Creating a GeoColor image from GOES-West/GOES-East L2 data using SatPy

After a significant amount of trial and error, I figured out how to use SatPy to create a GeoColor image from GOES satellites. I tried doing the same with Himawari L1 data, but the process kept running out of memory and crashing. Not sure why.

Installation

pip install satpy[all]
# in case that did not install everything:
pip install satpy[geotiff]
pip install satpy[rayleigh]
pip install satpy[generic_image]

Data download

If using L2 data, it needs to be ABI-L2-MCMIPC for CONUS, or equivalent for other regions. For example, navigate to AWS and download this file.

Generation

Having placed the file in, say, /home/user/abi, the following Python code should generate a GeoColor GeoTIFF image:

from satpy import Scene, find_files_and_readers

files = find_files_and_readers(base_dir="/home/user/abi",reader="abi_l2_nc")
scn = Scene(filenames=files)
scn.available_composite_names()
scn.load(scn.available_dataset_names())
scn.load(["geo_color"])
new_scn = scn.resample(scn["C01"].attrs["area"],resampler="nearest",reduce_data=False)
new_scn.save_dataset("geo_color",filename="geocolor.tif")

In theory, to reduce RAM usage, the following can be executed before generating the GeoColor image, but it didn’t seem to help anything in my case:

import dask

dask.config.set(num_workers=2)
dask.config.set({"array.chunk-size": "32MiB"})