html_url,issue_url,id,node_id,user,created_at,updated_at,author_association,body,reactions,performed_via_github_app,issue
https://github.com/pydata/xarray/issues/1351#issuecomment-291544663,https://api.github.com/repos/pydata/xarray/issues/1351,291544663,MDEyOklzc3VlQ29tbWVudDI5MTU0NDY2Mw==,1197350,2017-04-04T15:53:33Z,2017-04-04T15:53:33Z,MEMBER,"Why not combine all your model arrays into a single Dataset? Since they almost certainly share coordinates, this would make sense.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,219184224
https://github.com/pydata/xarray/issues/1351#issuecomment-291533765,https://api.github.com/repos/pydata/xarray/issues/1351,291533765,MDEyOklzc3VlQ29tbWVudDI5MTUzMzc2NQ==,1197350,2017-04-04T15:19:20Z,2017-04-04T15:19:20Z,MEMBER,"There is no such thing as a ""multidimensional DataArray"" in the xarray data model. DataArrays can have ""non-dimension coordinates"" which can be used for things like grouping. 

This would work
```python
new_array = DataArray(np.zeros((2,2)), dims=['x_logical','y_logical'])
new_array.coords['x_physical'] = x_physical
new_array.coords['y_physical'] = y_physical
new_array.groupby('x_physical').mean()
```

I don't know how to do this as a one-liner. Probably easier to create a Dataset rather than DataArray for this purpose
```python
ds = xr.Dataset({'testdata': (('y_logical', 'x_logical'), np.zeros((2,2)))},
                 coords={'x_physical': (('y_logical', 'x_logical'), np.ones((2,2))),
                         'y_physical': (('y_logical', 'x_logical'), np.ones((2,2)))})
```
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,219184224