Adds MAM4xx: Aerosol optics #2718
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Status Flag 'Pull Request AutoTester' - User Requested Retest - Label AT: RETEST will be reset after testing. |
|
Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects: Pull Request Auto Testing STARTING (click to expand)Build InformationTest Name: SCREAM_PullRequest_Autotester_Mappy
Jenkins Parameters
Build InformationTest Name: SCREAM_PullRequest_Autotester_Weaver
Jenkins Parameters
Using Repos:
Pull Request Author: odiazib |
susburrows
previously approved these changes
Apr 9, 2024
|
@mahf708 and @brhillman , can you please confirm if you feel your comments have been adequately addressed? Thanks! |
mahf708
previously approved these changes
Apr 9, 2024
|
Status Flag 'Pull Request AutoTester' - Jenkins Testing: 1 or more Jobs FAILED Note: Testing will normally be attempted again in approx. 2 Hrs. If a change to the PR source branch occurs, the testing will be attempted again on next available autotester run. Pull Request Auto Testing has FAILED (click to expand)Build InformationTest Name: SCREAM_PullRequest_Autotester_Mappy
Jenkins Parameters
Build InformationTest Name: SCREAM_PullRequest_Autotester_Weaver
Jenkins Parameters
SCREAM_PullRequest_Autotester_Mappy # 5166 FAILED (click to see last 100 lines of console output)SCREAM_PullRequest_Autotester_Weaver # 5514 PASSED (click to see last 100 lines of console output) |
DIFFs in the optics CIME case ( |
mahf708
previously approved these changes
Apr 10, 2024
singhbalwinder
previously approved these changes
Apr 10, 2024
brhillman
reviewed
Apr 10, 2024
| Kokkos::fence(); | ||
|
|
||
| //tau_w_g : aerosol asymmetry parameter * tau * w | ||
| const auto tau_g_sw = tau_g_sw_; |
There was a problem hiding this comment.
I think this should be tau_ssa_g_sw, not tau_g_sw
There was a problem hiding this comment.
In the fortran code, the name of this variable is tau_w_g; see here. I am okay changing this name from tau_g_sw to tau_ssa_g_sw or tau_g_sw to tau_w_g
There was a problem hiding this comment.
This comment // tau_w_g : aerosol asymmetry parameter * tau * w is from here. Is this comment correct?
There was a problem hiding this comment.
I don’t care so much about the _sw suffix, it’s the fact that the Fortran version has tau, w, and g to indicate that it’s a product of tau, w, and g, but the c++ version here only has tau and g in the name. So I read that and think this means it’s a product only of tau and g, which is not correct.
There was a problem hiding this comment.
I changed the name of this variables, from tau_g_sw to tau_ssa_g_sw. Thanks for checking this.
There was a problem hiding this comment.
Looks good! Thanks for clarifying this! It will make the next person who looks through this have an easier time understanding what is going on and what these fields represent!
brhillman
approved these changes
Apr 10, 2024
mahf708
approved these changes
Apr 10, 2024
kaizhangpnl
approved these changes
Apr 10, 2024
singhbalwinder
approved these changes
Apr 10, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
We are adding an aerosol optics interface code to call
MAM4xxaerosol optics code.To enable MAM4xx in EAMxx, we employ the cmake option
SCREAM_ENABLE_MAM. From this PR, this option is set by default toON. It will stayONprovided the impact on the build times is minimal. For global simulations, MAM4xx's aerosol optics calculations can be turned on by addingmam4_opticslabel at runtime in the scream namelist file.The aerosol optics interface computes quantities the
rrtmgpradiation model requires and also computes a diagnostic field aerosol optical depth (AOD).This interface invokes
mam4::aer_rad_props::aer_rad_props_swandmam4::aer_rad_props::aer_rad_props_lwto compute the short and long-wave contributions of the aerosol optics. Therefore, the core computations of the aerosol optics interface are part of theMAM4xxsubmodule. In particular,mam4::aer_rad_props::aer_rad_props_swcomputes the following fields:aero_g_sw,aero_ssa_sw,aero_tau_sw,aero_tau_forward, and the diagnosticaodvis(aerosol optical depth). Furthermore,mam4::aer_rad_props::aer_rad_props_lwcomputesaero_tau_lw.First, this interface includes the necessary machinery to read NetCDF files for aerosol optics using the
scream::AtmosphereInputinterface. Therefore, we added additional tags in FieldTag for the new dimensions required by the aerosol optical NetCDF files. In particular:num_modes: number of total modes.nrefindex_real: number of real refractive index.nrefindex_im: number of imaginary refractive index.ncoef_number: number of coefficients.mode: number of modes in NetCDF file.Note that the difference between
num_modesandmodeis one corresponds to the total number of modes in MAM, i.e., 4, and the second one is the total number of modes in the NetCDF file, e.g., 1.In addition, in the original NetCDF files from E3SMv2, the dimension name of the total number of short waves was
swband, and the total number of long waves waslwband. Because EAMxx has already defined these dimensions but with a different name, we report this issue: FieldLayout with new FieldTag #2602. To overcome this issue, we rename these dimensions in the NetCDF files to match the ones in EAMxx.We reported an issue with scream::AtmosphereInput (AtmosphereInput and non-time dependent variables #2601) when non-time dependent variables of more than two dimensions are employed. As a temporary fix, we modified the aerosol optical NetCDF files by adding a time variable in all variables with more than one dimension.
Testing
We added three standalone tests and one CIME test in EAMxx.
scream/components/eamxx/tests/single-process/mam4_optics: single-process test where onlymam4_opticsis executed.scream/components/eamxx/tests/multi-process/physics_only/mam_optics_rrtmgp. A multi-process test where aerosol optics is executing withrrtmgp.scream/components/eamxx/tests/multi-process/dynamics_physics/homme_shoc_mam_optics_cld_p3_rrtmgp: A multi-process test where aerosol optics is executing withhomme,dynamics,shoc,cld,p3, andrrtmgp. We used the coupled testtests/coupled/homme_shoc_cld_spa_p3_rrtmgpas the base and replaced spa with mam_optics.ne4pg2global model test that uses EAMxx's test mods to invoke mam4xx's aerosol optics codes by modifying scream namelist (see Figure 1).Figure 1. Aerosol optical depth computed by
EMAxx-MAM4xx.Note that these tests check for the correct execution of optics code and nan. Thus, the output's test is not compared with reference values. However, validation tests for checking that C++ code produces equal values that the mam4 Fortran code is part of the mam4xx.
mam4_opticsis name of the atmosphere process for the aerosol optics in EMAxx. Thus, the input.yaml file of each test must contain the following information to enable this process:Testing in MAM4xx.
We have implemented a validation approach in mam4xx that relies on generated validation files in Skywalker format, allowing us to efficiently compare values generated by an E3SM case and the standalone test in mam4xx.
A set of validation files is produced for each Fortran subroutine, and the corresponding test is implemented in mam4xx.
For aerosol optics, we performed 12 tests.
binterp_ts_355calc_diag_spec_ts_355calc_refin_complex_ts_355_lwcalc_refin_complex_ts_355_swcalc_volc_ext_ts_355modal_size_parameters_ts_355_ismethod2_falsemodal_size_parameters_ts_355_ismethod2_truemodal_aero_sw_ts_355modal_aero_lw_ts_355calc_parameterized_ts_355update_aod_spec_ts_355aer_rad_props_lw_ts_355aer_rad_props_sw_ts_355volcanic_cmip_sw_ts_355In these tests
ts_355corresponds to time step 355. Implementation of these tests is here.Compilation and run wall-times.
Enabling MAM4xx and excepting four EMAxx-MAM4xx tests increases ~ 8% (90 seconds) the compilation and run wall-times in weaver for EMAxx. Wall-time of
./scripts/test-all-scream -m weaver -c SCREAM_ENABLE_MAM=OFF -t dbgis 1155 second, i.e., MAM4xx is not enabled. Wall-time of./scripts/test-all-scream -m weaver -c SCREAM_ENABLE_MAM=ON -t dbgis 1245 seconds i.e., MAM4xx is enabled. In both cases, the compilation-run times are for a NVIDIA V100 (GPU).Label Time Summary when MAM4xx is enabled.
Label Time Summary when MAM4xx is not enabled.
Volcanic aerosols.
The following variables:
ssa_cmip6_sw,af_cmip6_sw,ext_cmip6_sw,ext_cmip6_lw_are obtained from netcdf files in mam4 (Fortran version). In mam4xx, we have set these variables to zero and will rely on eamxx to read/interpolate them; however, this work will be part of a future PR.New initial condition files with aerosols and gases
New initial condition (IC) files with aerosols and gas mass mixing ratios have been added. The files are added to the E3SM ANL data server so that it can downloaded to any machine where the tests are run.
The new IC file for the CIME simulations is:
inputdata/atm/scream/init/screami_mam4xx_ne4np4L72_c20240208.ncThe new IC for the unit tests is:
inputdata/atm/scream/init/scream_unit_tests_aerosol_optics_ne2np4L72_20220822.nc