Confirming impacts of convective gustiness

[megandevlan]

Starting a space to confirm the impacts of convective gustiness in later model runs (essentially looking at 104 vs. 109).

[megandevlan]

More plots to come, but at first glance it looks like we're still seeing some consistent impacts of gustiness. LHFLX biases (vs ERA5) improve with gustiness in the West Pacific during DJF especially (some improvements in Indian Ocean then too). Some regions get worse still, but overall things are looking about as expected.

[megandevlan]

Also relative to ERA5, precipitation biases look pretty similar to what we've seen before in cases with/without gustiness - general improvements in a DJF dry bias in the Indian Ocean, but a worse wet bias in parts of the West Pacific.

[megandevlan]

One of the encouraging things we saw early on with the convective gust addition was an improvement in some ENSO characteristics. We still see some signs that the gusts contribute to a better ENSO representation (in a few metrics, at least). The duration of ENSO events in particular is unrealistically long in run 109:

[megandevlan]

For reference, ADF diagnostics of run with vs. without convective gustiness is here: https://webext.cgd.ucar.edu/BLT1850/b.e30_beta02.BLT1850.ne30_t232.109/atm/