snow_ims

Shows how to download and process IMS Snow Cover data with Python.

The main script is download_plot_data.run, which calls the Python script plot_data.py where the data is read and some plots produced on different projections. These are defined for convenience in map_projection_ims.py

The main utilities to read the data are contained in ims_reader.py. Please note the following.

• Coordinates for 4 km resolution data (imslon_4km.bin and imslat_4km.bin) are already downloaded and placed in the same folder to avoid having to read from the (slow) server every time
• The 4 km daily data is read directly from the server using pandas as it is not too big: we can avoid having to download and process it beforehand.
• Coordinates for the 1 km resolution data are already downloaded and saved as pickle for the same reason. Furthermore they are already subsetted on the Eurasian continent (see image below) as the original files are too big to be saved (8 GB in total!)
• The 1 km daily data is downloaded and processed BEFORE launching the Python script in download_plot_data.run so as to avoid having to load it into Python, thus saving memory and CPU. We do in a quite clever way using only sed and cut: sed -ne '9031,17030p' ims_file.asc | cut -c11001-19000 > ims_file_subsetted_1km.asc. If you find an easier way...do let me know!