Hi Alex,
I would use Python over shell scripts. Arguably for some tasks you end up writing more code but I personally think the syntax is much nicer. You can call command line tools using the subprocess module (
https://docs.python.org/3/library/subprocess.html)
For example:
import subprocess
subprocess.call('find -name *MTL.txt | parallel arcsi.py -s ls8 -f KEA --stats -p RAD SREFSTDMDL --aeropro Maritime --atmospro MidlatitudeSummer --aot ' + aot + '--outbasename {.}_$aot.kea -o ../03_ARCSI_aot_manual -i {} > verbose2.txt',shell=True)
There is also a tool with ARCSI to generate a shell script with the ARCSI command for all files in a directory. You could write a Python script to call this for different AOT values (either the command line tool or Python function) and then use GNU Parallel to run everything in parallel using:
cat arcsi_commands.sh | parallel -n 4
Note the n parameter to specify a maximum of 4 cores. As ARCSI does a lot of file reading / writing you might find if you try to run to many instances at once how fast you can read and write data to the disk becomes a limiting factor.
In response you your original question, I think you need to use integers in your while loop, so a bit hacky but something like the following should work.
aot=5
while [ $aot -lt 100 ]; do
aotfloat=`echo "scale=2; $aot / 100" | bc`
echo $aotfloat
aot=$(($aot+10))
done
Thanks,
Dan