Coupled runs: 033, 034, 035, 036, 037, 038

[cecilehannay]

This thread is about the latest coupled runs:

• 033 (b.e23_alpha16b.BLT1850.ne30_t232.033 #356): include reverting tuning F to retrieve a decent tape recorder
• 034 (b.e23_alpha16b.BLT1850.ne30_t232.034 #361): same as 033 + microp_aero_wsub_scale = 1.0->0.75
• 035 (b.e23_alpha16b.BLT1850.ne30_t232.035 #362): same as 033 + microp_aero_wsub_scale = 1.0->0.5 + clubb_gamma_coef = 0.3 +clubb_gamma_coefb = 0.3
• 036 (b.e23_alpha16b.BLT1850.ne30_t232.036 #366): same as 035 + hard coded floor value of wsub is also scaled by 0.5 (wmixmin = 0.05)
• 037 (b.e23_alpha16b.BLT1850.ne30_t232.037 #367): same as 035 + microp_aero_wsub_scale = 0.5->0.25 + hard coded floor value of wsub is set to wmixmin = 0.025
• 038 (): same as 036 + microp_aero_wsub_min = 0.2->0.

033 was -1 W/m2 out of balance. We tried to correct that in 034 and 035.

Looking at the plot:

RESTOM in 035 seems still too cold. I am not sure it is worth to extent it further.
RESTOM in 034 looks a bit better. We could extent it a bit longer.

Thoughts: @PeterHjortLauritzen @adamrher ?

[cecilehannay]

Sorry, I inverted 34 and 35 in the text but I fixed it now.

[adamrher]

I agree with extending 34. Let me look at the clouds in 35 ...

[adamrher]

Run 34 SWCF compared to 33 and obs:

Unfortunately, I don't like where these clouds are headed. And in particular the further degradation of the DYCOMS region. The thinning of clouds at mid-latitudes is welcome though.

Run 35 SWCF compared to 33 and obs:

I do like where this is headed. The lowering of the gamma (033->0.30) brings back some DYCOMS clouds while reducing the wscale_factor (1.0->0.5) thins clouds over the storm tracks. Obviously this is not a tenable configurations owing to the -1 W/m2 RESTOM. But I like trading off the cloud forcing, thickening the MarineSc and thinning the storm track clouds -- that improves our biases. I'm wondering if we could further lower the wscale_factor=0.5->0.25 (?) or else maybe we could reduce the fall speed of warm rain?

For reference, the LWCF in 33, 34 and 35, is 24.22 W/m2, 23.87 W/m2 and 23.76, respectively. All of these are acceptable, IMO.

[cecilehannay]

Updated timeseries:

[cecilehannay]

After reviewing the results of 033, 034 and 035, we started 2 new coupled simulations: 036 and 037

[cecilehannay]

Updated timeseries including the 2 new runs:

36 produced bfb results and we stopped that run. I also ran the ADF for 34, 35, and 37. You can find the links in each run issue.

[PeterHjortLauritzen]

Tape recorder plots for coupled simulations:

[tilmes]

@cecilehannay
BURDENDUST to BURDENDUSTdn;
BURDENPOM to BURDENPOMdn;
BURDENBC to BURDENBCdn;
BURDENSEASALT to BURDENSEASALTdn;
BURDENSO4 to BURDENSO4dn;
BURDENSOA to BURDENSOAdn;
if available: AODVISstdn (you already have AODVISdn)