diff --git a/transfers/migrate_nmbgmr_site_names.py b/transfers/migrate_nmbgmr_site_names.py
new file mode 100644
index 00000000..2325b383
--- /dev/null
+++ b/transfers/migrate_nmbgmr_site_names.py
@@ -0,0 +1,114 @@
+"""
+One-time data migration: populate NMBGMR site names as ThingIdLink records.
+
+The legacy Location.csv has a SiteNames column with the human-readable site
+name assigned by NMBGMR (e.g. "Zwager domestic", "Pendaries Village Well #1").
+This value was never transferred into the ThingIdLink table, so the site_name
+property on Thing always returned None.
+
+This script is idempotent: it skips any (thing_id, NMBGMR, alternate_id) row
+that already exists.
+
+Usage (from repo root, with venv active):
+    python -m transfers.migrate_nmbgmr_site_names
+"""
+
+import logging
+
+import pandas as pd
+from sqlalchemy import insert, select, tuple_
+
+from db import Thing, ThingIdLink
+from db.engine import session_ctx
+from transfers.util import get_transfers_data_path
+
+logging.basicConfig(level=logging.INFO, format="%(levelname)s %(message)s")
+logger = logging.getLogger(__name__)
+
+ALTERNATE_ORGANIZATION = "NMBGMR"
+RELATION = "same_as"
+RELEASE_STATUS = "public"
+
+
+def run():
+    csv_path = get_transfers_data_path("nma_csv_cache/Location.csv")
+    logger.info("Reading %s", csv_path)
+
+    df = pd.read_csv(csv_path, dtype=str, usecols=["LocationId", "SiteNames"])
+    df = df[
+        df["SiteNames"].notna()
+        & (df["SiteNames"] != "NULL")
+        & (df["SiteNames"].str.strip() != "")
+    ].copy()
+    df["SiteNames"] = df["SiteNames"].str.strip()
+    logger.info("%d rows with a non-empty SiteNames value", len(df))
+
+    with session_ctx() as session:
+        # Match on LocationId -> nma_pk_location rather than PointID -> name.
+        # PointID is not unique across all Location rows; LocationId (the UUID
+        # primary key from NM_Aquifer) has higher fidelity. Suggested by
+        # jacob-a-brown in PR #668.
+        location_ids = df["LocationId"].tolist()
+        thing_id_by_location_id: dict[str, int] = {
+            location_id: thing_id
+            for location_id, thing_id in session.execute(
+                select(Thing.nma_pk_location, Thing.id).where(
+                    Thing.nma_pk_location.in_(location_ids)
+                )
+            ).all()
+        }
+        logger.info(
+            "%d / %d LocationIds matched a Thing in the database",
+            len(thing_id_by_location_id),
+            len(df),
+        )
+
+        # Build candidate rows.
+        candidates: list[dict] = []
+        for row in df.itertuples(index=False):
+            thing_id = thing_id_by_location_id.get(row.LocationId)
+            if thing_id is None:
+                continue
+            candidates.append(
+                {
+                    "thing_id": thing_id,
+                    "relation": RELATION,
+                    "alternate_id": row.SiteNames,
+                    "alternate_organization": ALTERNATE_ORGANIZATION,
+                    "release_status": RELEASE_STATUS,
+                }
+            )
+
+        # Skip rows that already exist (idempotent).
+        existing_keys: set[tuple[int, str, str]] = set(
+            session.execute(
+                select(
+                    ThingIdLink.thing_id,
+                    ThingIdLink.alternate_organization,
+                    ThingIdLink.alternate_id,
+                ).where(ThingIdLink.alternate_organization == ALTERNATE_ORGANIZATION)
+            ).all()
+        )
+        logger.info(
+            "%d NMBGMR ThingIdLink rows already in the database", len(existing_keys)
+        )
+
+        rows_to_insert = [
+            r
+            for r in candidates
+            if (r["thing_id"], r["alternate_organization"], r["alternate_id"])
+            not in existing_keys
+        ]
+        logger.info("%d new rows to insert", len(rows_to_insert))
+
+        if not rows_to_insert:
+            logger.info("Nothing to do.")
+            return
+
+        session.execute(insert(ThingIdLink), rows_to_insert)
+        session.commit()
+        logger.info("Done. Inserted %d NMBGMR site name links.", len(rows_to_insert))
+
+
+if __name__ == "__main__":
+    run()