Standardize notebook table-relationship documentation cells

hms_smoke_data_centers.ipynb

@@ -1184,23 +1184,35 @@
    "id": "22",
    "metadata": {},
    "source": [
-    "## Tables Created\n",
+    "## Tables Created by This Notebook and Their Relationships\n",
     "\n",
-    "This notebook creates four PostGIS tables for NOAA HMS smoke exposure analysis. The tables are designed to separate source observations, raw geometries, long-form data-center exposure, and the final per-site summary.\n",
+    "### Tables Created / Maintained\n",
+    "1. `public.hms_smoke_days`\n",
+    "- One row per observed HMS product day (daily denominator table).\n",
     "\n",
-    "| Table | Grain | Purpose |\n",
-    "|---|---|---|\n",
-    "| `public.hms_smoke_days` | One row per observed HMS product day | Denominator table for daily percentages, including days with zero smoke polygons. Stores `smoke_date`, source metadata, and `feature_count`. |\n",
-    "| `public.hms_smoke_daily` | One row per HMS smoke polygon | Raw smoke plume geometry table. Stores `smoke_date`, satellite/time fields, normalized `density`, `density_rank`, source metadata, and `geom`. |\n",
-    "| `public.data_center_hms_smoke_dc_day` | One row per `(master_id, smoke_date)` | Long-form daily exposure table for every data center on every observed HMS day. `max_density_rank = 0` means observed no smoke; `1`, `2`, and `3` represent light/unspecified, medium, and heavy smoke exposure. |\n",
-    "| `public.data_center_hms_smoke_exposure` | One row per `master_id` | Final per-data-center summary table joinable to `public.master_data_centers`. Includes location fields, observation status, smoke-period dates, exposure-day counts, percentage metrics, worst/mean density, and longest streak metrics. |\n",
+    "2. `public.hms_smoke_daily`\n",
+    "- One row per smoke polygon geometry from HMS source products.\n",
     "\n",
-    "Recommended use:\n",
+    "3. `public.data_center_hms_smoke_dc_day`\n",
+    "- One row per `(master_id, smoke_date)` with daily smoke exposure classification.\n",
     "\n",
-    "- Use `public.data_center_hms_smoke_exposure` for most site-level analysis and ranking.\n",
-    "- Use `public.data_center_hms_smoke_dc_day` for time-series analysis, seasonal summaries, or custom thresholds.\n",
-    "- Use `public.hms_smoke_daily` when you need the original smoke plume geometries for mapping or spatial QA.\n",
-    "- Use `public.hms_smoke_days` whenever calculating percentages so no-smoke observed days remain in the denominator."
+    "4. `public.data_center_hms_smoke_exposure`\n",
+    "- One row per `master_id` with summary smoke-exposure metrics.\n",
+    "\n",
+    "### Key Relationships\n",
+    "- `public.hms_smoke_days (smoke_date)`\n",
+    "  - 1-to-many -> `public.hms_smoke_daily (smoke_date)`\n",
+    "\n",
+    "- `public.master_data_centers (master_id)`\n",
+    "  - 1-to-many -> `public.data_center_hms_smoke_dc_day (master_id, smoke_date)`\n",
+    "  - 1-to-1 (effective) -> `public.data_center_hms_smoke_exposure (master_id)`\n",
+    "\n",
+    "- `public.data_center_hms_smoke_dc_day`\n",
+    "  - many-to-1 summary rollup -> `public.data_center_hms_smoke_exposure`\n",
+    "\n",
+    "### Rerun Notes\n",
+    "- Designed for repeat refreshes as additional HMS days become available.\n",
+    "- Summary exposure table is recomputed from daily source/bridge tables so results stay consistent after reloads."
    ]
   }
  ],