What happened: I have a script that downloads a file, reads and copies it to memory with ds.copy(deep=True), and then removes the downloaded file from disk. In 0.16.1, I get an error "No such file or directory" when I try to read the data from the deep-copied Dataset as if the Dataset was not actually copied into memory.

What you expected to happen: In 0.16.0 and earlier, the variable data is available (ds.varName.data) after it is copied into memory even after the original file was removed. But this doesn't work anymore in 0.16.1.

Minimal Complete Verifiable Example:

```python import xarray as xr import os import urllib.request

Get sample NetCDF file

url = 'https://www.unidata.ucar.edu/software/netcdf/examples/tos_O1_2001-2002.nc' FILE = 'tos_O1_2001-2002.nc' urllib.request.urlretrieve(url, FILE)

Open the NetCDF file

ds1 = xr.open_dataset(FILE)

Make a copy of the Dataset

ds2 = ds1.copy(deep=True)

and close the original

ds1.close()

remove the NetCDF file

os.remove(FILE)

Read the copied dataset

ds2 ```

Anything else we need to know?: Output for xarray v0.16.0

Output for xarray v0.16.1 FileNotFoundError: [Errno 2] No such file or directory: ...tos_O1_2001-2002.nc'

Environment:

Output of <tt>xr.show_versions()</tt> for xarray 0.16.0

INSTALLED VERSIONS ------------------ commit: None python: 3.8.5 | packaged by conda-forge | (default, Sep 16 2020, 17:19:16) [MSC v.1916 64 bit (AMD64)] python-bits: 64 OS: Windows OS-release: 10 machine: AMD64 processor: Intel64 Family 6 Model 142 Stepping 12, GenuineIntel byteorder: little LC_ALL: None LANG: None LOCALE: English_United States.1252 libhdf5: 1.10.6 libnetcdf: 4.7.4 xarray: 0.16.0 pandas: 1.1.2 numpy: 1.19.1 scipy: 1.5.0 netCDF4: 1.5.4 pydap: None h5netcdf: None h5py: 2.10.0 Nio: None zarr: None cftime: 1.2.1 nc_time_axis: None PseudoNetCDF: None rasterio: None cfgrib: 0.9.8.4 iris: None bottleneck: None dask: None distributed: None matplotlib: 3.3.2 cartopy: 0.18.0 seaborn: None numbagg: None pint: 0.16 setuptools: 49.6.0.post20200917 pip: 20.2.3 conda: None pytest: None IPython: 7.18.1 sphinx: None

Output of <tt>xr.show_versions()</tt> for xarray 0.16.1

INSTALLED VERSIONS ------------------ commit: None python: 3.8.5 | packaged by conda-forge | (default, Sep 16 2020, 17:19:16) [MSC v.1916 64 bit (AMD64)] python-bits: 64 OS: Windows OS-release: 10 machine: AMD64 processor: Intel64 Family 6 Model 142 Stepping 12, GenuineIntel byteorder: little LC_ALL: None LANG: None LOCALE: English_United States.1252 libhdf5: 1.10.6 libnetcdf: 4.7.4 xarray: 0.16.1 pandas: 1.1.2 numpy: 1.19.1 scipy: 1.5.0 netCDF4: 1.5.4 pydap: None h5netcdf: None h5py: 2.10.0 Nio: None zarr: None cftime: 1.2.1 nc_time_axis: None PseudoNetCDF: None rasterio: None cfgrib: 0.9.8.4 iris: None bottleneck: None dask: None distributed: None matplotlib: 3.3.2 cartopy: 0.18.0 seaborn: None numbagg: None pint: 0.16 setuptools: 49.6.0.post20200917 pip: 20.2.3 conda: None pytest: None IPython: 7.18.1 sphinx: Nonexarray: 0.16.0 pandas: 1.1.2 numpy: 1.19.1 scipy: 1.5.0 netCDF4: 1.5.4 pydap: None h5netcdf: None h5py: 2.10.0 Nio: None zarr: None cftime: 1.2.1 nc_time_axis: None PseudoNetCDF: None rasterio: None cfgrib: 0.9.8.4 iris: None bottleneck: None dask: None distributed: None matplotlib: 3.3.2 cartopy: 0.18.0 seaborn: None numbagg: None pint: 0.16 setuptools: 49.6.0.post20200917 pip: 20.2.3 conda: None pytest: None IPython: 7.18.1 sphinx: None