jcla490 / landfire-python Goto Github PK

Thank you for this API! It has worked great to grab data layers for some modelling simulation efforts a student of mine is attempting. I have a question you might be able to answer, but it is not a bug or feature request for your code. I apologize.

We are using a command-line version of FARSITE (eg forked from the 2016 "linux" code release). The goal is batch runs for creation of summary statistics over parameter sets. The issue is that this version of the fire growth model is hard-coded to only use old style LCP files not GeoTIFF. I had thought conversion would be trivial with gdal, but I was mistaken.

With command line gdal_translate -ot INT16 -of LCP test.tif test.lcp I get memory-related crashes of the gdal libraries under multiple installations of the gdal libraries (eg ubuntu-gis-stable or unstable or distro version). I've not filed any gdal bug report because I'm not sure that I'm not doing something stupid.

result gdal_translate call:

Input file size is 102, 102
double free or corruption (out)
Aborted (core dumped)

The problem is the similar using gdal through R terra package or python gdal interface (either crashes or errors that may indicate exception handling in the API). I've tried passing all the gdal options listed at https://gdal.org/drivers/raster/lcp.html to terra::writeRaster in R. Perhaps you know offhand how this is best done?