observing -32767 values after remaping
Added by Idris Wada over 5 years ago
Good day,
I used the remabil function to regrid a 90th percentile temp file derived from ERA5 temperature data from its native resolution to 0.5,1, and 2.5 degrees resolution and I noticed the -32767 values on some specific lon/lat throughout the time-series. Is this a bug or an error from my side? the version of my cdo is 1.9.4. I used the code remapbil,0.5grid.txt for the 0.5 degrees. I used this area sellonlatbox,4.4,12.08,8.34,12.7 to create the template.nc file for the remap. attached is my 90thtemp file, and the 3 .txt files and a sample of the error I observed.
thank you for you time.
| temp90thdegC-daily_ERA5_with_timesteps1981-2010.nc (25 MB) temp90thdegC-daily_ERA5_with_timesteps1981-2010.nc | 90th percentile temp file | ||
| 2.5by2.5_grid.txt (270 Bytes) 2.5by2.5_grid.txt | 2.5 grid file | ||
| 1by1_grid.txt (267 Bytes) 1by1_grid.txt | 1.0 grid.txt | ||
| 0.5grid.txt (275 Bytes) 0.5grid.txt | 0.5.txt | ||
| temp90th-timeseries-remap100.txt (11.5 MB) temp90th-timeseries-remap100.txt | output.txt sample error |
Replies (2)
RE: observing -32767 values after remaping - Added by Karin Meier-Fleischer over 5 years ago
Hi Idris,
its not a bug. The value -32767 is the _FillValue (missing value) of the variable t2m.
ncdump -h temp90thdegC-daily_ERA5_with_timesteps1981-2010.nc
netcdf temp90thdegC-daily_ERA5_with_timesteps1981-2010 {
dimensions:
time = UNLIMITED ; // (10950 currently)
bnds = 2 ;
longitude = 37 ;
latitude = 16 ;
variables:
double time(time) ;
time:standard_name = "time" ;
time:long_name = "time" ;
time:bounds = "time_bnds" ;
time:units = "hours since 1900-01-01 00:00:00.0" ;
time:calendar = "365_day" ;
time:axis = "T" ;
double time_bnds(time, bnds) ;
float longitude(longitude) ;
longitude:standard_name = "longitude" ;
longitude:long_name = "longitude" ;
longitude:units = "degrees_east" ;
longitude:axis = "X" ;
float latitude(latitude) ;
latitude:standard_name = "latitude" ;
latitude:long_name = "latitude" ;
latitude:units = "degrees_north" ;
latitude:axis = "Y" ;
float t2m(time, latitude, longitude) ;
t2m:long_name = "2 metre temperature" ;
t2m:units = "K" ;
t2m:_FillValue = -32767.f ;
t2m:missing_value = -32767.f ;
cdo sinfon temp90thdegC-daily_ERA5_with_timesteps1981-2010.nc
File format : NetCDF2
-1 : Institut Source T Steptype Levels Num Points Num Dtype : Parameter name
1 : unknown unknown v instant 1 1 592 1 F32 : t2m
Grid coordinates :
1 : lonlat : points=592 (37x16)
longitude : 4 to 13 by 0.25 degrees_east
latitude : 13.25 to 9.5 by -0.25 degrees_north
Vertical coordinates :
1 : surface : levels=1
Time coordinate : 10950 steps
RefTime = 1900-01-01 00:00:00 Units = hours Calendar = 365_day Bounds = true
YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss
1981-01-01 11:30:00 1981-01-02 11:30:00 1981-01-03 11:30:00 1981-01-04 11:30:00
1981-01-05 11:30:00 1981-01-06 11:30:00 1981-01-07 11:30:00 1981-01-08 11:30:00
1981-01-09 11:30:00 1981-01-10 11:30:00 1981-01-11 11:30:00 1981-01-12 11:30:00
...
The minimum latitude is 9.5 while the grid descriptor files start already at minimum latitude 8.5 respectively 8.75. Exactly for this reason you get missing values when you do remap.

-Karin
| compare.png (155 KB) compare.png |
RE: observing -32767 values after remaping - Added by Idris Wada over 5 years ago
thank you foryour response, i really appraciate.