UPDATE

location_data/test_data_driven_subdivision.py

@@ -2,7 +2,11 @@ from datetime import date
 
 from django.test import TestCase
 
-from location_data.data_driven_subdivision import sync_block_subdivision_with_result
+from location_data.data_driven_subdivision import (
+    EmptyObservationDatasetError,
+    build_clustering_dataset,
+    sync_block_subdivision_with_result,
+)
 from location_data.models import (
     AnalysisGridCell,
     AnalysisGridObservation,
@@ -133,3 +137,39 @@ class DataDrivenSubdivisionSyncTests(TestCase):
             self.subdivision.metadata["data_driven_subdivision"]["cluster_count"],
             2,
         )
+
+    def test_build_clustering_dataset_raises_clear_error_when_all_selected_features_are_null(self):
+        cell = AnalysisGridCell.objects.create(
+            soil_location=self.location,
+            block_subdivision=self.subdivision,
+            block_code="block-1",
+            cell_code="cell-null",
+            chunk_size_sqm=900,
+            geometry=self.boundary,
+            centroid_lat="35.689200",
+            centroid_lon="51.389200",
+        )
+        observation = AnalysisGridObservation.objects.create(
+            cell=cell,
+            run=self.run,
+            temporal_start=date(2025, 1, 1),
+            temporal_end=date(2025, 1, 31),
+            metadata={"job_refs": {"ndvi": "job-1"}},
+        )
+
+        with self.assertLogs("location_data.data_driven_subdivision", level="ERROR") as captured:
+            with self.assertRaisesRegex(
+                EmptyObservationDatasetError,
+                "Upstream processing completed but no usable feature values were persisted.",
+            ):
+                build_clustering_dataset(
+                    observations=[observation],
+                    selected_features=["ndvi", "ndwi", "lst_c", "soil_vv_db"],
+                    run=self.run,
+                    location=self.location,
+                )
+
+        joined = "\n".join(captured.output)
+        self.assertIn("No usable observations available for clustering", joined)
+        self.assertIn('"run_id": {}'.format(self.run.id), joined)
+        self.assertIn('"region_id": {}'.format(self.location.id), joined)