Ice shelf-specific forcing depths by angus-g · Pull Request #113 · ACCESS-NRI/MOM6

angus-g · 2026-10-26T07:30:29Z

Allow specification of different depths for a) the calculation of the surface boundary layer that feeds into the melt parameterisation; b) the application of the melt fluxes. There are three new parameters:

SHELF_FORCING_DEPTH: the equivalent to MINIMUM_FORCING_DEPTH applied in shelf regions;
HMIX_SHELF_PROP; the equivalent to HMIX_SFC_PROP (for T/S) applied in shelf regions;
HMIX_UV_SHELF_PROP; the equivalent to HMIX_UV_SFC_PROP (for velocity) applied in shelf regions.

Supersedes #78.

The boundary layer used for the melt parameterisation must match the depth over which the forcing is applied. The only controls for these are global: HMIX_SFC_PROP/HMIX_UV_SFC_PROP control the boundary layer for computing the parameterisation, and MINIMUM_FORCING_DEPTH controls the depth over which the melt is applied. We introduce new optional parameters to applyBoundaryFluxesInOut() to modify IforcingDepthScale to use a different `shelf_forcing_depth` according to the fraction of the cell covered by ice shelf.

This is the analogue to MINIMUM_FORCING_DEPTH, applied under ice shelves. It is only read when an ice shelf coverage field is provided to the diabatic driver initialisation routine.

This only affects the path when using the non-legacy ALE diabatic driver. In this case, the frac_shelf_h array is passed through, and used to apply a different forcing depth as required.

This adds HMIX_SHELF_PROP and HMIX_UV_SHELF_PROP, which are analogues to the `SFC` parameters. These determine the boundary layer depth for near-shelf temperature, salinity and velocities, which are used for calculation of melt parameterisation.

This now takes frac_shelf_h into account when computing depth_ml used for calculating the surface state.

dougiesquire

Thanks @angus-g. A few questions, comments and suggestions.

Also I think these changes need to be included: 9a6fa99, e7ecd85?

dougiesquire · 2026-10-27T01:09:35Z

@@ -4147,6 +4186,14 @@ subroutine extract_surface_state(CS, sfc_state_in)
          sfc_state%v(i,J) = 0.0
        enddo
        do k=1,nz ; do i=is,ie
+          if (use_iceshelves) then
+            frac_shelf_v = 0.0
+            if (G%areaT(i,j) + G%areaT(i,j+1) > 0.0) &
+              frac_shelf_v = (   CS%frac_shelf_h(i,j) * G%areaT(i,j) &
+                               + CS%frac_shelf_h(i,j+1) * G%areaT(i,j+1))&
+                             / (G%areaT(i,j) + G%areaT(i,j+1))
+            depth_ml = frac_shelf_v * CS%Hmix_UV_shelf + (1. - frac_shelf_v) * CS%Hmix_UV
+          endif
          hv = 0.5 * (h(i,j,k) + h(i,j+1,k)) * H_rescale
          if (depth(i) + hv < depth_ml) then
            dh = hv
@@ -4171,6 +4218,14 @@ subroutine extract_surface_state(CS, sfc_state_in)
          sfc_state%u(I,j) = 0.0
        enddo
        do k=1,nz ; do I=is-1,ie
+          if (use_iceshelves) then
+            frac_shelf_u = 0.0
+            if (G%areaT(i,j) + G%areaT(i+1,j) > 0.0) &
+              frac_shelf_u = (   CS%frac_shelf_h(i,j) * G%areaT(i,j) &
+                               + CS%frac_shelf_h(i+1,j) * G%areaT(i+1,j))&
+                             / (G%areaT(i+1,j) + G%areaT(i+1,j))
+            depth_ml = frac_shelf_u * CS%Hmix_UV_shelf + (1. - frac_shelf_u) * CS%Hmix_UV
+          endif


I think the loops' OpenMP directives needs updating for the new variables.

dougiesquire · 2026-10-27T01:11:00Z

+            if (G%areaT(i,j) + G%areaT(i+1,j) > 0.0) &
+              frac_shelf_u = (   CS%frac_shelf_h(i,j) * G%areaT(i,j) &
+                               + CS%frac_shelf_h(i+1,j) * G%areaT(i+1,j))&
+                             / (G%areaT(i+1,j) + G%areaT(i+1,j))


Suggested change

/ (G%areaT(i+1,j) + G%areaT(i+1,j))

/ (G%areaT(i,j) + G%areaT(i+1,j))

dougiesquire · 2026-10-27T01:41:45Z

+          if (use_iceshelves) then
+            frac_shelf_v = 0.0
+            if (G%areaT(i,j) + G%areaT(i,j+1) > 0.0) &
+              frac_shelf_v = (   CS%frac_shelf_h(i,j) * G%areaT(i,j) &
+                               + CS%frac_shelf_h(i,j+1) * G%areaT(i,j+1))&
+                             / (G%areaT(i,j) + G%areaT(i,j+1))
+            depth_ml = frac_shelf_v * CS%Hmix_UV_shelf + (1. - frac_shelf_v) * CS%Hmix_UV
+          endif


I'm confused here. Say we have HMIX_UV_SHELF_PROP = 0 and we're at a point fully under the shelf (ie frac_shelf_v = 1). That would give depth_ml = 0, so dh=0, so sfc_state%v stays 0. So doesn't the default HMIX_UV_SHELF_PROP = 0 actually set the surface velocity to zero, rather than doing no averaging.

Do we need to widen the if (CS%Hmix_UV>0.) then guard around this loop?

Actually just widening the guard probably wouldn't be enough, but you get the idea

Also, because this sits inside the if (CS%Hmix_UV>0.) then, HMIX_UV_SHELF_PROP does nothing unless HMIX_UV_SFC_PROP > 0. So it's not possible to average velocities under ice-shelves but not in the open ocean.

dougiesquire · 2026-10-27T01:59:12Z

+              frac_shelf_u = (   CS%frac_shelf_h(i,j) * G%areaT(i,j) &
+                               + CS%frac_shelf_h(i+1,j) * G%areaT(i+1,j))&
+                             / (G%areaT(i+1,j) + G%areaT(i+1,j))
+            depth_ml = frac_shelf_u * CS%Hmix_UV_shelf + (1. - frac_shelf_u) * CS%Hmix_UV


Would writing like this be better for preserving answers when HMIX_UV_SHELF_PROP == HMIX_UV_SFC_PROP:

Suggested change

depth_ml = frac_shelf_u * CS%Hmix_UV_shelf + (1. - frac_shelf_u) * CS%Hmix_UV

depth_ml = CS%Hmix_UV + frac_shelf_u * (CS%Hmix_UV_shelf - CS%Hmix_UV)

dougiesquire · 2026-10-27T02:01:53Z

+              frac_shelf_v = (   CS%frac_shelf_h(i,j) * G%areaT(i,j) &
+                               + CS%frac_shelf_h(i,j+1) * G%areaT(i,j+1))&
+                             / (G%areaT(i,j) + G%areaT(i,j+1))
+            depth_ml = frac_shelf_v * CS%Hmix_UV_shelf + (1. - frac_shelf_v) * CS%Hmix_UV


Would writing like this be better for preserving answers when HMIX_UV_SHELF_PROP == HMIX_UV_SFC_PROP:

Suggested change

depth_ml = frac_shelf_v * CS%Hmix_UV_shelf + (1. - frac_shelf_v) * CS%Hmix_UV

depth_ml = CS%Hmix_UV + frac_shelf_v * (CS%Hmix_UV_shelf - CS%Hmix_UV)

dougiesquire · 2026-10-27T02:33:35Z

  type(diabatic_CS),                          pointer       :: CS       !< module control structure
  type(stochastic_CS),                        pointer       :: stoch_CS !< stochastic control structure
  type(Wave_parameters_CS),                   pointer       :: Waves    !< Surface gravity waves
+  real, dimension(:,:),             optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]


Suggested change

real, dimension(:,:), optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]

real, dimension(SZI_(G),SZJ_(G)), optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]

dougiesquire · 2026-10-27T02:34:24Z

  type(oda_incupd_CS),     pointer       :: oda_incupd_CSp   !< pointer to the ocean data assimilation incremental
                                                             !! update module control structure
  type(int_tide_CS),       pointer       :: int_tide_CSp     !< pointer to the internal tide structure
+  real, dimension(:,:), optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]


Suggested change

real, dimension(:,:), optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]

real, dimension(SZI_(G),SZJ_(G)), optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]

Also, could we pass a logical here instead of the full array (Or just read ICE_SHELF directly from param_file)?

dougiesquire · 2026-10-27T02:39:59Z

                 optional, pointer       :: MLD_h !< Mixed layer thickness for brine plumes [H ~> m or kg m-2]
+  real, dimension(SZI_(G),SZJ_(G)), &
+                  optional, intent(in)   :: frac_shelf_h !< Fractional ice shelf coverage [nondim]
+  real, optional, intent(in)             :: shelf_forcing_depth !< Flux depth for shelf regions


Needs units:

Suggested change

real, optional, intent(in) :: shelf_forcing_depth !< Flux depth for shelf regions

real, optional, intent(in) :: shelf_forcing_depth !< Flux depth for shelf regions [H ~> m or kg m-2].

dougiesquire · 2026-10-27T02:41:49Z

  type(group_pass_type) :: tmp_pass_uv_T_S_h, pass_uv_T_S_h

  real    :: Hmix_z, Hmix_UV_z ! Temporary variables with averaging depths [Z ~> m]
+  real    :: Hmix_shelf_z, Hmix_UV_shelf_z


Suggested change

real :: Hmix_shelf_z, Hmix_UV_shelf_z

real :: Hmix_shelf_z, Hmix_UV_shelf_z ! Temporary variables with ice-shelf averaging depths [Z ~> m]

dougiesquire · 2026-10-27T02:43:28Z

  logical :: use_iceshelves
  character(240) :: msg
  integer :: turns    ! Number of quarter turns
+  real :: frac_shelf_u, frac_shelf_v


Suggested change

real :: frac_shelf_u, frac_shelf_v

real :: frac_shelf_u, frac_shelf_v ! Ice shelf area fraction at u and v points [nondim]

angus-g added 5 commits October 14, 2026 15:47

Read the SHELF_FORCING_DEPTH parameter

c4a0000

This is the analogue to MINIMUM_FORCING_DEPTH, applied under ice shelves. It is only read when an ice shelf coverage field is provided to the diabatic driver initialisation routine.

Pass ice shelf fraction into diabatic routine

65c4e34

This only affects the path when using the non-legacy ALE diabatic driver. In this case, the frac_shelf_h array is passed through, and used to apply a different forcing depth as required.

Add shelf variants of HMIX

42334c2

This adds HMIX_SHELF_PROP and HMIX_UV_SHELF_PROP, which are analogues to the `SFC` parameters. These determine the boundary layer depth for near-shelf temperature, salinity and velocities, which are used for calculation of melt parameterisation.

Split surface calculation between shelf and open ocean

8a93ac8

This now takes frac_shelf_h into account when computing depth_ml used for calculating the surface state.

angus-g requested a review from dougiesquire October 26, 2026 07:30

dougiesquire reviewed Oct 27, 2026

View reviewed changes

	/ (G%areaT(i+1,j) + G%areaT(i+1,j))
	/ (G%areaT(i,j) + G%areaT(i+1,j))

	depth_ml = frac_shelf_u * CS%Hmix_UV_shelf + (1. - frac_shelf_u) * CS%Hmix_UV
	depth_ml = CS%Hmix_UV + frac_shelf_u * (CS%Hmix_UV_shelf - CS%Hmix_UV)

	depth_ml = frac_shelf_v * CS%Hmix_UV_shelf + (1. - frac_shelf_v) * CS%Hmix_UV
	depth_ml = CS%Hmix_UV + frac_shelf_v * (CS%Hmix_UV_shelf - CS%Hmix_UV)

	real, dimension(:,:), optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]
	real, dimension(SZI_(G),SZJ_(G)), optional, intent(in) :: frac_shelf_h !< Fractional ice shelf coverage [nondim]

	real, optional, intent(in) :: shelf_forcing_depth !< Flux depth for shelf regions
	real, optional, intent(in) :: shelf_forcing_depth !< Flux depth for shelf regions [H ~> m or kg m-2].

	real :: Hmix_shelf_z, Hmix_UV_shelf_z
	real :: Hmix_shelf_z, Hmix_UV_shelf_z ! Temporary variables with ice-shelf averaging depths [Z ~> m]

	real :: frac_shelf_u, frac_shelf_v
	real :: frac_shelf_u, frac_shelf_v ! Ice shelf area fraction at u and v points [nondim]

Conversation

angus-g commented Oct 26, 2026

Uh oh!

dougiesquire left a comment

Choose a reason for hiding this comment

Uh oh!

dougiesquire Oct 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

dougiesquire Oct 27, 2026 •

edited

Loading