Agenda for 14 March 2022 meeting

[ccrook]

In this meeting I would like to continue progressing towards a final version of the abstract specification document. To this end I propose we discuss the following topics:

1. [x] Should the document include all proposed features or is it better to defer currently unused/unsupported features to a later version. In particular the features that could be potentially deferred are:
   • [x] geocentric addition of displacements to coordinates (currently included)
   • [x] geocentric bilinear interpolation (currently included)
   • [x] flagging grid cells affected by surface discontinuity (faulting). Following a lengthy discussion of this I have summarised the options.
2. [x] Revisit time functions. Because this document will be specifying these independently of another source they need to be right. This is different to (for example) bilinear interpolation on a grid which we are including but not uniquely defining. Having rewritten the refactored formulae and description in the informative annexe there is some inconsistency that I feel could be improved.
3. [x] What else does this specification need to be complete? Examples? What else?

[ccrook]

• [x] Action - remove geocentric methods. Leave note alerting to potential issues close to pole.

• [x] Action - flagging grid cells. Initial approach is not to include in normative specification, but to add discussion in informative section. More a field for research than for specification at this stage.

• [x] Action - leave time functions unchanged, review definition of func ref epoch in GGXF

• [x] Action - complete changes from this meeting. Submit final document for review (Kevin and Richard agreed in first instance). Find out from Keith R what we need to do to progress to official specification.

[ccrook]

The definition of function reference epoch in GGXF specification (v0.91, 2/3/2022) is

Epoch at which a deformation time function has value zero. A constant value may be added to a time function to set it to zero at this epoch. Some deformation time function types require a reference epoch for their definition. Each time function using a reference epoch has its own epoch; deformation time functions within a single ggxfGroup may have different reference epochs.

This is consistent with the usage in this specification and does not need amending IMO.

[ccrook]

All actions now completed