Process of converting the decads into months about spi HOT 2 CLOSED

The idea of this approach is to get more frequent monitoring for SPI 1 - 72, and updated every dekad. To answer your question why we need to extract dekad data into separate files, see below illustration.

SPI script will read the netcdf input file based on time dimension. It means if we leave the data as is (without extracting) and would like to calculate SPI3, then the script will use data: Jan 2020 D3, Feb 2020 D1, Feb 2020 D2 (see row 13 column DEF on above picture), and this is not correct. I hope its clear now.

Well if you are confused with my approach using dekad data, maybe you could try using monthly data - https://data.chc.ucsb.edu/products/CHIRPS-2.0/global_monthly/netcdf/chirps-v2.0.monthly.nc and skip step 3.5 and 3.7.

from spi.

Thank you for explaining the process so well! This is a much improved approach than the ARSET one.

from spi.