Surface momentum flux over sea in NorESM

[oyvindseland]

I received the following question from Tuomas Naakka

Felix Pithan is leading a study where climate models will be compared against MOSAiC data and he has asked a list of variables for the comparison. One those variables is surface momentum flux on sea ice. I have not found surface momentum flux components as an output from CICE and therefore I would like to kindly ask your help with issue. My question is:

Are surface momentum flux components calculated in CICE or is it possible to use surface momentum fluxes from CAM used instead? Are surface moment flux component from CAM equivalent to surface momentum flux which is directed from atmosphere to sea ice? @tto061

[tto061]

Hi, CAM's TAUX/Y diagnostics are obtained straight from the coupler with consistent time sampling, so they do reflect the surface stress at the surface (with the sign inverted). However, they are aggregated to the atmospheric grid, so over sea ice with concentrations <1 they are an average of ocean-surface and sea-ice-surface stress, plus any land fraction if present, and therefore not representative of the stress on sea-ice alone. In general, TAUX=LANDFRAC*wsx_lnd_i+(1-LANDFRAC)*(ICEFRAC*wsx_ice_i+(1-ICEFRAC)*wsx_ocn_i) where "_i" means conservatively interpolated onto CAM's grid. So e.g. away from land, you could conceivably back out the surface stress on ice from the surface stress diagnosed in the ocean model and the sea-ice concentration, as wsx_ice_i=(TAUX-(1-ICEFRAC)*wsx_ocn_i)/ICEFRAC Needless to say, this would still be all on the wrong grid, potentially averaged over ice-free and icy points. I'm a bit surprised that relevant standard CICE diagnostic fields "f_strairx" and "f_strairy" are actually not activated as default required output in the NorESM simulations. This seems be a default namelist setting that should be corrected. Thomas Toniazzo MISU Stockholms university SE-106 91 Stockholm Sverige

…

On 2023-03-24 15:29, oyvindseland wrote: I received the following question from Tuomas Naakka Felix Pithan is leading a study where climate models will be compared against MOSAiC data and he has asked a list of variables for the comparison. One those variables is surface momentum flux on sea ice. I have not found surface momentum flux components as an output from CICE and therefore I would like to kindly ask your help with issue. My question is: Are surface momentum flux components calculated in CICE or is it possible to use surface momentum fluxes from CAM used instead? Are surface moment flux component from CAM equivalent to surface momentum flux which is directed from atmosphere to sea ice? @tto061 <https://github.com/tto061> — Reply to this email directly, view it on GitHub <#433>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADZGLJGJEMSAWLLHNAABOXLW5WVUXANCNFSM6AAAAAAWGUM6XI>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[JensBDebernard]

In the CMIP runs air-ice stresses are given by the variables sistrxdtop : long_name = "x component of atmospheric stress on sea ice" sistrydtop : long_name = "y component of atmospheric stress on sea ice" They contain the correct ice-area weighted value agreed on in SIMIP, in contrast to "f_strairx" and "f_strairy". These stress fields should exist in all default NorESM experiments, however, I have not checked the CMIP database if they are cmorized. These fields are only available as monthly averages due to the storage constraints we faced on daily output. Jens Boldingh Debernard Norwegian Meteorological Institute

…

On Fri, 24 Mar 2023 at 16:31, tto061 ***@***.***> wrote: Hi, CAM's TAUX/Y diagnostics are obtained straight from the coupler with consistent time sampling, so they do reflect the surface stress at the surface (with the sign inverted). However, they are aggregated to the atmospheric grid, so over sea ice with concentrations <1 they are an average of ocean-surface and sea-ice-surface stress, plus any land fraction if present, and therefore not representative of the stress on sea-ice alone. In general, TAUX=LANDFRAC*wsx_lnd_i+(1-LANDFRAC)*(ICEFRAC*wsx_ice_i+(1-ICEFRAC)*wsx_ocn_i) where "_i" means conservatively interpolated onto CAM's grid. So e.g. away from land, you could conceivably back out the surface stress on ice from the surface stress diagnosed in the ocean model and the sea-ice concentration, as wsx_ice_i=(TAUX-(1-ICEFRAC)*wsx_ocn_i)/ICEFRAC Needless to say, this would still be all on the wrong grid, potentially averaged over ice-free and icy points. I'm a bit surprised that relevant standard CICE diagnostic fields "f_strairx" and "f_strairy" are actually not activated as default required output in the NorESM simulations. This seems be a default namelist setting that should be corrected. Thomas Toniazzo MISU Stockholms university SE-106 91 Stockholm Sverige On 2023-03-24 15:29, oyvindseland wrote: > I received the following question from Tuomas Naakka > > Felix Pithan is leading a study where climate models will be compared > against MOSAiC data and he has asked a list of variables for the > comparison. One those variables is surface momentum flux on sea ice. I > have not found surface momentum flux components as an output from CICE > and therefore I would like to kindly ask your help with issue. My > question is: > > Are surface momentum flux components calculated in CICE or is it > possible to use surface momentum fluxes from CAM used instead? Are > surface moment flux component from CAM equivalent to surface momentum > flux which is directed from atmosphere to sea ice? @tto061 > <https://github.com/tto061> > > — > Reply to this email directly, view it on GitHub > <#433>, or unsubscribe > < https://github.com/notifications/unsubscribe-auth/ADZGLJGJEMSAWLLHNAABOXLW5WVUXANCNFSM6AAAAAAWGUM6XI >. > You are receiving this because you were mentioned.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub <#433 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTH3UVQLCAJBVAX6SLFU4TW5W43NANCNFSM6AAAAAAWGUM6XI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>