Haversine Distance Calculator

Measure the great-circle geodesic distance between coordinates, determine bearings, and compute midpoints.

Quick-Select Landmarks

Geographic Coordinates

Point A

Latitude

Parsed: 40.71280°

Longitude

Parsed: -74.00600°

Point B

Latitude

Parsed: 51.50740°

Longitude

Parsed: -0.12780°

Planetary Preset (Radius)

Geodesic Path Visualization

Distance Outcomes

Kilometers:5570.22 km

Miles:3461.17 mi

Nautical Miles:3007.68 nm

Navigation Coordinates

Compass Bearing:51.2° (NE)

Geographic Midpoint

Lat: 52.36844°, Lon: -41.29031°

DMS Support: Enter raw values like 40° 42' 46" N directly into coordinate fields. The parser handles degree, minute, and second notations automatically.

Understanding Haversine Geodesic Math

The Haversine formula measures spherical distances:d = 2R × arcsin(√[sin²(Δφ/2) + cos(φ1)cos(φ2)sin²(Δλ/2)]).

Bearings represent navigation angles between geographic nodes. Midpoints calculate the midpoint on a great-circle path.

Built and maintained by Meet Shah · Last updated 1 August 2026

What this tool is used for

Getting the great-circle distance between two coordinate pairs.
Checking a distance a mapping API returned.
Calculating a straight-line distance for a service radius.
Comparing two locations' separation without a routing engine.
Building a distance filter for a dataset of coordinates.

Frequently Asked Questions

What does the haversine formula compute?: Great-circle distance — the shortest path over a sphere's surface, which is what an aircraft flies rather than the straight line through the earth. It uses the half-versed-sine of the central angle, then multiplies by the radius.
Why haversine rather than the spherical law of cosines?: Numerical stability at short distances. The law of cosines needs the arc-cosine of a value extremely close to 1 for nearby points, where floating-point precision collapses — errors of kilometres on a metre-scale distance. Haversine stays well-conditioned everywhere.
Which Earth radius should I use?: 6,371 km is the mean, and the default. The Earth is an oblate spheroid — 6,378 km at the equator, 6,357 at the poles — so a spherical model carries up to about 0.5% error. For sub-metre accuracy you need Vincenty's formulae on an ellipsoid instead.
What is the bearing the tool also reports?: The initial compass heading from the first point to the second. On a great circle it changes continuously — an aircraft flying from London to Tokyo starts heading roughly north and ends heading south — so the initial bearing is not the direction you hold for the whole route.
Does this account for terrain or roads?: No. It is the distance a bird flies over a smooth sphere. Road distance is typically 15–40% longer depending on the network, and elevation change adds more, so a haversine figure is a floor rather than an estimate of travel.

Common errors and gotchas

Reading the result as a travel distance, when it is a straight line over a sphere and ignores roads entirely.
Swapping latitude and longitude, which is the most common coordinate error there is.
Passing degrees to a formula expecting radians.
Assuming a spherical Earth is accurate enough, when Vincenty's formula matters over long distances.
Using a decimal sign convention inconsistently, so a southern latitude comes out northern.