Using MODIS albedo as input in WRF with Noah-MP

[seti]

Dear WRF Support Team,

I am using WRFV4.8.0 with the Noah-MP land surface model (sf_surface_physics = 4), and I would like to replace the default albedo with satellite-derived MODIS albedo for a 5-day simulation.

My plan is to download MODIS albedo corresponding to my simulation period, regrid it to the WRF domains, and use these values to modify the albedo field in the WRF input files.

However, I am not sure about the correct procedure for Noah-MP. In particular, I would appreciate your advice on the following questions:

1. Is directly replacing the ALBEDO field in wrfinput_d0* with MODIS-derived albedo an appropriate approach when using Noah-MP?
2. If I modify ALBEDO in wrfinput, will Noah-MP actually use these prescribed values during the simulation, or will it recalculate albedo after initialization?
3. If Noah-MP recalculates albedo, what is the recommended method to ensure that my MODIS-derived albedo is actually used throughout the 5-day simulation?
4. Do I need to modify ALBBCK, ALBEDO12M, or another variable instead of (or in addition to) ALBEDO?
5. Are any specific namelist options required, such as usemonalb, sst_update, or settings in the &noah_mp section?
6. Since Noah-MP uses opt_alb for the ground/snow surface albedo treatment (opt_alb = 1 for BATS and opt_alb = 2 for CLASS), how does this option interact with externally prescribed MODIS albedo? Would either option overwrite or modify the MODIS albedo during the simulation?

My simulation is for a winter period, so snow is present in parts of the domain. Therefore, I would also like to understand whether MODIS surface albedo can be prescribed directly in snow-covered grid cells, or whether Noah-MP will recompute the effective surface albedo based on its snow-albedo parameterization.

My main goal is to use time-specific MODIS albedo instead of the default/climatological albedo while keeping the treatment physically consistent with Noah-MP.

I would greatly appreciate any guidance on the recommended workflow.

Best regards,

 

[seti]

Dear WRF Support Team,

I am using WRFV4.8.0 with the Noah-MP land surface model (sf_surface_physics = 4), and I would like to replace the default albedo with satellite-derived MODIS albedo for a 5-day simulation.

My plan is to download MODIS albedo corresponding to my simulation period, regrid it to the WRF domains, and use these values to modify the albedo field in the WRF input files.

However, I am not sure about the correct procedure for Noah-MP. In particular, I would appreciate your advice on the following questions:

1. Is directly replacing the ALBEDO field in wrfinput_d0* with MODIS-derived albedo an appropriate approach when using Noah-MP?
2. If I modify ALBEDO in wrfinput, will Noah-MP actually use these prescribed values during the simulation, or will it recalculate albedo after initialization?
3. If Noah-MP recalculates albedo, what is the recommended method to ensure that my MODIS-derived albedo is actually used throughout the 5-day simulation?
4. Do I need to modify ALBBCK, ALBEDO12M, or another variable instead of (or in addition to) ALBEDO?
5. Are any specific namelist options required, such as usemonalb, sst_update, or settings in the &noah_mp section?
6. Since Noah-MP uses opt_alb for the ground/snow surface albedo treatment (opt_alb = 1 for BATS and opt_alb = 2 for CLASS), how does this option interact with externally prescribed MODIS albedo? Would either option overwrite or modify the MODIS albedo during the simulation?

My simulation is for a winter period, so snow is present in parts of the domain. Therefore, I would also like to understand whether MODIS surface albedo can be prescribed directly in snow-covered grid cells, or whether Noah-MP will recompute the effective surface albedo based on its snow-albedo parameterization.

My main goal is to use time-specific MODIS albedo instead of the default/climatological albedo while keeping the treatment physically consistent with Noah-MP.

I would greatly appreciate any guidance on the recommended workflow.

Best regards,

@Ming Chen

 

[cenlinhe]

Hi, to answer your questions:
1. are you using daily MODIS albedo timeseries or just one MODIS albedo 2-D map for your 5-day simulations? If it is time series, you will need to put it in the wrflow_input file as a lower boundary input data and follow the SST time series read-in code to pass the albedo time series to land model. If it is just a 2-D time-invariant map, then you can put it in wrfinput file and read in it and pass it to NoahMP following the existing land type or other static variables.

2. Noah-MP indeed dynamically calculates surface albedo every timestep via canopy radiative transfer. You will need to modify the source code inside Noah-MP to replace the computed surface albedo with MODIS value after the default albedo calculation is done. However, the difficult thing is that the energy balance calculation inside NoahMP is always associated with the canopy radiative transfer and dynamic albedo calculation. It is very challenging to apply your MODIS albedo and at the same time to keep the energy balance calculation consistent. Right now, I do not have a perfect solution for this.

3. Noah-MP does not use ALBBCK, ALBEDO12M, or other default Albedo variables in wrfinput. you can create a new variable.

4. no other specific namelist options because you will need to modify the source code to enable this capability.

5. opt_alb is only for snow albedo. The opt_alb parameterization updates snow albedo every timestep which is further used to combine with snow cover and bare soil albedo to get the total ground albedo. the ground albedo is then used by canopy radiative transfer model to compute the total surface albedo.