NTSB investigations, 2015

duckdb -csv -c "WITH phase_names(main, name) AS (VALUES (100, 'Prior to flight'), (150, 'Standing'), (200, 'Pushback/towing'), (250, 'Taxi'), (300, 'Takeoff'), (350, 'Initial climb'), (400, 'Enroute'), (450, 'Maneuvering'), (500, 'Approach'), (550, 'Landing'), (600, 'Emergency descent'), (650, 'Uncontrolled descent'), (700, 'Post-impact'), (750, 'After landing'), (800, 'Other'), (990, 'Unknown')),
    crosswalk(code, main, name) AS (VALUES (500, 150, 'Standing'), (501, 150, 'Standing - pre-flight'), (502, 150, 'Standing - starting engine(s)'), (503, 150, 'Standing - engine(s) operating'), (504, 150, 'Standing - engine(s) not operating'), (505, 150, 'Standing - idling rotors'), (510, 250, 'Taxi'), (511, 200, 'Taxi - pushback/tow'), (512, 250, 'Taxi - to takeoff'), (513, 250, 'Taxi - from landing'), (514, 250, 'Taxi - aerial'), (520, 300, 'Takeoff'), (521, 300, 'Takeoff - roll/run'), (522, 350, 'Takeoff - initial climb'), (523, 300, 'Takeoff - aborted'), (530, 400, 'Climb'), (531, 400, 'Climb - to cruise'), (540, 400, 'Cruise'), (541, 400, 'Cruise - normal'), (542, 400, 'Maneuvering - holding (IFR)'), (550, 400, 'Descent'), (551, 400, 'Descent - normal'), (552, 600, 'Descent - emergency'), (553, 650, 'Descent - uncontrolled'), (560, 500, 'Approach'), (561, 500, 'Approach - VFR pattern - downwind'), (562, 500, 'Approach - VFR pattern - base turn'), (563, 500, 'Approach - VFR pattern - base leg/base to final'), (564, 500, 'Approach - VFR pattern - final approach'), (565, 500, 'Go-around (VFR)'), (566, 500, 'Approach - IAF to FAF/outer marker (IFR)'), (567, 500, 'Approach - FAF/outer marker to threshold (IFR)'), (568, 500, 'Circling (IFR)'), (569, 500, 'Missed approach (IFR)'), (570, 550, 'Landing'), (571, 550, 'Landing - flare/touchdown'), (572, 550, 'Landing - roll'), (573, 550, 'Landing - aborted'), (574, 550, 'Emergency landing'), (575, 550, 'Emergency landing after takeoff'), (576, 600, 'Emergency descent/landing'), (580, 450, 'Maneuvering'), (581, 450, 'Maneuvering - aerial application'), (582, 450, 'Maneuvering - turn to reverse direction'), (583, 450, 'Maneuvering - turn to landing area (emergency)'), (590, 450, 'Hover'), (591, 450, 'Hover - in ground effect'), (592, 450, 'Hover - out of ground effect'), (600, 800, 'Other'), (610, 990, 'Unknown')),
    current_codes AS (
      SELECT e.case_mkey, TRY_CAST(e.\"tier1Number\" AS INTEGER) AS code,
        row_number() OVER (PARTITION BY e.case_mkey ORDER BY coalesce(e.\"isDefiningEvent\", false) DESC, e.position) AS n
      FROM 'https://openairdocs.org/data/latest/ntsb_events.parquet' e
      WHERE e.aircrafts_position = 1 AND e.\"tier1Number\" GLOB '[1-9][0-9]*'
        AND (e.\"isDefiningEvent\" OR e.case_mkey IN (SELECT \"mKey\" FROM 'https://openairdocs.org/data/latest/ntsb_case_detail.parquet' WHERE \"isPre1982\"))
    ),
    phases AS (
      SELECT case_mkey, 'current' AS scheme, code, CASE WHEN code >= 990 THEN 990 ELSE code // 50 * 50 END AS main
      FROM current_codes WHERE n = 1
      UNION ALL
      SELECT o.case_mkey, 'older', o.\"phaseOfFlightCode\", w.main
      FROM 'https://openairdocs.org/data/latest/ntsb_occurrences.parquet' o JOIN crosswalk w ON w.code = o.\"phaseOfFlightCode\"
      WHERE o.aircrafts_position = 1 AND o.\"occurrenceNumber\" = 1 AND o.case_mkey NOT IN (SELECT case_mkey FROM current_codes)
    ),
    case_phases AS (
      SELECT p.*, c.\"eventDate\", c.\"highestInjuryLevel\" = 'Fatal' AS fatal, a.aircraft_model_code
      FROM phases p JOIN 'https://openairdocs.org/data/latest/ntsb_case_detail.parquet' c ON c.\"mKey\" = p.case_mkey
      LEFT JOIN 'https://openairdocs.org/data/latest/ntsb_aircrafts.parquet' a ON a.case_mkey = p.case_mkey AND a.position = 1
      WHERE c.\"eventDate\" >= DATE '1962-01-01' AND p.main IN (SELECT main FROM phase_names)
    )
  SELECT c.\"ntsbNumber\", c.\"eventDate\", c.\"eventType\", c.\"eventCity\", c.\"eventStateOrRegion\", a.n_number, a.\"aircraftMake\", a.\"aircraftModel\", c.\"highestInjuryLevel\", c.\"totalFatal\", c.\"completionStatus\", c.report_url
  FROM 'https://openairdocs.org/data/latest/ntsb_case_detail.parquet' c LEFT JOIN 'https://openairdocs.org/data/latest/ntsb_aircrafts.parquet' a ON a.case_mkey = c.\"mKey\" AND a.position = 1
  WHERE year(c.\"eventDate\") = 2015
    AND c.\"mKey\" IN (SELECT case_mkey FROM case_phases WHERE main = 550)
  ORDER BY c.\"eventDate\" DESC, c.\"ntsbNumber\" DESC"

Prints the whole list as CSV. It reads the release’s Parquet files over HTTPS, so there’s nothing to download first. Install DuckDB to run it.

WITH phase_names(main, name) AS (VALUES (100, 'Prior to flight'), (150, 'Standing'), (200, 'Pushback/towing'), (250, 'Taxi'), (300, 'Takeoff'), (350, 'Initial climb'), (400, 'Enroute'), (450, 'Maneuvering'), (500, 'Approach'), (550, 'Landing'), (600, 'Emergency descent'), (650, 'Uncontrolled descent'), (700, 'Post-impact'), (750, 'After landing'), (800, 'Other'), (990, 'Unknown')),
  crosswalk(code, main, name) AS (VALUES (500, 150, 'Standing'), (501, 150, 'Standing - pre-flight'), (502, 150, 'Standing - starting engine(s)'), (503, 150, 'Standing - engine(s) operating'), (504, 150, 'Standing - engine(s) not operating'), (505, 150, 'Standing - idling rotors'), (510, 250, 'Taxi'), (511, 200, 'Taxi - pushback/tow'), (512, 250, 'Taxi - to takeoff'), (513, 250, 'Taxi - from landing'), (514, 250, 'Taxi - aerial'), (520, 300, 'Takeoff'), (521, 300, 'Takeoff - roll/run'), (522, 350, 'Takeoff - initial climb'), (523, 300, 'Takeoff - aborted'), (530, 400, 'Climb'), (531, 400, 'Climb - to cruise'), (540, 400, 'Cruise'), (541, 400, 'Cruise - normal'), (542, 400, 'Maneuvering - holding (IFR)'), (550, 400, 'Descent'), (551, 400, 'Descent - normal'), (552, 600, 'Descent - emergency'), (553, 650, 'Descent - uncontrolled'), (560, 500, 'Approach'), (561, 500, 'Approach - VFR pattern - downwind'), (562, 500, 'Approach - VFR pattern - base turn'), (563, 500, 'Approach - VFR pattern - base leg/base to final'), (564, 500, 'Approach - VFR pattern - final approach'), (565, 500, 'Go-around (VFR)'), (566, 500, 'Approach - IAF to FAF/outer marker (IFR)'), (567, 500, 'Approach - FAF/outer marker to threshold (IFR)'), (568, 500, 'Circling (IFR)'), (569, 500, 'Missed approach (IFR)'), (570, 550, 'Landing'), (571, 550, 'Landing - flare/touchdown'), (572, 550, 'Landing - roll'), (573, 550, 'Landing - aborted'), (574, 550, 'Emergency landing'), (575, 550, 'Emergency landing after takeoff'), (576, 600, 'Emergency descent/landing'), (580, 450, 'Maneuvering'), (581, 450, 'Maneuvering - aerial application'), (582, 450, 'Maneuvering - turn to reverse direction'), (583, 450, 'Maneuvering - turn to landing area (emergency)'), (590, 450, 'Hover'), (591, 450, 'Hover - in ground effect'), (592, 450, 'Hover - out of ground effect'), (600, 800, 'Other'), (610, 990, 'Unknown')),
  current_codes AS (
    SELECT e.case_mkey, TRY_CAST(e."tier1Number" AS INTEGER) AS code,
      row_number() OVER (PARTITION BY e.case_mkey ORDER BY coalesce(e."isDefiningEvent", false) DESC, e.position) AS n
    FROM 'https://openairdocs.org/data/latest/ntsb_events.parquet' e
    WHERE e.aircrafts_position = 1 AND e."tier1Number" GLOB '[1-9][0-9]*'
      AND (e."isDefiningEvent" OR e.case_mkey IN (SELECT "mKey" FROM 'https://openairdocs.org/data/latest/ntsb_case_detail.parquet' WHERE "isPre1982"))
  ),
  phases AS (
    SELECT case_mkey, 'current' AS scheme, code, CASE WHEN code >= 990 THEN 990 ELSE code // 50 * 50 END AS main
    FROM current_codes WHERE n = 1
    UNION ALL
    SELECT o.case_mkey, 'older', o."phaseOfFlightCode", w.main
    FROM 'https://openairdocs.org/data/latest/ntsb_occurrences.parquet' o JOIN crosswalk w ON w.code = o."phaseOfFlightCode"
    WHERE o.aircrafts_position = 1 AND o."occurrenceNumber" = 1 AND o.case_mkey NOT IN (SELECT case_mkey FROM current_codes)
  ),
  case_phases AS (
    SELECT p.*, c."eventDate", c."highestInjuryLevel" = 'Fatal' AS fatal, a.aircraft_model_code
    FROM phases p JOIN 'https://openairdocs.org/data/latest/ntsb_case_detail.parquet' c ON c."mKey" = p.case_mkey
    LEFT JOIN 'https://openairdocs.org/data/latest/ntsb_aircrafts.parquet' a ON a.case_mkey = p.case_mkey AND a.position = 1
    WHERE c."eventDate" >= DATE '1962-01-01' AND p.main IN (SELECT main FROM phase_names)
  )
SELECT c."ntsbNumber", c."eventDate", c."eventType", c."eventCity", c."eventStateOrRegion", a.n_number, a."aircraftMake", a."aircraftModel", c."highestInjuryLevel", c."totalFatal", c."completionStatus", c.report_url
FROM 'https://openairdocs.org/data/latest/ntsb_case_detail.parquet' c LEFT JOIN 'https://openairdocs.org/data/latest/ntsb_aircrafts.parquet' a ON a.case_mkey = c."mKey" AND a.position = 1
WHERE year(c."eventDate") = 2015
  AND c."mKey" IN (SELECT case_mkey FROM case_phases WHERE main = 550)
ORDER BY c."eventDate" DESC, c."ntsbNumber" DESC

Run it at the DuckDB prompt, or in any tool that reads Parquet over HTTPS. More ways to query the release are on Use the data.