| ... | ... | @@ -3,8 +3,16 @@ In some situations it can be hard to obtain convergence with *Exo-REM*. The foll |
|
|
|
|
|
|
|
Reminder: *Exo-REM* is not made to simulate highly-irradiated planets (equilibrium temperature > 2000 K) and non-H2 dominated atmopsheres (metallicity > 1000 times the solar metallicity).
|
|
|
|
|
|
|
|
## General advice
|
|
|
|
- When searching for convergence, it is often useful to start any new run with your previously calculated temperature profile.
|
|
|
|
- Start with `eddy_mode = 'constant'` and `n_clouds = 0`. If you are far from convergence, you want to avoid any source of instability. You can always change these parameters once you have obtained an acceptable solution.
|
|
|
|
- Most of the time, you will need `n_non_adiabatic_iterations` to be > 0, and you always need to have `n_iterations` > `n_non_adiabatic_iterations`.
|
|
|
|
- At first, keep `retrieval_tolerance = 0`. This parameter can help you gain time when you make small changes in your simulation parameters, but using it is risky when you are far from convergence.
|
|
|
|
- Do not increase too much `retrieval_flux_error_bottom` and especially `retrieval_flux_error_top`. You risk to converge to a bad solution. |
|
|
\ No newline at end of file |
|
|
|
- Do not increase too much `retrieval_flux_error_bottom` and especially `retrieval_flux_error_top`. You risk to converge to a bad solution. `retrieval_flux_error_bottom = 1e-3` and `retrieval_flux_error_top = 1e-5` are good starting points. Consider 1e-2 as a soft top limit for both parameters.
|
|
|
|
- Do not decrease too much `weight_apriori`. You need to allow the temperature profile to vary enough. Very roughly, if you are far from convergence and at the first iteration and the temperature profile vary at each level by less than 20%, `weight_apriori` is probably too small.
|
|
|
|
- Always have `n_burn_iterations` at least equal to `n_iterations`, unless you really have a hard time to find convergence.
|
|
|
|
|
|
|
|
## Cookbook when in trouble
|
|
|
|
When you have trouble to reach convergence, you can do, in that order:
|
|
|
|
1. Decrease `weight_apriori` roughly by one order of magnitude. If you crash, decrease by another order of magnitude. Repeat until you can reach the end of your run. Then, Increase `weight_apriori` again to 10. Repeat until you are satisfied.
|
|
|
|
2. |
|
|
\ No newline at end of file |
