The revised _srv_srtmcoverage.sh script seems to work and the plot looks correct. The point was update the _srtmtiles.nc grid (in the cache) to instead of just having two values (0 = no SRTM, 1 = SRTM) it will have three: (0 = no SRTM, 1 = SRTM land, 2 = SRTM with ocean) so that we know when we need to include the 15s earth_relief filler. Once I revert 6.3 to do the same as 6.2 regarding determining if we need ocean, then we can test with this new grid. We need the new tile grid to work for 6.2 and earlier and I think it will since we only checked if the grid was zero or not. 6.3 can then add more checks to distinguish 1 from 2. So there will be
A later PR to this repo that updates the srtm_tiles.nc grid (after testing it).
Two PRs to GMT to first reinstall previous behavior then an update to use the fancier tile grid.
The revised _srv_srtmcoverage.sh script seems to work and the plot looks correct. The point was update the _srtmtiles.nc grid (in the cache) to instead of just having two values (0 = no SRTM, 1 = SRTM) it will have three: (0 = no SRTM, 1 = SRTM land, 2 = SRTM with ocean) so that we know when we need to include the 15s earth_relief filler. Once I revert 6.3 to do the same as 6.2 regarding determining if we need ocean, then we can test with this new grid. We need the new tile grid to work for 6.2 and earlier and I think it will since we only checked if the grid was zero or not. 6.3 can then add more checks to distinguish 1 from 2. So there will be