diff --git a/backend/app/services/vector_feature_service.py b/backend/app/services/vector_feature_service.py
index 4742bba2..4795e4f4 100644
--- a/backend/app/services/vector_feature_service.py
+++ b/backend/app/services/vector_feature_service.py
@@ -209,8 +209,8 @@ class VectorFeatureService:
                 intersection_area = func.ST_Area(
                     func.ST_Transform(func.ST_Intersection(VectorFeature.geometry, envelope), 31370)
                 )
-                value_expression = numeric_value * intersection_area / func.nullif(source_area, 0.0)
-                is_estimate = True
+                coverage_ratio = intersection_area / func.nullif(source_area, 0.0)
+                value_expression = numeric_value * coverage_ratio
             aggregate_value = (
                 db.query(func.coalesce(func.sum(value_expression), 0.0))
                 .filter(*selection_filter)
@@ -218,6 +218,16 @@ class VectorFeatureService:
                 .scalar()
             )
             metric_value = float(aggregate_value or 0.0)
+            if method == "area_weighted_sum":
+                partial_feature_count = (
+                    db.query(func.count(VectorFeature.id))
+                    .filter(*selection_filter)
+                    .filter(coverage_ratio < 0.999999)
+                    .scalar()
+                )
+                is_estimate = bool(partial_feature_count)
+                if not is_estimate and config.get("warning_only_when_estimate", True):
+                    warning = None
         elif method != "feature_count":
             raise AppError(
                 code="INVALID_SELECTION_AGGREGATION",
diff --git a/backend/tests/test_sprint187_temporal_map_foundation.py b/backend/tests/test_sprint187_temporal_map_foundation.py
index 09ec3530..e6d7dda5 100644
--- a/backend/tests/test_sprint187_temporal_map_foundation.py
+++ b/backend/tests/test_sprint187_temporal_map_foundation.py
@@ -38,6 +38,14 @@ class ScalarSession:
         return ScalarQuery(self.value)
 
 
+class SequenceScalarSession:
+    def __init__(self, values: list[float]):
+        self.values = iter(values)
+
+    def query(self, *args):  # noqa: ANN002, ARG002
+        return ScalarQuery(next(self.values))
+
+
 class VersionQuery:
     def __init__(self, latest: DatasetVersion | None):
         self.latest = latest
@@ -183,6 +191,40 @@ def test_selection_area_aggregation_returns_hectares_without_loading_all_feature
     assert result["feature_count"] == 40
 
 
+def test_population_area_weighting_is_exact_for_full_features_and_estimated_for_partial_features() -> None:
+    dataset = temporal_dataset(project_id=uuid4(), observed_year=2025, metric_method="area_weighted_sum")
+    dataset.source_metadata["selection_aggregation"].update(
+        {
+            "property": "population_total",
+            "label": "Inwoners",
+            "unit": "inwoners",
+            "warning": "Partial-sector estimate",
+            "warning_only_when_estimate": True,
+        }
+    )
+    bbox = {"min_x": 5.0, "min_y": 51.1, "max_x": 5.2, "max_y": 51.3, "crs": "EPSG:4326"}
+
+    full = VectorFeatureService.summarize_features_by_bbox(
+        SequenceScalarSession([38_675.0, 0]),
+        dataset=dataset,
+        bbox=bbox,
+        total_feature_count=49,
+    )
+    partial = VectorFeatureService.summarize_features_by_bbox(
+        SequenceScalarSession([1_250.5, 2]),
+        dataset=dataset,
+        bbox=bbox,
+        total_feature_count=3,
+    )
+
+    assert full["metric_value"] == 38_675.0
+    assert full["is_estimate"] is False
+    assert full["warning"] is None
+    assert partial["metric_value"] == 1_250.5
+    assert partial["is_estimate"] is True
+    assert partial["warning"] == "Partial-sector estimate"
+
+
 def test_temporal_compare_returns_delta_and_canonical_change_payload(monkeypatch) -> None:
     project_id = uuid4()
     earlier = temporal_dataset(project_id=project_id, observed_year=2021)
@@ -244,6 +286,7 @@ def test_temporal_frontend_and_official_operator_contracts_exist() -> None:
     assert "Vergelijk periode" in workspace
     assert "/temporal/compare" in temporal_api
     assert "Statbel" in population and "area_weighted_sum" in population
+    assert '"identity_stable": False' in population
     assert "HistLandgebruik" in landuse and "intersection_area" in landuse
     assert "<wfs:GetFeature" in landuse and "session.post(" in landuse
     assert 'lambda value: value.startswith("weg"), "weg*"' in landuse
diff --git a/docs/CODEX_EXECUTION_LOG.md b/docs/CODEX_EXECUTION_LOG.md
index 3e5c1ad1..86465eb7 100644
--- a/docs/CODEX_EXECUTION_LOG.md
+++ b/docs/CODEX_EXECUTION_LOG.md
@@ -7803,3 +7803,5 @@ Live deployment correction:
 - Vector upload persistence is now atomic across Dataset, DatasetVersion and VectorFeature rows, with storage cleanup on rollback. This prevents the failed-indexing orphan state observed during the live import.
 - Tower could read the historical WFS capabilities but its gateway rejected XML `FILTER` content embedded in a GET query. The operator now sends the same read-only WFS 2.0 GetFeature request as XML POST; a live two-feature forest page returned successfully before redeployment.
 - The 1778 edition classifies roads as `weg`, while later editions can use road subclasses. The exact server-side road filter now uses `weg*`, covering the documented class family without admitting unrelated themes.
+- Live population comparison returned 37,015 inhabitants for 2021 and 38,675 for 2025 (+1,660 / +4.48%). Full covered sectors now report exact source totals; only cut sectors are labelled area-weighted estimates.
+- Population sector identity is explicitly unstable across annual geometry editions, so source boundary/code changes are not presented as added/removed population objects.
diff --git a/docs/DATA_SOURCES.md b/docs/DATA_SOURCES.md
index c757258c..d7d032e4 100644
--- a/docs/DATA_SOURCES.md
+++ b/docs/DATA_SOURCES.md
@@ -42,6 +42,10 @@ Complete sectors use their published population total. A rectangle that cuts
 through a sector uses an explicitly labelled area-weighted estimate; the
 source does not justify a more precise intra-sector distribution.
 
+Statistical-sector codes and boundaries can change between editions. GeoIntel
+therefore compares population metrics but does not present sector additions,
+removals or boundary changes as population-object lineage.
+
 ### Mol historical land use
 
 `scripts/provision_mol_historical_landuse.py` uses the official Digitaal
diff --git a/scripts/provision_mol_population_history.py b/scripts/provision_mol_population_history.py
index fdd2b168..ce5f45c0 100644
--- a/scripts/provision_mol_population_history.py
+++ b/scripts/provision_mol_population_history.py
@@ -225,14 +225,14 @@ def upload_snapshot(
         "nis_code": MUNICIPALITY_NIS_CODE,
         "attribution": ATTRIBUTION,
         "license": "CC BY 4.0",
-        "identity_stable": True,
-        "comparison_property": "population_total",
+        "identity_stable": False,
+        "identity_limitation": "Statistical-sector codes and boundaries can change between annual editions.",
         "selection_aggregation": {
             "method": "area_weighted_sum",
             "property": "population_total",
             "label": "Inwoners",
             "unit": "inwoners",
-            "is_estimate": True,
+            "warning_only_when_estimate": True,
             "warning": "Bevolking binnen een gedeeltelijke statistische sector is oppervlaktegewogen en blijft een schatting.",
         },
     }