Update documentation including namelist tables #322

apcraig · 2020-06-05T17:54:00Z

PR checklist

Since CICE-Consortium/CICE@83686a3f (Implement box model test from 2001 JCP paper (CICE-Consortium#151), 2018-10-22) one can set 'ktherm = -1' in the CICE namelist to disable the thermodynamics. This was added specifically for the "Box 2001" test case (configuration/scripts/options/set_nml.box2001 in CICE) which also sets 'calc_strair' to false. 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone, since the modifications to CICE_RunMod::ice_step in CICE-Consortium/CICE@83686a3f were not copied to the Icepack driver in icedrv_RunMod::ice_step. A simple implementation in Icepack would be to add the CICE driver logic added in CICE-Consortium/CICE@83686a3f in the Icepack driver. However, as explained below things are not as simple. It can be useful to disable thermodynamics but run the model with 'calc_strair = .true.'. This currently does not work as could be expected in CICE, since the atmospheric stresses are computed in icepack_atmo::icepack_atm_boundary, which is called from icepack_therm_vertical::icepack_step_therm1, itself called from ice_step_mod::step_therm1, which is only called if 'ktherm >= 0' (since CICE-Consortium/CICE@83686a3f). This leads to atmospheric stress being always zero under 'calc_strair = .true.', 'ktherm = -1'. In order to allow computing the atmospheric stresses in this case, add some logic the icepack_therm_vertical::icepack_step_therm1 to run some parts of the subroutine only if 'ktherm >= 0'. Namely, let frzmlt_bottom_lateral, set_sfcflux, thermo_vertical, update_aerosol, update_isotope and compute_ponds_{cesm,lvl,topo} only be called if 'ktherm >=0'. Conversely, always run neutral_drag_coeffs, icepack_atm_boundary, increment_age, update_FYarea and merge_fluxes. This commit does changes behaviour for potential users of Icepack as a library who initialize Icepack with 'ktherm=-1', 'calc_strair=.true.' and call 'icepack_step_therm1', as this now will yield non-zero atmospheric stresses. However, We do not anticipate this change to break a lot of users, since anyway setting 'ktherm=-1' before this commit had no effect whatsoever. The new behaviour aruably makes more sense physically. An upcoming commit will implement support for setting 'ktherm=-1' in the Icepack driver. Since we modify the indentation of several lines to keep the code readable, this commit is best viewed using 'git diff --ignore-all-space'.

As noted in the previous commit, 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone. Update icedrv_RunMod::ice_step to call some subroutines only if the thermodynamics is enabled ('ktherm >= 0'). Namely, let prep_radiation, biogeochemistry, step_therm2 and step_radiation only be called when thermodynamics is enabled, while letting the rest of the subroutines always be called. As noted in the previous commit, 'step_therm1' in particular has to be always called since the computation of the atmospheric stresses are done in that subroutine.

Since CICE-Consortium/CICE@83686a3f (Implement box model test from 2001 JCP paper (CICE-Consortium#151), 2018-10-22) one can set 'ktherm = -1' in the CICE namelist to disable the thermodynamics. This was added specifically for the "Box 2001" test case (configuration/scripts/options/set_nml.box2001 in CICE) which also sets 'calc_strair' to false. 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone, since the modifications to CICE_RunMod::ice_step in CICE-Consortium/CICE@83686a3f were not copied to the Icepack driver in icedrv_RunMod::ice_step. A simple implementation in Icepack would be to add the CICE driver logic added in CICE-Consortium/CICE@83686a3f in the Icepack driver. However, as explained below things are not as simple. It can be useful to disable thermodynamics but run the model with 'calc_strair = .true.'. This currently does not work as could be expected in CICE, since the atmospheric stresses are computed in icepack_atmo::icepack_atm_boundary, which is called from icepack_therm_vertical::icepack_step_therm1, itself called from ice_step_mod::step_therm1, which is only called if 'ktherm >= 0' (since CICE-Consortium/CICE@83686a3f). This leads to atmospheric stress being always zero under 'calc_strair = .true.', 'ktherm = -1'. In order to allow computing the atmospheric stresses in this case, add some logic the icepack_therm_vertical::icepack_step_therm1 to run some parts of the subroutine only if 'ktherm >= 0'. Namely, let the advanced snow physics, frzmlt_bottom_lateral, set_sfcflux, thermo_vertical, update_aerosol, update_isotope and compute_ponds_{cesm,lvl,topo} only be called if 'ktherm >=0'. Conversely, always run neutral_drag_coeffs, icepack_atm_boundary, increment_age, update_FYarea and merge_fluxes. This commit does changes behaviour for potential users of Icepack as a library who initialize Icepack with 'ktherm=-1', 'calc_strair=.true.' and call 'icepack_step_therm1', as this now results in most of the thermodynamics being disabled, and will yield non-zero atmospheric stresses. However, we do not anticipate this change to break a lot of users, since anyway setting 'ktherm=-1' before this commit had no effect whatsoever in Icepack. The new behaviour aruably makes more sense physically. An upcoming commit will implement support for setting 'ktherm=-1' in the Icepack driver. Since we modify the indentation of several lines to keep the code readable, this commit is best viewed using 'git diff --ignore-all-space'.

As noted in the previous commit, 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone. Update icedrv_RunMod::ice_step to call some subroutines only if the thermodynamics is enabled ('ktherm >= 0'). Namely, let prep_radiation, biogeochemistry, step_therm2 and step_radiation only be called when thermodynamics is enabled, while letting the rest of the subroutines always be called. As noted in the previous commit, 'step_therm1' in particular has to be always called since the computation of the atmospheric stresses are done in that subroutine.

Since CICE-Consortium/CICE@83686a3f (Implement box model test from 2001 JCP paper (CICE-Consortium#151), 2018-10-22) one can set 'ktherm = -1' in the CICE namelist to disable the thermodynamics. This was added specifically for the "Box 2001" test case (configuration/scripts/options/set_nml.box2001 in CICE) which also sets 'calc_strair' to false. 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone, since the modifications to CICE_RunMod::ice_step in CICE-Consortium/CICE@83686a3f were not copied to the Icepack driver in icedrv_RunMod::ice_step. A simple implementation in Icepack would be to add the CICE driver logic added in CICE-Consortium/CICE@83686a3f in the Icepack driver. However, as explained below things are not as simple. It can be useful to disable thermodynamics but run the model with 'calc_strair = .true.'. This currently does not work as could be expected in CICE, since the atmospheric stresses are computed in icepack_atmo::icepack_atm_boundary, which is called from icepack_therm_vertical::icepack_step_therm1, itself called from ice_step_mod::step_therm1, which is only called if 'ktherm >= 0' (since CICE-Consortium/CICE@83686a3f). This leads to atmospheric stress being always zero under 'calc_strair = .true.', 'ktherm = -1'. In order to allow computing the atmospheric stresses in this case, add some logic the icepack_therm_vertical::icepack_step_therm1 to run some parts of the subroutine only if 'ktherm >= 0'. Namely, let frzmlt_bottom_lateral, set_sfcflux, thermo_vertical, update_aerosol, update_isotope and compute_ponds_{cesm,lvl,topo} only be called if 'ktherm >=0'. Conversely, always run neutral_drag_coeffs, icepack_atm_boundary, increment_age, update_FYarea and merge_fluxes. This commit does changes behaviour for potential users of Icepack as a library who initialize Icepack with 'ktherm=-1', 'calc_strair=.true.' and call 'icepack_step_therm1', as this now will yield non-zero atmospheric stresses. However, We do not anticipate this change to break a lot of users, since anyway setting 'ktherm=-1' before this commit had no effect whatsoever. The new behaviour aruably makes more sense physically. An upcoming commit will implement support for setting 'ktherm=-1' in the Icepack driver. Since we modify the indentation of several lines to keep the code readable, this commit is best viewed using 'git diff --ignore-all-space'.

As noted in the previous commit, 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone. Update icedrv_RunMod::ice_step to call some subroutines only if the thermodynamics is enabled ('ktherm >= 0'). Namely, let prep_radiation, biogeochemistry, step_therm2 and step_radiation only be called when thermodynamics is enabled, while letting the rest of the subroutines always be called. As noted in the previous commit, 'step_therm1' in particular has to be always called since the computation of the atmospheric stresses are done in that subroutine.

Since CICE-Consortium/CICE@83686a3f (Implement box model test from 2001 JCP paper (CICE-Consortium#151), 2018-10-22) one can set 'ktherm = -1' in the CICE namelist to disable the thermodynamics. This was added specifically for the "Box 2001" test case (configuration/scripts/options/set_nml.box2001 in CICE) which also sets 'calc_strair' to false. 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone, since the modifications to CICE_RunMod::ice_step in CICE-Consortium/CICE@83686a3f were not copied to the Icepack driver in icedrv_RunMod::ice_step. A simple implementation in Icepack would be to add the CICE driver logic added in CICE-Consortium/CICE@83686a3f in the Icepack driver. However, as explained below things are not as simple. It can be useful to disable thermodynamics but run the model with 'calc_strair = .true.'. This currently does not work as could be expected in CICE, since the atmospheric stresses are computed in icepack_atmo::icepack_atm_boundary, which is called from icepack_therm_vertical::icepack_step_therm1, itself called from ice_step_mod::step_therm1, which is only called if 'ktherm >= 0' (since CICE-Consortium/CICE@83686a3f). This leads to atmospheric stress being always zero under 'calc_strair = .true.', 'ktherm = -1'. In order to allow computing the atmospheric stresses in this case, add some logic the icepack_therm_vertical::icepack_step_therm1 to run some parts of the subroutine only if 'ktherm >= 0'. Namely, let frzmlt_bottom_lateral, set_sfcflux, thermo_vertical, update_aerosol, update_isotope, drain_snow and compute_ponds_{cesm,lvl,topo} only be called if 'ktherm >=0'. Conversely, always run neutral_drag_coeffs, icepack_atm_boundary, increment_age, update_FYarea and merge_fluxes. This commit does changes behaviour for potential users of Icepack as a library who initialize Icepack with 'ktherm=-1', 'calc_strair=.true.' and call 'icepack_step_therm1', as this now will yield non-zero atmospheric stresses. However, We do not anticipate this change to break a lot of users, since anyway setting 'ktherm=-1' before this commit had no effect whatsoever. The new behaviour aruably makes more sense physically. An upcoming commit will implement support for setting 'ktherm=-1' in the Icepack driver. Note that we use one line 'if' statements and avoid adjusting indentation in order to reduce the risk of merge conflicts when this commit is rebased onto a newer version of CICE.

As noted in the previous commit, 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone. Update icedrv_RunMod::ice_step to call some subroutines only if the thermodynamics is enabled ('ktherm >= 0'). Namely, let prep_radiation, biogeochemistry, step_therm2 and step_radiation only be called when thermodynamics is enabled, while letting the rest of the subroutines always be called. As noted in the previous commit, 'step_therm1' in particular has to be always called since the computation of the atmospheric stresses are done in that subroutine.

Since CICE-Consortium/CICE@83686a3f (Implement box model test from 2001 JCP paper (CICE-Consortium#151), 2018-10-22) one can set 'ktherm = -1' in the CICE namelist to disable the thermodynamics. This was added specifically for the "Box 2001" test case (configuration/scripts/options/set_nml.box2001 in CICE) which also sets 'calc_strair' to false. 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone, since the modifications to CICE_RunMod::ice_step in CICE-Consortium/CICE@83686a3f were not copied to the Icepack driver in icedrv_RunMod::ice_step. A simple implementation in Icepack would be to add the CICE driver logic added in CICE-Consortium/CICE@83686a3f in the Icepack driver. However, as explained below things are not as simple. It can be useful to disable thermodynamics but run the model with 'calc_strair = .true.'. This currently does not work as could be expected in CICE, since the atmospheric stresses are computed in icepack_atmo::icepack_atm_boundary, which is called from icepack_therm_vertical::icepack_step_therm1, itself called from ice_step_mod::step_therm1, which is only called if 'ktherm >= 0' (since CICE-Consortium/CICE@83686a3f). This leads to atmospheric stress being always zero under 'calc_strair = .true.', 'ktherm = -1'. In order to allow computing the atmospheric stresses in this case, add some logic the icepack_therm_vertical::icepack_step_therm1 to run some parts of the subroutine only if 'ktherm >= 0'. Namely, let frzmlt_bottom_lateral, set_sfcflux, thermo_vertical, update_aerosol, update_isotope, drain_snow and compute_ponds_{cesm,lvl,topo} only be called if 'ktherm >=0'. Conversely, always run neutral_drag_coeffs, icepack_atm_boundary, increment_age, update_FYarea and merge_fluxes. This commit does changes behaviour for potential users of Icepack as a library who initialize Icepack with 'ktherm=-1', 'calc_strair=.true.' and call 'icepack_step_therm1', as this now will yield non-zero atmospheric stresses. However, We do not anticipate this change to break a lot of users, since anyway setting 'ktherm=-1' before this commit had no effect whatsoever. The new behaviour aruably makes more sense physically. An upcoming commit will implement support for setting 'ktherm=-1' in the Icepack driver. Note that we use one line 'if' statements and avoid adjusting indentation in order to reduce the risk of merge conflicts when this commit is rebased onto a newer version of CICE.

As noted in the previous commit, 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone. Update icedrv_RunMod::ice_step to call some subroutines only if the thermodynamics is enabled ('ktherm >= 0'). Namely, let prep_radiation, biogeochemistry, step_therm2 and step_radiation only be called when thermodynamics is enabled, while letting the rest of the subroutines always be called. As noted in the previous commit, 'step_therm1' in particular has to be always called since the computation of the atmospheric stresses are done in that subroutine.

Since CICE-Consortium/CICE@83686a3f (Implement box model test from 2001 JCP paper (CICE-Consortium#151), 2018-10-22) one can set 'ktherm = -1' in the CICE namelist to disable the thermodynamics. This was added specifically for the "Box 2001" test case (configuration/scripts/options/set_nml.box2001 in CICE) which also sets 'calc_strair' to false. 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone, since the modifications to CICE_RunMod::ice_step in CICE-Consortium/CICE@83686a3f were not copied to the Icepack driver in icedrv_RunMod::ice_step. A simple implementation in Icepack would be to add the CICE driver logic added in CICE-Consortium/CICE@83686a3f in the Icepack driver. However, as explained below things are not as simple. It can be useful to disable thermodynamics but run the model with 'calc_strair = .true.'. This currently does not work as could be expected in CICE, since the atmospheric stresses are computed in icepack_atmo::icepack_atm_boundary, which is called from icepack_therm_vertical::icepack_step_therm1, itself called from ice_step_mod::step_therm1, which is only called if 'ktherm >= 0' (since CICE-Consortium/CICE@83686a3f). This leads to atmospheric stress being always zero under 'calc_strair = .true.', 'ktherm = -1'. In order to allow computing the atmospheric stresses in this case, add some logic the icepack_therm_vertical::icepack_step_therm1 to run some parts of the subroutine only if 'ktherm >= 0'. Namely, let frzmlt_bottom_lateral, set_sfcflux, thermo_vertical, update_aerosol, update_isotope, drain_snow and compute_ponds_{cesm,lvl,topo} only be called if 'ktherm >=0'. Conversely, always run neutral_drag_coeffs, icepack_atm_boundary, increment_age, update_FYarea and merge_fluxes. This commit does changes behaviour for potential users of Icepack as a library who initialize Icepack with 'ktherm=-1', 'calc_strair=.true.' and call 'icepack_step_therm1', as this now will yield non-zero atmospheric stresses. However, We do not anticipate this change to break a lot of users, since anyway setting 'ktherm=-1' before this commit had no effect whatsoever. The new behaviour aruably makes more sense physically. An upcoming commit will implement support for setting 'ktherm=-1' in the Icepack driver. Note that we use one line 'if' statements and avoid adjusting indentation in order to reduce the risk of merge conflicts when this commit is rebased onto a newer version of CICE.

As noted in the previous commit, 'ktherm = -1' is documented as disabling the thermodynamics in the Icepack documentation since 2b27a78 (Update documentation including namelist tables (CICE-Consortium#322), 2020-06-05), although this capability does not exist in Icepack standalone. Update icedrv_RunMod::ice_step to call some subroutines only if the thermodynamics is enabled ('ktherm >= 0'). Namely, let prep_radiation, biogeochemistry, step_therm2 and step_radiation only be called when thermodynamics is enabled, while letting the rest of the subroutines always be called. As noted in the previous commit, 'step_therm1' in particular has to be always called since the computation of the atmospheric stresses are done in that subroutine.

apcraig added 2 commits June 5, 2020 10:31

update documentation including namelist tables

d7e63a8

update documentation

dfa6245

apcraig requested review from eclare108213 and duvivier June 5, 2020 17:54

apcraig added 2 commits June 5, 2020 11:10

update documentation

a115455

update version

b975252

apcraig merged commit 2b27a78 into CICE-Consortium:master Jun 5, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update documentation including namelist tables #322

Update documentation including namelist tables #322

apcraig commented Jun 5, 2020 •

edited

Loading

Update documentation including namelist tables #322

Update documentation including namelist tables #322

Conversation

apcraig commented Jun 5, 2020 • edited Loading

PR checklist

apcraig commented Jun 5, 2020 •

edited

Loading