Is it possible to write a .csv file from a xarray.Dataset in python?

Question

I have been using the python package xgrads to parse and read a descriptor file with a suffix .ctl which describes a raw binary 3D dataset, provided by GrADS (Grid Analysis and Display System), a widely used software for easy access, manipulation, and visualization of earth science data. I have been using the following code to read the binary data into a xarray.Dataset .

from xgrads import open_CtlDataset
dset = open_CtlDataset('./ur2m_eta40km_2001011312.ctl')

# print all the info in ctl file
print(dset)

<xarray.Dataset>
Dimensions:  (time: 553, lat: 36, lon: 30)
Coordinates:
  * time     (time) datetime64[ns] 2001-01-13T12:00:00 ... 2001-05-31T12:00:00
  * lat      (lat) float32 -21.2 -20.8 -20.4 -20.0 -19.6 ... -8.4 -8.0 -7.6 -7.2
  * lon      (lon) float32 -47.8 -47.4 -47.0 -46.6 ... -37.4 -37.0 -36.6 -36.2
Data variables:
    ur2m     (time, lat, lon) float32 dask.array<chunksize=(1, 36, 30), meta=np.ndarray>
Attributes:
    comment:  Relative Humidity 2m
    storage:  99
    title:    File
    undef:    1e+20
    pdef:     None

This .ctl file comprises forecast results of humidity, estimated over a predefined area at each 6 hours, from 2001-01-13 12:00:00 hs to 2001-05-31 12:00:00 hs. Plotting the results for the first time step (2001-01-13T12:00:00) I got this:

ds['ur2m'][0,...].plot()

I would like to know if it is possible to create tabular data from this xarray.Dataset and export it as a single .csv or .txt file, following the data structure below:

 long   lat ur2m time variable            datetime
-47.8 -21.2    0    1     ur2m 2001-01-13 12:00:00
-47.4 -21.2    0    1     ur2m 2001-01-13 12:00:00
-47.0 -21.2    0    1     ur2m 2001-01-13 12:00:00
-46.6 -21.2    0    1     ur2m 2001-01-13 12:00:00
  ...   ...  ...  ...     <NA>        ...     <NA>
-37.4  -7.2    0  553     ur2m 2001-05-31 12:00:00
-37.0  -7.2    0  553     ur2m 2001-05-31 12:00:00
-36.6  -7.2    0  553     ur2m 2001-05-31 12:00:00
-36.2  -7.2    0  553     ur2m 2001-05-31 12:00:00

The original data are available here

Answer 1

Try this: Convert netcdf to dataframe

df = ds.to_dataframe()

Save dataframe to csv

df = df.to_csv('df.csv')

Answer 2

I had a similar problem, did this and it worked.

Turn into xarray.DataArray

darray = ds.to_array()

Turn into dataset

df = darray.to_dataset("variable")

Export as csv

df.to_csv("mydata.csv")

Then you can import your csv with pandas and work on the columns to get your final dataset.