Effective ice crystal radius - mo_psrad_cloud_optics.f90
Posted by Alice Henkes 10/17/2023 https://redmine.hammoz.ethz.ch/issues/889
While we were working on coupling COSPv2 to ICON-HAM, we noticed that the values of Effective ice crystal radius outputs from MODIS simulator are substantially different from MODIS COSP Level-3 dataset (MCD06COSP). The global plots can be seen attached.
A possible reason was found in the routine mo_psrad_cloud_optics.f90. Maybe this problem is related to the %icnc_instant, which is not vertically reverse- lines 289 --306:
#ifdef ATM_PHY_ECHAM_SUBMODELS
IF ( nic_cirrus .NE. 0 ) THEN
DO jk=1,klev
re_crystals2d(:,jk) = eff_ice_crystal_radius(kbdim, kproma, ziwc(:,jk), activ_field(jg)%icnc_instant(:,jk, krow))
DO jl=1,kproma
re_crystals2d(jl,jk)=MAX))
END DO
END DO
ELSE
DO jk=1,klev
DO jl=1,kproma
re_crystals = MAX(reimin,&
MIN**0.216_wp))
re_crystals2d(jl,jk) = re_crystals
END DO
END DO
END IF
#endif
We suggest to flip the icnc vertically for the calculation of the re_crystals2d (line 292). We did some tests after vertically reverse and the simulation of Effective ice crystal radius and IWP from the MODIS COSP simulator were improved.