source/sink for added tracer in CCM3. Why 2*dtime?

[cchung@fiji_ucsd_edu]

I have added a tracer to the CCM3, and am now modifying the source/sink for the added tracer.

The CCM3 code that allows for the source/sink for the added tracer is in tphysac.F as follows:
" do m=1,pcnst
do k=1,plev
do i=1,plon
qm1(i,k,m) = qm1(i,k,m) + ztodt*srcsnk(i,k,m)
end do
end do
end do"
where qm1(:,:,2) is being modified by "ztodt*srcsnk(i,k,2)". According to tphysac.F, ztodt is 2.0 * dtime. I don't understand why the factor 2 has to be applied. Can somebody explain to me? I know that the tracer field gets updated every model time step (unlike the radiation code).

I am coupling Rasch's aerosol wet deposition code to tphysac.F in this experiment. If I don't understand the factor 2.0, I will probably underestimate the wet deposition by a factor of 2.0. Sincerely,

 

[pjr]

I have added a tracer to the CCM3, and am now modifying the source/sink for the added tracer.

The CCM3 code that allows for the source/sink for the added tracer is in tphysac.F as follows:
" do m=1,pcnst
do k=1,plev
do i=1,plon
qm1(i,k,m) = qm1(i,k,m) + ztodt*srcsnk(i,k,m)
end do
end do
end do"
where qm1(:,:,2) is being modified by "ztodt*srcsnk(i,k,2)". According to tphysac.F, ztodt is 2.0 * dtime. I don't understand why the factor 2 has to be applied. Can somebody explain to me? I know that the tracer field gets updated every model time step (unlike the radiation code).

I am coupling Rasch's aerosol wet deposition code to tphysac.F in this experiment. If I don't understand the factor 2.0, I will probably underestimate the wet deposition by a factor of 2.0. Sincerely,

The issue is that the standard CAM3 configuration uses the eulerian spectral dynamics core. For that core, the solutions are advanced from time step n-1 to time step n+1, ie two time steps. This issue is discussed and described in excruciating detail in section two of the model description found at
http://www.ccsm.ucar.edu/models/atm-cam/docs/description/

Phil