[Medin_standards] FW: Vertical CRS

[Lowry, Roy K]

Well worth sharing with the list!

From: James Rapaport [mailto:James.Rapaport at seazone.com]
Sent: 30 June 2010 15:16
To: Lowry, Roy K
Subject: RE: Vertical CRS

Hi Roy,

Using that code will not break anything. Logically it's incorrect though because it refers to a coordinate system, rather than a coordinate reference system. The ISO 19111 definition (ish) of CRS is:

Coordinate system that has its position, scale and orientation defined with respect to an object; this is accomplished through a datum<about:blank#szsl:0018>. In the context of spatial systems, the object is usually the Earth. However, it could be an object moving relative to the Earth, for example, an aeroplane carrying an aerial camera.

The key thing is that a CRS therefore, contains a definition of the coordinate system (which could be urn:ogc:def:cs:EPSG::6498) and the datum (in your case instantaneous sea level).

What I propose is that I contact OGP Surveying and Positioning and enquire about the validity (in a geodetic sense) of the examples I sent you and enquire more generally about whether such CRSs could be registered. Bear in mind that the EPSG database is for the Oil and Gas industry - its scope is national level geodetic parameters (e.g. OS National Grid) and those with a direct relevance to Oil and Gas. Therefore, there are no guarantees that they would accept these CRSs.

A geodetic parameter registry containing marine vertical CRSs is needed. Would MEDIN fund one? I wonder if there's scope for collaborating with OGP (EPSG)... Setting up the registry isn't the thing - it's having geodetic experts on hand to quality control the contents.

In the meantime you could do this (see attached XML for full encoding - example from INSPIRE so not quite MEDIN valid):

<gmd:extent>
        <gmd:EX_Extent>
          <gmd:verticalElement>
            <gmd:EX_VerticalExtent>
              <gmd:minimumValue>
                <gco:Real>42</gco:Real>
              </gmd:minimumValue>
              <gmd:maximumValue>
                <gco:Real>94</gco:Real>
              </gmd:maximumValue>
              <gmd:verticalCRS>
                <gml:VerticalCRS gml:id="SeaSurfaceDepthCrs">
                  <gml:identifier codeSpace="MEDIN">SeaSurfaceDepthCrs</gml:identifier>
                  <gml:name>Sea surface depth</gml:name>
                  <gml:remarks>Depths measured relative to the sea surface.</gml:remarks>
                  <gml:scope>Used for oceanographic surveying.</gml:scope>
                  <gml:verticalCS xlink:href="urn:ogc:def:cs:EPSG::6498"/>
                  <gml:verticalDatum>
                    <gml:VerticalDatum gml:id="SeaSurfaceDatum">
                      <gml:identifier codeSpace="MEDIN">SeaSurfaceDatum</gml:identifier>
                      <gml:name>Sea Surface Datum</gml:name>
                      <gml:remarks>Heights relative to the sea surface</gml:remarks>
                      <gml:scope>Oceanographic survey.</gml:scope>
                      <gml:anchorDefinition>Instantaneous position of the sea surface.</gml:anchorDefinition>
                    </gml:VerticalDatum>
                  </gml:verticalDatum>
                </gml:VerticalCRS>
              </gmd:verticalCRS>
            </gmd:EX_VerticalExtent>
          </gmd:verticalElement>
        </gmd:EX_Extent>
      </gmd:extent>

Cheers,

James

________________________________
From: medin_standards-bounces at biwebs1.nerc-liv.ac.uk [mailto:medin_standards-bounces at biwebs1.nerc-liv.ac.uk] On Behalf Of Lowry, Roy K
Sent: 30 June 2010 13:04
To: medin_standards at mailman.nerc-liv.ac.uk
Subject: [Medin_standards] FW: Vertical CRS
Hi James,

Thought this should be shared with the list and not just me. The encodings look OK to me, although these things are way outside my comfort zone.  However, it doesn't solve my current problem of what I put into sub-element 14.3 of the MEDIN documents we're creating. If we could follow Steve's suggestion and get these registered with EPSG then that would be problem solved.  Is that possible?

In the meantime, if I populated with urn:ogc:def:cs:EPSG::6498, which I assume at least specifies that z increases downwards, can you see any issues?

Cheers, Roy.

From: James Rapaport [mailto:James.Rapaport at seazone.com]
Sent: 29 June 2010 13:41
To: Lowry, Roy K
Subject: RE: Vertical CRS

Hi Roy,

I can't find anything registered (i.e. in EPSG or otherwise). I've cooked up a couple of XML files which you should be able to slot into an ISO 19139 record:

SeaSurfaceDepthCrs.xml contains a definition of a positive down CRS.

SeaSurfaceHeightCrs.xml contains a definition of a positive up CRS.

I think they are ok - at least I think they'll contain enough information suitable for metadata. What do you think?

James

________________________________
From: medin_standards-bounces at biwebs1.nerc-liv.ac.uk [mailto:medin_standards-bounces at biwebs1.nerc-liv.ac.uk] On Behalf Of Lowry, Roy K
Sent: 29 June 2010 10:56
To: medin_standards at mailman.nerc-liv.ac.uk
Subject: [Medin_standards] Vertical CRS
Dear All,

Does anybody have any suggestion what the vertical CRS (MEDIN element 14.3) should be for depths measured relative to an instantaneous sea level (i.e. zero is the sea surface at the place and time of the measurement)?

Cheers, Roy.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.nerc-liv.ac.uk/pipermail/medin_standards/attachments/20100630/8602723e/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dataset_verticalCRS.xml
Type: application/xml
Size: 16773 bytes
Desc: dataset_verticalCRS.xml
Url : http://mailman.nerc-liv.ac.uk/pipermail/medin_standards/attachments/20100630/8602723e/attachment.xml