Skip to content
GitLab
Projects Groups Topics Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
  • Sign in
  • startR startR
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributor statistics
    • Graph
    • Compare revisions
  • Issues 29
    • Issues 29
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 7
    • Merge requests 7
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • CI/CD
    • Repository
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • Earth SciencesEarth Sciences
  • startRstartR
  • Issues
  • #81
Closed
Open
Issue created Dec 03, 2020 by aho@ahoMaintainer

Wrong time attributes of Hadgem3 data

Hi @nperez

I found this problem with Hadgem3 (both monthly and daily data). The retrieved time attribute doesn't seem correct. You can see the results below. For daily data, dates[1] is "1998-09-02 12:00:00 UTC" while it should be "2000-11-01 12:00:00 UTC". For monthly data, dates[1] is "2000-06-07 UTC" while it should be 2000-11-16 UTC. The returned data values are correct, only the time attribute is wrong. I also tested with startR_v1.0.3 and it returned the same results.

I check the netCDF files and the values are correct, so the problem is not in the files. Other Hadgem3 files (e.g., sdate = 1960; var = 'tasmin') also have the same problem, but MPI-ESM1.2-HR daily data is fine. (e.g., path_mpi_esm <- '/esarchive/exp/CMIP6/dcppA-hindcast/mpi-esm1-2-hr/cmip6-dcppA-hindcast_i1p1/DCPP/MPI-M/MPI-ESM1-2-HR/dcppA-hindcast/r1i1p1f1/day/$var$/gn/v20200101/$var$_day_MPI-ESM1-2-HR_dcppA-hindcast_s$sdate$-r1i1p1f1_gn_$fyear$.nc')

Do you have any clue? I can look into Start() code later (right now I'm doing the decadal prediction tests we discussed yesterday.)

Cheers,
An-Chi

  #/// daily data ///
  path_hadgem3 <- paste0('/esarchive/exp/CMIP6/dcppA-hindcast/hadgem3-gc31-mm/',
                         'cmip6-dcppA-hindcast_i1p1/DCPP/MOHC/HadGEM3-GC31-MM/',
                         'dcppA-hindcast/r1i1p1f2/day/$var$/gn/v20200101/',
                         '$var$_day_HadGEM3-GC31-MM_dcppA-hindcast_s$sdate$-r1i1p1f2_gn_$fyear$.nc')
  sdate <- '2000'
  fyear_hadgem3 <- paste0(sdate, '1101-', as.numeric(sdate) + 10, '1230')

  data <- Start(dat = path_hadgem3,
                var = 'tasmax',
                sdate = sdate,
                fyear = fyear_hadgem3, #indices(1),
                time = 'all',
                lat = indices(1),
                lon = indices(1),
                return_vars = list(lat = 'dat', lon = 'dat', time = 'sdate'),
                num_procs = 1, 
                retrieve = TRUE)

  dates <- attr(data, 'Variables')$common$time

dim(dates)
sdate  time
    1  3660
dates[1,1:5]  # WRONG!!!!
[1] "1998-09-02 12:00:00 UTC" "1998-09-03 12:00:00 UTC"
[3] "1998-09-04 12:00:00 UTC" "1998-09-05 12:00:00 UTC"

  #/// monthly data ///
  path_hadgem3 <- paste0('/esarchive/exp/CMIP6/dcppA-hindcast/hadgem3-gc31-mm/',
                         'cmip6-dcppA-hindcast_i1p1/DCPP/MOHC/HadGEM3-GC31-MM/',
                         'dcppA-hindcast/r1i1p1f2/Amon/$var$/gr/v20200316/',
                         '$var$_Amon_HadGEM3_dcppA-hindcast_s$sdate$_r1i1p1f2_gr_200011-201103.nc')

  data <- Start(dat = path_hadgem3,
                var = 'tas',
                sdate = '2000',
                time = 'all',
                latitude = indices(1),
                longitude = indices(1),
                return_vars = list(latitude = 'dat', longitude = 'dat', time = 'sdate'),
                num_procs = 1, 
                retrieve = TRUE)

  dates <- attr(data, 'Variables')$common$time
  dates[1,1:5]  #WRONG!!
[1] "2000-06-07 UTC" "2000-07-07 UTC" "2000-08-06 UTC" "2000-09-05 UTC"
[5] "2000-10-05 UTC"
Edited Dec 03, 2020 by aho
Assignee
Assign to
Time tracking