opticalflowleo 0.8.1

Optical Flow derived winds from overlapping LEO granules

windflow_leo

A package for deriving atmospheric winds from overlapping granules of satellites in a LEO (low earth orbit) train formation, using dense feature tracking of CrIS retrievals of specific humidity (JPSS) or AVHRR imagery (MetOp).

This package runs inference on a pre-trained model based on WindFlow: Dense feature tracking of atmospheric winds with deep optical flow:

Vandal, T., Duffy, K., McCarty, W., Sewnath, A., & Nemani, R. (2022). Dense feature tracking of atmospheric winds with deep optical flow, Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.

Note: this project uses git submodules. Use this command to clone:

git clone --recursive https://gitlab.ssec.wisc.edu/rink/windflow_leo.git

To install:
conda env create -f environment.yml
or
conda env create -f environment.yml -n new_environment_name
conda activate new_environment_name
pip install opticalflowleo