Standard names: Clarification of definition of "depth"

[martinjuckes]

Proposer's name Martin Juckes.

Date 16 Dec 2025

This is a question of clarification, not a proposal for a new term.

- Term depth

- Description Current description: "Depth is the vertical distance below the surface."

- Units m

The question is whether there is a precise interpretation of "surface" intended here? Is it, for instance, the top of the liquid column (i.e. the interface between sea water and ice or atmosphere)? Or is it intended to be flexible?

Within the CMIP7 data request, sea surface in sea_surface_height_above_geoid is interpreted as top of a liquid water equivalent column rather than the top of the liquid water column.

[github-actions]

Thank you for your proposal. These terms will be added to the cfeditor (http://cfeditor.ceda.ac.uk/proposals/1) shortly. Your proposal will then be reviewed and commented on by the community and Standard Names moderator.

[JonathanGregory]

Dear @martinjuckes

As briefly stated in the guidelines,

The surface called "surface" means the lower boundary of the atmosphere.

I believe that we intend to use it consistently like that, except that seasonal surface_snow is on top of the surface, rather than beneath it. In regions of sea ice, the surface is the top of the sea ice. Perhaps we ought to add the definition of surface to every standard name which mentions it.

The phrase sea_surface is isn't entirely consistent e.g. sea_surface_temperature is not the temperature exactly at the liquid water surface, but a bulk property. I think it does make sense to interpret sea_surface_height_above_geoid in the way you state: as referring to the level of the top of the liquid water equivalent column rather than the top of the liquid water column (and not the bottom of the atmosphere or the top of any floating ice). This is what's advised by the paper on sea level terminology followed by the IPCC AR6, which says:

In ocean areas with floating ice, the liquid surface boundary is the bottom of the ice. For those areas, we define the SSH as the liquid-water equivalent sea surface which the liquid would have if the ice were replaced by an equal mass of sea water of the density of the surface water in its vicinity.

Best wishes

Jonathan

[ChrisBarker-NOAA]

The question is whether there is a precise interpretation of "surface" intended here? Is it, for instance, the top of the liquid column (i.e. the interface between sea water and ice or atmosphere)? Or is it intended to be flexible?

And another option: "depth" can be the distance from the mean water level, depending on how smart the system is about knowing the water level fluctuations :-(

Example: in particle tracking models, the particles are at a depth -- but the model is probably not using a reference frame that follows the water surface.

Another complication is that the water surface (even without ice) varies on different timescales -- e.g. tidal frequency (hours) is common, but also wave frequency (seconds) -- which is meant in a given context?

Tricky stuff!

[JonathanGregory]

@ChrisBarker-NOAA. Yes, these words are used with several meanings. In CF standard names, unqualified depth always means depth below the unqualified surface. If it's depth below any other surface or level, it's stated explicitly e.g. depth_below_geoid. Also, following the sea-level terminology paper, sea_surface means an instantaneous or time-varying quantity (possibly some temporal statistic described by cell_methods), while mean_sea_level means a long-term datum. The latter is unavoidably vague, since sea level varies wrt the geoid on all timescales, as you say.

[taylor13]

Just to note that the term "surface" in the standard name of an ocean variable is, I think, usually the upper liquid surface of the water (not the "lower boundary of the atmosphere." For example, the description of sea_surface_density reads "Sea surface density is the density of sea water near the surface (including the part under sea-ice, if any)."

Perhaps we should consider noting that in the guidelines quoted above (#296 (comment)).

[JonathanGregory]

Yes, sorry not to be clear in what I wrote to Martin and Chris. The plain surface means the bottom of the atmosphere. The sea_surface isn't exactly a subset of the surface, though. It has various meanings. Sometimes it's the bottom of the atmosphere, but not always. In sea_surface_height, the liquid-water-equivalent is the best definition, I think, and that's different from sea_surface_temperature, for example.

Do you think that sea_surface_density and sea_surface_salinity refer to a bulk property of the same "upper layer" as sea_surface_temperature does?

[martinjuckes]

Hi @JonathanGregory , @taylor13 ,

The oceanographers requesting data from CMIP7 have asked for sea_surface_salinity, along with pH and velocity, at the top of the liquid column. We don't appear to have sea_surface_density in the request.

Many variables, particularly air-sea fluxes, are explicitly defined to be at the lower boundary of the atmosphere: e.g. surface_downward_heat_flux_in_sea_water.

For many, that standard name appears to be deliberately vague: e.g. sea_surface_salinity is salinity close to the sea surface. This appears to differ from atmospheric usage in which near surface variables use the same standard names as bulk variables. E.g. near-surface air temperature is air_temperature.

So I think we can conclude from Jonathan's comment #296 (comment) that reference surface for depth is not defined in CF and we would need to add a definition in CMIP7 if we want clarity and consistency.

[JonathanGregory]

Dear @martinjuckes

Many variables, particularly air-sea fluxes, are explicitly defined to be at the lower boundary of the atmosphere: e.g. surface_downward_heat_flux_in_sea_water.

Because it's in_sea_water, I think that example (and others like it) must be at the top of the liquid water, not the bottom of the atmosphere, in areas of sea ice. What does the CMIP7 request say about this?

In the past there has been a lot of discussion about SST. Like sea_surface_salinity, sea_surface_temperature is deliberately vague. As I'm sure you know, it's been measured in lots of ways historically, and that's the reason. I wonder whether SSS and SST are supposed to correspond. There are other more precise quantities than SST, with foundation, skin and subskin, but such concepts haven't been proposed for SSS.

I suppose we should review names with sea_surface and insert appropriate clarification in the definitions, and also clarify what the "surface" is in the definition of depth, as you say, especially over the sea.

Best wishes

Jonathan

[martinjuckes]

HI @JonathanGregory , the surface_downward_heat_flux_in_sea_water standard names includes the statement 'The surface called "surface" means the lower boundary of the atmosphere.'. There are many similar cases. If your view is correct, these should be corrected, but they appear to be long-standing definitions.

The CMIP variable hfds for surface_downward_heat_flux_in_sea_water does have the statement 'This is the net flux of heat entering the liquid water column through its upper surface (excluding any "flux adjustment") .' So perhaps we do need to consider an adjustment to the standard name definitions?

For other CMIP variables, such as dpco2 (surface_carbon_dioxide_partial_pressure_difference_between_sea_water_and_air) there is no explicit guidance, but as it is a difference between sea water and atmospheric partial pressures I suspect that they may want the liquid water interface to the atmosphere.

Regarding depth: yes, I think clarification is needed. In the short term we can clarify what we intend in the CMIP7 data request. For CF we may need to add additional more specific names. The difference will be quite substantial under floating ice shelves.

[JonathanGregory]

Dear Martin

the surface_downward_heat_flux_in_sea_water standard names includes the statement 'The surface called "surface" means the lower boundary of the atmosphere.'. There are many similar cases. If your view is correct, these should be corrected, but they appear to be long-standing definitions.

I don't doubt that they are, but I expect that means it wasn't thought about when writing the definition. Indeed, hfds should be the heat entering the water, and thus includes heat fluxes from the sea ice.

This will take some work to get clear!

Jonathan

[ChrisBarker-NOAA]

Back to the original question.

• Term depth

• Description Current description: "Depth is the vertical distance below the surface."

• Units m

The question is whether there is a precise interpretation of "surface" intended here?

I suspect (those with a longer history, correct me if I'm wrong) that this is a early standard name, from a time when we were less precise in our definitions. perhaps a bit more text should be added here.

Jonathan pointed out that:

unqualified depth always means depth below the unqualified surface.

which means that which surface is used depends on context.

Indeed, I'm focused on oceanography, but presume one would use depth for how deep below the ground a sample was taken, yes?

Which is unfortunate, but maybe unavoidable (and practical). Maybe some extra text, which I'm having trouble coming up with, but:

"Depth is the vertical distance below the surface. The specific surface depends on the context of the variable used"

Thinking about my use cases, we use depth in two ways:

1. to define the vertical grid in an oceanographic model
2. to define the depth the a particle is at in a particle tracking model.

In both cases, the "surface" is a bit vague -- it's what the model uses, and may or may not be the geoid, or local MWL, or .....

But this is (often) clear from the context, so that's fine.

maybe there's a need for something like depth_below_sea_surface or depth_below_mean_water_level

[looping iin that we might not want to use "sea" when it could be a lake or ...]

[JonathanGregory]

Dear @ChrisBarker-NOAA, @martinjuckes, Karl @taylor13

We will need to check the standard names which mention surface and insert some clarification in the description where required. I don't believe the existing standard names are intentionally vague, though. I think that surface (not sea_surface) always means the bottom of the atmosphere. That's the most obvious meaning, isn't it. If you say that something "hits the surface" or "lands on the surface", the impact is with whatever medium underlies the atmosphere. I notice that we also have some ground_level names, and surface heat flux in_snow and in_sea_ice. The need for these confirms that surface means the top of the snow or sea ice, and not the bottom of the snow on land (that's the ground).

As discussed above, there is may be some ambiguity about sea_surface, however, in existing standard names, specifically about whether it's at the top of floating ice and snow, or the bottom of it. Again, some of them need clarification in their descriptions, as Martin's initial question shows.

If there are use-cases for deliberately vaguer terms, we can consider adding them, of course. Likewise, as discussed elsewhere, we can consider standard names for inland_water_surface and sea_or_inland_water_surface if needed.

I continue to think the answer about sea_surface_height is that it refers to the liquid-water-equivalent surface, and that depth in the ocean means below the liquid-water-equivalent sea surface. That seems the correct choice to me because (a) in many or most OGCMs the sea ice doesn't actually occupy any of the volume of the top layer of the ocean, as far as ocean tracer conservation is concerned, (b) for ocean dynamics, this is the most useful definition of depth, considering that the hydrostatic pressure depends on the liquid-water-equivalent depth, including floating ice, not the liquid water alone, (c) for sea-level altimetry, that's what "sea surface" means, because global mean sea level is related to the liquid-water-equivalent volume of the ocean, including the volume of the floating ice if melted.

Best wishes

Jonathan