In *this note*, I show results of several simulations, some with or without sponges. One of them (`exp2_l`) in particular has zero dissipation in the middle layer and is configured with a sponge layer. This simulation was run with a weaker wind run than in recent simulations (MAG=100 instead of MAG=500 and 1000). The first diagnostic of this simulation shows that, at least in the middle layer, the Eulerian-mean flow stabilizes after about day 2000 (lower panels of Fig. 1).

I show here the Eulerian-mean flow, the Lagrangian-mean flow and the LPV analysis for two different periods: around day 2000 (average over two cycles starting at day 2015 and 2065) and around day 4200 (average over two cycles starting at day 4215 and 4265).

Conclusions are:

- The LPV analysis works well (outside the sponge layer at least). Note that there is no panel for the dissipative term; it is exactly zero here in this case (because it depends on KH which is set to zero).
- There is not much difference between the two periods of analysis.
- Because the time rate of change of QL is not zero, we do not have zero VL to be zero and we may need to perform the analysis over many more cycles in order to cancel the time rate of change.

Computed with `theory_test_several_cycles_script.m` in `RESEARCH/MODELISATION/HIM/studies/diss_train_of_eddies/exp2/exp2_l/analysis_1d` on the main disk on `ipu1`. The Matlab files are `diag_VC_two_cycles_100day_long_day2015_day2065_exp2_l.mat` and `diag_VC_two_cycles_100day_long_day4215_day4265_exp2_l.mat` in that same directory.