MPAS Fixed SST vs Daily-Updated SST

[Snow_Rain_Cloud]

Dear technical support,

I configured a set of experiments to look at SST differences between two experiments with config_sst_update = true and false.

Please find attached figure F000_sst.png the skintemp difference in one run initialized at 00Z 21 Oct 2021. The second figure, F006_sst.png is the same run but for a 6-h forecast. The differences along the coast and sharp gradients seem to appear, when SST has not been updated yet - which will happen at f024.

Do you have any ideas on why this happens? Apologies for the poor figure quality and colors – our server is under maintenance so I cannot re-create plots.
I am using 60-15 km meshes, 55 levels, mesoscale reference suite, and version 8.2.2.
Thanks so much for your time!

Minghua

 

[Ming Chen]

Hi Minghua,

Thank you for raising this issuee. Would you please upload your namelist.atmopshere ?

Also, please upload sea ice field at the initial time and at 6-h FCST time for me to take a look? Thanks.

 

[Snow_Rain_Cloud]

Hi Ming,

Hoping you are doing well! Please find attached my namelists and the plot (for XICE variables (in shades).

 

[Snow_Rain_Cloud]

&nhyd_model
config_time_integration_order = 2
config_dt = 90.0
config_start_time = '2021-10-21_00:00:00'
config_run_duration = '5_00:00:00'
config_split_dynamics_transport = true
config_number_of_sub_steps = 2
config_dynamics_split_steps = 3
config_horiz_mixing = '2d_smagorinsky'
config_visc4_2dsmag = 0.05
config_scalar_advection = true
config_monotonic = true
config_coef_3rd_order = 0.25
config_epssm = 0.1
config_smdiv = 0.1
/
&damping
config_zd = 22000.0
config_xnutr = 0.2
/
&limited_area
config_apply_lbcs = false
/
&io
config_pio_num_iotasks = 0
config_pio_stride = 1
/
&decomposition
config_block_decomp_file_prefix = 'x4.535554.graph.info.part.'
/
&restart
config_do_restart = false
/
&printout
config_print_global_minmax_vel = true
config_print_detailed_minmax_vel = false
/
&IAU
config_IAU_option = 'off'
config_IAU_window_length_s = 21600.
/
&physics
config_sst_update = true
config_sstdiurn_update = false
config_deepsoiltemp_update = false
config_radtlw_interval = '00:30:00'
config_radtsw_interval = '00:30:00'
config_bucket_update = 'none'
config_physics_suite = 'mesoscale_reference'
/
&soundings
config_sounding_interval = '3:00:00'
/
&physics_lsm_noahmp
config_noahmp_iopt_dveg = 4
config_noahmp_iopt_crs = 1
config_noahmp_iopt_btr = 1
config_noahmp_iopt_runsrf = 3
config_noahmp_iopt_runsub = 3
config_noahmp_iopt_sfc = 1
config_noahmp_iopt_frz = 1
config_noahmp_iopt_inf = 1
config_noahmp_iopt_rad = 3
config_noahmp_iopt_alb = 1
config_noahmp_iopt_snf = 1
config_noahmp_iopt_tksno = 1
config_noahmp_iopt_tbot = 2
config_noahmp_iopt_stc = 1
config_noahmp_iopt_gla = 1
config_noahmp_iopt_rsf = 4
config_noahmp_iopt_soil = 1
config_noahmp_iopt_pedo = 1
config_noahmp_iopt_crop = 0
config_noahmp_iopt_irr = 0
config_noahmp_iopt_irrm = 0
config_noahmp_iopt_infdv = 1
config_noahmp_iopt_tdrn = 0
/

 

[Snow_Rain_Cloud]

The namelist is for the sst updated run. The only difference for fixed sst namelist is config_sst_update = false

 

[Ming Chen]

Hi Ming Hua,

Thanks for uploading the sea ice plots and namelist.atmosphere.

For the SST update issue, SST at open water points should remain same as in the initial condition until it reaches the update time. This is what I have seen in your figure F006_sst.png.

The same figure shows differences between 00 h and 06 h (before SST is updated) along the coast, which I believe is attributed to the existence of sea ice along the coast. In Noah LSM, sea ice is treated as 'land' points, and TSK at these points are recalculated based on sea ice physics. In this case, SST and SKINTEMP are no longer the same.

Hope this is helpful for you. Please let me know if you have more questions about this issue. Thanks.

 

[Snow_Rain_Cloud]

Thanks, Ming, for the timely reply. I understand that sea ice is present at higher latitudes and is treated as land data. However, differences also appear in the offshore areas of California and in tropical regions, where I do not think sea ice is present.

Even though TSK is recalculated based on the sea-ice physics, the initial and boundary conditions, including the sea-ice fields, should be identical in the two runs before the updated time.

I am still puzzled about the source of the differences over land and sea-ice regions. Are any interpolation procedures applied during the intermediate steps between f000 and the first updated forecast hour?

 

[Ming Chen]

Hi Ming Hua,

I don't think there is any interpolation conducted during the model integration.

Would you please confirm that the variable shown in F006_sst.png is skintemp? Can you also upload the difference in SST at 06 h FCST? Is it convenient to get IVGTYP, ISLTYP, LANDSEA, XLAND at one specific point in the tropical coast where you see difference in SKINTEMP at 06 h FCST?

I ask this because I am suspicious that landmask and landuse type may be inconsistent along some coastal regions. However, I am not able to confirm this yet. We need to further look into this issue.

 

[Snow_Rain_Cloud]

Thanks so much, Ming. I'll create the figures after Derecho is back and get them back to you. Would you like me to put them on a ppt / pdf or individual plots as attachments?