Post History

It seems like what you want to do can be achieved by using
```python
data.assign_coords(t=[0.123])
```

The error message is extremely confusing in this respect, but it seems like the new value for the coordinate must have the same shape as the original coordinate value.
In this case, `data.t.shape` would output `(1, )`, which corresponds to the shape of a vector.
Therefore, you also have to specify a vector-like input.

Here, I used a list, but you could also use a numpy array or compute the new entry from the original value, e.g.
```python
data.assign_coords(t=data.t + 0.123)
```
Of course, this assumes that `data.t` is zero everywhere (and not just on the edges).