- Email the appropriate data owner/modelling group to inform the owner themselves of the issue. The contact details can be found in the netCDF global attributes, for example:
:contact = "email@example.com" ;
- Email the esg-issue mailing list (firstname.lastname@example.org) to inform ESGF, who will in turn notify the data owners.
- Email email@example.com with the issue you've found and we will attempt to verify the issue and, if the direct contact details are not available, will assist you to notify data owners/modelling group.
See https://es-doc.github.io/esdoc-errata-client/faq.html if you have more questions about the Errata service or please email firstname.lastname@example.org if you have any other questions about the CMIP data at NCI.
To ensure a clear and discoverable management of errata for CMIP6 data, an errata service has been developed by Earth System Documentation (ES-Doc) https://es-doc.org/cmip6-dataset-errata/. The errata service provides a standard means of documenting and reporting errata discovered in a dataset. Through the ES-Doc service, users are provided with a means to search and view the errata and version history of a dataset. The search service provide a means to search data errata history via either keyword faceted search, or PID (PID = persistent identiers and each PID will resolve to the list of affected datasets and the thus the related errata) search.
Details of all the errata that has affected data hosted at NCI is contained within the CMIP6 Dataset Errata page.
In addition to the reported issues listed below, an external errata log is also maintained PCMDI for CMIP5: https://pcmdi.llnl.gov/mips/cmip5/errata.html.
30-Apr-2018: Reporter: Kate Snow. Description: Issues were found with the location of data files under the ACCESS1-0, r3i1p1, historical data on rr3. Two separate issues were discovered:
- The ACCESS1-0, daily, historical, r3i1p1 sea ice data contains two versions with overlapping time periods. This affects all sea-ice variables: sic, sit, usi and vsi.
- The ACCESS1-0, historical, r3i1p1, monthly data contain files covering the time period of 2006-2020 when files within the "historical" experiment should only go to the end of 2005. This affects numerous variables.
Resolution: Data files not within the correct directory locations will be moved appropriately. Specifically:
- The sea-ice files with overlapping time periods will be separated into two separate versions. Files in /g/data/rr3/publications/CMIP5/output1/CSIRO-BOM/ACCESS1-0/historical/day/seaIce/day/r3i1p1/ that end in "19610101-19701231.nc", "19710101-19801231.nc", "19810101-19901231.nc", "19910101-20001231.nc" or "20010101-20051231.nc" will be moved from the v20140402 version to another version labelled v20130924, with the "latest" folder being correspondingly updated.
- Any files in /g/data/rr3/publications/CMIP5/output1/CSIRO-BOM/ACCESS1-0/historical/mon under the r3i1p1 ensemble that cover the years from 2006-2020 will be moved from the "historical" experiment directory to the g/data/rr3/publications/CMIP5/output1/CSIRO-BOM/ACCESS1-0/historicalExt/mon directory, under the ensemble folder r3i1p1.
- Once the above changes have been completed, the online published versions of the these ACCESS1-0 datasets will also be updated.
05-Mar-2018: Reporter: Original communication came from Samuel Somot at Meteo-France/CNRM. Description: From the communication "An error has been detected in some files of the simulation « historical_r1i1p1 » of the CNRM-CM5 CMIP5 model. The atmosphere data at the 6-hourly frequency on the model level (6hLev) for the variables ps, ta, ua, va, hus do not come for the member r1i1p1. We therefore advise not to use them." Full details here: Communication-Issue-Files_CNRM-CM5_historical_6hLev_en.pdf Note that the affected files also affect CORDEX simulations using those files as lateral boundary conditions. Full details in the attached.
Resolution: Tim Erwin to adjust the DRS path and move affected CNRM-CM5 CMIP5 files from the r1i1p1 folder to a new folder named historical_r1i1p1. A README is also to be added to the original r1i1p1 path to advice users of the issue.
13-Jan-2017: Reporter: Claire Trenham. Description: Gridspec historical fx files for ocean and seaIce do not contain valid global attributes for ACCESS1.0 or ACCESS1.3. The following files need to have valid metadata added to be published to ESGF:
Resolution: Claire to add metadata from ocean gridspec initially and then to both ocean and seaIce from ocean Volcello. This might not be entirely valid but it lets the data be published. Will return to ua6/authoritative with a new file version #. Simon Marsland authorised 12/1/2017 via email.
31-Oct-2016: Reporter: Laurence Garcia-Villada. Description: the year '1951' appears to be randomly found throughout the timeseries in the following file
Here is an output from CDO showyear command on the above file:
Warning (cdfScanVarAttributes) : NetCDF: Variable not found - areacella
1951 1953 1954 1955 1951 1956 1951 1959 1960 1961 1951 1963 1964 1951 1967 1968 1969 1970 1971 1972 1973 1951 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1951 1990 1991 1951 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 1951 2005
Paola: there's other replicas of same version with correct data DRSv2 and drstree are linking this one though, links should be changed to:
which corresponds exactly to the currently published v20160426 version
Resolution: Erroneous file has been removed and link changed to /g/data/ua6/unofficial-ESG-replica/tmp/tree/esgdata1.nccs.nasa.gov/thredds/fileServer/GISS/cmip5/historical/E2-H_historical_r2i1p1/tas_Amon_GISS-E2-H_historical_r2i1p1_195101-200512.nc which has the correct time dimension. - Tim
7-Sept-2016: Reporter: Paola Petrelli. Description: /g/data/ua6/authoritative contains the following broken symlink /g/data/ua6/authoritative/IPCC/CMIP5/CSIRO-BOM/ACCESS1-0/1pctCO2/mon/ocean/Omon/r1i1p1/files/friver_20121203/friver_Omon_ACCESS1-0_1pctCO2_r1i1p1_030001-043912.nc -> /projects/p66/CMIP5/published/CMIP5/output1/CSIRO-BOM/ACCESS1-0/1pctCO2/mon/ocean/Omon/r1i1p1/files/friver_20120115/friver_Omon_ACCESS1-0_1pctCO2_r1i1p1_030001-043912.nc. Resolution: This symlink points to an old structure from DCC. As the actual authoritative files is the v20120115 version (also present in the directory), this symlink has been removed by ClaireT.
8-July-2015: Reporter: Paola Petrelli and Will Hobbs. Description: Overlapping files for CanESM2 historical r1i1p1 sea ice concentration, one spanning 1850-2005, the other 1979-2005. The later was obstructing an analysis script from running. Resolution: The file was not deleted in case the IPCC AR5 time ranges turn out to be required for some reason, so this file was removed from the unofficial-esg-replica tree, and placed in unofficial-esg-replica/tmp/ceh599/CanESM2/. NOTE that there is now a dead link in the DRS tree as a result, and the DRS tree should be rebuilt.
10-Jun-2015: Reporter: Tim Bedin. Description: Damien Irving has a workflow tool which may be used to identify minor problems in data files, such as breaks in the time axis, incorrectly set missing_value flag, incorrectly masked values. Such problems (and any applied data corrections) should be reported on this page.