Prime

    Measurement is the broader abstraction this entry instantiates.

Hierarchy path (1) — routes to 1 parentless root

Neighborhood in Abstraction Space

Flattening sits in a sparse region of the domain-specific corpus (92nd percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (2769 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-25

Not to Be Confused With

  • Eccentricity.
  • Axis ratio or compression factor.
  • Triaxiality parameter.
  • Dynamical ellipticity.
  • Sphericity or roundness.
  • Equatorial bulge as a length.
  • Data-structure flattening.
  • Native-Category Flattening.

References

[1] John P. Snyder, Map Projections—A Working Manual, U.S. Geological Survey Professional Paper 1395 (1987), doi:10.3133/pp1395. registry ↩

[2] Charles F. F. Karney, “On Auxiliary Latitudes,” Survey Review 56, no. 395 (2024): 165–180, doi:10.1080/00396265.2023.2217604. registry ↩

[3] Friedrich W. Bessel, “The Calculation of Longitude and Latitude from Geodesic Measurements,” English trans. C. F. F. Karney and R. E. Deakin, Astronomische Nachrichten 331, no. 8 (2010): 852–861, doi:10.1002/asna.201011352. registry ↩

[4] Helmut Moritz, “Geodetic Reference System 1980,” Journal of Geodesy 74 (2000): 128–133, doi:10.1007/s001900050278. registry ↩