Erroneous discharge values while performing WRF-Hydro Coupled simulations

[Amit]

Hiii,

I am performing WRF-Hydro v5.0.3 uncouple and coupled simulations over a catchment. The results seems to be good while running in uncouple mode in terms of discharge, surface runoff etc. While I perform coupled simulations I get unusually high values of surface runoff and discharge (in terms of e+6) scatter all around the domain. I have tried both NOAH and NOAHMP (NOAHMP with different options). I don't get the reason behind it, if someone else have come across such problem and solved it then pls let me know. Following is the list of physics option for the coupled simulation (NOAHMP).


&physics
mp_physics = 6, 6, 3,
ra_lw_physics = 4, 4, 1,
ra_sw_physics = 4, 4, 1,
radt =25, 5, 30,
sf_sfclay_physics = 1, 1, 1,
sf_surface_physics = 2, 2, 2,
bl_pbl_physics = 1, 1, 1,
bldt = 0, 0, 0,
cu_physics = 1, 1, 0,
cudt = 5, 5, 5,
isfflx = 1,
ifsnow = 1,
!icloud = 1,
!icloud_bl = 0,
surface_input_source = 1,
num_soil_layers = 4,
sf_urban_physics = 0, 0, 0,
usemonalb = .true.,
sst_skin = 1,
!sst_update = 1,
maxiens = 1,
maxens = 3,
maxens2 = 3,
maxens3 = 16,
ensdim = 144,
NUM_LAND_CAT = 21,
!GRAV_SETTLING = 2
!CU_RAD_FEEDBACK = .true.
!KFETA_TRIGGER = 2
!ICLOUD_BL = 1,
LIGHTNING_OPTION = 3, 3,
LIGHTNING_DT = 75, 75,
LIGHTNING_START_SECONDS = 75, 75,
FLASHRATE_FACTOR = 2, 3.5,
CELLCOUNT_METHOD = 2, 2,
ICCG_METHOD = 2, 2,
DO_RADAR_REF = 1,

/
&noah_mp
dveg = 4,
opt_crs = 1,
opt_btr = 1,
opt_run = 3,
opt_sfc = 1,
opt_frz = 1,
opt_inf = 1,
opt_rad = 3,
opt_alb = 2,
opt_snf = 1,
opt_tbot = 2,
opt_stc = 3,
opt_gla = 2,
opt_rsf = 4,
opt_soil = 1,
opt_pedo = 1,
opt_crop = 0,
/

[Katelyn FitzGerald]

Hi Amit,

Have you compared the precipitation and other forcing fields used in your standalone and coupled simulations?  I'm curious if there might be an issue with some of the meteorological fields coming in from the atmospheric model.

I also noticed that you have opt_pedo turned on within the noah_mp namelist. I believe this is intended to be used when opt_soil is set to option 3.  Perhaps more importantly, it's not an option we (the development team at NCAR) have verified to work with WRF-Hydro.  You might find this document (https://ral.ucar.edu/sites/default/files/public/Noah-MPOptionsIndicatorsofusagewithWRFHydroNWM.pdf) describing support for various options useful.

I hope this helps.

Best,

Katelyn

[Amit]

Hii Katelyn,

Thanks for the reply, I have checked the meteorological fields it seems to be correct. The problem arises when I perform coupled simulations, because 3 day forecast of Rain using uncoupled WRFV-4.0.1 does not produce high amount of surface runoff as in case of coupled  (WRFV-4.0.1 and HydroV-5.0.3) simulations. As you have pointed out about the NOAH-MP options, I have updated the namelist based on it and provided other physics options as one of the example coupled cases. Even though I am getting these unusual runoff values. Following is the list of physics options and I have also attached screenshots of rain and surface runoff for coupled and uncoupled simulations. 

&physics

 physics_suite                  = 'CONUS',

 radt                                = 25,  30,

 mp_physics                    =  3,   3,

 ra_lw_physics                 =  1,   1,

 ra_sw_physics                =  1,   1,

 cudt                               =  0,   0,

 cu_physics                     =  0,   0,

 sf_sfclay_physics           =  1,   1,

 bl_pbl_physics                =  1,   1,

 sf_surface_physics         =  4,   4,

 icloud                             =  1,

 num_soil_layers              =  4,

 num_land_cat                 = 21,

 sf_urban_physics            =  0,

 /

 &noah_mp

 dveg     = 4,

 opt_crs  = 1,

 opt_btr  = 1,

 opt_run  = 3,

 opt_sfc  = 1,

 opt_frz  = 1,

 opt_inf  = 1,

 opt_rad  = 3,

 opt_alb  = 2,

 opt_snf  = 1,

 opt_tbot = 2,

 opt_stc  = 1,

 opt_gla  = 1,

 opt_rsf  = 1,

 /

[Katelyn FitzGerald]

You might look into the time stepping.  This works slightly differently in coupled (WRF is driving the LSM time step and it's typically much smaller) vs standalone simulations (HRLDAS/WRF-Hydro is driving the LSM time step).  The WRF and/or LSM time step needs to be a multiple of the channel and terrain routing time steps.  I believe the updated code on master in the NCAR/wrf_hydro_nwm_public has some additional checks on the various model time steps that may be useful.  

Something else that might be worth checking if you haven't already is the land cover.  It looks like you're using the 21 category MODIS classification system.  You'll want to make sure you're using the corresponding HYDRO_MODIS.TBL (rename this to HYDRO.TBL) for both your standalone and coupled simulations. However, I'm guessing this is not the source of the runoff issue.

[Katelyn FitzGerald]

Hi Amit,

I think I misunderstood your initial question.  When WRF-Hydro is coupled to WRF, the hydro components handle the surface runoff and this output variable from the land surface model is no longer valid.  This is likely why you're seeing the strange values.  You'll want to look in the output files for WRF-Hydro rather than the wrfout* files for runoff information.  The specific files and variables output will change depending on model configuration, but we do have a document detailing this on our documentation page (https://ral.ucar.edu/projects/wrf_hydro/technical-description-user-guide). 

Hope this helps.

Best,

Katelyn

[Amit]

Hi Katelyn,

Thanks for your reply, I was also wondering about that, the surface runoff output in wrfout* files are bit strange. Anyway with updated NOAM_MP options hydro output seems to be good. The streamflow and infiltration excess values are in range. Thanks for your suggestions.

Amit