GAMMA Questions

# convert GeoTIFF to dem and dem_par pair

dem_import USGS_13_n53w170_20210429.tif tmp1.dem USGS_13_n53w170.dem_par 0 1

# replace no-data values by very small values

replace_values tmp1.dem 0.0 0.00001 tmp2.dem 10812 0 2 1

# calculate height values relative to ellipsoid using EGM2008 (as an approximation of the North American Vertical Datum of 1988 (NAVD 88))

dem_import tmp2.dem USGS_13_n53w170.dem USGS_13_n53w170.dem_par 4 0 $DIFF_HOME/scripts/egm2008-5.dem $DIFF_HOME/scripts/egm2008-5.dem_par

# visualize

disdem_par USGS_13_n53w170.dem USGS_13_n53w170.dem_par

# Notes:

# - I couldn't find 3dep_n53w170.tif, but used USGS_13_n53w170_20210429.tif instead. The processing sequence should be the same,

# except possibly for the value to replace using "replace_values" and the dataset width.

# - Here, according to the GeoTIFF, the NAD83 reference coincides with WGS84, except for the slightly different ellipsoids used.

# - We use the EGM2008 geoid model to get heights values relatiave to the ellipsoid.

# This may result in probably negligible height inaccuracies (few centimeters). In case the height

# has to be extremely precise, a raster version of NAVD 88 should be used instead of EGM2008.

# - Note that recently, we've been using a lot the 30m Copernicus DEM: it's freely available, covers the whole world, is reliable,

# and its resolution is good enough for most satellite-based SAR data.