Model / geo · All operations

Measure distance between places

A variant of Work with geography

Compute the distance between two points or geometries, planar, great-circle or by road.

Status

In arcform

Not built yet

arcform does not do this yet. Its place on the roadmap

What other tools call it

Meridian's classification; an independent blind rating agreed on 95.7% of 324 rows (variant, version 3).

Alteryx Designer (1)
  • Distanceread 2026-09-27

    Compute point-to-point, point-to-edge or drive distance between two spatial objects

Bruin (1)
dbt-utils (1)
  • haversine_distanceread 2026-09-27

    Compute great-circle distance between two lat/lon points in miles or kilometres

Duckle (1)
  • Spatial Distanceread 2026-09-25

    distance from each row to a target geometry, planar or spheroid by CRS

Ibis (2)
  • GeoDistanceread 2026-09-27

    Minimum distance between two geometries

  • GeoMaxDistanceread 2026-09-27

    Largest 2D distance between two geometries

SQLMesh (1)
  • @HAVERSINE_DISTANCEread 2026-09-27

    Compute haversine distance between two lat/lon points in miles or kilometres

Background

Geographical distance, or geodetic distance, is the distance measured along the surface of the Earth, or the shortest arc length.[1] The haversine formula determines the great-circle distance between two points on a sphere given their longitudes and latitudes.[2] Vincenty's formulae calculate the distance between two points on a spheroid, and are more accurate than methods that assume a spherical Earth.[3]

When to use it

Because the Earth is nearly spherical, great-circle distance formulas applied to longitude and geodetic latitude are accurate to within about 0.5%.[4] In DuckDB's spatial extension, ST_Distance returns the planar distance between two geometries, while ST_Distance_Sphere returns the haversine distance in metres and supports only points.[5] In PostGIS, a geometry-type distance between longitude and latitude coordinates comes out in nominal degrees that do not correspond to any true angular difference between the points.[6] Never take a planar distance over raw longitude and latitude: use a sphere or ellipsoid function, or project the points into metres first. (Meridian's view)

The method

The haversine formula is a special case of the law of haversines, a formula in spherical trigonometry that relates the sides and angles of spherical triangles.[2] The first table of haversines in English was published by James Andrew in 1805, and the term haversine, a contraction of half-versed sine, was coined in 1835 by James Inman.[2]

Vincenty's formulae, developed by Thaddeus Vincenty in 1975, have been widely used in geodesy because they are accurate to within 0.5 mm on the Earth ellipsoid.[3] In DuckDB's spatial extension, ST_Distance_Spheroid uses the GeographicLib library to solve the inverse geodesic problem on an ellipsoidal model of the earth.[5]

Dijkstra's algorithm finds the shortest paths between nodes in a weighted graph, which may represent a road network, and was conceived by Edsger W. Dijkstra in 1956.[7]

Sources

  1. [1]Geographical distance - Wikipediasecondary source
  2. [2]Haversine formula - Wikipediasecondary source
  3. [3]Vincenty's formulae - Wikipediasecondary source
  4. [4]Great-circle distance - Wikipediasecondary source
  5. [5]Spatial Functions – DuckDBdocumentation
  6. [6]Chapter 4. Data Managementdocumentation
  7. [7]Dijkstra's algorithm - Wikipediasecondary source