Coordination Action for the integration of


Solar System Infrastructures and Science

Coordinates

How to describe coordinates is important in observational metadata but could alse be important in other parts of a data system. The issues of spatial coordinates depends on what is being measured and described; temporal and spectral coordinates should not have the same issues.

Spatial Coordinates

Give the diversity of the different types of science within the Solar System, it is not possible to define a single coordinate system than can be used in all cases. The coordinate system that is most suitable depends on the domain, the science topic and Solar System body in question.

For example, if when undertaking a study in geo-sciences, the coordinate system need to be related to the Earth. If the study of things on or near the surface, a coordinate system related to the rotation axis of the Earth may be appropriate; if the study os of the ionopshere or magnetopshere then the coordinate system that is related to the Earth's magnetic field is required. The further from the Earth's surface, the more the coordinate system needs to be related to the Earth-Sun line.

The HELIO project is heliophysics, studying the effects of the Sun on the Solar System. The project has produced a document describing the Spatial Coordinate Systems that it is using that chooses different coordinates systems for different part of the problem.

The IVOA has produced a set of recommendation on Space-Time Coordinate Metadata for the Virtual Observatory that contains some useful material but is more oriented towards astrophysics.

Temporal Coordinates

While the selection of appropriate spatial coordinates depends very much on what is being observed and the location of the observer, there ought to be fewer issues in relation to how time is expressed. However, for data sets that use humans rather than machines to supply the time information, the formats used for the time fields are proving to be very variable, particularly for event data.

All time information should be expressed in a format that is compliant with standards developed by the Consultative Committee for Space Data Systems (CCSDS). In addition, time should be Coordinated Universal Time (UTC) or one of its components such as International Atomic Time (TAI).

In some datasets, time is related to a particular epoch – in the case of a spacecraft this might be launch, start of the cruise phase, orbital insertion or some things similar. This is not a problem so long as this is provided as an additional parameter; if it replaces the proper time information then the data can be difficult to interpret for anyone that is not familiar with the data (c.f. assumed knowledge).

The same argument holds where dates are given as day of year: so long as this is supplied as an additional parameter there is no problem; if the information is mixed with the other elements in the time field then the temporal information becomes very difficult to assimilate in any automated fashion.

Spectral Coordinates

Depending on the type of observation and instrument, spectral information might be expressed in wavelength, frequency or energy.

Confusion can occur when the orders of magnitude associated with a value are different for different communities. For example, traditionally there is a decametric radio community in solar physics but a decimetre is no longer a normal fraction of a metre and another community may use a different term.

The units being used should always be properly described and as far as possible standard factors of 10 should be used; ensuring that this is done is one of the difficulties associated defining metadata (c.f. parameter annotation).