diff --git a/doc/why-xarray.rst b/doc/why-xarray.rst
index d0a6c591b29..25d558d99d5 100644
--- a/doc/why-xarray.rst
+++ b/doc/why-xarray.rst
@@ -62,9 +62,8 @@ The power of the dataset over a plain dictionary is that, in addition to
 pulling out arrays by name, it is possible to select or combine data along a
 dimension across all arrays simultaneously. Like a
 :py:class:`~pandas.DataFrame`, datasets facilitate array operations with
-heterogeneous data -- the difference is that the arrays in a dataset can not
-only have different data types, but can also have different numbers of
-dimensions.
+heterogeneous data -- the difference is that the arrays in a dataset can have 
+not only different data types, but also different numbers of dimensions.
 
 This data model is borrowed from the netCDF_ file format, which also provides
 xarray with a natural and portable serialization format. NetCDF is very popular
diff --git a/xarray/tests/test_backends.py b/xarray/tests/test_backends.py
index a5c42fd368c..f6254b32f4f 100644
--- a/xarray/tests/test_backends.py
+++ b/xarray/tests/test_backends.py
@@ -2163,6 +2163,7 @@ def test_encoding_unlimited_dims(self):
 
 @requires_h5netcdf
 @requires_netCDF4
+@pytest.mark.filterwarnings("ignore:use make_scale(name) instead")
 class TestH5NetCDFData(NetCDF4Base):
     engine = "h5netcdf"
 
@@ -2173,16 +2174,25 @@ def create_store(self):
 
     @pytest.mark.filterwarnings("ignore:complex dtypes are supported by h5py")
     @pytest.mark.parametrize(
-        "invalid_netcdf, warns, num_warns",
+        "invalid_netcdf, warntype, num_warns",
         [(None, FutureWarning, 1), (False, FutureWarning, 1), (True, None, 0)],
     )
-    def test_complex(self, invalid_netcdf, warns, num_warns):
+    def test_complex(self, invalid_netcdf, warntype, num_warns):
         expected = Dataset({"x": ("y", np.ones(5) + 1j * np.ones(5))})
         save_kwargs = {"invalid_netcdf": invalid_netcdf}
-        with pytest.warns(warns) as record:
+        with pytest.warns(warntype) as record:
             with self.roundtrip(expected, save_kwargs=save_kwargs) as actual:
                 assert_equal(expected, actual)
-        assert len(record) == num_warns
+
+        recorded_num_warns = 0
+        if warntype:
+            for warning in record:
+                if issubclass(warning.category, warntype) and (
+                    "complex dtypes" in str(warning.message)
+                ):
+                    recorded_num_warns += 1
+
+        assert recorded_num_warns == num_warns
 
     def test_cross_engine_read_write_netcdf4(self):
         # Drop dim3, because its labels include strings. These appear to be
@@ -2451,6 +2461,7 @@ def skip_if_not_engine(engine):
 
 
 @requires_dask
+@pytest.mark.filterwarnings("ignore:use make_scale(name) instead")
 def test_open_mfdataset_manyfiles(
     readengine, nfiles, parallel, chunks, file_cache_maxsize
 ):