Example python scripts to detect flash drought events from gridded soil moisture data
under /prep
- soilm_weighted.sh: to create soil moisture data at a specific depth (cdo)
- pentad.sh: to convert daily soil moisture data to pentad (cdo)
- compute_qval.py: to compute n-quantile of soil moisture over a given period (e.g. 20th or 40th percentile to define FD onset and termination)
- prep_quan.py: to compute soil moisture quantiles at every time steps
under /main
- main_drought.py: to detect drought events at each pixel (e.g. soil moisture decreases from 40th to 20th percentiles)
- fd_detect.py: to select only FD (e.g. soil moisture 40th->20th percentile within n-pentads)
The scripts are used in
- Flash drought drives rapid vegetation stress in arid regions in Europe, ERL, O and Park (2023) https://doi.org/10.1088/1748-9326/acae3a
- Global ecosystem responses to flash droughts are modulated by background climate and vegetation conditions. Commun Earth Environ, O and Park (2024) https://doi.org/10.1038/s43247-024-01247-4