[PROJ] Projection for Sandwell et al.'s topex topo and grav files?

Kurt Schwehr schwehr at gmail.com
Tue May 7 14:51:53 PDT 2019


Hi all,

I figured I should ask here too if anyone know what the correct projection
is for the Sandwell .img grids from http://topex.ucsd.edu.  I'm trying to
keep the files in their original projections as I switch them to geotiffs.
I've asked David Sandwell directly too if he knows.

This works, but warps the data first:

gmt img2grd -V -R-180/180/-80/80 topo_18.1.img  -Gtest2.grd -T1 -D
gdal_translate -a_srs EPSG:4326 test2.grd test2.tif
gdalinfo test2.tif  # Results look believable
gdal_translate topo-18-1-epsg4326.tif topo-18-1-epsg4326-deflate.tif -co
COMPRESS=DEFLATE -co PREDICTOR=3

Then imported into QGIS or Earth Engine as a normal user, things line up
pretty well.

    https://code.earthengine.google.com/f43c8b13bafa26fd8c7f83ce9a919f6e


I'd rather do it more like this:

gmt img2grd -V -R-180/180/-80/80 topo_18.1.img
-Gspherical-mercator-proj.grd -T1 -D -M
img2grd: Expects topo_18.1.img to be 21600 by 17280 pixels spanning
0/360.0/-80.738009/80.738009.
img2grd: To fit [averaged] input, your topo_18.1.img is adjusted to
-R-180/180/-80.0023237126/80.0023237126.
img2grd: The output grid size will be 21600 by 16752 pixels.
img2grd: Created 21600 by 16752 Mercatorized grid file.  Min, Max values
are -10914  8550

gives this which doesn't work as is:

gdalinfo spherical-mercator-proj.grd
Driver: netCDF/Network Common Data Format
Files: spherical-mercator-proj.grd
Size is 21600, 16752
Coordinate System is `'
Origin = (0.000000000000000,279.199999999999989)
Pixel Size = (0.016666666666667,-0.016666666666667)
Metadata:
  NC_GLOBAL#Conventions=COARDS, CF-1.5
  NC_GLOBAL#description=Spherical Mercator Projected with -Jm1
-R0/360/-80.0023237126/80.0023237126
  NC_GLOBAL#GMT_version=5.4.3 (r19528) [64-bit]

Should I be using one of 54004 or 41001?  e.g.
https://epsg.io/?q=spherical+mercator

Thanks!
-kurt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20190507/9dea0069/attachment.html>


More information about the PROJ mailing list