diff --git a/.changelog/6557.txt b/.changelog/6557.txt
new file mode 100644
index 00000000000..cbc01b426d9
--- /dev/null
+++ b/.changelog/6557.txt
@@ -0,0 +1,3 @@
+```release-note:enhancement
+bigquery: added `avro_options` field to `google_bigquery_table` resource
+```
diff --git a/google/resource_bigquery_table.go b/google/resource_bigquery_table.go
index 090d2d43bdc..5ff0426cc61 100644
--- a/google/resource_bigquery_table.go
+++ b/google/resource_bigquery_table.go
@@ -586,6 +586,22 @@ func resourceBigQueryTable() *schema.Resource {
 								},
 							},
 						},
+						// AvroOptions: [Optional] Additional options if sourceFormat is set to AVRO.
+						"avro_options": {
+							Type:        schema.TypeList,
+							Optional:    true,
+							MaxItems:    1,
+							Description: `Additional options if source_format is set to "AVRO"`,
+							Elem: &schema.Resource{
+								Schema: map[string]*schema.Schema{
+									"use_avro_logical_types": {
+										Type:        schema.TypeBool,
+										Required:    true,
+										Description: `If sourceFormat is set to "AVRO", indicates whether to interpret logical types as the corresponding BigQuery data type (for example, TIMESTAMP), instead of using the raw type (for example, INTEGER).`,
+									},
+								},
+							},
+						},
 
 						// IgnoreUnknownValues: [Optional] Indicates if BigQuery should
 						// allow extra values that are not represented in the table schema.
@@ -1324,6 +1340,9 @@ func expandExternalDataConfiguration(cfg interface{}) (*bigquery.ExternalDataCon
 	if v, ok := raw["hive_partitioning_options"]; ok {
 		edc.HivePartitioningOptions = expandHivePartitioningOptions(v)
 	}
+	if v, ok := raw["avro_options"]; ok {
+		edc.AvroOptions = expandAvroOptions(v)
+	}
 	if v, ok := raw["ignore_unknown_values"]; ok {
 		edc.IgnoreUnknownValues = v.(bool)
 	}
@@ -1370,6 +1389,10 @@ func flattenExternalDataConfiguration(edc *bigquery.ExternalDataConfiguration) (
 		result["hive_partitioning_options"] = flattenHivePartitioningOptions(edc.HivePartitioningOptions)
 	}
 
+	if edc.AvroOptions != nil {
+		result["avro_options"] = flattenAvroOptions(edc.AvroOptions)
+	}
+
 	if edc.IgnoreUnknownValues {
 		result["ignore_unknown_values"] = edc.IgnoreUnknownValues
 	}
@@ -1531,6 +1554,31 @@ func flattenHivePartitioningOptions(opts *bigquery.HivePartitioningOptions) []ma
 	return []map[string]interface{}{result}
 }
 
+func expandAvroOptions(configured interface{}) *bigquery.AvroOptions {
+	if len(configured.([]interface{})) == 0 {
+		return nil
+	}
+
+	raw := configured.([]interface{})[0].(map[string]interface{})
+	opts := &bigquery.AvroOptions{}
+
+	if v, ok := raw["use_avro_logical_types"]; ok {
+		opts.UseAvroLogicalTypes = v.(bool)
+	}
+
+	return opts
+}
+
+func flattenAvroOptions(opts *bigquery.AvroOptions) []map[string]interface{} {
+	result := map[string]interface{}{}
+
+	if opts.UseAvroLogicalTypes {
+		result["use_avro_logical_types"] = opts.UseAvroLogicalTypes
+	}
+
+	return []map[string]interface{}{result}
+}
+
 func expandSchema(raw interface{}) (*bigquery.TableSchema, error) {
 	var fields []*bigquery.TableFieldSchema
 
diff --git a/google/resource_bigquery_table_test.go b/google/resource_bigquery_table_test.go
index f5c40b050cb..9410427d9d3 100644
--- a/google/resource_bigquery_table_test.go
+++ b/google/resource_bigquery_table_test.go
@@ -599,6 +599,32 @@ func TestAccBigQueryTable_HivePartitioningCustomSchema(t *testing.T) {
 	})
 }
 
+func TestAccBigQueryTable_AvroPartitioning(t *testing.T) {
+	t.Parallel()
+	bucketName := testBucketName(t)
+	resourceName := "google_bigquery_table.test"
+	datasetID := fmt.Sprintf("tf_test_%s", randString(t, 10))
+	tableID := fmt.Sprintf("tf_test_%s", randString(t, 10))
+	avroFilePath := "./test-fixtures/bigquerytable/avro-generated.avro"
+
+	vcrTest(t, resource.TestCase{
+		PreCheck:     func() { testAccPreCheck(t) },
+		Providers:    testAccProviders,
+		CheckDestroy: testAccCheckBigQueryTableDestroyProducer(t),
+		Steps: []resource.TestStep{
+			{
+				Config: testAccBigQueryTableAvroPartitioning(bucketName, avroFilePath, datasetID, tableID),
+			},
+			{
+				ResourceName:            resourceName,
+				ImportState:             true,
+				ImportStateVerify:       true,
+				ImportStateVerifyIgnore: []string{"deletion_protection"},
+			},
+		},
+	})
+}
+
 func TestAccBigQueryTable_RangePartitioning(t *testing.T) {
 	t.Parallel()
 	resourceName := "google_bigquery_table.test"
@@ -1591,6 +1617,44 @@ EOH
 `, bucketName, datasetID, tableID)
 }
 
+func testAccBigQueryTableAvroPartitioning(bucketName, avroFilePath, datasetID, tableID string) string {
+	return fmt.Sprintf(`
+resource "google_storage_bucket" "test" {
+  name          = "%s"
+  location      = "US"
+  force_destroy = true
+}
+
+resource "google_storage_bucket_object" "test" {
+  name    = "key1=20200330/init.avro"
+  source = "%s"
+  bucket  = google_storage_bucket.test.name
+}
+
+resource "google_bigquery_dataset" "test" {
+  dataset_id = "%s"
+}
+
+resource "google_bigquery_table" "test" {
+  deletion_protection = false
+  table_id   = "%s"
+  dataset_id = google_bigquery_dataset.test.dataset_id
+
+  external_data_configuration {
+    source_format = "AVRO"
+    autodetect = true
+    source_uris= ["gs://${google_storage_bucket.test.name}/*"]
+
+    avro_options {
+      use_avro_logical_types = true
+    }
+
+  }
+  depends_on = ["google_storage_bucket_object.test"]
+}
+`, bucketName, avroFilePath, datasetID, tableID)
+}
+
 func testAccBigQueryTableRangePartitioning(datasetID, tableID string) string {
 	return fmt.Sprintf(`
   resource "google_bigquery_dataset" "test" {
diff --git a/google/test-fixtures/bigquerytable/avro-generated.avro b/google/test-fixtures/bigquerytable/avro-generated.avro
new file mode 100644
index 00000000000..c10ee312e51
Binary files /dev/null and b/google/test-fixtures/bigquerytable/avro-generated.avro differ
diff --git a/website/docs/r/bigquery_table.html.markdown b/website/docs/r/bigquery_table.html.markdown
index c13fc58a4bd..a71beeab998 100644
--- a/website/docs/r/bigquery_table.html.markdown
+++ b/website/docs/r/bigquery_table.html.markdown
@@ -170,6 +170,10 @@ in Terraform state, a `terraform destroy` or `terraform apply` that would delete
     partitioning on an unsupported format will lead to an error, as will providing
     an invalid specification. Structure is [documented below](#nested_hive_partitioning_options).
 
+* `avro_options` (Optional) - Additional options if `source_format` is set to  
+    "AVRO".  Structure is [documented below](#nested_avro_options).
+
+
 * `ignore_unknown_values` (Optional) - Indicates if BigQuery should
     allow extra values that are not represented in the table schema.
     If true, the extra values are ignored. If false, records with
@@ -261,6 +265,13 @@ in Terraform state, a `terraform destroy` or `terraform apply` that would delete
     can be either of `gs://bucket/path_to_table` or `gs://bucket/path_to_table/`.
     Note that when `mode` is set to `CUSTOM`, you must encode the partition key schema within the `source_uri_prefix` by setting `source_uri_prefix` to `gs://bucket/path_to_table/{key1:TYPE1}/{key2:TYPE2}/{key3:TYPE3}`.
 
+<a name="nested_avro_options"></a>The `avro_options` block supports:
+
+* `use_avro_logical_types` (Optional) - If is set to true, indicates whether  
+    to interpret logical types as the corresponding BigQuery data type  
+    (for example, TIMESTAMP), instead of using the raw type (for example, INTEGER).
+    
+
 <a name="nested_time_partitioning"></a>The `time_partitioning` block supports:
 
 * `expiration_ms` -  (Optional) Number of milliseconds for which to keep the