AMOIS and AMFIS Discussion 2026 08 14 - team-sx/iwxxm-sx GitHub Wiki

Details

2026-Aug-14, 06:00-07:00 UTC, Webex

Agenda

  1. Ilkka has kindly provided updated XML schema drafts and examples for AMOIS/AMFIS at https://github.com/spatineo/iwxxm-sx/tree/amf.
  2. Choy has (lately) pushed the a cleaned up version of IWXXM 2025-2 to GitHub to serve as the new baseline of IWXXM-SX.
    • More work has to be done:
      • The Common package has types not relevant to IWXXM-SX and will be removed in due course. Shall we move those SX only types to a new package to have a cleaner separation with legacy stuff?
      • We may want to start testing the use of separate namespaces for packages (3 packages are already implemented with iwxxm:MeteorologicalFeature type, viz WAFS SIGWX FC, VONA and QVACI).
      • How do we deal with external packages which we only want parts of a feature type?
    • Resolving issues mentioned in previous meeting:
      • If iwxxm:PhenomenonGeometry describes "product" coverage and iwxxm:PhenomenonProperty describes "product" characteristics, what OM&S concepts can we bring in?
      • How can we put the optional real and being inferred observation under iwxxm:PhenomenonProperty? Are we using full OM&S concepts here? Do we need a new element at the same level as iwxxm:PhenomenonGeometry to contain such information?

Notes

An instance of iwxxm:MeteorologicalFeature has the following structure:

<iwxxm:MeteorologicalFeature>
   <gml:identifier ... />
   <iwxxm:phenomenonCategory />   <-- Enumeration
   <iwxxm:phenomenonTime />
   <iwxxm:phenomenon />   <-- Code list (external)
   <iwxxm:phenomenonGeometry />
   <iwxxm:phenomenonProperty />
</iwxxm:MeteorologicalFeature>

Currently we have defined:

  1. "Geometry types" for use with iwxxm:phenomenonGeometry, which should cover most anticipated coverages for a general phenomenon, and we may want to extend it to include AIXM types used to describe aerodrome features like TDZ, etc.:

  2. We may want to extend "Property types" to cover description of all reusable measures (e.g. temperature, wind, etc.) as well as their alternative descriptions (e.g. TSML, etc.) and other characteristics (e.g. probability of occurrence, PDF/CDF, etc.).

    Note: During early development of IWXXM when we were directly applying O&M, we have created an external package METCE which is suppose to be the tool box to be used in om:Procedure. In "Property type" I incline to create an all-in-one measure specific feature type covering concepts on om:result, om:Procedure and so forth

  3. We may probably need a new element at the same level as iwxxm:phenomenonProperty, etc. to described the observation(s) inferring the result described in iwxxm:phenomenonGeometry and iwxxm:phenomenonProperty. The feature type can be another iwxxm:MeteorologicalFeature or based on (simplified) concepts from OS&M

  4. Ilkka brought up a point that the name of element iwxxm:phenomenonGeometry refers essentially to a spatial bounding box and is not suitable to be used with other spatial identifiers like aerodrome, runway zones, etc. He mentioned the term Feature Of Interest (FOI) which has a broader scope could be considered.

Next Meeting

Not scheduled

Attendees

  • BL Choy
  • Ilkka Rinne