I have coordinates in Adindan UTM30N (Clarke 1880) that I want to re-project into Burkina Faso Transverse Mercator (BFTM). I've created custom projections in QGIS and MapInfo for BFTM using some parameters I've found, but the BFTM coordinates I'm getting when transforming with my custom projection don't match some example coordinates that I have, so I'm wondering if my custom projections are incorrect. Does anyone have any experience of the BFTM projection or could someone please verify the custom projection definitions I've created, given the BFTM parameters below?

Here's the PROJ4 string I've used to create the projection in QGIS:

+proj=tmerc +lat_0=0 +lon_0=-1.5 +x_0=600000 +y_0=0 +ellps=GRS80 +units=m +no_defs +k_0=0.9996

Here's the MapInfo projection string I've created:

"BFTM_TM_GRS80", 8, 33, 7, -1.5, 0, 0.9996, 600000, 0

And here are the BFTM parameters I have:

"The characteristics of the planimetric coordinate system of Burkina Faso are: The ITRF 2008 reference having as ellipsoid the IAG GRS80 (Half Major Axis a = 6 378 137.00 and Flattening 1/f = 298.257222101); BFTM map projection (transverse cylindrical Mercator projection); The Central Meridian 1° 30' west; The coordinates of the center of the projection (X0 = 600,000.00 m, Y0 = 0.00 m); The scale factor: k0 = 0.9996."

These are 2 example coordinates I have:


These are the BFTM coordinates I get for the same 2 points when using my custom projections:


1 Answer 1


I suspect you're missing the transformation from Adindan to WGS 84. The Burkina Faso geographic coordinate reference system (GeoCRS) is based on ITRF2008, so relatively close to WGS 84. The EPSG registry lists these parameters for Adindan to WGS 84 in Burkina Faso:

X translation = -118.0
Y translation = -14.0
Z translation = 218.0

This transformation isn't very accurate, but hopefully will get you closer.

Try adding +towgs84=-118.0,-14.0,218.0 to your source data's proj definition.

Disclosure: I work for Esri and I'm a member of the subcommittee that maintains the EPSG geodetic registry.

  • Thanks for the help - that does get me a little closer, but as you say, still not very accurate, a few hundred metres away from the example XY point I have. Sorry if it's a stupid question but it's not enough to have the GRS80 ellipsoid parameter in the projection definition? I also need to have a parameter to convert from Adindan to ITRF2008 (or WGS84 as a substitute)?
    – rsf69
    Dec 18, 2021 at 19:52
  • Yes, Adindan is a classical datum and would have been defined using ground control points and its ellipsoid surface would be more aligned to the ground. There wasn't enough info to center the ellipsoid at the Earth's center. New datums/GeoCRS that use ITRFxx (and WGS84 realizations) are connected to GNSS data and use an Earth's center of mass origin. So that offset (among other differences) needs to be taken into account.
    – mkennedy
    Dec 19, 2021 at 20:13
  • Great, thanks for info, appreciated.
    – rsf69
    Dec 21, 2021 at 14:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.