Differences between T_S, T_G and T_SNOW – in #14: CCLM-CLM

in #14: CCLM-CLM

<p> Dear <span class="caps"> CCLM </span> - <span class="caps"> CLM </span> users, <br/> Our group has recently developed a fully-coupled ice-sheet – ocean -sea-ice – atmosphere – land model involving <span class="caps"> CCLM </span> _CLM. Our configuration covers the Antarctic. I would like to know what are the differences between the following variables outputted by <span class="caps"> COSMO </span> : T_S (temperature of surface), T_G (temperature at the boundary soil-atmosphere) and T_SNOW (temperature of snow surface). I did not find in the documentation how they are computed. I observe that T_S and T_G provide identical temperatures over the continental ice, but it is not true over the ocean (especially over the sea ice). I am also surprised to see differences up to 20°C in absolute value between T_S and T_SNOW (at most places, the surface is fully covered by snow). <br/> Thank you very much for your input, <br/> Sylvain </p>

  @sylvainmarchi in #3876541

<p> Dear <span class="caps"> CCLM </span> - <span class="caps"> CLM </span> users, <br/> Our group has recently developed a fully-coupled ice-sheet – ocean -sea-ice – atmosphere – land model involving <span class="caps"> CCLM </span> _CLM. Our configuration covers the Antarctic. I would like to know what are the differences between the following variables outputted by <span class="caps"> COSMO </span> : T_S (temperature of surface), T_G (temperature at the boundary soil-atmosphere) and T_SNOW (temperature of snow surface). I did not find in the documentation how they are computed. I observe that T_S and T_G provide identical temperatures over the continental ice, but it is not true over the ocean (especially over the sea ice). I am also surprised to see differences up to 20°C in absolute value between T_S and T_SNOW (at most places, the surface is fully covered by snow). <br/> Thank you very much for your input, <br/> Sylvain </p>

Differences between T_S, T_G and T_SNOW

SM wrote

Dear CCLM - CLM users,
Our group has recently developed a fully-coupled ice-sheet – ocean -sea-ice – atmosphere – land model involving CCLM _CLM. Our configuration covers the Antarctic. I would like to know what are the differences between the following variables outputted by COSMO : T_S (temperature of surface), T_G (temperature at the boundary soil-atmosphere) and T_SNOW (temperature of snow surface). I did not find in the documentation how they are computed. I observe that T_S and T_G provide identical temperatures over the continental ice, but it is not true over the ocean (especially over the sea ice). I am also surprised to see differences up to 20°C in absolute value between T_S and T_SNOW (at most places, the surface is fully covered by snow).
Thank you very much for your input,
Sylvain

View in channel
<p> I do not use the coupled version ans therefore have no experience how the parameters behave there. In the uncoupled version there is a sketch in the subroutine src_soil_multlay.f90: </p> <pre> ! Definition of temperature/water related variables in the soil model ! ! ! --------------T_snow-------- ! | ! fr_snow | 1 - fr_snow ! | ! ______________T_s__________|________________T_s_____________ ! //////////////////////////////////////////////////////////// ! ! Layer k=1 - - - - - T_SO(k) , W_SO(k), W_SO_ICE(k) - - - - ! ! ____________________________________________________________ ! ! . ! . ! . ! ! ------------------------------------------------------------ ! ! ! Layer k=ke_soil+1 - - - (climate layer) - - - - - - - - - - ! ! ! ____________________________________________________________ ! ! ! ! The surface temperature T_g is the snow fraction (fr_snow) ! weighted sum of T_snow and T_s! !============================================================================== </pre>

  @burkhardtrockel in #612d76c

<p> I do not use the coupled version ans therefore have no experience how the parameters behave there. In the uncoupled version there is a sketch in the subroutine src_soil_multlay.f90: </p> <pre> ! Definition of temperature/water related variables in the soil model ! ! ! --------------T_snow-------- ! | ! fr_snow | 1 - fr_snow ! | ! ______________T_s__________|________________T_s_____________ ! //////////////////////////////////////////////////////////// ! ! Layer k=1 - - - - - T_SO(k) , W_SO(k), W_SO_ICE(k) - - - - ! ! ____________________________________________________________ ! ! . ! . ! . ! ! ------------------------------------------------------------ ! ! ! Layer k=ke_soil+1 - - - (climate layer) - - - - - - - - - - ! ! ! ____________________________________________________________ ! ! ! ! The surface temperature T_g is the snow fraction (fr_snow) ! weighted sum of T_snow and T_s! !============================================================================== </pre>
BR replied

I do not use the coupled version ans therefore have no experience how the parameters behave there. In the uncoupled version there is a sketch in the subroutine src_soil_multlay.f90: 

!     Definition of temperature/water related variables in the soil model
!
!
!     --------------T_snow--------
!                                | 
!           fr_snow              |            1 - fr_snow
!                                |      
!     ______________T_s__________|________________T_s_____________
!     ////////////////////////////////////////////////////////////
!
!     Layer k=1 - - - - -  T_SO(k) , W_SO(k), W_SO_ICE(k) - - - -
!
!     ____________________________________________________________
!                                
!                                .
!                                .
!                                . 
!
!     ------------------------------------------------------------
!
!
!     Layer k=ke_soil+1 - - - (climate layer) - - - - - - - - - -
!
!
!     ____________________________________________________________
!
!
!
!     The surface temperature T_g is the snow fraction (fr_snow)
!     weighted sum of T_snow and T_s!
!==============================================================================
View in channel
<p> Thanks for replying Burkhardt. Given that the surface temperature T_s is the snow fraction weighted sum of T_snow and T_s and that each model cell in Antarctica is almost 100% covered by snow, one can expect T_g to be close to T_s then? </p>

  @sylvainmarchi in #7fed132

<p> Thanks for replying Burkhardt. Given that the surface temperature T_s is the snow fraction weighted sum of T_snow and T_s and that each model cell in Antarctica is almost 100% covered by snow, one can expect T_g to be close to T_s then? </p>
SM replied

Thanks for replying Burkhardt. Given that the surface temperature T_s is the snow fraction weighted sum of T_snow and T_s and that each model cell in Antarctica is almost 100% covered by snow, one can expect T_g to be close to T_s then?

View in channel
<p> Over areas covered 100% by snow I would expect T_G to be the same as T_SNOW. T_S is the temperature at the earth surface under the snow this can be much different than the temperature at the snow surface T_SNOW. </p>

  @burkhardtrockel in #a19d5e4

<p> Over areas covered 100% by snow I would expect T_G to be the same as T_SNOW. T_S is the temperature at the earth surface under the snow this can be much different than the temperature at the snow surface T_SNOW. </p>
BR replied

Over areas covered 100% by snow I would expect T_G to be the same as T_SNOW. T_S is the temperature at the earth surface under the snow this can be much different than the temperature at the snow surface T_SNOW.

View in channel