diff --git a/xarray/core/variable.py b/xarray/core/variable.py
index 496f956990f..53c1fd76f7b 100644
--- a/xarray/core/variable.py
+++ b/xarray/core/variable.py
@@ -215,14 +215,17 @@ def _possibly_convert_objects(values):
     """
     as_series = pd.Series(values.ravel(), copy=False)
     # When receiving objects which pd.Series can't resolve by its own
-    # we try astype-conversion to "us"-resolution for datetimes and pd.Timestamp.
+    # we try astype-conversion to "ns"-resolution for datetimes and pd.Timestamp.
     if (
         values.dtype.kind == "O"
         and as_series.dtype.kind == "O"
         and as_series.size > 0
-        and isinstance(as_series[0], datetime | pd.Timestamp)
+        and (
+            isinstance(as_series[0], datetime | pd.Timestamp)
+            or pd.api.types.is_datetime64_dtype(as_series[0])
+        )
     ):
-        as_series = as_series.astype("=M8[us]")
+        as_series = as_series.astype("=M8[ns]")
     result = np.asarray(as_series).reshape(values.shape)
     if not result.flags.writeable:
         # GH8843, pandas copy-on-write mode creates read-only arrays by default
diff --git a/xarray/tests/test_variable.py b/xarray/tests/test_variable.py
index 15a45b5d220..e3c55081d6a 100644
--- a/xarray/tests/test_variable.py
+++ b/xarray/tests/test_variable.py
@@ -1086,8 +1086,8 @@ def test_numpy_same_methods(self):
         "values, unit",
         [
             (np.datetime64("2000-01-01"), "s"),
-            (pd.Timestamp("2000-01-01T00"), "s"),
-            (datetime(2000, 1, 1), "us"),
+            (pd.Timestamp("2000-01-01T00"), "ns"),
+            (datetime(2000, 1, 1), "ns"),
         ],
     )
     def test_datetime64_conversion_scalar(self, values, unit):
@@ -1102,8 +1102,8 @@ def test_datetime64_conversion_scalar(self, values, unit):
         "values, unit",
         [
             (np.timedelta64(1, "D"), "s"),
-            (pd.Timedelta("1 day"), "us"),
-            (timedelta(days=1), "us"),
+            (pd.Timedelta("1 day"), "ns"),
+            (timedelta(days=1), "ns"),
         ],
     )
     def test_timedelta64_conversion_scalar(self, values, unit):
@@ -1126,12 +1126,12 @@ def test_0d_str(self):
     def test_0d_datetime(self):
         # todo: check, if this test is OK
         v = Variable([], pd.Timestamp("2000-01-01"))
-        assert v.dtype == np.dtype("datetime64[s]")
+        assert v.dtype == np.dtype("datetime64[ns]")
         assert v.values == np.datetime64("2000-01-01", "s")
 
     @pytest.mark.filterwarnings("ignore:Converting non-default")
     @pytest.mark.parametrize(
-        "values, unit", [(pd.to_timedelta("1s"), "us"), (np.timedelta64(1, "s"), "s")]
+        "values, unit", [(pd.to_timedelta("1s"), "ns"), (np.timedelta64(1, "s"), "s")]
     )
     def test_0d_timedelta(self, values, unit):
         # todo: check, if this test is OK
@@ -2679,7 +2679,7 @@ def test_datetime(self):
         actual = as_compatible_data(datetime(2000, 1, 1))
         assert np.asarray(expected) == actual
         assert np.ndarray is type(actual)
-        assert np.dtype("datetime64[us]") == actual.dtype
+        assert np.dtype("datetime64[ns]") == actual.dtype
 
     def test_tz_datetime(self) -> None:
         # todo: check, if this test is OK
@@ -3002,7 +3002,7 @@ def test_from_pint_wrapping_dask(self, Var):
         (np.array([np.datetime64("2000-01-01", "ns")]), "ns"),
         (np.array([np.datetime64("2000-01-01", "s")]), "s"),
         (pd.date_range("2000", periods=1), "ns"),
-        (datetime(2000, 1, 1), "us"),
+        (datetime(2000, 1, 1), "ns"),
         (np.array([datetime(2000, 1, 1)]), "ns"),
         (pd.date_range("2000", periods=1, tz=pytz.timezone("America/New_York")), "ns"),
         (
@@ -3079,7 +3079,7 @@ def test_pandas_two_only_datetime_conversion_warnings(
         (np.array([np.timedelta64(10, "ns")]), "ns"),
         (np.array([np.timedelta64(10, "s")]), "s"),
         (pd.timedelta_range("1", periods=1), "ns"),
-        (timedelta(days=1), "us"),
+        (timedelta(days=1), "ns"),
         (np.array([timedelta(days=1)]), "ns"),
         (pd.timedelta_range("1", periods=1).astype("timedelta64[s]"), "s"),
     ],