AMOIS and AMFIS Discussion 2026 09 30 - team-sx/iwxxm-sx GitHub Wiki

2026-Sep-30, 11:00-12:00 UTC, Webex

1. Recap of last discussion:
   • Observation WxObject will indicate the location/coverage where the observation is made in ROI (a generic geospatial object)
   • Forecast WxObject will indicate the target location/coverage of the forecast in ROI (an AIXM object)
   • Producers may give hints which target location/coverage (an AIXM object) an observation WxObject is applicable, and if a target location/coverage has more than one applicable WxObject its priority for application
   • All observation and forecast WxObjects will make reference to an aerodrome (an AIXM AirportHeliport object). In other words AMOIS and AMFIS are effectively clusters of observation and forecast WxObjects of an aerodrome
   • If the reference to an aerodrome can be relaxed (e.g. to a generic geospatial object), then it is possible to accommodate BoM and others' wish to use the same format for observations and forecasts made for a non-aeronautical location [This topic is beyond the scope of TT-AvData, but the technical approach to be taken will be shown below and hopefully it can shed some light for further discussions by other groups]
2. Topics for discussion in this meeting:

   • For AMOIS (as well as AMFIS) a schematron rule will be used to mandate the reference to be an AIXM AirportHeliport as well as other objects:

     <iwxxm:MeteorologicalFeature>
        <gml:identifier ... />
        <iwxxm:phenomenonCategory>AMOIS</iwxxm:phenomenonCategory>   <-- This is the place to indicate the kind of WxObject it is
        <iwxxm:phenomenonTime />
        <iwxxm:phenomenon />
        <iwxxm:phenomenonRegionOfInterest />
        <iwxxm:phenomenonProperty />
        <iwxxm:Reference>
           ... AIXM AirportHeliport object   <-- This is the place to indicate references to an aerodrome
        </iwxxm:Reference>
     </iwxxm:MeteorologicalFeature>
     

     <sch:pattern ...>
        <sch:rule ...>
           <sch:assert />   <-- xpath to check whether "AMOIS" and AIXM AirportHeliport object coexist
        </sch:rule>
     </sch:pattern>
     

   • Can observations make outside the "aerodrome boundary" (this is not the boundary used in present and recent weather but used to discriminate whether an observation or a forecast should categorized as inside or out of an aerodrome) be regarded as part of the aerodrome observations?

     

   • Representing wind information:

     • Single value

       <iwxxm:phenomenonProperty>
          <iwxxm:windSpeed uom="m/s">4.0</iwxxm:windSpeed>
          <iwxxm:windDirection uom="deg">240</iwxxm:windDirection>
          <iwxxm:AggregatedQuantity />   <-- Details on aggregation (e.g. 10 minute mean wind)
       </iwxxm:phenomenonProperty
       

     • Time series (Data is stored in a table. See example)

       <iwxxm:phenomenonProperty>
          <iwxxm:LinkableTimeseries>
             <tsml:defaultPointMetadata>
                <iwxxm:ValueRecord />   <-- Define the columns (fields) and rows (records)
                ...
             </tsml:defaultPointMetadata>
             <tsml:point><iwxxm:RecordTVP><iwxxm:value>11 200 16</iwxxm:value></iwxxm:RecordTVP></tsml:point>
             ...
          </iwxxm:LinkableTimeseries>
       </iwxxm:phenomenonProperty
       

     • Probability distribution (Data is stored in a table with additional attributes for individual row (record). See example)

       <iwxxm:phenomenonProperty>
          <iwxxm:LinkableTimeseries>
             <tsml:defaultPointMetadata>
                <iwxxm:ValueRecord />   <-- Define the columns (fields) and rows (records)
                ...
             </tsml:defaultPointMetadata>
             <tsml:point>
                <iwxxm:DistributionRecordTVP>
                   <iwxxm:mean>1.1 105 1.5</iwxxm:mean>
                   <iwxxm:predictionIntervalBoundary level="0.95" type="lower">1.0 90 1.2</iwxxm:predictionIntervalBoundary>
                   <iwxxm:predictionIntervalBoundary level="0.95" type="upper">1.2 115 1.7</iwxxm:predictionIntervalBoundary>
                   <iwxxm:predictionIntervalBoundary level="0.90" type="lower">0.9 87 1.0</iwxxm:predictionIntervalBoundary>
                   <iwxxm:predictionIntervalBoundary level="0.90" type="upper">1.5 116 1.9</iwxxm:predictionIntervalBoundary>
                </iwxxm:DistributionRecordTVP>
             </tsml:point>
             ...
          </iwxxm:LinkableTimeseries>
       </iwxxm:phenomenonProperty
       

• It was considered that "ROI" could have a better replacement.
• For clusters of aerodrome observation and forecast, there should not be a boundary defined in Annex 3/PANS-MET to decide whether they could be included in the clusters or not. The producer should decide themselves.
• Further discussions shall be made on reusing schemas for AMOIS and AMFIS in applications beyond their intended scopes, including the reporting of weather situations in locations away from an aerodrome.
• Presenting multiple phenomena in a single WxObject, in the same way as the time series of a weather parameter, should reduce the size of the resulting IWXXM instance. Having said that, this also violates the intention of using a WxObject to carry information of a single phenomena.

1. A wrap up for WG-WHL/1 will be provided for presentation in the workshop session of WG-WHL/1.
2. Further discussions and demonstrations will have to be made on (a) reuse of AMOIS/AMFIS containers and (b) presenting multiple phenomena in a single WxObject.

2026-Oct-14, 11:00-12:00 UTC, Webex

Name (Affiliation)	Attendance
Rosalind Lapsley (EUROCONTROL)	X
Dirk Zinkhan (DWD)	X
Svenja Koos (DWD)
Joonas Eklund (FMI)	X
Anu Lång (FMI)	X
Rauhala Mikko (FMI)
Ilkka Rinne (FMI)	X
Terhi Nikkanen (FMI)	X
Ashwin Naidu (BoM)
Patrick Desgagnes (NAVCAN)
Becky Kotten (FAA)	X
Tom Helms (FAA)	X
Michiko Ikeda (JMA)	X
Jan Korosi (IBL)	X
JJ Biel-Goebel (WISK)
Paul Hershberg (NOAA)
BL Choy (HKO)	X