Do BI and BO attributes also apply to :climatology variables

[TomLav]

Question

I hesitated to fill this as a defect, asking a question instead.

Some attributes (e.g. units, calendar, etc...) are of type BI. From the convention (Chapter 7.1, Cell Boundaries)

A boundary variable inherits the values of some attributes from its parent coordinate variable. If a coordinate variable has any of the attributes marked "BI" (for "inherit") in the "Use" column of Appendix A, Attributes, they are assumed to apply to its bounds variable as well. It is recommended that BI attributes not be included on a boundary variable. If a BI attribute is included, it must also be present in the parent variable, and it must exactly match the parent attribute’s data type and value.

This is clear.

However, as far as I can see, the concept of BI attributes is not re-stated in Chapter 7.4 Climatological Statistics. In that chapter, the boundary information is recorded with attribute :climatology instead of :bounds. The examples in Chapter 7.4 all show boundary variables without the "BI" attributes, so one can assume that the rule for BI attributes also applies here, but it is not clearly stated.

Early in the convention, in chapter 3.1 Units, one can read:

The units attribute is required for all variables that represent dimensional quantities (except for boundary variables defined in Section 7.1, "Cell Boundaries" and climatology variables defined in Section 7.4, "Climatological Statistics").

Thus:

1. Can we confirm that BI attributes also concern the climatology variables defined in Chapter 7.4?
2. Is this obvious enough in the convention or should we find a way to state this clearly?

[davidhassell]

Hi @TomLav,

1. Can we confirm that BI attributes also concern the climatology variables defined in Chapter 7.4?

I think that this is the case. The climatology variable certainly contains cell bounds, even if they are an encoding for "multiple time segments equivalent to multiple line geometry parts" (see the Coordinates part of the CF data model).

2. Is this obvious enough in the convention or should we find a way to state this clearly?

I think that we should state this clearly. Here's a go at this, in the second paragraph of 7.4. Climatological Statistics (suggested new text):

"... Intervals of climatological time are conceptually different from ordinary time intervals; a given interval of climatological time represents a set of subintervals which are not necessarily contiguous. To indicate this difference, a climatological time coordinate variable does not have a bounds attribute, instead it has a climatology attribute which names the climatological boundary variable. The climatological boundary variable must have dimensions (n,2), n being the dimension of the climatological time axis. The rules and recommendations for specifying climatological boundary variable attributes are the same as those for boundary variables in general, as described in Section 7.1 Cell Boundaries. Using the units and calendar of the time coordinate variable, element (i,0) of the climatology boundary variable specifies the beginning of the first subinterval and element (i,1) the end of the last subinterval used to evaluate the climatological statistics with index i in the time dimension. The time coordinates should be values that are representative of the climatological time intervals, such that an application which does not recognise climatological time will nonetheless be able to make a reasonable interpretation."

[TomLav]

Thank you @davidhassell. In addition to your suggestion of adding text to Section 7.4, maybe we could add a small thing in Annex A?

(suggested added text in italic bold)

For variable attributes, the possible values of "Use" are: C for variables containing coordinate data, D for data variables, M for geometry container variables, Q for quantization container variables, Do for domain variables, BI and BO for boundary and climatology variables (see Section 7.1, "Cell Boundaries" for the distinction between BI and BO), and - for variables with some other purpose.

[davidhassell]

Sounds good. May add a "boundary": BI and BO for boundary and climatology boundary variables

[TomLav]

Should I move forward and open an issue and associated PR (despite @davidhassell came with all the new text)? If yes, is this a defect or an enhancement?

[JonathanGregory]

I would label it as a defect issue, because climatology variables are the same as bounds in the data model, so this isn't really a change to the intention of the convention. It's debatable. What do you think, @davidhassell?

[davidhassell]

Hi,

@TomLav, it'd be great if you could start an issue and PR :)

@JonathanGregory, I agree that this should be a defect, for the reasons stated.

[JonathanGregory]

@TomLav opened conventions defect issue 593 to clarify this point. This discussion can be closed once that issue is accepted, which will occur if no-one raises any concerns on the issue before 24th March.

[JonathanGregory]

Conventions issue 593 has been merged, so I'm closing this discussion. Thanks, @TomLav and @davidhassell.