WRF-Chem ignoring values of DUST_X in wrfinput

[olafur rognvaldsson]

Hello everyone.

I'm running V4.1.2 of WRF(Chem) operationally for Africa using dust option nr. 3. I'm trying to cycle the DUST_X variables between forecasts but for some reason the model ignores the values found in the wrfinput_d01 file and always starts each new forecast from scratch (i.e. no dust at time-step zero).

Following are snippets from the registry.chem and registry.var_chem files

state real dust_1 ikjftb chem 1 - i0{12}rhusdf=(bdy_interp:dt) "DUST_1" "dust size bin 1: 0.5um effective radius" "ug/kg-dryair"

and

state real dust_1 ikjftb chem 1 - irhusdf=(bdy_interp:dt) "DUST_1" "dust size bin 1: 0.5um effective radius" "ug/kg-dryair"

In both cases the DUST_X variables are written to input, restart and history files. Is there some mechanism that forces the model to read said variables from the wrfinput-file that I may be missing???

Kind regards, Ólafur.

 

[olafur rognvaldsson]

Sighhh, turns out this is not a bug but a feature.

The wrf-chem dust behaves differently to wrf-chem volcano in the sense that one needs to create a dedicated input-file if one wants the model to "re-use" dust-values from earlier simulations. This calls for a simple change in the namelist.input file and we need to copy/link wrfinput_d01 to wrf_chem_input_d01 once we have cycled the dust-values from previous forecast.

So I guess the lesson of the day is: RTFM