Choose a SNAP version.
Install with conda.

Quickstart

For current SNAP 13 workflows:

mamba create -n snap13 -c sarforge -c conda-forge esa-snap-s1tbx-gpt=13.0.0
conda activate snap13
gpt -h

conda works too, but mamba usually resolves these large environments faster.

Select a SNAP version

Use a separate environment per SNAP version. That keeps Java, launchers, and SNAP module state predictable.

mamba create -n snap9  -c sarforge -c conda-forge esa-snap-s1tbx-gpt=9.0.0
mamba create -n snap10 -c sarforge -c conda-forge esa-snap-s1tbx-gpt=10.0.0
mamba create -n snap11 -c sarforge -c conda-forge esa-snap-s1tbx-gpt=11.0.0
mamba create -n snap12 -c sarforge -c conda-forge esa-snap-s1tbx-gpt=12.0.0
mamba create -n snap13 -c sarforge -c conda-forge esa-snap-s1tbx-gpt=13.0.0

Check the install

conda activate snap13
which gpt
which snap
echo "$CONDA_PREFIX"
gpt -h
gpt Calibration -h

Expected launcher paths:

$CONDA_PREFIX/opt/snap/bin/gpt
$CONDA_PREFIX/opt/snap/bin/snap

pyroSAR

pyroSAR detects SNAP by finding the snap launcher on PATH. This package preserves the standard SNAP layout, so no manual pyroSAR config should be needed.

mamba install -c conda-forge pyrosar

from pyroSAR.examine import ExamineSnap

snap = ExamineSnap()
print(snap.gpt)

For InSAR phase unwrapping, install snaphu separately:

mamba install -c conda-forge snaphu

Troubleshooting

If gpt starts but crashes during SNAP initialization, check that you are using the fixed build for the version:

conda list | grep esa-snap-s1tbx-gpt

Packaging-only fixes are delivered as higher conda build numbers. For example, 13.0.0-2 includes Linux/headless startup defaults.

If the solver keeps an old local cache, create a new environment or ask for the exact build:

mamba create -n snap13-fixed -c sarforge -c conda-forge "esa-snap-s1tbx-gpt=13.0.0=*"