Menu
Testing new features to include in CESM1.5.
These are the runs we did before coupling all the components together in preparation of cesm1.5
These runs were done in the fall 2015.
- Aerosol tuning
- Dust: assessing dust change seen in cesm1.5
- Turbulent mountain stress parameters
- CAM5.5 + CLM4.5
New topography
New topo from Peter: see plots of the new topo variables and the differences between new and "operational" topography
| ID | Case Description | Diagnostics | Details/Comments |
|---|---|---|---|
| control | control with clubb (cam5_4_07) | w/ obs |
|
| topo | New topography files |
w/ obs w/ control |
Convective scavenging
Status of the convective scavenging:
- cam5.5: the convective scavenging cannot be turned as is in cam5.5.
The convective scavenging has a contribution from the deep convecton and from the shallow convection scheme.
CLUBB doesn't produce the output needed to compute the contribution from the shallow convection to the convective scavenging.
We did a simulation using the deep convection contribution to scavenging but turning off shallow convection contribution
- cam5.4: the convective scavenging doesn't work in cam5.4. The model crashes in the land because snow grain radius of out of bounds.
- cam5.3: the convective scavenging has significant impact. It seems consistent with what Po-Lun did previously.
As a reference you can look at C87 on Po-Lun's simulations webpage
C87 is a 10-year run is done and the comparison with the control run (standard CAM5.2.10)
We did another experiment with cam5.3 to assess the contribution of shallow convection to the convective scavenging.
This is to validate what we did with CLUBB. The contribution of the shallow convection to convective scavenging doesn't seem that big.
| ID | Case Description | Diagnostics | Details/Comments |
|---|---|---|---|
| cam5.3 control | control with cam5.3 | w/ obs |
|
| cam5.3 scavenging | cam5.3: Turn on convective scavenging | Po-Lun simulation: C87 |
|
| cam5.3 scavenging_nosh | cam5.3: Turn on convective scavenging Only for the deep convection. Shallow convection is turned off (similar to what is done in clubb |
w/ obs w/ control w/ cam5.3_scavenging |
|
| cam5.4 control | control with cam5.4 | w/ obs |
|
| cam5.4 scavenging | cam5.4: Turn convective scavenging on |
|
crashes with in the land
because |
| cam5.5 control | control with clubb (cam5_4_07) |
|
|
| cam5.5 scavenging | Turn convective scavenging on Only for the deep convection. Shallow convection is turned off |
|
Aerosol tuning
Effort to improve the aerosol (lead by Simone)
- BC is too high in the tropics.
- We also try to decrease the
so4
| ID | Case Description | Diagnostics | Details/Comments |
|---|---|---|---|
| control | control with beta07 |
||
| bc | f_act_conv = 0.0->0.4_r8 |
if (m == modeptr_pcarbon) then |
|
| so4 | sol_factic_interstitial = 0.4_r8-> 0.5_r8 |
That should just increase the solubility of all the modes. | |
| bc_so4 | Combined both changes: f_act_conv = 0.0->0.4_r8 sol_factic_interstitial = 0.4_r8-> 0.5_r8 |
w/ so4 aero_diag |
Dust
Sorting out differences in the dust. There is an increase of DUST between cam5.5 done by Pete and cesm1.5
All of this seems to be related to the code change for TMS (cam5_4_12) and the other half is related to the land.
| ID | Case Description | Diagnostics | Details/Comments |
|---|---|---|---|
| cam5_4_11 | cam5_4_11 | w/ control AODDUST |
AODDUST (1 yr)=0.020 AODDUST (2 yr)=0.019 |
| cam5_4_12 | cam5_4_12 change in TMS code but keep the same tms parameters |
w/ cam5_4_11 AODDUST |
AODDUST (1 yr)=0.036 AODDUST (2 yr)=0.038 |
| cam5_4_12 + dust retuning |
cam5_4_12 + dust_emis_fact = 0.44D0 |
w/ cam5_4_11 AODDUST |
AODDUST (1 yr)=0.021 AODDUST (2 yr)= |
| cam5_4_24_clm4 | This is the cam tag we use for cesm1.5 but here it uses CLM4 cam5_4_24 + CLM4 |
w/ cam5_4_12 AODDUST |
AODDUST (1 yr)=0.038 AODDUST (5 yr)=0.039 |
| cam5_4_24_clm4cn | This is the cam tag we use for cesm1.5 but here it uses CLM4 + CN cam5_4_24 + CLM4 + CN |
w/ cam5_4_24_clm4 AODDUST |
AODDUST (1 yr)=0.042 AODDUST (2 yr)=0.040 |
| cam5_4_24_clm5cn | cesm1.5: cam5_4_24 + CLM5 | w/ cam5_4_24_clm4cn AODDUST |
AODDUST (1 yr)=0.053 AODDUST (5 yr)=0.048 |
Turbulent mountain stress sensitivity study
The introduction of turbulent mountain stress (tms), a missing stress process keying off sub-grid orographic scales between small and resolved scales, demonstrated many improvements in north-Atlantic and southern ocean surface stress biases in CAM5.
The tms parameterization include free parameters that have significant potential to influence the coupled system by regulating surface stresses over the ocean.
| Default | ||
| horomin | 1 | Minimum value of subgrid orographic height for mountain stress |
| orocnst | 1 | Converts from standard deviation to height |
| z0fac | 0.075 | Factor determining z_0 from orographic standard deviation |
| zomax | 100 | Maximum of Zo for orography |
| ID | Case Description | Diagnostics | Details/Comments |
|---|---|---|---|
| control | control with clubb (cam5_4_07) | w/ obs |
|
| tms_horomin_10 | horomin = 1 -> 10 |
|
|
| tms_horomin_0.5 | horomin = 1 ->0.5 |
|
|
| tms_oro_0.5 | orocnst = 1 -> 0.5 | w/ obs w/ control |
|
| tms_oro_2 | orocnst = 1 ->2 | w/ obs w/ control |
|
| tms_z0fac_0.004 | z0fac = 0.075 -> 0.004 | w/ obs w/ control |
|
| tms_z0fac_0.2 | z0fac = 0.075 -> 0.2 | w/ obs w/ control |
|
| tms_z0max_50 | z0max =100 -> 50 | w/ obs w/ control |
no impact |
| tms_z0max_10 | z0max =100 ->10 |
|
Assessing CAM5.5 + CLM4.5
| ID | Case Description | Diagnostics | Details/Comments |
|---|---|---|---|
| cam5_clm4 | F2000 compset cam5 + clm4 |
w/ obs |
All the runs are done with FC5. |
| cam5.4_clm4 | cam5.4 + clm4 |
w/ cam5-clm4 |
|
| cam5.5_clm4 | cam5.5 + clm4 |
w/ cam5.4-clm4 |
|
| cam5.5_clm4.5 | cam5.5 + clm4.5 |
|
|
| cam5.5_clm4.5_c0_lnd | cam5.5 + clm4.5 +c0_lnd = 0.0075 ->0.015 |
|
|
| cam5.5_clm4.5_topo | cam5.5 + clm4.5 + new topography |
|