Only one "distance from reference surface" allowed per instrument instance

[tomkralidis]

(moved from wmo-cop/wmo-oscar#28)

Instrument instances (uniquely identified by manufacturer, model and serial number) might have more than one sensors/inlets at different heights. Such cases cannot be represented in the current Schema and UI. To be corrected in the next version of the schema and on the UI.

[joergklausen]

WMDR XSD defines this currently as

<xs:element name="heightAboveLocalReferenceSurface" type="gml:MeasureType" minOccurs="0" maxOccurs="1">
	<xs:annotation>
		<xs:documentation>5-05 Vertical distance of sensor from specified reference surface, in the direction away from the earth's center. Positive values indicate above reference surface, negative values indicate below references surface (e.g., below ocean surface).</xs:documentation>
	</xs:annotation>
</xs:element>

as part of a wmdr:deployment. The requirement can be met in one of two ways:

1. change cardinality of element to [0..*]
2. Specify a convention in the note section to allow a comma-separated list of values.
   Option 1 is cleaner from a modeling point of view. Option 2 is presumably cheaper from an implementation point of view.

We need to agree on the way forward.
@tomkralidis @NunesL Comments?

[tomkralidis]

I would go for option 1. As a rule, we should try to avoid CSV'ing in lieu of proper cardinality.

[david-i-berry]

For marine wind (and air temperature) observations made on board ships we need to know the height of the sensors both above the deck on which they are installed but also the sea surface. Essentially, we have one sensor but two different reference surfaces.

[joergklausen]

For marine wind (and air temperature) observations made on board ships we need to know the height of the sensors both above the deck on which they are installed but also the sea surface. Essentially, we have one sensor but two different reference surfaces.

height of the sensor above sea level is part of the 3D coordinates of the sensor/instrument.
height of the sensor above reference surface is intended to describe the distance to the nearest surface that could influence the observation.

[david-i-berry]

Sorry, I should rephrase this. It is the height above the water surface the is required. We also have observations from water bodies above sea level and so we have 3 height requirements.

Height above sea level (important for pressure measurements)
Height above the deck upon which a sensor is installed (important for understanding measurement errors and biases).
Height above the water surface (important for flux calculations between the underlying water and the atmosphere).

[anthoninlize]

Following what @david-i-berry said above, in ship weather station metadata format only the height above the deck and the height above the water surface (ie height of the deck + height above the deck) are recorded. We don't have the height above sea level (which should then take into consideration tide coefficient?).
For other instruments, eg a mooring line, it is the height/depth with regard to the water surface again. Above water is positive, below water is negative (as seen on a mooring diagram).

We (OceanOPS) transmit to OSCAR/Surface the height above the water surface as the sensor height above local reference surface.