TODO.md

icon_globe

• Parallelize downloading and extracting of data by writing a function which is then called in the background
• Try to see whether it's faster to call open_mfdataset and load variables or instead read the datasets with open_dataset and then merge them together afterwards.
• Check if psyplot is faster in doing the plot instead than plain matplotlib with tricontourf.
• Move processing of data in the projection loop after the data has being subsetted