# Converting unknown coordinate (EPSG:3395?) to EPSG:3857

I'm attempting to convert a 7-digit coordinate of an unknown coordinate system I receive in this format:

``````{ lat: 4575651.267366762, lng: -9729851.06598594 }
``````

After a few hours of trial and error, I found that a conversion from `EPSG:3395` to my intended format `EPSG:3857` yielded the best results, but are still unfortunately much too far north. The intended location of the coordinate above is a few meters from:

``````37°58'23.1"N 87°24'15.3"W
``````

What is the coordinate system of the given 7-digit coordinate?

I think that coordinate is EPSG:3857.

Let's do some coordinate stuff in R:

``````st_as_text(pt)
## [1] "POINT (-9729851 4575651)"
``````

say it is 3857...

``````ptc = st_sfc(pt,crs=3857)
``````

where does it go in lat-long?

``````st_transform(ptc, "+init=epsg:4326")

##    POINT (-87.40474 37.97328)
``````

Reformat those coords to D:M:S

``````measurements::conv_unit(37.97328 ,from="dec_deg",to="deg_min_sec")
## [1] "37 58 23.8080000000191"

measurements::conv_unit(-87.40474 ,from="dec_deg",to="deg_min_sec")
## [1] "-87 24 17.064000000013"
``````

Compare with your DMS: `37°58'23.1"N 87°24'15.3"W` - close enough?

Projfinder is useful here: http://projfinder.com/ - put the coordinates of the unknown coordinate system in the box, zoom as close as you can to where you think that point should be, and hit the "find" button - it then finds all CRS in the near area, shows them in a list, and how far they are from the centre point of your map.

• Looks like I fooled myself! Thanks so much for your help! – Devon Germano May 16 at 17:57

The coordinates are in EPSG:3857, WGS 84 / Pseudo-Mercator.

EPSG:3395 is WGS 84 / World Mercator.

So what's the difference? EPSG:3857 uses a major auxiliary sphere for the Earth model. Its radius is equal to the semi-major axis of the WGS 84 ellipsoid, 6378137.0 m. That triggers the spherical equations of Mercator. EPSG:3395 uses the WGS 84 ellipsoid and the ellipsoidal Mercator equations.

The resulting differences are in the Y / Northing values only.

• This helps clarify my mistake, thank you! – Devon Germano May 16 at 18:02