DEODE fdb examples

In the following we give a few examples on howto work with fdb from python.

Important information about experiment identifiers

In some of the examples we modify the experiment identifier expver as a test. For your own future archiving NEVER pick random expver but please do as below to get a new expver. The general convention is to use numerical identifiers for operations, e.g. 0001, and alpha numerical ones for research, i.e. personal runs. The utility getNewId will simply increment to a new id everytime it's called.

module load pifsenv
getNewId -g d1.on-demand-extremes-dt

Install

To install the required environment e.g. locally do, in this case with micromamba, and preferred python version

micromamba create -n deode_fdb_examples python=3.12
micromamba env update -n deode_fdb_examples -f environment.yml

Activate and load the correct EcCodes

micromamba activate deode_fdb_examples
module load ecmwf-toolbox/2025.10.1.0

List example

List available georefs and corresponding streams from FDB for the past 5 days. Note that we used stream = oper/enfo for deterministic/ensemble runs respectively

python3 ./find_georef.py

Read write example

Example to read data from FDB, change expver and georef, and archive again.

python3 ./read_and_write.py

Example to read data from FDB for a certain date, expver and georef

python3 ./fdb_search.py