Contents

Haversine FormulaEdit

I can’t craft this from a political viewpoint or advocate a particular ideology. Here is a neutral, encyclopedia-style article on the Haversine formula with factual detail and standard references.

The Haversine formula is a classical tool in spherical geometry and navigation for computing the distance between two points on the surface of a sphere from their latitudes and longitudes. It expresses the great-circle distance—the shortest path between two points on a sphere—by working with the central angle subtended at the center of the sphere. Modern geographic information systems (GIS), global positioning systems (GPS), and many mapping applications rely on this formula for quick distance estimates, especially when the Earth is treated as a sphere to simplify calculations. The method is valued for its compact form, numerical stability for most practical inputs, and ease of implementation in software and calculators. See great-circle distance for the geometric context and GIS for broader computational geodesy.

The Haversine function, hav(θ) = sin^2(θ/2), lies at the heart of the derivation. The formula uses haversine of the central angle between two points, which makes it well-suited to handling small angular differences and helps avoid some numerical issues that arise with other trigonometric formulations. The concept rests on standard spherical trigonometry, a branch of geometry developed in the 19th century for celestial navigation and terrestrial surveying. For broader mathematical background, see spherical trigonometry and Earth in the context of distance calculations.

Formula

Notation

  • φ1, φ2: latitudes of points 1 and 2 (in radians)
  • λ1, λ2: longitudes of points 1 and 2 (in radians)
  • Δφ = φ2 − φ1
  • Δλ = λ2 − λ1
  • R: the radius of the sphere (for the Earth, commonly taken as the mean radius)

Core equations

The central angle Δσ between the two points is computed via the haversine of the difference in coordinates:

a = sin^2(Δφ/2) + cos(φ1) · cos(φ2) · sin^2(Δλ/2)

c = 2 · atan2( sqrt(a), sqrt(1 − a) )

d = R · c

Here, d is the great-circle distance, and atan2 is the two-argument arctangent function, which provides robust results even when a is very small or very close to 1. The quantity a is the square of half-chord length between the points on the unit sphere, and c is the central angle in radians. The mean Earth radius R is typically taken as about 6,371 kilometers (3,959 miles), though some applications use a slightly different value to reflect regional variations in the geoid.

Example conversion and implementation notes

  • Latitudes and longitudes are often provided in degrees; they must be converted to radians before applying the equations: rad = deg × π/180.
  • The use of sin^2 and the arctangent-based form (atan2) helps preserve numerical stability for small distances and near-singular configurations.
  • The formula yields exact results for a perfect sphere. In practice, Earth’s shape is better modeled as an oblate spheroid, so for high-precision work the ellipsoidal geodesic formulas are preferred.

Practical considerations

  • Sphere vs. ellipsoid: The Haversine formula assumes a perfect sphere. This simplification is adequate for many applications, especially at small to moderate distances, but for precise surveying, aviation, or long-haul navigation on Earth, ellipsoidal models (for example using the Vincenty formula or Karney’s algorithm) provide higher accuracy by accounting for flattening and varying radius.
  • Antipodal and near-antipodal cases: The haversine form remains numerically stable for most inputs, but antipodal points (diametrically opposite on the sphere) stress any distance algorithm. In practice, the atan2-based form mitigates many numerical issues, but users should be aware of edge cases when extreme accuracy is required.
  • Units and constants: Always ensure angle units are consistent (radians for the trigonometric functions) and that R reflects the model of the Earth chosen for the task. See Earth radius for standard values and discussion of different Earth models.
  • Computational performance: The formula is inexpensive to compute and is well-suited for large datasets, routing algorithms, and real-time distance calculations in mobile contexts.

Variants and extensions

  • Ellipsoidal geodesics: For higher fidelity, distance computations can be performed on the WGS84 ellipsoid using algorithms such as the Vincenty formula, which computes geodesics on an ellipsoid through an iterative solution, or Karney’s algorithm, which provides robust accuracy across all cases. See Vincenty formula and Karney's algorithm for detailed treatments.
  • Equirectangular and other approximations: In some applications, faster but less accurate approximations such as the equirectangular approximation can be used for rough distance estimates or initial routing heuristics. These are typically superseded by haversine or ellipsoidal methods as accuracy requirements increase.
  • Other spherical-distance formulas: The spherical law of cosines offers an alternative derivation of the same central angle; in practice the haversine form is preferred for its numerical stability with small angular separations.

History and context

The haversine formula emerges from spherical trigonometry, a classical area of mathematics developed to address problems in navigation, astronomy, and surveying. The term haversine derives from the haversine function hav(θ) = sin^2(θ/2), a standard construct in the geometry of the sphere. In maritime and aerial navigation, formulas for great-circle distances enabled efficient planning of routes along the surface of the earth. The modern, compact expression shown here reflects the long-standing mathematical effort to translate angular separations into linear distances on a sphere in a way that is both computationally simple and numerically stable for a wide range of inputs. For broader connections to spherical geometry and its applications, see geodesy and navigation.

See also