search

ES-DOC / esdoc-docs

ES-DOCS documentation such as presentations
5 stars 1 forks source link

Create and review a Conformance Workflow diagram #144

Closed murphysj closed 6 years ago

murphysj commented 8 years ago

Some things to consider:

a) Who will create the CIM2 classes b) How will the documents be generated (e.g. the Questionnaire??) c) What other documents will conformances link to and what other documents (e.g. experiments) will link to conformances. d) where in the pipeline will baseline cim document info be added (e.g. UID)

charliepascoe commented 8 years ago

First thoughts on Conformances conformancethoughts

charliepascoe commented 8 years ago

Link to conformance stuff in the CIM https://github.com/ES-DOC/esdoc-cim/blob/master/v2/schema/activity_classes.py

charliepascoe commented 8 years ago

Note that when I refer to "requirement" I mean any kind of numerical requirement which generally means forcingConstraint but can also refer to other kinds of requirements such as ensemble_requirement or the general requirement class.

murphysj commented 8 years ago

Charlotte,

Question 1: Are the requirement standalone cim documents or are they just attributes of the experiment?

When we talk about linking documents, what this really means is putting a construct in the "XML" of one document that points to the the ID of another.

I am trying to figure out where the IDs to the conformances will live.

Question 2: when you say spreadsheet, do you image esdoc pre-populating a spreadsheet for each modeling group to fill in that would only have their stuff?

On Thu, Jun 23, 2016 at 8:39 AM, Charlotte notifications@github.com wrote:

Note that when I refer to "requirement" I mean any kind of numerical requirement which generally means forcingConstraint but can also refers to other kinds of requirements such as ensemble_requirement or the general requirement class.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ES-DOC/esdoc-docs/issues/144#issuecomment-228071148, or mute the thread https://github.com/notifications/unsubscribe/ABcVAP4vuv2znRdtLK8yL7fa66Vz8XYbks5qOpqkgaJpZM4IfhUO .

Sylvia Murphy NESII/CIRES/NOAA Earth System Research Laboratory 325 Broadway, Boulder CO 80305 Time Zone: U.S. Mountain Web: http://www.esrl.noaa.gov/nesii/ Email: sylvia.murphy@noaa.gov Phone: 303-497-7753

charliepascoe commented 8 years ago

Answer 1. Requirements are stand-alone CIM documents and not simply attributes of experiments.
Each requirement will have a doc-id associated with it, as will each conformance.

Answer 2. As fas as CIM 2 stands at the moment a conformance document has a description and a status. The description is free text. The status is selected from the enumeration: Conformed / Partially Conformed / Not Conformed / Not Applicable.

It doesn't really matter how we capture this information but it seems to me that a spreadsheet would keep things simple. But conformances could just as easily be auto-generated by the groups, or captured in a web-form that we provide, or from a google doc spreadsheet...

The important thing is to ensure that we link the Requirement docs with their associated Conformance docs.

davidhassell commented 8 years ago

I've just being looking at Mark Elkington's proposal on this from ~1 year ago: https://files.slack.com/files-pri/T0LCK0NKV-F196XR3QF/forcingsnotes.pdf

Should we be updating the CIM for the items listed in section "Simulation Forcings Applied" (pages 3 and 4)?

murphysj commented 8 years ago

@davidhassell

I think this document should be reviewed line by line on a call...hopefully the next one on 6/28. We should be confirming that the current experimental requirements take some of these things into account.

davidhassell commented 8 years ago

conformance_workflow_v1

A first picture of a workflow

davidhassell commented 8 years ago

conformance_workflow_v2 Added that Models should be listed in an attribute of the Conformance document.

davidhassell commented 8 years ago

For the record, the forcings table 12.1 from IPCC AR5 is table_12.1.pdf

Charlotte - I think that I might be missing some information. E.g for "Pre-Industrial Land Use", where can I find out exactly what the "constant pre-industrial land-use forcing" has to actually be?

Thanks, David

charliepascoe commented 8 years ago

The forcing that needs to be conformed to should be described by the forcing constraint (requirement). In the specific case of land use forcing, CMIP6 provides a historical land use data set, but no specific pre-industrial data set. So I've linked to the reference for the historical land use data set in the pre-industrial land use forcing constraint. I have assumed that the historical data begins in pre-industrial period (1850 or there about). By the time we get to the end of the review process all the requirements should be clearly specified.

davidhassell commented 8 years ago

OK, that makes sense. Thanks.

In order to auto-generate table 12.1, I'm getting the feeling that we will need a conformance specialisation for each numerical requirement, so that table 12.1 questions like "Was CO2 forcing applied (Y/N)" can be answered. I don't think answers to such questions can be inferred from the forcing constraints (?). This idea is still embryonic.