Cassini ToolsKenya cadastral grid converter
Converter Map How it works Accuracy CRS string Open converter
Cassini-Soldner · Clarke 1858 · Arc 1960

Turn Kenya's legacy Cassini coordinates into UTM you can trust.

A rigorous Cassini-Soldner projection, refined with a national network of surveyed toposheet corners, converts cadastral grid coordinates — in metres or feet — to UTM Arc 1960 and back, at metre-level accuracy.

See the method
634ground control points
±3 mmedian accuracy
4Cassini belts · 35–41°E
2-wayCassini ⇄ UTM

Coordinate converter

Paste coordinates, or load a sample.
Copy output
Spatial preview

See your points on the map.

Converted points plot live on an interactive map alongside the Survey of Kenya toposheet grid and the ground-control network. Toggle layers and switch between satellite and street basemaps.

Converted points Control points Toposheet grid
Run a conversion above to plot points. Toggle the toposheet grid and click any sheet for its number and name.
The method

Two layers: a rigorous projection, then real-world control.

Kenya's pre-1950 cadastral surveys used the Cassini-Soldner projection on the Clarke 1858 ellipsoid, measured in British feet, with belts centred on 35°, 37°, 39° and 41°E. Converting to modern UTM Arc 1960 needs more than a formula — the historical grid has local distortions that only surveyed control can capture.

01 · PROJECTION

Rigorous Cassini-Soldner

Ellipsoidal forward and inverse Cassini-Soldner on Clarke 1858, in feet, per belt — then a geodetic transfer into Arc 1960 (Clarke 1880), each belt on its correct UTM zone (35°E→36, 37/39/41°E→37).

Clarke 1858 · a = 20 926 348 ft
02 · CONTROL

National corner network

Hundreds of published toposheet corners — each carrying both Cassini and UTM values — are cleaned, de-duplicated and turned into a residual field that corrects the projection locally.

634 control points · 4 belts
03 · BLEND

Distance-aware correction

Near control, the correction runs at full strength for metre-level results. Far from any control it fades smoothly back to the pure projection, so you never get a wild extrapolation.

Inverse-distance · smooth taper

Validated against the surveyed grid.

±3 mmedian error (leave-one-out on sheet corners)
±37 mRMSE across all belts
<1 mtypical near a control point
<1 cmCassini → UTM → Cassini round-trip

Accuracy is highest where surveyed control is dense. The leave-one-out test removes each corner and predicts it from its neighbours — a deliberately pessimistic measure, since real conversions fall inside the network rather than at its removed edges. Where control is sparse the result gracefully approaches the rigorous projection (~200 m bias vs the legacy grid) instead of failing. Source: national Survey of Kenya toposheet corner coordinates.

For GIS & CAD

A ready-to-use CRS definition.

Drop this PROJ string into QGIS, GDAL or any PROJ-based tool to define a Cassini belt as a custom CRS. Choose your central meridian: