I now have the complete column list from the CSV header. Let me also confirm the column descriptions from the metadata I already retrieved. I have enough to answer. Here's the review of the columns for this dataset. ## BTS ATRI Freight Mobility Initiative: County-to-county truck travel time (2022) — Experimental **Source:** [data.gov catalog page](https://catalog.data.gov/dataset/bts-atri-freight-mobility-initiative-county-to-county-truck-travel-time-2022-experimental) · underlying Socrata resource `d7b8-pmxm` on [data.bts.gov](https://data.bts.gov/api/views/d7b8-pmxm/columns.json) The dataset has **11 columns** (confirmed from the CSV header of the live resource). Note: the catalog page itself does **not** list column names — it only links to the columns metadata endpoint, and the `data.transportation.gov` view ID `d7b8-pmxm` now returns "not found"; the working copy lives on `data.bts.gov`. | # | Column (field name) | Type | Description | |---|---|---|---| | 1 | Origin CTFIPS (`orig_county_fips`) | number | Origin county FIPS code | | 2 | Origin State (`orig_state`) | text | Origin county state (2-letter) | | 3 | Origin County (`orig_county_name`) | text | Origin county name | | 4 | Destination CTFIPS (`dest_county_fips`) | number | Destination county FIPS code | | 5 | Destination State (`dest_state`) | text | Destination county state | | 6 | Destination County (`dest_county_name`) | text | Destination county name | | 7 | Year (`year`) | number | Calendar year (2022) | | 8 | Pair Moves (`pair_moves`) | text | Volume of truck movements for the county pair, bucketed (e.g. `<1K`) | | 9 | Travel Time 25th percentile (`tt_percentile_25`) | number | 25th-percentile travel time (seconds) | | 10 | Travel Time 50th percentile (`tt_percentile_50`) | number | Median travel time (seconds) | | 11 | Travel Time 75th percentile (`tt_percentile_75`) | number | 75th-percentile travel time (seconds) | **Key observations:** - The table is a **county-pair origin/destination matrix**: three origin fields (FIPS, state, name) and three destination fields, plus a year. - Travel times are given as **three percentiles** (25th, 50th, 75th), matching the dataset description's methodology of "calculating 25th, 50th, and 75th travel times for all trucks observed moving between those counties." - `pair_moves` is stored as **text/bucketed** (e.g. `<1K`), not a raw integer. - The dataset contains **3,807,134 rows** (per the column cardinality metadata) and is licensed public domain. **Caveats:** The catalog page's own metadata links point to `data.transportation.gov`, but that view ID is no longer resolvable there — the live data and column definitions are served from `data.bts.gov`. The catalog page also lists three download formats (JSON, XML, CSV) but does not enumerate columns on-page.