Group discussion > The INSPIRE data specifications review process

The INSPIRE data specifications review process

Hanna Koivula
2122 days ago

As a followup to my presentation in Paris:

The INSPIRE data specifications are now available for review and public consultation.


For GBIF national nodes & data providers the Data specifications for i) habitats and biotopes,  ii) biogeographical regions and iii) on species distribution are particularly relevant. Also "D2.9 Guidelines for the use of Observations & Measurements and Sensor Web Enablement-related standards in INSPIRE Annex II and III data specification development" should be looked carefully into.

I'll try to post more detailed comments tomorrow, before escaping to holidays...



Hanna Koivula
2029 days ago

Here a copy-pasted email from Nils Valland:

I refer to our brief meeting on this topic at the GBIF18-conference in Buenos Aires.

The TWGs, the Data Specifications Drafting Team and INSPIRE JRC Data Specifications Team delivered  in June 20, 2011 a draft version 2.0. of all the data specifications of themes in Annex II and III. For the time being there is an ongoing open consultation which will be closed by Oct 21 2011.  http://inspire.jrc.ec.europa.eu/index.cfm/newsid/10521 The data specification for Species distribution (enclosed) can be downloaded from this site. http://inspire.jrc.ec.europa.eu/index.cfm/pageid/201/consultation/45851 The consistency between a use case for local use and the data specification can be questioned, due to the interpretation of the definition of distribution.


The INSPIRE Directive defines Species Distribution as geographical distribution of occurrence of animal and plant species aggregated by grid, region, administrative unit or other analytical unit [Directive 2007/2/EC]. The grid cells may range from 1x1 m unlimited upwards. The interpretation of this definition is given in the text of the draft version. See the executive summary pp. VI to VII. The definition is not intended to cover the ‘raw’ field observation data. Therefore the TWG have taken occurrence to mean the spatial representation of a species at a specific location, rather than being equivalent to an observation. In the interpretation the term “aggregated” is used, which most commonly means to form into a class or cluster. It is closely related to (but not synonymous with) the term “amalgamated”, which means to combine to form one structure. Both terms are used throughout the document, as being suitable for describing the process of converting raw observations into a distribution of occurrence. Though the TWG has been loyal to the INSPIRE Directive definition is has been thoroughly debated since the consequences is that the data specification does not support the dissemination of high resolution primary data, e.g. from the GBIF-network or other sources if they exists. The data model allows though to present the underlying data providers and source databases in each grid cell or other spatial representation. For GBIF as a major contributor from a lot of academic institutions to biodiversity data in Europe ( for the time being more than 100 mill records), the consequence is  that good quality primary data with high resolution will not be available as primary data as such through the INSPIRE data portal. From national experiences it is clear that spatial species data for users at local scale (municipalities etc.) , they demand and benefit from detailed primary data in their management and decisionmaking. For national and international use cases the data specification is sufficiently detailed.


I think this issues is important for GBIF and the dissemination of our content into all major portals presenting biodiversity data for science and society. The consultations is organized through national responsible INSPIRE institutions. National institutions being potential providers of data should give there comment or proposal to their Member State Contact Point (MSCP). http://inspire.jrc.ec.europa.eu/index.cfm/pageid/481  Unfortunately I have not been able to retrieve the list or the URL  to this list of MSCP from the INSPIRE website or elsewhere. But they are found in the list of LMOs, Legally Mandated Organisations in INSPIRE http://inspire.jrc.ec.europa.eu/index.cfm/pageid/42/list/2  To be efficient, it is recommended to use the template (enclosed) for the consultation.

If you think this is important to other GBIF-nodes in Europe, please forward this message to them.

Nils Valland

Senior adviser

Norwegian Biodiversity Information Centre (Artsdatabanken)
Mailing address: Erling Skakkes gate 47, N-7491 Trondheim, Norway

Phone +47 73 59 23 01

Mobile +47 92 41 20 37



Steve Wilkinson
2027 days ago

Thanks Hanna
I was involved with this but effectively the battle was sort of lost before I arrived in that they were very focused on sharing "species distribution" rather than the raw data. This is not necessarily bad for GBIF I dont think. Firstly we have a mechanism that is more scalable and useful for exachnaging raw data than the INSPIRE one is (which definitely isnt scalable in my opinion). SO I think there will continue to be a need and role even though it is not INPSIRE compliant. Secondly we might imagine having a "wrapper" on top of a DwC file(s) or perhaps just IPT to output a distribtion in INSPIRE format (so that by using IPT to mobilise data you both have a scalable infrastructure for sharing raw data and meet your INSPIRE obligations. I feel this is more practical in the short term than trying to alter the INSPIRE spec.

Hanna Koivula
2027 days ago

Good point Steve. I agree, these metadata standards are for the "products" not datasets per se, which could be two different things. Of course it should be possible to obtain the data behind the product for deriving new maps for different purposes, but this is still different from raw data. For example our "map-product" might be an analysis derived from collection, observational and literature occurrences, so defining the resolution for the data is neither easy nor meaningful.  An other thing is, how and should we earmark datasets that are available in INSPIRE somehow also in GBIF metadata?

Steve Wilkinson
2026 days ago

Hi Hanna
Agreed. I think the metadata needs a quick look. In INSPIRE there are fields to hold links to getting the data (either as a WS or I think the raw data itself). I think this would need to be replicated in GBIF metedata so that both sources woudl include the link to download the raw data (from GBIF) and the link to view a distribution (as an INSPIRE compliant WS).

Paco Pando
2026 days ago

Dear Hanna et al.,

I was involved on the discussions of the INSPIRE Thematic Working Group before Steve, and in fact, lost this battle then.

<!--[if gte mso 9]> Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 <![endif]--><!--[if gte mso 9]> <![endif]--><!--[if gte mso 10]>

/* Style Definitions */
{mso-style-name:"Table Normal";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
font-family:"Times New Roman","serif";}


For your convenience I extract here the statement origin of the discussion:

• The definition in INSPIRE Directive proposal does not include individual observations or other point based data, but focuses on aggregated versions of data about geographical distribution of species. Aggregation can be at any level of resolution, e.g. in geographical grid systems divided into 100x100 meter grid or 50x50km grid cells. Possibly also point-based observations and isolines generation between observations should be accepted and included in INSPIRE. Possibly these can be defined as options in the "other analytical unit". Aggregation may also be interpreted not only as space-based aggregation, but time-based aggregation as well.

 …and my key points:

3.  The difference between  aggregated and non aggregated data blurs depending o  the level  of "aggregation" and the level of granularity you are working with.  In other words, the aggregation vs. Non aggregation issue is an artifact of the methodology and the point of view of the user/observer/compiler.

5.  to restrict species distribution data --under INSPIRE-- to "aggregared only" I believe  would be a mistake since those data are unsuitable for serious analysis  (climate change scenarios, invasiveness models, etc.).

 I slightly disagree with Steve here in the sense that I think it is strategically important that for GBIF that raw data are considered within INSPIRE, an a great advantage for the end user .




Cees Hof
2017 days ago

Hi all,

As the INSPIRE consultation deadline is tomorrow, do we need any last minute action here?

As Steve, Paco and Hanna have been, or are, all involved in this matter we should have enough expertise to reach a conclusion on this issue. 

If providing input means just contacting your national Member State Contact Point and forward your hesitations then providing input might still be feasible at this late time.



Hanna Koivula
2017 days ago

Dear all, (and thanks Cees for giving this last minute call!)

I would appreciate if any of you could raise problematic items and/or just name issues that you are going to comment. Even just very shortly! It would help greatly since we had to postpone the task and we'll work with the final comments tomorrow (and submit comment the very last minute...)

I'll post some issues during the day and I'm happy if someone else would do the same!


Hanna Koivula
2016 days ago

Here a list of paragraphs and short comments of issues, that we are going to comment about Bio-geographical Regions:

Executive summary:

  • How to include national classifications?
  • Unclear relation between the definitions and use cases

Paragraph 1-4, recommendation 1:

  • Paragraph on data quality is unclear UML Overview:

  • Mandatory European classification systems have gaps in boreal areas (how to include local classifications?)


  • It should be clarified when national or local color schemes are accepted

Hanna Koivula
2016 days ago

So, half an hour to go!

Here our comments on the Species Distribution Specifications:


  • How to provide European Geo-portal amalgamated, European wide datasets? (E.g. Atlas Flora Europaeae) Currently only national mechanisms are defined.

2.2. and 2.4.

  • Definition of Data Set Series; Different options, how to form dataset series or time series according to the datatsets and/or methods used, need to be defined more clearly.
  • In the Ispire GLOSSARY it is only described as "collection of data sets sharing the same product specification"


  • SpeciesDistributionCoverage; Paragraph under the note is very unclear
  • Linkages to single observations;  connection between SD and Environmental monitoring themes needs to be clarified. Will it be possible to offer a link to raw data in the original data source or metadata? (Not neccessarily in the Environmental Monitoring Facilities Theme or ISO 19157 format)
  • Shared metadata; Specify how metadata can be shared between datasets
  • Species distribution unit; Time based aggregation should be allowed
  • IR Requirement 3; The Grid_ETRS89-LAEA as defined in Regulation 1089/2010/EC shall be used when defining a rectified grid - This will will reduce the amount of usable older datasets that are available only in nationally used (maybe obsolite) grids.


I hope that we can continnue this conversation, even if the commentary is now closed.