Answer-Changing Features in CESM1_0

Answer-Changing Features in CESM1_0_3

The following changes in the cesm1_0_3 tag will result in most cases not producing bit-for-bit results when compared with cesm1_0_2.

Note that for scientifically supported compsets, all changes are climate preserving except for CAM5 compsets.

Compsets: B_2000_CN_CHEM, B_1850_CN_CHEM, B_1850-2000_CN_CHEM and F_1850_CN_CHEM

Climate-preserving changes: change horizontal interpolation of input datasets to generalize for unstructured grids and updates to dry deposition.
Non-climate-preserving changes: correction to the wet deposition list (include CH3OOH) and restore lightning NOx production.

Climate-preserving changes: change horizontal interpolation of input datasets to generalize for unstructured grids and updates to dry deposition.
Non-climate-preserving changes: update trop_mozart with latest MOZART4 mechanism and include short wavelength photolysis in tropospheric chemistry.

Climate-preserving changes: change in the co2vmr surface value used in the nlte_fomichev, change horizontal interpolation of input datasets to generalize for unstructured grids, WACCM compsets changed to use 'xactive_lnd' method and corrected the reaction: O1D + CCL4 -> 4*CL for WACCM compsets.
Non-climate-preserving changes: include CO2 reactions for WACCM to improve concentrations in upper regions, new treatment of stratospheric aerosols and correction to the wet deposition list (N2O5 removed from list).

Compsets: B_1850_WACCM_CN (waccm_mozart_v1) and B_1850-2005_WACCM_CN (waccm_mozart_v1)

Climate-preserving changes: change in the co2vmr surface value used in the nlte_fomichev, change horizontal interpolation of input datasets to generalize for unstructured grids, WACCM compsets changed to use 'xactive_lnd' method and corrected the reaction: O1D + CCL4 -> 4*CL for WACCM compsets.

F_SD_CAMCHEM (trop_mozart), F_2000_TROP_MOZART (trop_mozart), B_2000_WACCM_CN (waccm_mozart), B_1955-2005_WACCM_CN (waccm_mozart_v1), B_RCP2.6_WACCM_CN (waccm_mozart_v1) B_RCP4.5_WACCM_CN (waccm_mozart_v1), B_RCP8.5_WACCM_CN (waccm_mozart_v1), F_1955-2005_WACCM_CN (waccm_mozart), F_SD_WACCM (waccm_mozart), F_2000_WACCM_SC (waccm_ghg), F_SD_BAM (trop_bam), F_TROP_STRAT_CHEM (trop_strat_bam_v1).

Simulations at T31 for 1850 with the active-land model (B,E,F, and I compsets) answers will be greater-than-roundoff-level different from cesm1_0_2 solutions because of a new surface dataset and new initial conditions that were added in. The new T31 surface dataset uses a new glacier fraction dataset.
Answers MAY be different to roundoff for non-transient compsets because of the problem documented in the "List of minor issues for CLM" given under Known Problems for CESM1.0.2.
SNOWLIQ and SNOWICE will be different on history files because of a change to the averaging of these fields.

In all cases using the active-ocean model (B,C, and G compsets), answers will be greater-than-roundoff-level different from cesm1_0_2 solutions, because single-precision constants used in the overflows parameterization have been changed to double precision in cesm1_0_3.
In cases using the active-ocean ecosystem model (any compset using OCN_TRACER_MODULES="ecosys"), under certain rare conditions, solutions may not be bit-for-bit with those generated from previous CESM1 release tags, due to the improvements made in the robustness of the pH solver. Note that the the carbonate diagnostics fields (CO3, HCO3, H2CO3) will be slightly different from previous tags because the phlo_3d_init value is different from the previous tags.