Choose climate records with AI
Choose a climate data record by variable, dates, footprint and sampling interval before looking for individual files.
Choose climate records with AI when a familiar variable name brings up several satellite archives. A radiation series covering one Meteosat disc and a global precipitation product answer different questions even if their date ranges overlap. The ml-menu skill reads EUMETSAT’s record guide to suggest a collection and the product filter you will need later.
Suppose you are planning a long-run albedo study. Within one collection you may find daily, monthly and five-day products. The guide pairs the appropriate prefix with the variable name where documented. You can make a shortlist without downloading a sample or signing into the Data Store.
What it does
- Match a precipitation record to the geographical area and period your analysis needs.
- Distinguish monthly means from other products sharing the same collection.
- Spot a change of satellite or sampling frequency that affects a proposed comparison.
- Collect the documented variable name ahead of writing an analysis notebook.
- Identify native-format records before trying to read them as netCDF files.
Install and use
- Start with the built-in guide, which needs no login. Specify the variable, years, area and desired time interval.
- Read the suggested record and its caveats. Ask the agent to flag any gap or undocumented variable.
- If you want files, use the EUMDAC client and Data Store credentials. Confirm the number of products, size and local destination first.
FAQ
Source
Made by EUMETSAT. Original skill: ml-menu, published on GitHub under the CC-BY-4.0 licence. The files are unchanged.
View on GitHub