European Commission logo
INSPIRE Community Forum

INSPIRE and external radiation dose & radioactive fallout measurements

Hi all,

I'm involved with a project which includes publishing measurement results of external radiation dose rates (near real-time) and the nuclear fallout samples (weekly sampling) of Finland as INSPIRE datasets. We need to decide which INSPIRE thematic data models to use for publishing this data as an INSPIRE Download Service. 

As something related very much to the human health, both the radiation dose and the fallout measurements might naturally fall under the Human Health theme, but based in what I've learned by reading the HH Data Specification and the data model, this theme is mostly concerned with statistical data and thematic mapping, not so much with the continuous process of publishing the actual measurement data from an observation network.

So my current idea is to use the AC-MF theme data models for publishing the measurement values and the EF for publishing the information about the observation networks. In essence this means that we would be using the same data models already used for the meteorological and air quality observation data. Do you see this as a valid approach?

There seems to be few (if any) existing examples for publishing this kind of datasets within INSPIRE, even though I know that similar measurement networks exist in other EU member states. We would obviously like to harmonize with other data providers, so any examples of such projects would be more than welcome.

  • Alessandro SARRETTA

    By Alessandro SARRETTA

    Dear Ilkka,

    I've published your question also in the HH group and in the Marine and Atmosphere Cluster to possibly extend the discussion.

  • Keiran MILLARD

    By Keiran MILLARD

    As radiation is a physical or chemical property of the environment, it can be considered in either the OF or AF specification, depending on whether the radiation is in the water or atmosphere. 

  • Alessandro SARRETTA

    By Alessandro SARRETTA

    Dear Ilkka,

    your solution to use AC-MF (or OF in the the ocean domain) model for values and EF for networks seems the best one, given the information you provided.

    Could you please add more information on how (type of sensors and network) measurement are taken, to understand what part of EF model are you going to use?

    Regarding your sentence "I know that similar measurement networks exist in other EU member states", could you provide any more detail, so that I can try to contact them and understand whether they are interested in approaching INSPIRE?

    And as a follow up of this discussion, do you have any (even partial) result, like EF or AC-MF xml (or even better active and running service :-)) that can be shared? I think examples are one of the most needed outputs... :-)

    Thanks

  • Johanna Ott

    Hi Ilkka,

    How did you finally go ahead with this?

    I am currently harmonising data containing stations to measure the gamma dose rate in Germany and their measuring results. I mapped the stations to EF EnvironmentalMonitoringStations and I wonder if I can report the measuring results as OM_Observation objects in the same endpoint only using the EF schema (as the EF schema contains a link to OM) or if I need to create a second endpoint using a different schema (HH? AC?).

     

    Thanks and all the best

    Johanna

  • Katharina SCHLEIDT

    By Katharina SCHLEIDT

    Hi Johanna,

    While I don't know what Ilkka finally did, I'm agreed with all the above that EF is the core data spec for radiation measurements. Radiation measurement was also foreseen during the development of the EF spec, it was confirmed that the underlying data models are identical to air quality; I'd also assume that the mediaMonitored attribute would be set to "air" except for in edge cases where one is somehow measuring radiation in water or something else (compost).

    Also a note on application of AC/MF, and more generally on the specialized observation types available from the GCM:

    First of all, AC/MF has the honor of having the absolutely most compact schema, all you need to worry about are the observations.

    With the specialized observations, there was actually too much schema generated. This had to do with the update from O&M V1 to V2 during the data specification process. In O&M V1, hard-typing was used to specialize observations by resultType, in V2 in the xml encoding this was changed to soft-typing via the type attribute of the Observation to the types listed in the OGC registry, e.g.

    or one of it's siblings. This way, you save the effort of having to provide the specialized observation type

    Finally, this would be an ideal use case for SensorThings. Far easier to deploy and use!

    :)

    Kathi

     

  • Johanna Ott

    Hi Kathi,

    Thank you for your feedback!

    I decided to move on with the EF schema. Observations are still confusing me though. If the featureOfInterest of such an observation is the sample taken at one measuring point, the sampledFeature would be the air in total or am I wrong? If that is correctly understood how can I describe that in the sampledFeature element?

     

    All the best

    Johanna

  • Katharina SCHLEIDT

    By Katharina SCHLEIDT

    Hi Johanna,

    yes, identifying the FoI is often the tricky bit in O&M modelling. In air quality, we had long discussions on this issue, with various players pushing for reuse of the EF Feature (unfortunately an example in O&M V2, being removed in V3). Reasons NOT to reuse the EF:

    • Semantically, it's just wrong, you're not measuring the ozone concentration of the facility (building), you're measuring it in the air bubble around the intake
    • The location of the intake is often essential (air can be very different at 1 vs. 5m height), and a station can have multiple intakes. Providing this information correctly is thus essential

    For air quality, there's a lot of information on our best practices page:

    https://inspire.ec.europa.eu/forum/pages/view/30357/efom-best-practices

    I've also done a simple static example, all consistently linked within the GitHub repository:

    https://github.com/DataCoveEU/INSPIRE_EF/tree/master/AQ_Example

    Hope this helps!

    :)

    Kathi

  • Katharina SCHLEIDT

    By Katharina SCHLEIDT

    PS: the following posting on INSPIRE air quality data via SensorThings from the API4INSPIRE project could be inspirational: https://www.linkedin.com/pulse/ad-hoc-austrian-air-quality-api-kathi-schleidt/

  • Johanna Ott

    Hi Kathi,

    Thanks a lot for your input. Not re-using the EF makes sense to me.

    The link to the sampledFeature from the example (http://sweet.jpl.nasa.gov/2.2/realmAtmoBoundaryLayer.owl) won't resolve (same for the links to the INSPIRE e-reporting pages btw - e.g. http://inspireaq.jrc.ec.europa.eu/wiki/index.php/Main_Page). For the 52° north client, I am running into internal server errors. So I am currently not able to check what was used as sampledFeature in the example cases.

    I also tried checking the data using a rather old QGIS extension for reading the SOS data but did not get it to run. Any hints on what other clients can be used here?

     

    All the best

    Johanna

  • Katharina SCHLEIDT

    By Katharina SCHLEIDT

    Hi Johanna,

    assuming you mean "now using EF makes sense", not "not" ;)

    Thanks for pointing out that the references to the NASA Sweet ontology no longer resolve. I've now found the new version (sad that NASA JPL didn't retain the old URIs, guess they're above FAIR principles), updated in the github examples. This is in accordance with D2.9, that requests a higher level sampledFeature, sweet as default source:

    http://sweetontology.net/realmAtmoBoundaryLayer

    I'm not sure where the link to the JRC AQD Reporting Page is stored, can't help JRC taking this down, if you tell me where it is I'll just remove it from the example.

    To the issues with SOS readers, here the problem tends to be the various result types allowed by O&M Observations, especially if implemented as derived observation types (reason for my note on this)

    As for viewing, the newer versions of QGIS can now view natively to my memory, even resolve xlinks

    :)

    Kathi

Environmental Monitoring & Observations

Environmental Monitoring & Observations

Environmental Monitoring Facilities, Observations and Measurements