Problem with shortwave scheme

by **SeBi013** on Mon Feb 01, 2010 5:23 am

Hi all,

I'm actually testing NMM for solar energy forecasting and I'm trying some configurations in the physics options to obtain the better forecast in shortwave radiation (precisely the variable RSWIN)

The best forecast is when I use the Dudhia Scheme for the option "ra_sw_physics" (see below the entire namelist). But when I use this namelist, the variables which contain the cloud pourcentage ( CFRACL, CFRACM, and CFRACH) are at all time steps equal to zero.

This absence of cloud forecast is relatively paradoxal wheras the shortwave forecast is relatively correct.

Questions:
1/ Can somebody help me for this mistake ?
2/ Can somebody give me a good configuration to solar energy forecast ?

Thanks you

Sebastien D.

Code: Select all: &time_control run_days = 2, run_hours = 0, run_minutes = 0, run_seconds = 0, start_year = 2010,2010, start_month = 01,01, start_day = 13,13, start_hour = 00,00, start_minute = 00, 00, start_second = 00, 00, tstart = 00, end_year = 2010,2010, end_month = 01,01, end_day = 15,15, end_hour = 00,00, end_minute = 00, 00, end_second = 00, 00, interval_seconds = 10800, history_interval = 240, 6, frames_per_outfile = 1000,1000, restart = .false., restart_interval = 1440, reset_simulation_start = F, io_form_input = 2, io_form_history = 2, io_form_restart = 2, io_form_boundary = 2, io_form_auxinput1 = 2, auxinput1_inname = "met_nmm.d<domain>.<date>" debug_level = 1 / &domains time_step = 18, time_step_fract_num = 0 time_step_fract_den = 1 max_dom = 2, s_we = 1, 1, e_we = 56, 78, s_sn = 1, 1, e_sn = 92, 116, s_vert = 1, 1, e_vert = 28, 28, num_metgrid_levels = 27, dx = 0.054, 0.027, dy = 0.054, 0.027, grid_id = 1, 2, parent_id = 0, 1, i_parent_start = 1, 15, j_parent_start = 1, 29, parent_grid_ratio = 1, 3, parent_time_step_ratio = 1, 3, p_top_requested = 5000., ptsgm = 0., eta_levels = 1.000 , 0.992 , 0.990 , 0.989 , 0.987 , 0.985 , 0.982 , 0.978 , 0.964 , 0.946 , 0.922 , 0.894 , 0.860 , 0.817 , 0.766 , 0.707 , 0.644 , 0.576 , 0.507 , 0.444 , 0.380 , 0.324 , 0.273 , 0.188 , 0.152 , 0.093 , 0.038 , 0.000 , tile_sz_x = 0, tile_sz_y = 0, numtiles = 1, / &physics mp_physics = 5, 5, ra_lw_physics = 1, 1, ra_sw_physics = 1, 1, nrads = 100, 100, nradl = 100, 100, co2tf = 1, sf_sfclay_physics = 2, 2, sf_surface_physics = 2, 2, bl_pbl_physics = 2, 2, nphs = 10, 10, cu_physics = 4, 4, ncnvc = 10, 10, tprec = 3, 0.25, theat = 3, 0.25, tclod = 3, 0.25, trdsw = 3, 0.25, trdlw = 3, 0.25, tsrfc = 3, 0.25, pcpflg = .false., isfflx = 0, ifsnow = 0, icloud = 1, num_soil_layers = 4, mp_zero_out = 0, / &dynamics / &bdy_control spec_bdy_width = 1, specified = .true., nested = .false., / &grib2 / &namelist_quilt nio_tasks_per_group = 0, nio_groups = 1, /

by **jimmyc** on Mon Feb 01, 2010 6:08 pm

what namelist options for radiation were you using before?

by **SeBi013** on Fri Feb 05, 2010 11:46 am

Hi,

In fact, I've tested a lots of configurations.
Sometimes, I have pretty good RSWIN but I haven't cloud for example :

ra_lw_physics : 1
ra_sw_physics : 1
mp_physics : 5
cu_physics : 2
sf_sfclay_physics : 3
bl_pbl_physics : 3

... Remplace the "x" in the code below by these variables :

Code: Select all: &time_control run_days = 2, run_hours = 0, run_minutes = 0, run_seconds = 0, start_year = 2010,2010, start_month = 01,01, start_day = 13,13, start_hour = 00,00, start_minute = 00, 00, start_second = 00, 00, tstart = 00, end_year = 2010,2010, end_month = 01,01, end_day = 15,15, end_hour = 00,00, end_minute = 00, 00, end_second = 00, 00, interval_seconds = 10800, history_interval = 240, 6, frames_per_outfile = 1000,1000, restart = .false., restart_interval = 1440, reset_simulation_start = F, io_form_input = 2, io_form_history = 2, io_form_restart = 2, io_form_boundary = 2, io_form_auxinput1 = 2, auxinput1_inname = "met_nmm.d<domain>.<date>" debug_level = 1 / &domains time_step = 18, time_step_fract_num = 0 time_step_fract_den = 1 max_dom = 2, s_we = 1, 1, e_we = 56, 78, s_sn = 1, 1, e_sn = 92, 116, s_vert = 1, 1, e_vert = 28, 28, num_metgrid_levels = 27, dx = 0.054, 0.027, dy = 0.054, 0.027, grid_id = 1, 2, parent_id = 0, 1, i_parent_start = 1, 15, j_parent_start = 1, 29, parent_grid_ratio = 1, 3, parent_time_step_ratio = 1, 3, p_top_requested = 5000., ptsgm = 0., eta_levels = 1.000 , 0.992 , 0.990 , 0.989 , 0.987 , 0.985 , 0.982 , 0.978 , 0.964 , 0.946 , 0.922 , 0.894 , 0.860 , 0.817 , 0.766 , 0.707 , 0.644 , 0.576 , 0.507 , 0.444 , 0.380 , 0.324 , 0.273 , 0.188 , 0.152 , 0.093 , 0.038 , 0.000 , tile_sz_x = 0, tile_sz_y = 0, numtiles = 1, / &physics mp_physics = x,x ra_lw_physics = x,x ra_sw_physics = x,x nrads = 100, 100, nradl = 100, 100, co2tf = 1, sf_sfclay_physics = x,x sf_surface_physics = 2, 2, bl_pbl_physics = x,x nphs = 10, 10, cu_physics = x,x ncnvc = 10, 10, tprec = 3, 0.25, theat = 3, 0.25, tclod = 3, 0.25, trdsw = 3, 0.25, trdlw = 3, 0.25, tsrfc = 3, 0.25, pcpflg = .false., isfflx = 0, ifsnow = 0, icloud = 1, num_soil_layers = 4, mp_zero_out = 0, / &dynamics / &bdy_control spec_bdy_width = 1, specified = .true., nested = .false., r / &grib2 / &namelist_quilt nio_tasks_per_group = 0, nio_groups = 1, /

But sometimes, I have pretty good clouds variables but I haven't a good RSWIN for example :

Use the same namelist that before but with these modifications:

ra_lw_physics : 99
ra_sw_physics : 99
mp_physics : 5
cu_physics : 2
sf_sfclay_physics : 3
bl_pbl_physics : 3

I don't understand why sometimes the clouds are present in the wrfout and sometimes not.

In according to your experience in the WRF, which configuration would you recommend to me for the forecast of downward short wave ?

Thanks a lot

Sébastien D.

by **SeBi013** on Wed Feb 10, 2010 5:10 am

Hi,

I already search on other search engine, but I found nothing to solve my problem.

Problem with shortwave scheme

Problem with shortwave scheme

Re: Problem with shortwave scheme

Re: Problem with shortwave scheme

Re: Problem with shortwave scheme

Who is online