Convert Lat Long to Northing Easting

What northing and easting actually mean

Northing and easting are distances in metres measured from a defined origin, on a flat grid laid over a curved Earth. Easting is how far east of the origin a point sits; northing is how far north. Unlike latitude and longitude, which are angles, these are straightforward linear measurements — which is exactly why surveyors and engineers prefer them, because you can subtract two of them and get a distance.

The catch is that flattening a curved surface onto a grid requires a projection, and a projection needs an origin, and different countries and purposes chose different ones. So a northing and easting pair is meaningless without knowing which grid it belongs to. A pair reading 530000, 180000 is central London on the British National Grid; the same numbers in UTM zone 30 are in the sea off West Africa.

This page treats that ambiguity as the main event rather than a footnote. The converter above requires you to say which system and which datum you mean, and it labels the output with both, because an unlabelled coordinate pair is the single most common cause of the errors described below.

Which grid system are you actually using?

Before converting anything, establish which grid the numbers belong to. The clues are usually in the magnitude and the context.

UTM: northing runs from 0 at the equator to about 9,300,000 near the poles, and easting is always between roughly 160,000 and 834,000 because of the 500,000 m false origin. A six- or seven-digit northing paired with a six-digit easting, with a zone number attached, is UTM.

British National Grid: both values are six figures or fewer and the pair is often written with two letters, as in TQ 301 805. Anything with a two-letter prefix in Britain is BNG, not UTM.

US State Plane: values are often very large, and crucially may be in US survey feet rather than metres. A northing in the millions with a state name attached is usually State Plane.

MGRS: looks like UTM with an extra pair of letters, e.g. 30U XA 1234 5678. It is UTM underneath with a 100 km square identifier.

A local site grid: arbitrary origin chosen by the project, often with round numbers like 1000, 1000 at a site benchmark. These do not convert without the project’s own transformation parameters.

If a dataset arrives without this stated, the honest answer is that it cannot be converted reliably, and guessing is how points end up in the wrong country. Ask for the EPSG code — a single number that specifies system and datum together, such as 27700 for the British National Grid or 32630 for UTM zone 30 north on WGS84.

Datums: the part that quietly ruins conversions

A datum is the model of the Earth’s shape and its position relative to the ground that a coordinate is measured against. Two datums can give the same point noticeably different latitude and longitude values, because they disagree about where the reference surface sits.

The differences are not academic. WGS84 and OSGB36 describe positions in Britain that differ by up to roughly 120 metres. NAD27 and NAD83 can differ by as much as 200 metres in parts of the United States. WGS84 and NAD83 are close today — around one to two metres in the continental US — but they were designed to coincide and have drifted apart with tectonic motion, so the gap grows. ETRS89 is fixed to the Eurasian plate and moves away from WGS84 by roughly two and a half centimetres a year.

What makes this dangerous is that the failure is silent. Convert a WGS84 GPS reading into British National Grid coordinates while telling the software it is OSGB36, and you get a plausible-looking grid reference that is a hundred metres out. Nothing errors. The numbers have the right number of digits and fall inside the right range. This is why the converter above states the datum on the output and why it is the first control on the page rather than the last.

Offsets are approximate and vary by location within each region; treat them as orders of magnitude, never as fixed corrections.

UTM zones and latitude bands

UTM divides the world into 60 zones, each six degrees of longitude wide, numbered eastward from the antimeridian. Zone 1 covers 180°W to 174°W; zone 30 covers 6°W to 0°, which is why most of Britain sits in zone 30. The zone number tells the projection which central meridian to use.

Each zone is also divided into latitude bands eight degrees tall, lettered C at 80°S through X at 84°N. The letters I and O are skipped to avoid confusion with 1 and 0, and X is stretched to twelve degrees so the system reaches Svalbard. Together they give a designator like 30U for London or 18T for New York.

The formula for the zone is simple enough to check by hand: take the longitude, add 180, divide by 6, take the whole number, add 1. The central meridian of any zone is 6 times the zone number minus 183. The converter shows both calculations with your own numbers substituted, so an unexpected zone is immediately verifiable rather than something to trust.

How the conversion works

Converting latitude and longitude to a grid means projecting from a curved reference surface onto a plane. UTM and the British National Grid both use a Transverse Mercator projection, which wraps a cylinder around the Earth touching a chosen meridian and unrolls it. Close to that central meridian distortion is tiny; far away it grows, which is why the world is cut into narrow zones in the first place.

Two constants shift the results into convenient positive numbers. A false easting of 500,000 metres is added so that points west of the central meridian do not go negative — which is why any point exactly on its zone’s central meridian has an easting of exactly 500,000. In the southern hemisphere a false northing of 10,000,000 metres is added for the same reason, so a northing of 5,000,000 means one thing north of the equator and something quite different south of it.

The maths itself is a series expansion and is not something to hand-roll. Any production implementation should use a maintained projection library and reference the transformation by its EPSG code, because the accuracy of a hand implementation degrades at zone edges and high latitudes in ways that are hard to spot in testing.

Precision, digits and grid references

Because northings and eastings are metres, precision is legible directly from the number of digits — unlike degrees, where you have to know the latitude to interpret a decimal place. Dropping digits from the right coarsens the reference in clean powers of ten.

The British grid-reference convention formalises this. TQ alone identifies a 100 km square; TQ 30 80 narrows to 1 km; TQ 301 805 to 100 m; and TQ 30123 80567 to a single metre. Each pair of added digits divides the square by ten in both directions, which is why grid references always have an even number of digits after the letters.

Converting more than one point

Almost every guide to coordinate conversion assumes a single point, and almost every real task involves a list. A set of survey stations, a column of site addresses geocoded overnight, a batch of sample locations from a field trip — the moment there is more than one, a web converter that handles one at a time stops being useful and people move to a spreadsheet or a GIS package.

The bulk tab on this page exists for that gap. Paste one coordinate per line, or paste CSV and map the columns, and every row converts under the same system and datum with the result available as CSV. The system and datum are stated once and applied consistently, which also removes the most common bulk-conversion error: mixing datums partway through a dataset because two sources were combined without checking.

Two practical cautions. First, converting a list does not make the source coordinates trustworthy — if half came from a phone GPS and half from a land registry extract, they are in different datums before you start and converting them together silently averages a hundred-metre disagreement into your data. Second, a round trip is not a test of accuracy; converting to a grid and back returns almost exactly what you put in even when the datum is wrong, because the same wrong assumption is applied in both directions.

Who uses this and why