BUG: Preserve timezone in numpy_dtype for ArrowDtype

Author: Koookadooo

pandas/core/dtypes/dtypes.py

@@ -2276,21 +2276,17 @@ def name(self) -> str:  # type: ignore[override]
     def numpy_dtype(self) -> np.dtype:
         """Return an instance of the related numpy dtype"""
         if pa.types.is_timestamp(self.pyarrow_dtype):
-            # pa.timestamp(unit).to_pandas_dtype() returns ns units
-            # regardless of the pyarrow timestamp units.
-            # This can be removed if/when pyarrow addresses it:
-            # https://github.com/apache/arrow/issues/34462
+            # Preserve timezone information if present
+            if self.pyarrow_dtype.tz is not None:
+                # Use PyArrow's to_pandas_dtype method for timezone-aware types
+                return self.pyarrow_dtype.to_pandas_dtype()
+            # Fall back to naive datetime64 for timezone-naive timestamps
             return np.dtype(f"datetime64[{self.pyarrow_dtype.unit}]")
         if pa.types.is_duration(self.pyarrow_dtype):
-            # pa.duration(unit).to_pandas_dtype() returns ns units
-            # regardless of the pyarrow duration units
-            # This can be removed if/when pyarrow addresses it:
-            # https://github.com/apache/arrow/issues/34462
             return np.dtype(f"timedelta64[{self.pyarrow_dtype.unit}]")
         if pa.types.is_string(self.pyarrow_dtype) or pa.types.is_large_string(
             self.pyarrow_dtype
         ):
-            # pa.string().to_pandas_dtype() = object which we don't want
             return np.dtype(str)
         try:
             return np.dtype(self.pyarrow_dtype.to_pandas_dtype())

pandas/tests/dtypes/test_dtypes.py

@@ -24,6 +24,7 @@
     DatetimeTZDtype,
     IntervalDtype,
     PeriodDtype,
+    ArrowDtype,
 )
 
 import pandas as pd
@@ -1103,6 +1104,32 @@ def test_update_dtype_errors(self, bad_dtype):
         with pytest.raises(ValueError, match=msg):
             dtype.update_dtype(bad_dtype)
 
+class TestArrowDtype(Base):
+    @pytest.fixture
+    def dtype(self):
+        """Fixture for ArrowDtype."""
+        import pyarrow as pa
+        return ArrowDtype(pa.timestamp("ns", tz="UTC"))
+
+    def test_numpy_dtype_preserves_timezone(self, dtype):
+        # Test timezone-aware timestamp
+        assert dtype.numpy_dtype == dtype.pyarrow_dtype.to_pandas_dtype()
+
+    def test_numpy_dtype_naive_timestamp(self):
+        import pyarrow as pa
+        arrow_type = pa.timestamp("ns")
+        dtype = ArrowDtype(arrow_type)
+        assert dtype.numpy_dtype == pa.timestamp("ns").to_pandas_dtype()
+
+    @pytest.mark.parametrize("tz", ["UTC", "America/New_York", None])
+    def test_numpy_dtype_with_varied_timezones(self, tz):
+        import pyarrow as pa
+        arrow_type = pa.timestamp("ns", tz=tz)
+        dtype = ArrowDtype(arrow_type)
+        if tz:
+            assert dtype.numpy_dtype == arrow_type.to_pandas_dtype()
+        else:
+            assert dtype.numpy_dtype == pa.timestamp("ns").to_pandas_dtype()
 
 @pytest.mark.parametrize(
     "dtype", [CategoricalDtype, IntervalDtype, DatetimeTZDtype, PeriodDtype]