diff --git a/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorUpdaterFactory.java b/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorUpdaterFactory.java
index 918f21716f455..31a1957b4fb91 100644
--- a/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorUpdaterFactory.java
+++ b/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorUpdaterFactory.java
@@ -109,24 +109,32 @@ public ParquetVectorUpdater getUpdater(ColumnDescriptor descriptor, DataType spa
           // For unsigned int64, it stores as plain signed int64 in Parquet when dictionary
           // fallbacks. We read them as decimal values.
           return new UnsignedLongUpdater();
-        } else if (isTimestamp(sparkType) &&
+        } else if (sparkType == DataTypes.TimestampType &&
           isTimestampTypeMatched(LogicalTypeAnnotation.TimeUnit.MICROS)) {
-          validateTimestampType(sparkType);
           if ("CORRECTED".equals(datetimeRebaseMode)) {
             return new LongUpdater();
           } else {
             boolean failIfRebase = "EXCEPTION".equals(datetimeRebaseMode);
             return new LongWithRebaseUpdater(failIfRebase, datetimeRebaseTz);
           }
-        } else if (isTimestamp(sparkType) &&
+        } else if (sparkType == DataTypes.TimestampType &&
           isTimestampTypeMatched(LogicalTypeAnnotation.TimeUnit.MILLIS)) {
-          validateTimestampType(sparkType);
           if ("CORRECTED".equals(datetimeRebaseMode)) {
             return new LongAsMicrosUpdater();
           } else {
             final boolean failIfRebase = "EXCEPTION".equals(datetimeRebaseMode);
             return new LongAsMicrosRebaseUpdater(failIfRebase, datetimeRebaseTz);
           }
+        } else if (sparkType == DataTypes.TimestampNTZType &&
+          isTimestampTypeMatched(LogicalTypeAnnotation.TimeUnit.MICROS)) {
+          validateTimestampNTZType();
+          // TIMESTAMP_NTZ is a new data type and has no legacy files that need to do rebase.
+          return new LongUpdater();
+        } else if (sparkType == DataTypes.TimestampNTZType &&
+          isTimestampTypeMatched(LogicalTypeAnnotation.TimeUnit.MILLIS)) {
+          validateTimestampNTZType();
+          // TIMESTAMP_NTZ is a new data type and has no legacy files that need to do rebase.
+          return new LongAsMicrosUpdater();
         } else if (sparkType instanceof DayTimeIntervalType) {
           return new LongUpdater();
         }
@@ -195,12 +203,11 @@ boolean isTimestampTypeMatched(LogicalTypeAnnotation.TimeUnit unit) {
       annotation.getUnit() == unit;
   }
 
-  void validateTimestampType(DataType sparkType) {
+  private void validateTimestampNTZType() {
     assert(logicalTypeAnnotation instanceof TimestampLogicalTypeAnnotation);
-    // Throw an exception if the Parquet type is TimestampLTZ and the Catalyst type is TimestampNTZ.
+    // Throw an exception if the Parquet type is TimestampLTZ as the Catalyst type is TimestampNTZ.
     // This is to avoid mistakes in reading the timestamp values.
-    if (((TimestampLogicalTypeAnnotation) logicalTypeAnnotation).isAdjustedToUTC() &&
-      sparkType == DataTypes.TimestampNTZType) {
+    if (((TimestampLogicalTypeAnnotation) logicalTypeAnnotation).isAdjustedToUTC()) {
       convertErrorForTimestampNTZ("int64 time(" + logicalTypeAnnotation + ")");
     }
   }
@@ -1149,10 +1156,6 @@ private static boolean isLongDecimal(DataType dt) {
     return false;
   }
 
-  private static boolean isTimestamp(DataType dt) {
-    return dt == DataTypes.TimestampType || dt == DataTypes.TimestampNTZType;
-  }
-
   private static boolean isDecimalTypeMatched(ColumnDescriptor descriptor, DataType dt) {
     DecimalType d = (DecimalType) dt;
     LogicalTypeAnnotation typeAnnotation = descriptor.getPrimitiveType().getLogicalTypeAnnotation();
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetQuerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetQuerySuite.scala
index 43103db522bac..2ba01eea51e20 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetQuerySuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetQuerySuite.scala
@@ -255,6 +255,18 @@ abstract class ParquetQuerySuite extends QueryTest with ParquetTest with SharedS
     }
   }
 
+  test("SPARK-46466: write and read TimestampNTZ with legacy rebase mode") {
+    withSQLConf(SQLConf.PARQUET_REBASE_MODE_IN_WRITE.key -> "LEGACY") {
+      withTable("ts") {
+        sql("create table ts (c1 timestamp_ntz) using parquet")
+        sql("insert into ts values (timestamp_ntz'0900-01-01 01:10:10')")
+        withAllParquetReaders {
+          checkAnswer(spark.table("ts"), sql("select timestamp_ntz'0900-01-01 01:10:10'"))
+        }
+      }
+    }
+  }
+
   test("Enabling/disabling merging partfiles when merging parquet schema") {
     def testSchemaMerging(expectedColumnNumber: Int): Unit = {
       withTempDir { dir =>