diff --git a/doc/whats-new.rst b/doc/whats-new.rst
index 85e73e1b7e8..68b2d738073 100644
--- a/doc/whats-new.rst
+++ b/doc/whats-new.rst
@@ -172,8 +172,10 @@ Documentation
   By `Matthias Riße <https://github.com/risebell>`_.
 - Apply ``black`` to all the code in the documentation (:pull:`4012`)
   By `Justus Magin <https://github.com/keewis>`_.
-- Narrative documentation now describes :py:meth:`map_blocks`. :ref:`dask.automatic-parallelization`.
+- Narrative documentation now describes :py:meth:`map_blocks`: :ref:`dask.automatic-parallelization`.
   By `Deepak Cherian <https://github.com/dcherian>`_.
+- Add documentation for the parameters and return values of :py:meth:`DataArray.sel`.
+  By `Justus Magin <https://github.com/keewis>`_.
 
 Internal Changes
 ~~~~~~~~~~~~~~~~
diff --git a/xarray/core/dataarray.py b/xarray/core/dataarray.py
index 3451ff14c8f..44773e36e30 100644
--- a/xarray/core/dataarray.py
+++ b/xarray/core/dataarray.py
@@ -1076,6 +1076,19 @@ def sel(
         """Return a new DataArray whose data is given by selecting index
         labels along the specified dimension(s).
 
+        In contrast to `DataArray.isel`, indexers for this method should use
+        labels instead of integers.
+
+        Under the hood, this method is powered by using pandas's powerful Index
+        objects. This makes label based indexing essentially just as fast as
+        using integer indexing.
+
+        It also means this method uses pandas's (well documented) logic for
+        indexing. This means you can use string shortcuts for datetime indexes
+        (e.g., '2000-01' to select all values in January 2000). It also means
+        that slices are treated as inclusive of both the start and stop values,
+        unlike normal Python indexing.
+
         .. warning::
 
           Do not try to assign values when using any of the indexing methods
@@ -1088,6 +1101,45 @@ def sel(
           Assigning values with the chained indexing using ``.sel`` or
           ``.isel`` fails silently.
 
+        Parameters
+        ----------
+        indexers : dict, optional
+            A dict with keys matching dimensions and values given
+            by scalars, slices or arrays of tick labels. For dimensions with
+            multi-index, the indexer may also be a dict-like object with keys
+            matching index level names.
+            If DataArrays are passed as indexers, xarray-style indexing will be
+            carried out. See :ref:`indexing` for the details.
+            One of indexers or indexers_kwargs must be provided.
+        method : {None, 'nearest', 'pad'/'ffill', 'backfill'/'bfill'}, optional
+            Method to use for inexact matches:
+
+            * None (default): only exact matches
+            * pad / ffill: propagate last valid index value forward
+            * backfill / bfill: propagate next valid index value backward
+            * nearest: use nearest valid index value
+        tolerance : optional
+            Maximum distance between original and new labels for inexact
+            matches. The values of the index at the matching locations must
+            satisfy the equation ``abs(index[indexer] - target) <= tolerance``.
+        drop : bool, optional
+            If ``drop=True``, drop coordinates variables in `indexers` instead
+            of making them scalar.
+        **indexers_kwargs : {dim: indexer, ...}, optional
+            The keyword arguments form of ``indexers``.
+            One of indexers or indexers_kwargs must be provided.
+
+        Returns
+        -------
+        obj : DataArray
+            A new DataArray with the same contents as this DataArray, except the
+            data and each dimension is indexed by the appropriate indexers.
+            If indexer DataArrays have coordinates that do not conflict with
+            this object, then these coordinates will be attached.
+            In general, each array's data will be a view of the array's data
+            in this DataArray, unless vectorized indexing was triggered by using
+            an array indexer, in which case the data will be a copy.
+
         See Also
         --------
         Dataset.sel