Source: /home/coolhand/html/datavis/data_trove/data/quirky/shipwrecks.json
Saturn profiled 6,914 rows across 14 columns. The stats below are deterministic and machine-readable; the prose is a language-model interpretation of those stats (opt-in, added after the fact, never sees raw rows).
!pip install saturn-dissect
import subprocess
subprocess.run([
"saturn", "analyze", "/home/coolhand/html/datavis/data_trove/data/quirky/shipwrecks.json",
"--findings", "data-trove-shipwrecks.json",
"--llm", "anthropic:default",
])This dataset is an OpenStreetMap-derived catalogue of 6,914 shipwrecks and related maritime hazards mapped globally. The most important thing to explore first is the `type` and `seamark_type` columns, which reveal that the overwhelming majority (~73-78%) of entries are labelled simply 'shipwreck' or 'wreck', with a long tail of submarines, aircraft, barges, and other vessels worth examining. A secondary point of interest is the high null rates across many descriptive fields — `heritage` (99.8% null), `year_sunk` (99.5% null), and `wikipedia` (95.5% null) — meaning rich contextual data exists for only a tiny fraction of wrecks, and the dataset is far more useful as a spatial inventory than a historical record. The `access` column, where populated, shows most accessible wrecks are open ('yes'), but a meaningful share require permits or are private, which could interest dive-site analysts.
citing: type.top_value · type.top_rate · seamark_type.top_value · seamark_type.top_rate · heritage.null_rate · year_sunk.null_rate · wikipedia.null_rate · access.null_rate · access.top_value · access.top_rate · row_count · lat.min · lat.max
saturn.schema() · 14 columns
| column | kind | n | null% | unique | alerts |
|---|---|---|---|---|---|
| name | text | 6,914 | 0.0% | 6,841 | near_unique |
| lat | numeric | 6,914 | 0.0% | 6,902 | outliers |
| lon | numeric | 6,914 | 0.0% | 6,910 | outliers |
| year_sunk | categorical | 6,914 | 99.5% | 36 | long_tail null_rate |
| type | categorical | 6,914 | 0.0% | 31 | long_tail |
| wikipedia | categorical | 6,914 | 95.5% | 307 | long_tail null_rate |
| wikidata | categorical | 6,914 | 94.8% | 353 | long_tail null_rate |
| description | categorical | 6,914 | 94.9% | 304 | long_tail null_rate |
| heritage | categorical | 6,914 | 99.8% | 4 | long_tail null_rate |
| access | categorical | 6,914 | 92.6% | 8 | null_rate |
| depth | categorical | 6,914 | 77.4% | 502 | null_rate |
| seamark_type | categorical | 6,914 | 6.6% | 15 | |
| osm_id | numeric | 6,914 | 0.0% | 6,914 | |
| osm_type | categorical | 6,914 | 0.0% | 2 |
| value | count | share |
|---|---|---|
| shipwreck | 5081 | 73.5% |
| wreck | 1345 | 19.5% |
| ship | 381 | 5.5% |
| barge | 27 | 0.4% |
| submarine | 18 | 0.3% |
| aircraft | 17 | 0.2% |
| plane | 10 | 0.1% |
| boat | 4 | 0.1% |
| vehicle | 3 | 0.0% |
| motor_vehicle | 3 | 0.0% |
| schooner | 2 | 0.0% |
| car | 2 | 0.0% |
| sailboat | 2 | 0.0% |
| battleship | 2 | 0.0% |
| steamer | 1 | 0.0% |
| airplane | 1 | 0.0% |
| freightcar | 1 | 0.0% |
| train | 1 | 0.0% |
| paddle steamer | 1 | 0.0% |
| motorbike | 1 | 0.0% |
| value | count | share |
|---|---|---|
| wreck | 5055 | 73.1% |
| dangerous | 598 | 8.6% |
| non-dangerous | 358 | 5.2% |
| distributed_remains | 306 | 4.4% |
| hulk | 56 | 0.8% |
| hull_showing | 46 | 0.7% |
| shoreline_construction | 14 | 0.2% |
| mast_showing | 8 | 0.1% |
| obstruction | 7 | 0.1% |
| harbour | 2 | 0.0% |
| restricted_area | 1 | 0.0% |
| plane | 1 | 0.0% |
| beacon_special_purpose | 1 | 0.0% |
| landmark | 1 | 0.0% |
| no | 1 | 0.0% |
| value | count | share |
|---|---|---|
| yes | 341 | 4.9% |
| no | 73 | 1.1% |
| permit | 27 | 0.4% |
| private | 27 | 0.4% |
| unknown | 20 | 0.3% |
| permissive | 17 | 0.2% |
| customers | 3 | 0.0% |
| foot | 1 | 0.0% |
| bin | count |
|---|---|
| -77.42 – -73.44 | 1 |
| -73.44 – -69.45 | 0 |
| -69.45 – -65.46 | 0 |
| -65.46 – -61.47 | 1 |
| -61.47 – -57.48 | 1 |
| -57.48 – -53.49 | 20 |
| -53.49 – -49.5 | 39 |
| -49.5 – -45.51 | 41 |
| -45.51 – -41.52 | 50 |
| -41.52 – -37.53 | 107 |
| -37.53 – -33.54 | 235 |
| -33.54 – -29.55 | 110 |
| -29.55 – -25.56 | 56 |
| -25.56 – -21.57 | 118 |
| -21.57 – -17.58 | 67 |
| -17.58 – -13.59 | 28 |
| -13.59 – -9.597 | 30 |
| -9.597 – -5.607 | 72 |
| -5.607 – -1.617 | 73 |
| -1.617 – 2.373 | 67 |
| 2.373 – 6.363 | 40 |
| 6.363 – 10.35 | 180 |
| 10.35 – 14.34 | 105 |
| 14.34 – 18.33 | 85 |
| 18.33 – 22.32 | 108 |
| 22.32 – 26.31 | 84 |
| 26.31 – 30.3 | 75 |
| 30.3 – 34.29 | 149 |
| 34.29 – 38.28 | 529 |
| 38.28 – 42.27 | 748 |
| 42.27 – 46.26 | 608 |
| 46.26 – 50.25 | 494 |
| 50.25 – 54.24 | 1302 |
| 54.24 – 58.23 | 846 |
| 58.23 – 62.22 | 212 |
| 62.22 – 66.21 | 85 |
| 66.21 – 70.2 | 103 |
| 70.2 – 74.19 | 39 |
| 74.19 – 78.18 | 5 |
| 78.18 – 82.17 | 1 |
| value | count | share |
|---|---|---|
| 12.4 | 11 | 0.2% |
| 16 | 11 | 0.2% |
| 18 | 11 | 0.2% |
| 15.5 | 11 | 0.2% |
| 19.2 | 11 | 0.2% |
| 1.1 | 10 | 0.1% |
| 17.4 | 10 | 0.1% |
| 15.6 | 10 | 0.1% |
| 7 | 10 | 0.1% |
| 14 | 10 | 0.1% |
| 5 | 10 | 0.1% |
| 15.1 | 10 | 0.1% |
| 6.4 | 9 | 0.1% |
| 9 | 9 | 0.1% |
| 8 | 9 | 0.1% |
| 19 | 9 | 0.1% |
| 15.2 | 9 | 0.1% |
| 20 | 9 | 0.1% |
| 16.4 | 9 | 0.1% |
| 18.5 | 9 | 0.1% |
| column | kind | null % |
|---|---|---|
| name | text | 0.0% |
| lat | numeric | 0.0% |
| lon | numeric | 0.0% |
| year_sunk | categorical | 99.5% |
| type | categorical | 0.0% |
| wikipedia | categorical | 95.5% |
| wikidata | categorical | 94.8% |
| description | categorical | 94.9% |
| heritage | categorical | 99.8% |
| access | categorical | 92.6% |
| depth | categorical | 77.4% |
| seamark_type | categorical | 6.6% |
| osm_id | numeric | 0.0% |
| osm_type | categorical | 0.0% |
| lat | lon | osm_id | |
|---|---|---|---|
| lat | +1.00 | -0.27 | +0.16 |
| lon | -0.27 | +1.00 | +0.16 |
| osm_id | +0.16 | +0.16 | +1.00 |
This column contains the names of individual shipwrecks, as confirmed by dominant top words: 'shipwreck' (5032 occurrences across 6914 rows), 'wreck', 'ss', 'uss', and 'hms'. With 6841 unique values out of 6914 rows and a near-zero null rate, it is essentially a name/label field — but the 73 duplicates (1.06% duplicate rate) are mildly surprising and may indicate the same wreck is referenced under the same name in multiple records. Lengths cluster tightly (median 20, p95 21 characters) with a long tail reaching 153, suggesting most names are concise vessel names while a minority carry extended descriptions.
Treatment: Use as a display label; investigate 73 duplicates for deduplication or record linkage before treating as a unique identifier.
saturn.columns["name"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 0 (0.0%) |
| unique | 6,841 |
| len_min | 2 |
| len_max | 153 |
| len_mean | 18.35 |
| len_median | 20 |
| len_p95 | 21 |
| word_mean | 2.058 |
| word_median | 2 |
| n_empty | 0 |
| n_duplicates | 73 |
| duplicate_rate | 0.01056 |
| vocab_size | 7,602 |
| readability_flesch_mean | 73.37 |
| emoji_rate | 0 |
| url_rate | 0 |
| one_word_rate | 0.0849 |
| allcaps_rate | 0.01403 |
| boilerplate_rate | 0 |
| alert: near_unique | 98.9% of rows are unique strings |
name.
| chars | count |
|---|---|
| 2 – 6 | 173 |
| 6 – 10 | 481 |
| 10 – 13 | 493 |
| 13 – 17 | 305 |
| 17 – 21 | 3911 |
| 21 – 25 | 1393 |
| 25 – 28 | 71 |
| 28 – 32 | 35 |
| 32 – 36 | 16 |
| 36 – 40 | 12 |
| 40 – 44 | 10 |
| 44 – 47 | 5 |
| 47 – 51 | 3 |
| 51 – 55 | 0 |
| 55 – 59 | 1 |
| 59 – 62 | 1 |
| 62 – 66 | 1 |
| 66 – 70 | 1 |
| 70 – 74 | 0 |
| 74 – 78 | 1 |
| 78 – 81 | 0 |
| 81 – 85 | 0 |
| 85 – 89 | 0 |
| 89 – 93 | 0 |
| 93 – 96 | 0 |
| 96 – 100 | 0 |
| 100 – 104 | 0 |
| 104 – 108 | 0 |
| 108 – 111 | 0 |
| 111 – 115 | 0 |
| 115 – 119 | 0 |
| 119 – 123 | 0 |
| 123 – 127 | 0 |
| 127 – 130 | 0 |
| 130 – 134 | 0 |
| 134 – 138 | 0 |
| 138 – 142 | 0 |
| 142 – 145 | 0 |
| 145 – 149 | 0 |
| 149 – 153 | 1 |
This column represents geographic latitude values, spanning from -77.42° (near Antarctica) to 82.17° (high Arctic), with 6,902 unique values across 6,914 rows. The mean (33.15°) sits notably below the median (43.85°), driven by a left skew of -1.42 — indicating a cluster of records in mid-to-high northern latitudes with a pull from southern hemisphere or equatorial observations. Roughly 12.5% of values (864 rows) are flagged as outliers, likely corresponding to polar or deep southern hemisphere coordinates that deviate from the dominant northern mid-latitude band.
Treatment: Retain as-is for geospatial modelling; consider pairing with longitude and binning into geographic regions to handle the skewed distribution and outlier polar values.
saturn.columns["lat"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 0 (0.0%) |
| unique | 6,902 |
| min | -77.42 |
| max | 82.17 |
| mean | 33.15 |
| median | 43.85 |
| std | 29.88 |
| q1 | 26.58 |
| q3 | 53.87 |
| iqr | 27.29 |
| skew | -1.417 |
| kurtosis | 0.8666 |
| n_outliers | 864 |
| outlier_rate | 0.125 |
| zero_rate | 0 |
| alert: outliers | 12.5% rows beyond 1.5 IQR |
lat. Vertical dash marks the median.
| bin | count |
|---|---|
| -77.42 – -73.44 | 1 |
| -73.44 – -69.45 | 0 |
| -69.45 – -65.46 | 0 |
| -65.46 – -61.47 | 1 |
| -61.47 – -57.48 | 1 |
| -57.48 – -53.49 | 20 |
| -53.49 – -49.5 | 39 |
| -49.5 – -45.51 | 41 |
| -45.51 – -41.52 | 50 |
| -41.52 – -37.53 | 107 |
| -37.53 – -33.54 | 235 |
| -33.54 – -29.55 | 110 |
| -29.55 – -25.56 | 56 |
| -25.56 – -21.57 | 118 |
| -21.57 – -17.58 | 67 |
| -17.58 – -13.59 | 28 |
| -13.59 – -9.597 | 30 |
| -9.597 – -5.607 | 72 |
| -5.607 – -1.617 | 73 |
| -1.617 – 2.373 | 67 |
| 2.373 – 6.363 | 40 |
| 6.363 – 10.35 | 180 |
| 10.35 – 14.34 | 105 |
| 14.34 – 18.33 | 85 |
| 18.33 – 22.32 | 108 |
| 22.32 – 26.31 | 84 |
| 26.31 – 30.3 | 75 |
| 30.3 – 34.29 | 149 |
| 34.29 – 38.28 | 529 |
| 38.28 – 42.27 | 748 |
| 42.27 – 46.26 | 608 |
| 46.26 – 50.25 | 494 |
| 50.25 – 54.24 | 1302 |
| 54.24 – 58.23 | 846 |
| 58.23 – 62.22 | 212 |
| 62.22 – 66.21 | 85 |
| 66.21 – 70.2 | 103 |
| 70.2 – 74.19 | 39 |
| 74.19 – 78.18 | 5 |
| 78.18 – 82.17 | 1 |
This column contains geographic longitude values, spanning the full valid range from -179.28° to 179.45° and covering both hemispheres. The mean (3.07°) and median (8.32°) are both modestly east of the Prime Meridian, suggesting a concentration of records in Europe/Africa, while the wide IQR of 58.74° and std of 69.12° confirm global scatter. Notably, 806 rows (11.66%) are flagged as outliers, likely corresponding to locations in the Americas or Pacific — not erroneous values, but genuine geographic extremes relative to the modal cluster.
Treatment: Use as-is or pair with latitude for spatial modelling; consider projecting to radians or embedding via geohash for ML pipelines.
saturn.columns["lon"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 0 (0.0%) |
| unique | 6,910 |
| min | -179.3 |
| max | 179.4 |
| mean | 3.067 |
| median | 8.322 |
| std | 69.12 |
| q1 | -40.75 |
| q3 | 17.99 |
| iqr | 58.74 |
| skew | 0.5093 |
| kurtosis | 0.9211 |
| n_outliers | 806 |
| outlier_rate | 0.1166 |
| zero_rate | 0 |
| alert: outliers | 11.7% rows beyond 1.5 IQR |
lon. Vertical dash marks the median.
| bin | count |
|---|---|
| -179.3 – -170.3 | 54 |
| -170.3 – -161.3 | 17 |
| -161.3 – -152.4 | 7 |
| -152.4 – -143.4 | 5 |
| -143.4 – -134.4 | 4 |
| -134.4 – -125.5 | 15 |
| -125.5 – -116.5 | 266 |
| -116.5 – -107.5 | 15 |
| -107.5 – -98.57 | 3 |
| -98.57 – -89.6 | 39 |
| -89.6 – -80.63 | 161 |
| -80.63 – -71.66 | 521 |
| -71.66 – -62.7 | 174 |
| -62.7 – -53.73 | 212 |
| -53.73 – -44.76 | 147 |
| -44.76 – -35.79 | 158 |
| -35.79 – -26.82 | 39 |
| -26.82 – -17.85 | 31 |
| -17.85 – -8.886 | 111 |
| -8.886 – 0.08197 | 631 |
| 0.08197 – 9.05 | 1144 |
| 9.05 – 18.02 | 1435 |
| 18.02 – 26.99 | 411 |
| 26.99 – 35.96 | 272 |
| 35.96 – 44.92 | 92 |
| 44.92 – 53.89 | 101 |
| 53.89 – 62.86 | 61 |
| 62.86 – 71.83 | 8 |
| 71.83 – 80.8 | 31 |
| 80.8 – 89.76 | 7 |
| 89.76 – 98.73 | 7 |
| 98.73 – 107.7 | 23 |
| 107.7 – 116.7 | 34 |
| 116.7 – 125.6 | 44 |
| 125.6 – 134.6 | 48 |
| 134.6 – 143.6 | 44 |
| 143.6 – 152.5 | 109 |
| 152.5 – 161.5 | 69 |
| 161.5 – 170.5 | 163 |
| 170.5 – 179.4 | 201 |
This column records the year (or date) a vessel was sunk, but it is almost entirely empty — 99.46% of the 6,914 rows are null, leaving only about 38 non-null values. Among those, the formats are wildly inconsistent: bare years ('1942', '1854'), full dates in multiple formats ('30 June 1890', 'June 7, 1928', '1937-09-02'), partial dates ('1963-02'), and even a range ('1643..1663'), making normalisation non-trivial. With 36 unique values across ~38 populated rows the column is near-unique relative to its populated set, and the top value '1942' appears only twice.
Treatment: Parse and normalise to a standard year integer after regex-based format detection; treat as sparse metadata and do not use as a primary feature without imputation strategy given 99.46% nulls.
saturn.columns["year_sunk"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 6,877 (99.5%) |
| unique | 36 |
| top_value | 1942 |
| top_rate | 0.05405 |
| cardinality | 36 |
| entropy | 5.155 |
| entropy_ratio | 0.9972 |
| alert: long_tail | 35 singleton categories |
| alert: null_rate | 99.5% null |
year_sunk.
| value | count | share |
|---|---|---|
| 1942 | 2 | 0.0% |
| 30 June 1890 | 1 | 0.0% |
| 1854 | 1 | 0.0% |
| 1971 | 1 | 0.0% |
| 1937-09-02 | 1 | 0.0% |
| 1963-02 | 1 | 0.0% |
| 1643..1663 | 1 | 0.0% |
| 1982 | 1 | 0.0% |
| June 7, 1928 | 1 | 0.0% |
| 1435 | 1 | 0.0% |
| 1920-12-16 | 1 | 0.0% |
| 1490s | 1 | 0.0% |
| ~1700 | 1 | 0.0% |
| 20 April 1943 | 1 | 0.0% |
| 25 May 1963 | 1 | 0.0% |
| 1710 | 1 | 0.0% |
| 1915 | 1 | 0.0% |
| 1909 | 1 | 0.0% |
| 1951 | 1 | 0.0% |
| 1952 | 1 | 0.0% |
This column classifies underwater or maritime wreck sites by vessel/object type, with 31 distinct categories across 6,914 records and no nulls. The distribution is heavily dominated by 'shipwreck' (73.5% of records) and 'wreck' (19.5%), together accounting for over 93% of all entries — the remaining 29 categories share just ~6.5%, confirming the long-tail alert. The near-redundancy between 'shipwreck', 'wreck', and 'ship' (plus 'boat', 'barge') suggests inconsistent taxonomy that may need consolidation before modelling.
Treatment: Consolidate overlapping categories (e.g. 'shipwreck'/'wreck'/'ship') into a canonical taxonomy, then one-hot or target-encode for modelling.
saturn.columns["type"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 0 (0.0%) |
| unique | 31 |
| top_value | shipwreck |
| top_rate | 0.7349 |
| cardinality | 31 |
| entropy | 1.166 |
| entropy_ratio | 0.2353 |
| alert: long_tail | 17 singleton categories |
type.
| value | count | share |
|---|---|---|
| shipwreck | 5081 | 73.5% |
| wreck | 1345 | 19.5% |
| ship | 381 | 5.5% |
| barge | 27 | 0.4% |
| submarine | 18 | 0.3% |
| aircraft | 17 | 0.2% |
| plane | 10 | 0.1% |
| boat | 4 | 0.1% |
| vehicle | 3 | 0.0% |
| motor_vehicle | 3 | 0.0% |
| schooner | 2 | 0.0% |
| car | 2 | 0.0% |
| sailboat | 2 | 0.0% |
| battleship | 2 | 0.0% |
| steamer | 1 | 0.0% |
| airplane | 1 | 0.0% |
| freightcar | 1 | 0.0% |
| train | 1 | 0.0% |
| paddle steamer | 1 | 0.0% |
| motorbike | 1 | 0.0% |
This column stores Wikipedia article links associated with dataset entities (ships and aircraft), formatted as language-prefixed slugs (e.g., 'en:SS Edmund Fitzgerald', 'fr:Armorique (navire)'). The null rate is extremely high at 95.47%, meaning only ~313 of 6,914 rows have any Wikipedia reference. Among populated values, cardinality is very high (307 unique values across ~313 non-null rows), with the top value appearing only 4 times — indicating near-unique coverage and a long-tail distribution. A language mix is present (English 'en:', French 'fr:', Arabic 'ar:'), which could complicate any downstream lookup or joining logic.
Treatment: Use as an optional enrichment link; do not use in modelling due to 95.47% nulls and near-unique cardinality; parse language prefix if language-specific resolution is needed.
saturn.columns["wikipedia"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 6,601 (95.5%) |
| unique | 307 |
| top_value | en:SS Edmund Fitzgerald |
| top_rate | 0.01278 |
| cardinality | 307 |
| entropy | 8.245 |
| entropy_ratio | 0.998 |
| alert: long_tail | 303 singleton categories |
| alert: null_rate | 95.5% null |
wikipedia.
| value | count | share |
|---|---|---|
| en:SS Edmund Fitzgerald | 4 | 0.1% |
| fr:Armorique (navire) | 2 | 0.0% |
| en:Curtiss C-46 Commando | 2 | 0.0% |
| en:USS Amesbury | 2 | 0.0% |
| en:SS America (1939) | 1 | 0.0% |
| ar:سفينة زيستل جورم | 1 | 0.0% |
| en:BOS 400 | 1 | 0.0% |
| en:New Carissa | 1 | 0.0% |
| en:MV Cita | 1 | 0.0% |
| en:SS Richard Montgomery | 1 | 0.0% |
| en:Kroombit Tops National Park#Crash site | 1 | 0.0% |
| en:Astron (ship) | 1 | 0.0% |
| en:SS Yongala | 1 | 0.0% |
| et:Raketa (laev, 1949) | 1 | 0.0% |
| en:USNS General Hoyt S. Vandenberg (T-AGM-10) | 1 | 0.0% |
| en:USS Oriskany (CV-34) | 1 | 0.0% |
| en:USS Massachusetts (BB-2) | 1 | 0.0% |
| en:Water Witch (schooner) | 1 | 0.0% |
| en:Burlington Bay Horse Ferry | 1 | 0.0% |
| en:Champlain II | 1 | 0.0% |
This column stores Wikidata entity identifiers (Q-codes), linking dataset records to Wikidata knowledge graph entries. The most striking signal is the extreme null rate of 94.78%, meaning only ~360 of 6,914 rows carry a Wikidata link at all. Among the 353 unique Q-codes present, the distribution is nearly flat — the top value 'Q1286267' appears only 4 times, entropy ratio is 0.998, and the long-tail alert confirms almost no repeated values — suggesting each populated row points to a distinct entity with minimal reuse.
Treatment: Use as an optional foreign key to enrich records via Wikidata API lookup; do not use as a feature directly given 94.78% null rate.
saturn.columns["wikidata"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 6,553 (94.8%) |
| unique | 353 |
| top_value | Q1286267 |
| top_rate | 0.01108 |
| cardinality | 353 |
| entropy | 8.446 |
| entropy_ratio | 0.9979 |
| alert: long_tail | 347 singleton categories |
| alert: null_rate | 94.8% null |
wikidata.
| value | count | share |
|---|---|---|
| Q1286267 | 4 | 0.1% |
| Q959696 | 2 | 0.0% |
| Q215692 | 2 | 0.0% |
| Q2862787 | 2 | 0.0% |
| Q1145708 | 2 | 0.0% |
| Q11675753 | 2 | 0.0% |
| Q463091 | 1 | 0.0% |
| Q32276 | 1 | 0.0% |
| Q115709756 | 1 | 0.0% |
| Q14213801 | 1 | 0.0% |
| Q2877353 | 1 | 0.0% |
| Q7006376 | 1 | 0.0% |
| Q6719379 | 1 | 0.0% |
| Q41771616 | 1 | 0.0% |
| Q7394285 | 1 | 0.0% |
| Q7420193 | 1 | 0.0% |
| Q1359321 | 1 | 0.0% |
| Q4811601 | 1 | 0.0% |
| Q1424289 | 1 | 0.0% |
| Q1618842 | 1 | 0.0% |
This column contains free-text descriptions of maritime wrecks or nautical features, with entries referencing WWII-era vessels, jetties, fishing boats, and abandoned craft. The most striking signal is the 94.87% null rate — nearly the entire dataset lacks a description — making this column nearly unusable at scale. Among the 304 unique values across 6,914 rows, entropy is very high (8.05, ratio 0.976), indicating wide diversity in phrasing, and a language mix is evident (e.g., French 'Chaloupe abandonnée à terre' alongside English entries).
Treatment: Exclude from modelling due to 94.87% null rate; if used, tokenize and embed the 5.13% populated values, and flag language mixing before NLP processing.
saturn.columns["description"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 6,559 (94.9%) |
| unique | 304 |
| top_value | WWII era concrete fuel barge converted into breakwater |
| top_rate | 0.03944 |
| cardinality | 304 |
| entropy | 8.052 |
| entropy_ratio | 0.9763 |
| alert: long_tail | 282 singleton categories |
| alert: null_rate | 94.9% null |
description.
| value | count | share |
|---|---|---|
| WWII era concrete fuel barge converted into breakwater | 14 | 0.2% |
| Wrecks | 7 | 0.1% |
| WWII concrete barge sunk as part of jetty, partially covered by jetty and fill | 5 | 0.1% |
| Location is based on divers hand drawn maps. Due to the wreak breaking up and salvage, the wreak is scattered over a large area. | 4 | 0.1% |
| Partially sunken ships | 4 | 0.1% |
| Concrete petrol barge sunk as part of breakwater | 4 | 0.1% |
| Wrecks of Zulu fishing boats | 3 | 0.0% |
| Chaloupe abandonnée à terre | 3 | 0.0% |
| WWII era concrete fuel barge sunk as part of jetty foundation | 3 | 0.0% |
| Armada Ship | 2 | 0.0% |
| remains of sunken wooden boats | 2 | 0.0% |
| Hundido el 3 de julio de 1898 durante la batalla naval de Santiago de Cuba en la Guerra Hispano-Cubana-Norteamericana. | 2 | 0.0% |
| 09/09/2006 : Epave en bois, longue de 20 mètres, large de 4 mètres et haute de 3 mètres. | 2 | 0.0% |
| Steamer | 2 | 0.0% |
| Iron-hulled barque | 2 | 0.0% |
| On shore wreck of a small abandoned wooden ship. | 2 | 0.0% |
| Épave | 2 | 0.0% |
| Dojście do wraków w zasadzie wolne. Jednak mogą wystąpić sytuacje gdy będzie to utrudnione lub niemożliwe. | 2 | 0.0% |
| Wrecked sealing vessel | 2 | 0.0% |
| Staten Island boat graveyard | 2 | 0.0% |
This column appears to encode a 'heritage' flag or classification with only 4 distinct values ('1', '2', 'no', 'yes'), suggesting a binary or ordinal attribute that may have been inconsistently encoded across sources. The critical finding is a null rate of 99.81%, meaning only 13 of 6,914 rows have any value at all — rendering this column nearly useless for modelling. Among those 13 non-null values, '2' dominates at 76.9%, while 'no', 'yes', and '1' each appear only once, indicating a mixed encoding scheme (numeric vs. boolean strings) on an already negligible sample.
Treatment: Drop this column; 99.81% null rate and only 13 non-null observations make it statistically unusable.
saturn.columns["heritage"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 6,901 (99.8%) |
| unique | 4 |
| top_value | 2 |
| top_rate | 0.7692 |
| cardinality | 4 |
| entropy | 1.145 |
| entropy_ratio | 0.5726 |
| alert: long_tail | 3 singleton categories |
| alert: null_rate | 99.8% null |
heritage.
| value | count | share |
|---|---|---|
| 2 | 10 | 0.1% |
| no | 1 | 0.0% |
| yes | 1 | 0.0% |
| 1 | 1 | 0.0% |
This column appears to encode access permission or restriction tags for geographic features (likely OpenStreetMap-style data), with values such as 'yes', 'no', 'permit', 'private', 'permissive', and 'customers'. The striking finding is a 92.64% null rate — only 509 of 6,914 rows carry a value — meaning this attribute is almost entirely absent from the dataset. Among the non-null values, 'yes' dominates heavily at 66.99% of populated rows, suggesting most tagged features have open access.
Treatment: Flag extreme sparsity (92.64% nulls); treat nulls as a distinct 'untagged' category or drop column if missingness renders it uninformative for modelling.
saturn.columns["access"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 6,405 (92.6%) |
| unique | 8 |
| top_value | yes |
| top_rate | 0.6699 |
| cardinality | 8 |
| entropy | 1.647 |
| entropy_ratio | 0.549 |
| alert: null_rate | 92.6% null |
access.
| value | count | share |
|---|---|---|
| yes | 341 | 4.9% |
| no | 73 | 1.1% |
| permit | 27 | 0.4% |
| private | 27 | 0.4% |
| unknown | 20 | 0.3% |
| permissive | 17 | 0.2% |
| customers | 3 | 0.0% |
| foot | 1 | 0.0% |
This column represents a numeric depth measurement (likely in meters or similar units) stored as a categorical string, with values ranging from small decimals like '1.1' to integers like '19.2'. The most striking signal is a null rate of 77.36%, meaning only ~1,556 of 6,914 rows carry a value — a severe missingness that warrants investigation into whether it is structurally absent (e.g., not applicable to certain record types) or a data quality issue. Among populated rows, cardinality is very high (502 unique values) with an entropy ratio of 0.956, indicating nearly uniform spread and essentially no dominant depth value — the top value '12.4' appears only 11 times. The column should be cast to numeric before any modelling use.
Treatment: Cast to float, investigate structural vs. random missingness before imputing or dropping nulls, then use as a numeric feature.
saturn.columns["depth"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 5,349 (77.4%) |
| unique | 502 |
| top_value | 12.4 |
| top_rate | 0.007029 |
| cardinality | 502 |
| entropy | 8.579 |
| entropy_ratio | 0.9563 |
| alert: null_rate | 77.4% null |
depth.
| value | count | share |
|---|---|---|
| 12.4 | 11 | 0.2% |
| 16 | 11 | 0.2% |
| 18 | 11 | 0.2% |
| 15.5 | 11 | 0.2% |
| 19.2 | 11 | 0.2% |
| 1.1 | 10 | 0.1% |
| 17.4 | 10 | 0.1% |
| 15.6 | 10 | 0.1% |
| 7 | 10 | 0.1% |
| 14 | 10 | 0.1% |
| 5 | 10 | 0.1% |
| 15.1 | 10 | 0.1% |
| 6.4 | 9 | 0.1% |
| 9 | 9 | 0.1% |
| 8 | 9 | 0.1% |
| 19 | 9 | 0.1% |
| 15.2 | 9 | 0.1% |
| 20 | 9 | 0.1% |
| 16.4 | 9 | 0.1% |
| 18.5 | 9 | 0.1% |
This column contains a nautical/maritime classification type for seamarks, most likely drawn from an OpenStreetMap or similar marine charting schema. The distribution is severely dominated by 'wreck' at 78.3% of 6,914 rows, with the next largest category 'dangerous' at only 8.6%, giving a low entropy ratio of 0.307. The mix of subtypes (hull_showing, mast_showing, distributed_remains, hulk) suggests these are sub-classifications of wrecks that could have been normalized into a hierarchy rather than a flat taxonomy. The 6.6% null rate warrants attention if completeness matters for navigation safety contexts.
Treatment: One-hot or ordinal encode; consider grouping wreck subtypes (hull_showing, mast_showing, hulk, distributed_remains) into a parent 'wreck' hierarchy before modelling, and impute or flag the 6.6% nulls.
saturn.columns["seamark_type"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 459 (6.6%) |
| unique | 15 |
| top_value | wreck |
| top_rate | 0.7831 |
| cardinality | 15 |
| entropy | 1.2 |
| entropy_ratio | 0.307 |
seamark_type.
| value | count | share |
|---|---|---|
| wreck | 5055 | 73.1% |
| dangerous | 598 | 8.6% |
| non-dangerous | 358 | 5.2% |
| distributed_remains | 306 | 4.4% |
| hulk | 56 | 0.8% |
| hull_showing | 46 | 0.7% |
| shoreline_construction | 14 | 0.2% |
| mast_showing | 8 | 0.1% |
| obstruction | 7 | 0.1% |
| harbour | 2 | 0.0% |
| restricted_area | 1 | 0.0% |
| plane | 1 | 0.0% |
| beacon_special_purpose | 1 | 0.0% |
| landmark | 1 | 0.0% |
| no | 1 | 0.0% |
This column is an OpenStreetMap (OSM) object identifier — a large integer surrogate key assigned by the OSM platform to geographic features. Every one of the 6,914 rows has a distinct value with zero nulls, confirming it functions purely as a unique identifier. The value range (13 M to ~13.7 B) and flat distribution (kurtosis −1.45, near-uniform spread across a ~8.4 B IQR) are consistent with OSM's incrementally assigned ID space across different data vintages. No outliers are flagged and the mild positive skew (0.44) suggests a slight concentration of older, lower-numbered IDs.
Treatment: Retain as a join/lookup key to OSM data; drop from any model feature set as it carries no predictive signal.
saturn.columns["osm_id"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 0 (0.0%) |
| unique | 6,914 |
| min | 1.306e+07 |
| max | 1.371e+10 |
| mean | 5.365e+09 |
| median | 3.145e+09 |
| std | 4.464e+09 |
| q1 | 1.348e+09 |
| q3 | 9.788e+09 |
| iqr | 8.44e+09 |
| skew | 0.4355 |
| kurtosis | -1.453 |
| n_outliers | 0 |
| outlier_rate | 0 |
| zero_rate | 0 |
osm_id. Vertical dash marks the median.
| bin | count |
|---|---|
| 1.306e+07 – 3.555e+08 | 262 |
| 3.555e+08 – 6.979e+08 | 451 |
| 6.979e+08 – 1.04e+09 | 454 |
| 1.04e+09 – 1.383e+09 | 614 |
| 1.383e+09 – 1.725e+09 | 699 |
| 1.725e+09 – 2.068e+09 | 184 |
| 2.068e+09 – 2.41e+09 | 137 |
| 2.41e+09 – 2.752e+09 | 46 |
| 2.752e+09 – 3.095e+09 | 213 |
| 3.095e+09 – 3.437e+09 | 623 |
| 3.437e+09 – 3.78e+09 | 54 |
| 3.78e+09 – 4.122e+09 | 37 |
| 4.122e+09 – 4.464e+09 | 58 |
| 4.464e+09 – 4.807e+09 | 44 |
| 4.807e+09 – 5.149e+09 | 48 |
| 5.149e+09 – 5.492e+09 | 40 |
| 5.492e+09 – 5.834e+09 | 49 |
| 5.834e+09 – 6.177e+09 | 90 |
| 6.177e+09 – 6.519e+09 | 140 |
| 6.519e+09 – 6.861e+09 | 94 |
| 6.861e+09 – 7.204e+09 | 67 |
| 7.204e+09 – 7.546e+09 | 40 |
| 7.546e+09 – 7.889e+09 | 40 |
| 7.889e+09 – 8.231e+09 | 64 |
| 8.231e+09 – 8.573e+09 | 43 |
| 8.573e+09 – 8.916e+09 | 59 |
| 8.916e+09 – 9.258e+09 | 374 |
| 9.258e+09 – 9.601e+09 | 135 |
| 9.601e+09 – 9.943e+09 | 125 |
| 9.943e+09 – 1.029e+10 | 131 |
| 1.029e+10 – 1.063e+10 | 20 |
| 1.063e+10 – 1.097e+10 | 45 |
| 1.097e+10 – 1.131e+10 | 73 |
| 1.131e+10 – 1.166e+10 | 37 |
| 1.166e+10 – 1.2e+10 | 928 |
| 1.2e+10 – 1.234e+10 | 85 |
| 1.234e+10 – 1.268e+10 | 88 |
| 1.268e+10 – 1.302e+10 | 141 |
| 1.302e+10 – 1.337e+10 | 54 |
| 1.337e+10 – 1.371e+10 | 28 |
This column encodes the OpenStreetMap geometry type, distinguishing between point features ('node') and linear/polygonal features ('way'). With only 2 distinct values across 6,914 rows and zero nulls, it is a clean binary categorical. The distribution is moderately skewed: 'node' accounts for 72.3% (5,000 rows) versus 'way' at 27.7% (1,914 rows), which is consistent with OSM datasets where point POIs outnumber way geometries.
Treatment: One-hot encode or map to binary flag (node=1, way=0) before modelling; consider whether geometry type meaningfully differs from other features in the pipeline.
saturn.columns["osm_type"].stats
| stat | value |
|---|---|
| n | 6,914 |
| nulls | 0 (0.0%) |
| unique | 2 |
| top_value | node |
| top_rate | 0.7232 |
| cardinality | 2 |
| entropy | 0.8511 |
| entropy_ratio | 0.8511 |
osm_type.
| value | count | share |
|---|---|---|
| node | 5000 | 72.3% |
| way | 1914 | 27.7% |
click to copy
@misc{saturn-data-trove-shipwrecks-2026,
author = {Steuber, Luke},
title = {Saturn reading: data trove shipwrecks},
year ={2026},
howpublished = {\url{https://dr.eamer.dev/saturn/view/data-trove-shipwrecks}},
note = {Profiled with saturn-dissect v0.2.0, prompt saturn-insight-v2, model anthropic:default},
}
Steuber, L. (2026). Saturn reading: data trove shipwrecks. Source: /home/coolhand/html/datavis/data_trove/data/quirky/shipwrecks.json. Profiled with saturn-dissect v0.2.0 (saturn-insight-v2, anthropic:default). Retrieved from https://dr.eamer.dev/saturn/view/data-trove-shipwrecks