NCAR / CARMA_dev

Discussions on CARMA development and descriptions of model simulations
3 stars 0 forks source link

B-case WACCM-MA CARMA 2deg Simulation #43

Open tilmes opened 5 months ago

tilmes commented 5 months ago

Purpose of the simulation: Test first WACCM-MA CARMA 2deg B-case simulation to produce a Pre-Industrial Control and Historical simulation as a model to run climate simulations (into the future)!

I gradually changed the clubb_gamma coefficient from 0.28 to 0.25

CodeBase: /glade/work/fvitt/cesm/carmats16_cesm2.2/

Case dir: /glade/work/tilmes/derecho/carma/b.e21.B1850WmaCARMA.f19_g17.carmats16_cesm2.2.001

Currently 72 year: RESTOM:

image

Diagnostics: CARMA vs WACCM-MA MAM 2deg year 52-62: https://webext.cgd.ucar.edu/B1850WmaCARMA/atm/b.e21.B1850WmaCARMA.f19_g17.carmats16_cesm2.2.001_52_62_vs_b.e21.BWma1850.f19_g17.CMIP6-piControl-WACCM-MA-2deg.001_300_349/html_img/mean_diag_LatLon.html

tilmes commented 4 months ago

Several diagnostics were performed by John Fasullo: https://webext.cgd.ucar.edu/Multi-Case/CMAT/CVDP/CVDP-B1850WmaCARMA/output/ Showing that CARMA has too many bright clouds, indicating that the clubb_gamma value was adjusted too much to counter the warming. Suggestions by John, to first perform F2000-case to get the model into a 0.8W/m2 imbalance and then start from a different / warmer ocean to spin up the model.

Two FWma2000 cases were performed, one using the default club_gamma value 0.28 and one using a larger club_gamma value, making the clouds less reflective:

https://acomstaff.acom.ucar.edu/tilmes/amwg/carma/FWmaCARMA2000climo.f19_f19_mg17.carma_trop_strat16.002_1_8_vs_FWmaCARMA2000climo.f19_f19_mg17.carma_trop_strat16.001_1_8/website/index.html

FWma2000 restom plots in /glade/work/tilmes/CESM/plots clubb_gamma_0.28:

image

clubb_gamma_0.30

image

Conclusions, another case will be performed with clubb_gamma_0.29, to get closer to 0.8W/m2

image
tilmes commented 2 months ago

Next, a F1850 simulation had been performed, using the same setting as for the F2000 case but with F1850 conditions: problem, RESTOM is larger than that from the F2000 case:

image

The ADF diagnostics are here: https://acomstaff.acom.ucar.edu/tilmes/amwg/carma/FWmaCARMA1850climo.f19_f19_mg17.carma_trop_strat16.003_1_8_vs_FWmaCARMA2000climo.f19_f19_mg17.carma_trop_strat16.003_1_8/website/html_img/mean_diag_LatLon.html

tilmes commented 2 months ago

Three more F2000 cases are performed to increase clubb gamma:

image image image