WIS 2 Topic hierarchy – Additional Level 9 types for surface observations

[StuartJMatthews]

Overview of proposal*

The current options for surface based observations (https://codes.wmo.int/wis/topic-hierarchy/earth-system-discipline/weather/surface-based-observations) are very limited.

The limited options means it is not clear how all the various observation types wanting to use the WIS 2 topic hierarchy can proceed.

This issue highlights some of the barriers to using the current options at level 9 and requests guidance on how to proceed and clarity on the purpose of Level 9 below surface based observations.

Amendment details*

Additions to surface-based-observations topic OR approve not using the bottom level (9) of the hierarchy.

Add topics (level 9) to the surface-based-observations

Requestor(s)*

Stuart Matthews, UK Met Office on behalf of the EUMETNET Observations programmes.

EUMETNET Programmes distribute observations data from several observing networks on behalf of its Members.
EUMETNET Programmes requesting changes...

• E-Profile (vertical profiling instruments),
• OPERA (weather radar),
• E-ABO (Aircraft observations),
• E-GVAP (GNSS derived water vapour) and
• Management Programme (high temporal and spatial resolution land surface observations)

Consultations

EUMETENT Expert Teams for the relevant observing networks and the EUMETNET Expert team on Federated European Meteorological Data Infrastructure (FEMDI)

Describe WIS2 data publication plans associated with the proposal

(identify relevant datasets, WIS2 Nodes, discovery metadata publication)

Datasets related to the EUMETNET Programmes (wind profilers, microwave radiometers, ceilometers and LIDAR, weather radar (single site, national composites, pan-European composite), AMDAR, etc.

New WIS nodes will be create for these datasets. These nodes will be hosted by the UK Met Office and MET Norway. Additional EUMETNET Members may also share EUMETNET observations in the future.

Discovery metadata for all these datasets is being created. Drafts are available.

Comments

Limited options, at level 9, for surface-based-observations

I am raising this ‘Issue’ on behalf of several EUMETENET activities that wish to share observations via WIS 2.0 and therefore use the wis2 topic hierarchy.

Within the topic hierarchy, below surface-based-observations, there are current very few approved attributes, and the approved attributes appear to be encoded GTS type products.

We are aware several EUMETNET Members have requested additions below surface-based-observations, but we don’t have visibility of these requests.

EUMETNET are now nearly ready to distribute observations data, from our Members observing networks, via WIS 2.0. Unfortunately, due to the lack of approved surface-based-observations, we must use ‘experimental’ within the hierarchy.

We want to avoid using ‘experimental’ as there would be a breaking change for users when the topics are approved.

Also, the purpose of level 9 is not clear. Should level 9 only contain encoded messages, or can we request observing networks to be added to level 9?

If we can request observing networks to be added, at what level of granularity should we request? For example, for aircraft-based observation (ABO) should we request ‘ABO’ to be added, or should we request more detailed descriptions for the actual type of ABO data. For example, should we request AMDAR, Mode-S, TAMDAR, etc. to be added, instead of ABO?

An additional option we considered, but were not sure what allowed, was to not include level 9 in the hierarchy, or possibly adding an 'undeclared' attribute.

Due to my uncertainty of the purpose of Level 9 and how granular we should be, I have avoided making a long list of proposals here. I would though be very happy to work with you on resolution this issue.

[amilan17]

@StuartJMatthews thank you for this request and your questions. It is not recommended to include networks as topics. This information can be described in the associated discovery metadata record.

I invite you to review the guidance on proposing topics: https://wmo-im.github.io/wis2-cookbook/cookbook/wis2-cookbook-DRAFT.html#_defining_and_proposing_topic_for_wmo_earth_system_disciplines

[StuartJMatthews]

@amilan17 . Thanks!
In the guidance there’s a good example of how a user might subscribe to all SYNOP messages. I.e.,
cache/a/wis2/+/core/data/weather/surface-weather-observation/synop

I was thinking level 9 could be used similarly to subscribe to different observation types, not only GTS type bulletins (e.g., /synop, /temp, etc)

For example, if a user wanted to subscribe to all radar data. A user could subscribe to something like...
origin/a/wis2/+/data/recommended/weather/surface-weather-observation/radar

Otherwise the user would need to subscribe to all surface-weather-observations and then filter out the radar data by another method.

This looks like the way to go. I just read the section on Facilitate client side filtering in the guidance and [(https://github.com/wmo-im/wis2-notification-message/issues/110)].

I'll do some more reading about using properties object of the notification message.

[StuartJMatthews]

@amilan17. I think this ticket can now be closed. For EUMETNET products distributed via WIS 2.0, we will omit level 9 completely. We'll use the metadata id within the MQTT message to allow users to filter for the observation type their require. We will also advise EUMETNET Members to do likewise for surface based observations, apart from the BUFR encoded products already listed here. https://codes.wmo.int/wis/topic-hierarchy/earth-system-discipline/weather/_surface-based-observations

[StuartJMatthews]

@amilan17 and @tomkralidis . After reading some of the tickets in ready for fast track. E.g., #137 it isn't clear to me now how to proceed. In #137 its suggest the 'leaf' below surface-based-observation should be used but then the comment in this ticket says we shouldn't add observing networks to level 9. As mentioned in this ticket it also isn't clear how granular we should be. E.g., Should we use ABO (Aircraft base obs) or should we use AMDAR, TAMDAR, etc. I'm also concerned about #182 is it suggests placing AMDAR data below weather/aviation. I've put my concerns about this in #182.

[StuartJMatthews]

Given the recommendation in #137 and EUMETNET are ready to share data via WIS 2.0 that are not made available via GTS, I would like to suggest following as leaves below surface-based-observations.

• abo (this is related to topic hierarchy needed for aircraft reports (AMDAR) #182) for AMADR, TAMDAR, AFIRS, Mode-S, etc.
• verfical-profiles for wind profilers, lidar, VAD radar, mircowave radiometers, etc.
• weather-radar for single site (volumetric) and composite/mosaic products
• surface-observation for high temporal and spatial resolution observations from land surface stations
• gnss for integrated water vapour observations derived from GNSS satelite delay.

Again, I wasn't exactly sure about the names to be applied, or how granular to be. As can be seen, the above suggestions can be very easily sub-divided and sub-divided further.

More observing networks at level 9 below surface-based-observations are likely to be needed. The above will allow EUMETNET to share the observations they have available now and are ready to distribute via WIS 2.0.

[amilan17]

@StuartJMatthews
I put your proposal into the table below with suggested definitions.

Earth system discipline	Sub-discipline	New topic name	Topic description
weather	surface-based-observations	abo	Aircraft-based observations, e.g., AMDAR, TAMDAR, AFIRS, Mode-S, etc.
weather	surface-based-observations	vertical-profiles	Vertical profiles, such as wind profilers, lidar, VAD radar, microwave radiometers, etc.
weather	surface-based-observations	weather-radar	Weather radar, e.g., single site (volumetric) and composite/mosaic products
weather	surface-based-observations	surface-observation	Surface observations, e.g., high temporal and spatial resolution observations from land surface stations
weather	surface-based-observations	gnss	Global navigation satellite system, e.g., integrated water vapour observations derived from GNSS satellite delay

[golfvert]

Following an email by @StuartJMatthews and my answer by email, @amilan17 can you make sure the proposal above is approved by domain experts, and when done, pass it on to ET-Metadata ? Formal approval is for FT2026-1. Thanks.

[tomkralidis]

TT-WISMD 2025-10-06

• ACTION @amilan17 to analyze and provide feedback