...
| Panel | ||||||
|---|---|---|---|---|---|---|
| ||||||
The different macros available for this exercise are very similar to those in previous exercises.
For this exercise, use the icons in the row labelled 'Experiments'. These work in a similar way to the previous exercises. ens_exps_rmse.mv : this will produce RMSE plumes for all the above experiments and the operational ensemble. ens_exps_to_an.mv : this produces 4 plots showing the ensemble spread from the OpenIFS experiments compared to the analysis. ens_exps_to_an_spag.mv : this will produce spaghetti maps for a particular parameter contour value compared to the analysis. ens_part_to_all.mv : this allows the spread & mean of a subset of the ensemble members to be compared to the whole ensemble. Note that the larger geographical area, mapType=1, will only work for MSLP due to data volume restrictions. |
| Info |
|---|
For these tasks the Metview icons in the row labelled 'ENS' can also be used to plot the different experiments (e.g. stamp plots). Please see the comments in those macros for more details. |
Task 1. RMSE plumes
Use the ens_exps_rmse.mv icon and plot the RMSE curves.
...
Use the ens_exps_to_an.mv icon and plot the ensemble spread for the different experiments.
Q. How does change changing the representation of uncertainty affect the spread?
Q. Which experiment gives the better spread?
Q. Is it possible to determine whether initial uncertainty or model uncertainty is more or less important in the forecast error?
...