写入 netcdf 时重命名 xarray 变量恢复为旧变量

Question

I am trying to rename a variable in some netcdf files. I am able to accomplish this using xr.Dataset.rename() . However, when I write that out to a netcdf file it has the old variable name that I was trying to replace. Here is the output from the original:

xr.open_dataset('/Volumes/Ext HDD 1/Python_data/crop_means/argentinacorn_tp_mean.nc')

#rename and save files
import glob
import xarray as xr
precip_files=glob.glob('/Volumes/Ext HDD 1/Python_data/crop_means/*tp*.nc')

for i in precip_files:

    data=xr.open_dataset(i)
    data2=data.rename_vars(name_dict={'tp_mean':'total_precip'})
    data2.to_netcdf('/Volumes/Ext HDD 1/Python_data/crop_means2/'+i.split('/')[-1].split('_'). 
    [0]+'_total_precip_mean.nc')

When loading in one of the new files this is the output:

xr.open_dataset('/Volumes/Ext HDD 1/Python_data/crop_means2/argentinacorn_total_precip_mean.nc')

They are exactly the same. If I just run the code without writing it out, this is what outputs:

data=xr.open_dataset('/Volumes/Ext HDD 1/Python_data/crop_means/argentinacorn_tp_mean.nc')
data=data.rename_vars(name_dict={'tp_mean':'total_precip'})

Answer 1

Using data.rename instead of data.rename_vars will solve this issue.

    data2=data.rename(name_dict={'tp_mean':'total_precip'})
    data2.to_netcdf(...)