1-01-07 New code list "Matrix"

[fstuerzl]

Summary and Purpose The biogeophysical context of an observed variable according to the WIGOS Metadata Standard is expressed by the "matrix" or medium in which a species is measured. The proposed new code list should contain matrices from all domains (atmosphere, earth, ocean, outer space, terrestrial) with useful definitions.

Final Proposal ~ last update 9 June	notation	name	description	tags
aerosol	Aerosol	Suspension of fine solid particles or liquid droplets in air or another gas. [William C. Hinds, Aerosol Technology]	atmosphere
aquifer	Aquifer	A subsurface layer of water-bearing permeable rock, soil, or unconsolidated solid material.	terrestrial
cloudParticle	Cloud particle	Ice particle or liquid droplet present in a cloud. The particles of water, either drops of liquid water or ice crystals, which form a cloud. [International meteorological vocabulary (WMO-No. 182). 1992 (2nd ed.)]	atmosphere
dryDeposition	Dry deposition	The sum of atmospheric trace gases and particulate matter removed from the atmosphere by sedimentation, turbulence or diffusion.	atmosphere
gasPhase	Gas phase	Physical state of matter, also called vapour phase. Molecules are sufficiently far apart to overcome attractive forces and to fill all available space almost uniformly.	atmosphere, earth, marine, terrestrial, space
groundwater	Groundwater	Precipitation that has infiltrated the soil beyond the surface and is present in empty spaces underground.	terrestrial, hydrological
interstitialAir	Interstitial air	Air trapped in small cavities in a solid matrix, e.g. snow, ice, soil, bedrock or similar.	terrestrial, cryosphere
interstitialWater	Interstitial water	Water contained in the voids of porous or fractured media. [International Glossary of Hydrology (WMO-No. 385). 2012.]	terrestrial, cryosphere, hydrological
liquidPhase	Liquid phase	A physical state of matter. In liquids, molecules flow easily around each other but are held together by attractive forces. Liquids assume the shape of their containers or appear as droplets in an aerosol.	atmosphere, earth, marine, terrestrial
particlePhase	Particle phase	Liquid or solid particles, or liquid particles with a solid core suspended in a matrix in a different aggregation state.	atmosphere, terrestrial, marine, space
snow	Snow	A porous, sintered material consisting of an ice structure and a continuously connected pore space, which contains air and potentially liquid water.	terrestrial, atmosphere, cryosphere, hydrological
soil	Soil	Unconsolidated material, containing living and inert organic and mineral components, lying directly on the surface of the Earth and which enables vegetation to grow	terrestrial, cryosphere, hydrological
suspendedMatter	Suspended matter	Solid matter suspended in a liquid or gas, or droplets of liquid suspended in gas. In solids, molecules are arranged in a regular pattern, firmly held in place and only able to vibrate within a limited space.	terrestrial, cryosphere, hydrological, marine
totalDeposition	Total deposition	The sum of dry and wet deposition.	atmosphere, terrestrial, marine
wetDeposition	Wet deposition	The sum of atmospheric trace gases and particulate matter removed from the atmosphere after suspension in atmospheric water and washed out through rain, snow or fog.	atmosphere, terrestrial, marine
permafrost	Permafrost	Subsurface material that remains continuously at or below 0 °C throughout at least two consecutive years.	terrestrial, cryosphere

Reason Provide a consolidated vocabulary for matrices

[champikagallage]

There is no similar concept for marine/ocean. Instead Ocean may use Physical/Biogeochemical and Biological categories.

[joergklausen]

See https://github.com/wmo-im/wmds/wiki/Tags-for-observed-variables for discussion.

[gaochen-larc]

From Seinfeld and Pandis (page 47 - 48): "Whereas an aerosol is technically defined as a suspension of fin solid or liquid particles in gas, common usage refers to the aerosol as the particulate component only (Table 2.19"

[markusfiebig]

From Willeke & Baron, page 3: "An aerosol is an assembly of liquid or solid particles suspended in a gaseous medium (e.g. air) long enough to enable observation and measurement." I know the Seinfeld & Pandis definition. They likely wanted to avoid discussions.

[markusfiebig]

From, Hinds, Aerosol Technology: "An aerosol (abbreviation of "aero-solution") is a suspension of fine solid particles or liquid droplets in air or another gas."

[gaochen-larc]

I propose to add two matrix entries: snowpack and snow interstitial air. These two items should be linked to "terrestrial" or "cryosphere" if we decide to add... Comments?

[jbianchi81]

What should be the matrix for "Ground water (amount per unit area)"? Ground water resides in the interstitial space between particles of Earth's crust (soil, regolith or bedrock). Maybe we could use something like "crust's interstitial water" or just use "liquid phase". Also, what should be the matrix for land surface properties such as "Lake area", "Land cover", "Land Surface Topography" and the like (which may be represented as map layers)? Maybe we could use "Surface".

[joergklausen]

RE: ground water: I would go with the term aquifer for the matrix as defined, e.g., by National Geographic : An aquifer is a body of rock and/or sediment that holds groundwater. Groundwater is the word used to describe precipitation that has infiltrated the soil beyond the surface and collected in empty spaces underground.

IMHO, soil should also be a matrix, whereby water in soil is not ground water.

I support snowpack, snow interstitial air, and I would add soil interstitial air.

[gaochen-larc]

I support soil interstitial air.

[jbnowak-larc]

I, too, support snowpack, snow interstitial air, and I would add soil interstitial air.

[dkubisti]

Short comment on the definition of dry deposition: Whilst this definition was taken from WMO/GAW, it describes only transport to the surface by gravity (free fall). Important as well is turbulence and diffusion. A better description might be: The transport and removal of atmospheric trace gases and particulate matter by sedimentation, turbulence and diffusion?

[fstuerzl]

In the \Atmosphere\POPs part there are variables with attributes such as "in total precipitation", "in precipitation" and "in wet precipitation", which can't be precisely mapped to a matrix in the table above.

@markusfiebig, could you help with that?

attribute	Matrix info on EBAS	corresponding Matrix in new code list
in total precipitation	precip_tot: Measurements refer to a chemical or physical property in precipitation including additional information about the amount that has been absorbed on the precipitation funnel wall (concentration in precipitation + amount absorbed on the funnel). Used by Germany when reporting Persistent Organic Pollutants data.	?
in precipitation	precip: Measurements refer to a chemical or physical property in precipitation (rain or snow).	?
in wet precipitation	-	Liquid phase?

[markusfiebig]

@fstuerzl, just double checked the above definitions. The granularity of the matrix information in EBAS is a bit finer as compared to the one proposed for this list. "in precipitation" refers to concentrations of substances in precipitation, "in wet precipitation" refers to fluxes of substances in precipitation to the ground, i.e. different quantity types. "in total precipitation" refers to the method of including deposits from the funnel walls into the sample. For our purposes, all of these would be "Wet deposition".

[markusfiebig]

One comment concerning the definition of "Wet deposition". The definition only refers to dissolved gases, not to particles scavenged by precipitation. Should we rephrase the definition to: "The process whereby atmospheric gases mix with suspended water in the atmosphere, or particles are scavenged by precipitation, and are then washed out through rain, snow or fog."

[jbnowak-larc]

I agree with the inclusion of turbulence and diffusion in the definition of dry deposition and the inclusion of particles in the definition of wet deposition. These addition make both definitions more complete.

[markusfiebig]

Should we define "Cloud particle" more precisely? Something like: Particle having grown in size to pass the thermodynamic barrier with respect to water or ice supersaturation.

[fierz]

@joergklausen @gaochen-larc @rodicanitu

1. Inclusion of "snowpack" and "snow interstitial air" 
I strongly support the introduction of the new domain 'Cryosphere' as discussed in issue #277. In that context I wonder whether the components of the cryosphere (see description under #277) should not be used as matrices for that domain. For example, 'snow' would be one of those, and not 'snowpack' as proposed above. From my cryospheric perspective this would make more sense as I presume here one understands the material snow under snowpack. However, I cannot really see the need for a matrix 'snow interstitial air'. If snow becomes a matrix, I'd rather suggest to use tags like 'pore space' and 'interstitial air' - if the latter is still needed. The two tags could also be used for soils avoiding numerous quite similar 'matrices'. In addition one could also more easily define snow temperature and snow density with tags 'temperature' and 'density'. I may not have well understood these concepts though and thus be totally wrong. Open for discussion. 


2. "Particle phase" 
After reading a few times the definition for that matrix I still wonder whether 'bulk solid material' fits in there. What is meant by this: the solid phase of a material (larger than a particle), or soil, or snow on the ground? In the latter two cases, it would be advisable to use them as matrices on their own instead of a 'pseudo'-phase particle.

3. "Cloud particle"
 The current definition reads as ice would be a cloud droplet. I'd suggest 'Ice crystals and liquid cloud droplets' to define that matrix.

I think this discussion may also show that these overlapping spheres are looked at from different scientific perspectives. Definitions and concepts in the overlaps should thus be clear and understandable by all involved parties.

[gaochen-larc]

I also prefer 'Ice crystals and liquid cloud droplets', as proposed by @fierz. As for the "Particle phase", I am also not sure the meaning of "bulk" in this context. I think the definition stands with the last part "or of bulk solid material". Maybe I missed something here...

There have been a lot work to illustrate the unusual chemical processes occurred only in the snow interstitial air.

[amilan17]

to do:

• update domains columns to match the new domain lis (because it's a useful reference)

[joergklausen]

I take the blame for not having consolidated and prepared a branch. It is time to do that now. I'll see if I can get some help from @fstuerzl . One addition to the list that came up, to be discussed, is 'permafrost'. I personally prefer to include this term in the list of tags, rather than matrices, because it can apply to several matrices.

[joergklausen]

Short comment on the definition of dry deposition: Whilst this definition was taken from WMO/GAW, it describes only transport to the surface by gravity (free fall). Important as well is turbulence and diffusion. A better description might be: The transport and removal of atmospheric trace gases and particulate matter by sedimentation, turbulence and diffusion? While I agree with @dkubisti on the more precise description of the processes involved, the general problem with many comments above is that they do not explicitly define a 'matrix'. I will attempt to come up with a consolidated list.

[joergklausen]

@amilan17 Please prepare branch. @fierz, @gaochen-larc @markusfiebig @champikagallage Please confirm branch once established.

[rodicanitu]

@joergklausen - please add "marine" tag to "permafrost"

[gaochen-larc]

Question: could one use multiple matrix terms for a variable? I came across measurement of total water, i.e., the sum of water vapor, liquid water, and ice water.

[fierz]

@joergklausen

While I agree with @dkubisti on the more precise description of the processes involved, the general problem with many comments above is that they do not explicitly define a 'matrix'. I will attempt to come up with a consolidated list.

I fully agree it is not easy to define a matrix properly (see my comment above) and I am looking forward to the consolidated list by @joergklausen. However, I wonder whether the tags (a nice mechanism indeed to group variables 😉 ), including nesting/sub-tags, could not do a large part of the work, allowing for less matrix definitions/descriptions. I think of 'permafrost', 'marine' + 'permafrost', 'snow', and 'snow' + 'cover' as valuable tags that would not need to be a matrix too!?

[joergklausen]

Question: could one use multiple matrix terms for a variable? I came across measurement of total water, i.e., the sum of water vapor, liquid water, and ice water.

Technically, this could work with ISO19156 OM_Observation ‚parameter‘, which is the intended mechanism for the implementation. Although I thought of restricting it to cardinality 1 for each name, in order to make it simpler for parsers that need to deal with this eventually.

My idea of specifying a matrix, however, is not to list several, but rather to constrain. For your example, why would you even want to list them instead of just referring to the domain ‚atmosphere‘?

[gaochen-larc]

My idea of specifying a matrix, however, is not to list several, but rather to constrain. For your example, why would you even want to list them instead of just referring to the domain ‚atmosphere‘?

This will work. Somehow, I thought matrix is required.

[amilan17]

@joergklausen Can you confirm that this will be worked on for FT22-2 and that it's not ready FT22-1?

[joergklausen]

@ferrighi @jbianchi81 @gaochen-larc @JohnEyre Please comment and/or confirm

[JohnEyre]

@joergklausen @ferrighi @jbianchi81 I've now read the discussion on this issue. I don't think it raises any problems in terms of achieving consistency with OSCAR/Requirements (#190) or more generally with the work on WIGOS/RRR. Would it be helpful for WIGOS/RRR also to adopt the "matrix" concept? I don't know, but I will alert them to this possibility.
The choice of the name "matrix" is not intuitively obvious to me. Why was this chosen?

[joergklausen]

@JohnEyre The term 'matrix' originates from analytical chemistry AFAIK. I would say, yes, O/R should adopt it. This doesn't necessarily imply that all variable names in O/R need to be represented in various columns. I think it is perfectly okay if an application decides to combine various elements into one variable name, but it is helpful IMHO if these elements are all drawn from agreed vocabularies. For example, we might have a vocabulary for state variables and quantities with elements 'temperature' and 'nitrate'; we might also have a vocabulary for matrices with the element 'soil' in it; we might further include 'root zone' in the vocabulary on observed layers. From this, a variable 'nitrate concentration in the root zone of soil' can be constructed, and an application might well manage this with a dedicated variable id. In terms of the WIGOS metadata representation, the various elements of this name may have to be encoded separately or not.

[jbianchi81]

It looks like the proposed definition for soil actually defines aquifer. I propose to change it for the following (from International Glossary of Hydrology p. 315)

• Unconsolidated material, containing living and inert organic and mineral components, lying directly on the surface of the Earth and which enables vegetation to grow

[fstuerzl]

Thanks for pointing that out, @jbianchi81.

I've created the new table in the branch: https://github.com/wmo-im/wmds/commit/5c4eefe8da86dbed4cde6e9c9863d7e6bbde22a6

[amilan17]

@fstuerzl I think that WMO Nr. 182 should be referenced like this:

International meteorological vocabulary (WMO-No. 182), 1992 (2nd ed.)

[ferrighi]

minor comments on the branch:

• permafrost name should be Permafrost
• total deposition and permafrost definitions are missing the quote
• tag Earth should be earth

[fstuerzl]

@amilan17, @ferrighi, I've updated the branch and the proposal accordingly.

[joergklausen]

Approved by the team

[amilan17]

https://github.com/wmo-im/wmds/wiki/2022-05-19-TT-WIGOSMD-22 notes: approved by the team

[fierz]

I understand the issue is 'validated' and I apologise for my annoying slow response time. Nevertheless, I strongly suggest the definition for the matrix Snow to be revisited! What is currently in the table clearly refers to 'Snowfall' but not to 'Snow'. Indeed, this does not fit by any means snow on the ground, which is also referred to in the matrices Interstitial air and Interstitial water. As a snow scientist I would propose to use the following description:

A highly porous, sintered material made up of a continuous ice structure and a continuously connected pore space, forming together the snow microstructure. (International Classification for Seasonal Snow on the Ground)

I realise the above definition does not include 'atmospheric' snow, that is snow crystal and flakes (not snowfall). But are those not included in the matrix Cloud particle ?

[joergklausen]

@fierz Better late than sorry, and we haven't crossed the finish line quite yet ... a week or so is left. I agree with your interpretation of the current definition. And I also think the intention of this matrix is snow on the ground, not individual particles falling from the sky. Your proposed description appears indeed more useful, then. However, I could not find this definition in the International Classification for Seasonal Snow on the Ground you refer to. So, either we accept your description as coming from an expert, or else we need the correct reference. In any case, a second confirmation would be helpful. Can you contact someone and ask them to make a statement here, please? Lastly, it would be useful to see a few examples of observed quantities where this matrix is relevant. I myself can think of 'temperature of snow', 'density of snow', 'mass concentration of black carbon in snow', 'albedo of snow'.

[fierz]

@joergklausen Happy to see we agree. The definition I provided is the second sentence of the first paragraph of '1. Features of deposited snow' in the ICSSG. Anyway, we will provide a confirmation by other experts asap (@rodicanitu: can we discuss this shortly, please?) Regarding observed quantities, 'liquid water content of snow', 'snow chemistry', 'optical grain size of snow' aka 'specific surface area of snow', 'isotopes in snow', etc. could be added. Please let me know whether you need more ore more detailed descriptions here.

[joergklausen]

@fierz Found it :-) I am happy with this definition, as said, some others should also chime in. Regarding additional observed variables, these must be proposed using a separate issue. You are most welcome to open one if these variables are of concern.

[fierz]

@joergklausen OK, great, we'll get a confirmation by other experts. I think will come back with additional 'snow' variables - mostly the above - as soon as we have a Task Team working on a revision of the snow chapter in the GIMO.

[fierz]

@joergklausen & @rodicanitu: I discussed the description for snow with other experts and after some bouncing mails between Samuel Morin (SG-CRYO/INFCOM & MeteoFrance), Chris Derksen (Task Team on Cryosphere Observing Requirements/GCW & ECCC) and myself, we confirm that the following fits best: A porous, sintered material consisting of an ice structure and a continuously connected pore space, which contains air and potentially liquid water.

[amilan17]

@joergklausen please see updates to this branch, If you approve, please move to 'ready for FT' column.

[fierz]

@amilan17 @joergklausen That description does not fit the tag "Atmosphere" anymore, except for snow-atmosphere interactions. Up to you to decide whether to keep it or not.

[joergklausen]

@fierz The description doesn't exclude the presence of snow (flakes) in the air as far as I am concerned, I would leave this tag.

@amilan17 I confirm the branch with one purely editorial modification. Please update the issue accordingly to reflect the branch.

[amilan17]

ready for FT