European Commission logo
INSPIRE Community Forum

Using EEA reporting grids

Has anyone any experience of using the EEA grids as the reference for HB and SD reporting? Especially in the marine environment? https://www.eea.europa.eu/data-and-maps/data/eea-reference-grids-2

  • Iurie MAXIM

    Hi Keiran,

    I have experience in using EEA grids for HB and SD and I was part of the TWG for HB, SD and BR.

    Maybe the word reporting that you used is quite confusing because within INSPIRE there is no such "reporting" obligation and as it can be confused with article 12 Birds Directive and article 17 Habitats Directive reporting obligation, where the EEA grids are used for the distribution and range of species and habitats.

    Together with one of my colleagues we created for ETC/BD and EEA the Range Tool that determine the range of a species/habitat based on the occurrence or distribution of species/habitat.

    Related to your second part of the question, it is not important if the grid cells are in the terrestrial or marine environment, maybe except if there are some specific habitats that are let's say terrestrial-marine (i.e.: estuaries).

    Do you have a specific question ?

    Iurie Maxim

  • Ciprian SAMOILĂ

    By Ciprian SAMOILĂ

    Hi Iurie,

    Is the Range Tool that you are referring to above available online for public download?

    Best,

    Ciprian Samoila

  • Iurie MAXIM

    Hi Ciprian,

    Indeed the Range Tool I am referring is available for public download. It creates dissolved grided distribution and range based on occurrence or on undissolved distribution data.

    It can be found here: https://bd.eionet.europa.eu/activities/Reporting/Article_17/Reporting_Tool/Reporting_Tool_Software

    For implementation of HB and SD best current solution is to create polygons/multipart polygons for each species/habitat based on EEA grids by merging adjacent grid-cells. So simply create polygons having the vertices in the grid cells corners.

    If the GML will be provided in ATOM in LAEA5210 ETRS89 projection, than you can minimize the number of vertices by putting only those that are at the extremities, so not in all the exterior grid-cell corners, but I do not advise you to do so because if someone would like to reproject the the geometry in another projection, than only the vertices will be reprojected and the lines will not be curved, so for long distances (as it may be the case especially in the sea for MS that have large marine areas), the shape and the area will be not correct. If providing the data trough WFS or if you are aware of the issue described above, then best is to generate the polygons with the vertices in all grid cells, even if they will be colinear in LAEA5210, but once reprojected, the area and shape will still be quite correct.

    Do not provide undissolved grid-cells as they will be quite large for download and will not be displayed nicely in the view-services. When I worked at the ETC/BD for article 17 reporting, I asked MS to provide the distribution maps as disolved/merged gridd-cells in order to avoid large data to be sent and for allowing data to be displayed. To better understand have a look at http://www.arcgis.com/home/webmap/viewer.html?url=http://bio.discomap.eea.europa.eu/arcgis/rest/services/Article17/Article17_Distribution_WM/MapServer&source=sd and display the habitats distribution (uncheck species distribution and check distribution) and then filter the data by chosing "code" "is" "1110". Zoom near shores and will see how the geometries should look like (for habitat type 1110-Sandbanks).

    I do not advise to use xlinks, nor local, nor remote xlinks to grid cells IDs.

    If  local xlinks will be used, the files will become huge for download and data will not be usable in any existing GIS application, nor even for viewing. There is no GIS software that can display geometries encoded as xlinks, or at least I am not aware that there are some software that can do this by using resolveDepth parameter. If one knows such a software, please share this information.

    If remote xlinks will be used, it should be known that there is currently no WFS application that resolve remote xlinks. If someone knows a WFS that resolves remote xlinks, pleae share this information. Once there will be such applications still there will be the problem of displaying the grid cells.

    Undissolved grid cells are not looking nice unless they have no outline and have solid fill, transparent or not, but still this is not the best solution to display grid cell vector data. After many testing the best solution to display species/habitats distribution grid data is to use line hatched dissolved grid cells with outline and no solid fill.

    Range Tool can be used for generating dissolved grid cell polygonal distribution to be used as the geometry in GML, Manual of the extension can be found here https://bd.eionet.europa.eu/activities/Reporting_Tool/Documents/Range_Tool_User_Manual_18092012.pdf The tool works with ESRI ArcGIS Desktop, so if the software is not available, the extension cant be used. We planned to develop a similar tool for qGIS.

    Iurie

  • Keiran MILLARD

    By Keiran MILLARD

    The key issue relates to the fact that offshore (marine) habitats are typically captured according to WGS84 (EPSG:4326) and therefore a transformation is need to the EEA grids. I am just wondering if anyone has a production workflow that does this and have any issues been identified?

  • Iurie MAXIM

    Hi,

    Just use the range tool and will solve your problem. The tool was designed to cover this aspect as well. You should use the biogeographical regions as well în order to constrain the distribution of marine species and habitats to remain only în marine grid cells.
    Read the manual and run the tool. Member States used it for reporting species and habitats distribution and range.
    The transformation of UTM grid cells into LAEA5210 grid cells is imperfect anything you will do.
    You have two choices: either use the UTM grid cells as poligons or as point (centroids of grid cells). First will produce a larger distribution (more grid cells number than the.number of grid cells în the input), second option with centroids will create a smaller or eqial distribution than that IN the input. Decide with is better. It can be that some are better for some species/habitats, some for other.

    Iurie

  • Keiran MILLARD

    By Keiran MILLARD

    Hi Lurie, Thank-you very much for the advice. Keiran

Biodiversity & Area Management

Biodiversity & Area Management

If themes like Protected Sites, Area Management/Restriction/Regulation Zones and Reporting Units, Habitats and Biotopes, Species Distribution, Bio-geographical Regions matters to you, join these groups!