# xarray.Dataset.bfill

#### Dataset.bfill(dim, limit=None)[[source]](https://github.com/pydata/xarray/blob/main/xarray/core/dataset.py#L6820-L6883)

Fill NaN values by propagating values backward

*Requires bottleneck.*

* **Parameters:**
  * **dim** (`Hashable`) – Specifies the dimension along which to propagate values when
    filling.
  * **limit** ([`int`](https://docs.python.org/3/builtins/functions.html#int) or [`None`](https://docs.python.org/3/builtins/constants.html#None), *optional*) – The maximum number of consecutive NaN values to backward fill. In
    other words, if there is a gap with more than this number of
    consecutive NaNs, it will only be partially filled. Must be greater
    than 0 or None for no limit. Must be None or greater than or equal
    to axis length if filling along chunked axes (dimensions).

### Examples

<div class="try_examples_button_container"><button class="try_examples_button" onclick="window.tryExamplesShowIframe('8fd88070-8914-4928-9b64-43b9f2899da5','164cb416-0190-4284-a0a3-1a503a8d73a4','374f48e9-3247-4706-bea6-45de1d6a86e9','../lite/tree/../notebooks/index.html?path=62f22790_fc48_444a_87fc_20572c9c2a56.ipynb','None')">Try it in your browser!</button></div>
```pycon
>>> time = pd.date_range("2023-01-01", periods=10, freq="D")
>>> data = np.array(
...     [1, np.nan, np.nan, np.nan, 5, np.nan, np.nan, 8, np.nan, 10]
... )
>>> dataset = xr.Dataset({"data": (("time",), data)}, coords={"time": time})
>>> dataset
<xarray.Dataset> Size: 160B
Dimensions:  (time: 10)
Coordinates:
  * time     (time) datetime64[us] 80B 2023-01-01 2023-01-02 ... 2023-01-10
Data variables:
    data     (time) float64 80B 1.0 nan nan nan 5.0 nan nan 8.0 nan 10.0
```

# filled dataset, fills NaN values by propagating values backward

```pycon
>>> dataset.bfill(dim="time")
<xarray.Dataset> Size: 160B
Dimensions:  (time: 10)
Coordinates:
  * time     (time) datetime64[us] 80B 2023-01-01 2023-01-02 ... 2023-01-10
Data variables:
    data     (time) float64 80B 1.0 5.0 5.0 5.0 5.0 8.0 8.0 8.0 10.0 10.0
```

# Limit the backward filling to a maximum of 2 consecutive NaN values

```pycon
>>> dataset.bfill(dim="time", limit=2)
<xarray.Dataset> Size: 160B
Dimensions:  (time: 10)
Coordinates:
  * time     (time) datetime64[us] 80B 2023-01-01 2023-01-02 ... 2023-01-10
Data variables:
    data     (time) float64 80B 1.0 nan 5.0 5.0 5.0 8.0 8.0 8.0 10.0 10.0
```

<div id="374f48e9-3247-4706-bea6-45de1d6a86e9" class="try_examples_outer_iframe  hidden"><div class="try_examples_button_container"><button class="try_examples_button" onclick="window.tryExamplesHideIframe('8fd88070-8914-4928-9b64-43b9f2899da5','374f48e9-3247-4706-bea6-45de1d6a86e9')">Go Back</button><button class="try_examples_button" onclick="window.openInNewTab('8fd88070-8914-4928-9b64-43b9f2899da5','374f48e9-3247-4706-bea6-45de1d6a86e9')">Open In Tab</button></div><div id="164cb416-0190-4284-a0a3-1a503a8d73a4" class="jupyterlite_sphinx_iframe_container"></div></div><script>document.addEventListener("DOMContentLoaded", function() {window.loadTryExamplesConfig("../try_examples.json");});</script>
* **Returns:**
  [`Dataset`](xarray.Dataset.html.md#xarray.Dataset)

#### SEE ALSO
[`Dataset.ffill`](xarray.Dataset.ffill.html.md#xarray.Dataset.ffill)