Hugging Face's logo

Datasets:
prasad-gade05
/
Satellite_Traffic_Conjunction_Risk

Tasks:
Other
Modalities:
Tabular
Text
Formats:
parquet
Size:
100K - 1M
Tags:
space
orbital-debris
satellites
conjunction-analysis
sgp4
celestrak
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
Dataset Viewer
Auto-converted to Parquet Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
model: string
date: timestamp[ns]
forecast: double
ci_lower: double
ci_upper: double
mape_holdout_pct: double
crossing_99999: null
-- schema metadata --
pandas: '{"index_columns": [], "column_indexes": [], "columns": [{"name":' + 930
to
{'band_start': Value('int64'), 'lower_km': Value('float64'), 'upper_km': Value('float64'), 'regime': Value('string'), 'object_count': Value('int64'), 'debris_share': Value('float64'), 'shell_hhi': Value('float64'), 'conj_events': Value('float64'), 'mean_rel_speed': Value('float64'), 'cluster': Value('int32'), 'risk_label': Value('string')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2840, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/parquet/parquet.py", line 220, in _generate_tables
                  yield Key(file_idx, batch_idx), self._cast_table(pa_table)
                                                  ~~~~~~~~~~~~~~~~^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/parquet/parquet.py", line 156, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              model: string
              date: timestamp[ns]
              forecast: double
              ci_lower: double
              ci_upper: double
              mape_holdout_pct: double
              crossing_99999: null
              -- schema metadata --
              pandas: '{"index_columns": [], "column_indexes": [], "columns": [{"name":' + 930
              to
              {'band_start': Value('int64'), 'lower_km': Value('float64'), 'upper_km': Value('float64'), 'regime': Value('string'), 'object_count': Value('int64'), 'debris_share': Value('float64'), 'shell_hhi': Value('float64'), 'conj_events': Value('float64'), 'mean_rel_speed': Value('float64'), 'cluster': Value('int32'), 'risk_label': Value('string')}
              because column names don't match

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Orbital Commons — Satellite Traffic & Conjunction Risk Analytics Dataset

Processed and enriched datasets derived from CelesTrak's public GP / SATCAT / SOCRATES Plus feeds, produced by the Orbital Commons Medallion pipeline on 2026-08-22. Every table ships as Parquet with explicit dtypes.

Why this dataset is different

  1. OMM-native, six-digit-ID complete. Ingested exclusively via OMM JSON/CSV, so it includes satellites cataloged after 2026-07-11 ("SARAMAGO", NORAD ID
    1. that legacy-TLE pipelines silently drop.
  2. Pre-computed conjunction-event log with reported Pc. 148,985 screened events from one full SOCRATES Plus run with TCA, miss distance, closing speed, max probability and dilution - ready for analysis without re-running STK.
  3. Danger-zone clustering. 25 km altitude bands scored by a K-Means model over crowding, debris share, ownership concentration and conjunction load.
  4. ARIMA growth forecast. Weekly catalog-size forecast to 2031 with 95% CI and holdout MAPE.
  5. Documented catalog-overflow gap. The public catalog (70,355 rows) vs the official counter (100,403): the withheld 7xxxx-9xxxx block is quantified.

Tables

File Rows Grain Highlights
dim_space_object.parquet 70,355 one object name, intl designator, owner state, nation, operator attribution, type (PAY/R/B/DEB), ops status, RCS m2, launch/decay dates, mean altitude, regime band, GP freshness
fact_conjunction_events.parquet 148,985 event primary/secondary objects + names, TCA UTC, miss km, closing speed km/s, max probability, dilution km, regimes of both sides
fact_catalog_growth.parquet 25,435 day since 1957 cumulative cataloged/decayed/on-orbit, daily deltas, numbering-format era
fact_spatial_density.parquet 74 run_date x 25 km band counts by type, density per 1000 km3, owner-share HHI, clutter ratio
fact_orbital_inventory.parquet 29 run_date x regime x type object and active-payload counts
analytics_band_clusters.parquet 73 25 km band K-Means cluster + risk label (Quiet/Moderate/Busy/Critical), features
analytics_catalog_forecast.parquet 260 forecast week ARIMA(1,1,1) point forecast + 95% CI, holdout MAPE
analytics_foster_topn.parquet 50 top-risk event our Foster/Chan Pc vs SOCRATES reported Pc

Quickstart 

import pandas as pd
obj = pd.read_parquet("dim_space_object.parquet")
con = pd.read_parquet("fact_conjunction_events.parquet")

top = con.nlargest(10, "max_probability")
print(top[["primary_name", "secondary_name", "min_range_km", "max_probability"]])

Column notes

  • NORAD_CAT_ID / object_id: integer catalog number (includes six-digit IDs >= 100000).
  • object_type: PAY payload, R/B rocket body, DEB debris, UNK unknown.
  • regime: VLEO <350km, LEO-Constellation 350-650, LEO-SSO 650-950, Legacy-Debris 950-1500, Upper-LEO 1500-2000, MEO, GEO, HEO/Deep-Space.
  • max_probability: collision probability reported by SOCRATES Plus (STK/CAT).
  • shell_hhi: Herfindahl index of OWNER-state shares within a band (0-1).

Provenance & citation

Source data: CelesTrak (celestrak.org), Dr T.S. Kelso; SATCAT/SOCRATES/GP are products of the US Space Surveillance Network via CelesTrak. Please cite CelesTrak when using this dataset, e.g.:

Kelso, T.S. "CelesTrak." https://celestrak.org/ (accessed August 2026).

Processing pipeline: Orbital Commons (github.com//orbital_commons), Medallion architecture; SGP4 propagation via python-sgp4/Skyfield.

Limitations

  • Public SATCAT excludes the unpublished 7xxxx-9xxxx catalog-number block; row-level records cover ~70k of ~100k officially assigned numbers.
  • Conjunction table reflects ONE SOCRATES run (7-day lookahead); it is not a historical archive.
  • Pc re-derivations use spherical-covariance assumptions (see methodology tab of the companion dashboard).
Downloads last month
33
Number of rows:
245,278
Total file size:
10.1 MB