Dear all,
I am working on a paper in which SPI trends are evaluated with the CHIRPS precipitation dataset.
https://www.nature.com/articles/sdata201566
This one is based by integrating remote sensing and ground data.
Has it any sense to validate it with a rainfall station?
How to check if the station is among the ones used for the dataset generation?
Dear all,
I found the answer by writing directly to Chris Funk, the developer of the database.
Evaluating with stations is okay, as long as you assume that a good product won't have a perfect relationship, because of the whole point-versus-area estimation issue.
It is better to validate the database by not using the stations used for its generation.
You can find lists of the stations in the folder below. These files show the 'standard' inputs from the GHCN, GTS, GSOD etc. sources and 'extra' stations provided by collaborators.
ftp://ftp.chg.ucsb.edu/pub/org/chg/products/CHIRPS-2.0/diagnostics/list_of_stations_used/monthly/
- Badges
- Science topic
- Similar topics
- Geoscience
- Remote Sensing