janstey ceda-cc and cfcheck report for CanESM5 v20190429

[mollymacrae]

The following errors were raised by the ceda-cc quality checks for the control/s20180125-r3i1p2f1/ test sample:

[mollymacrae]

Global attribute: mip_era should be changed to 'SNAPSI' not 'CMIP6'

[mollymacrae]

Variable siconc is not in table 6hrPt, is this meant to be sinconca?

[mollymacrae]

cell_methods attribute erros: The cell_methods attribute needs to be changed to match the MIP tables for the following variables

mrso, mrsos, snd, snw: need to change cell_methods attribute to cell_methods="area: mean where land time: point" e.g. for mrso: ERROR: Variable [mrso] has incorrect attributes: cell_methods="area: mean where land time: mean" [correct: "area: mean where land time: point"]

rlut, tas,uas, vas: need to change cell_methods attribute to cell_methods="area: mean time: point" e.g. for rlut: ERROR: Variable [rlut] has incorrect attributes: cell_methods="area: mean time: mean" [correct: "area: mean time: point"]

sithick: need to change cell_methods attribute to cell_methods="area: time: point where sea_ice" ERROR: Variable [sithick] has incorrect attributes: cell_methods="area: mean where sea_ice time: mean" [correct: "area: time: point where sea_ice"]

tntnd, utendepfd, utendnd, utendogw, utendvtem, utendwtem, vtendogw, xgwdparam, ygwdparam: need to change cell_methods attribute to cell_methods="longitude: mean time: mean" e.g. for tntnd: ERROR: Variable [tntnd] has incorrect attributes: cell_methods="longitude: mean time: point time: mean" [correct: "longitude: mean time: mean"]

tos: need to change cell_methods attribute to cell_methods= "area: mean where sea time: point" ERROR:Variable [tos] has incorrect attributes: cell_methods="area: mean where sea time: mean" [correct: "area: mean where sea time: point"]

[mollymacrae]

units: the units attribute for utendvtem and vtendwtem needs to be changed to units="m s-2" e.g. ERROR: Variable [utendvtem] has incorrect attributes: units="m s-1 d-1" [correct: "m s-2"]

[aph42]

Variable siconc is not in table 6hrPt, is this meant to be sinconca?

cell_methods attribute erros: The cell_methods attribute needs to be changed to match the MIP tables for the following variables

As discussed in the CEDA meeting on Oct 19, SNAPSI is happy to accept siconc in lieu of sinconca, and the cell methods should describe what was done/output.

[JamesAnstey]

units: the units attribute for utendvtem and vtendwtem needs to be changed to units="m s-2" e.g. ERROR: Variable [utendvtem] has incorrect attributes: units="m s-1 d-1" [correct: "m s-2"]

The units attribute has been changed to m s-2 for these. I plotted the data to confirm that the units of these fields are in fact m s-2, i.e. the units of the actual data are correct and it was the metadata that was wrong.

Just to add, the m s-1 d-1 units I had in an older version of the SNAPSI tables probably propagated from the CMIP6 cmor tables, which still have the incorrect units for utendwtem and utendvtem. The DynVarMIP data request in Gerber and Manzini 2016, Table 2 requested m s-2 for these fields.

[JamesAnstey]

Global attribute: mip_era should be changed to 'SNAPSI' not 'CMIP6'

I've set mip_era = "SNAPSI" in all files

[JamesAnstey]

Variable siconc is not in table 6hrPt, is this meant to be sinconca?

cell_methods attribute erros: The cell_methods attribute needs to be changed to match the MIP tables for the following variables

As discussed in the CEDA meeting on Oct 19, SNAPSI is happy to accept siconc in lieu of sinconca, and the cell methods should describe what was done/output.

If it becomes possible to provide siconca I'll let you know, but thanks for accepting siconc.

cell_methods in the files reflects what the data actually are, and here's the list of variables for which cell_methods differs from the specification in the SNAPSI tables (version 0.9.4):

6hrPt: mrso, mrsos, rlut, sithick, snd, snw, tas, tos, uas, vas
6hrZ: tntnd, utendepfd, utendogw, utendnd, utendvtem, utendwtem, vtendogw, xgwdparam, ygwdparam