Personal tools
You are here: Home Overview Best Practices Best Practices for Vertical Datums
Document Actions

Best Practices for Vertical Datums

by David Forrest last modified 2008-06-23 10:31

This page attempts to describe the issues behind identifying and representing vertical positions for marine sensor data.

The below is a very drafty collection of thoughts on vertical datums for oceanographic measurements - DavidForrest


Vertical Datums and Marine Sensor DataDescribing the vertical position of a marine sensor is complicated by the lack of fixed and geolocatable reference systems in many marine environments.  Bouys float on the free surface of the ocean, while moorings and tripods sit underneath a layer of water with uncertain density.  Referencing floating platforms to an ellipsoid or geoid tends to ignore the time-varying surface of the sea, and determining the depth of the sea for a bottom-mounted sensor tends to ignore the uncertainty in the

Common sitings for marine data

Sensors for marine systems can be located on several types of platforms:

  1. Fixed sensor, geolocatable in three dimensions (e.g.: NOAA Tide Gauge with NAVD88
  2. Fixed sensor, geolocatable in two dimensions (e.g., NOAA Tide Gauge with no NAVD88)
  3. Fixed sensor sitting on the bottom (e.g. Sediment tripod
  4. Floating sensor suspended from the surface (e.g. the temperature sensor at 1m below the waterline of NDBC 44004 )
  5. Floating sensor suspended from the bottom (e.g.: Polar Ice Thickness Sensor

Sea Level-based systems

Systems or platforms which are ultimately referenced to the a body floating on the surface of the water should specify their measurements as based from surface.    The surface may move up and down, and its location may not be known, however the measurements with respect to the surface would be well known.  For example, consider the air temperature at 4m above the waterline of NDBC bouy 44004, or the  temperature 1m below the waterline.   Tides, weather, and waves may move the bouy up or down by several meters, while the readings will still be above or below the surface of the water.

In this case, we should use a blended CRS with a WGS84 horizontal datum (EPSG:4326) and EPSG:5113 to specify the free surface of the water.

NAVD88

Fixed stations which can be vertically located with respect to NAVD88 should reference NAVD88 as their datum. 

Data referenceable to the North American Vertical Datum of 1988  should use something like EPSG:5703.

Conversions between NAVD88 and WGS84 can be performed with http://www.ngs.noaa.gov/GEOID/

Bottom-based systems


Systems on the sea floor or tethered to the sea floor by fixed amounts are problematic.  If at all possible, these should be worked into a WGS-84 location, but the precision of the location of the measurements may be lost in the translation to a more uncertain coordinate system.  For example, if bottom boundary layer scientists are studying the bed in a set of 1cm slices above the bed, and the instrument is located in 100m of water, the depth may not be determinable to 1 cm accuracy, while the differences are important.  In this case, a 'meters above bed' datum for the measurement is most representative of the data.  However, there does not seem to be an EPSG code for bottom-mounted coordinate reference systems.


OGC Namespaces for combined CRSs

Per the OCG spec Definition identifier URNs in OGC namespace (07-092r1) v. 1.1.2, pp 15-16, section 7.5.1.d, we can form combined CRSs in OGC's namespace like this:
  • urn:ogc:def:crs,crs:EPSG:6.15:4326,crs:EPSG:6.15:5113 -- WGS84 with sea level
  • urn:ogc:def:crs,crs:EPSG:6.15:4326,crs:EPSG:6.15:5703 -- WGS84 with NAVD88


OOSTethys.org: hosted by GoMOOS | powered by Plone