juliensimon/esa-rosetta-observations

---
license: cc-by-4.0
pretty_name: "ESA Rosetta Observations"
language:
- en
description: "Complete observation metadata catalog from the ESA Rosetta mission to Comet 67P/Churyumov-Gerasimenko (8,214,033 observations across 15 instruments). Static dataset from the ESA Planetary Science Archive."
task_categories:
- tabular-classification
tags:
- space
- comet
- 67p
- rosetta
- philae
- esa
- planetary-science
- open-data
- tabular-data
- parquet
size_categories:
- 10M<n<100M
configs:
- config_name: default
data_files:
- split: train
path: data/**/*.parquet
default: true
- config_name: rosina
data_files:
- split: train
path: data/rosina/*.parquet
- config_name: osiris
data_files:
- split: train
path: data/osiris/*.parquet
- config_name: rpc
data_files:
- split: train
path: data/rpc/*.parquet
- config_name: alice
data_files:
- split: train
path: data/alice/*.parquet
- config_name: cosima
data_files:
- split: train
path: data/cosima/*.parquet
- config_name: virtis
data_files:
- split: train
path: data/virtis/*.parquet
- config_name: navcam
data_files:
- split: train
path: data/navcam/*.parquet
- config_name: midas
data_files:
- split: train
path: data/midas/*.parquet
---

# ESA Rosetta Observations

<div align="center">
<img src="banner.jpg" alt="Rosetta spacecraft approaching Comet 67P/Churyumov-Gerasimenko" width="400">
<p><em>Credit: NASA/ESA</em></p>
</div>

*Part of the [Solar System Datasets](https://huggingface.co/collections/juliensimon/solar-system-datasets-67dbfa3057e38241e7ea2aee) and [Planetary Science Datasets](https://huggingface.co/collections/juliensimon/planetary-science-datasets-69c24cb17e3db14fc30f0716) collections on Hugging Face.*

Complete observation metadata catalog from the **ESA Rosetta** mission to Comet 67P/Churyumov-Gerasimenko — **8,214,033**
observations across 15 instruments.

## Dataset description

Rosetta was ESA's groundbreaking mission to Comet 67P/Churyumov-Gerasimenko. Launched in 2004, it became the first spacecraft to orbit a comet (August 2014) and deployed the Philae lander for the first-ever soft landing on a comet nucleus (November 2014). The mission studied the comet's nucleus, coma, and plasma environment in unprecedented detail over a 2-year escort phase, witnessing the comet's approach to and retreat from the Sun. Rosetta's ROSINA mass spectrometer detected molecular oxygen and glycine (an amino acid) in the coma — findings with profound implications for our understanding of Solar System chemistry and the origin of water and prebiotic molecules on Earth. OSIRIS captured stunning images of 67P's bilobed nucleus revealing active pits, cliffs, and dust jets. The mission concluded with a controlled descent onto the comet surface on September 30, 2016.

This dataset contains the full observation metadata from the ESA Planetary Science Archive (PSA), conforming to the EPN-TAP standard. Each row represents one observation or data granule, with timing, spatial coverage, instrument parameters, and access URLs.

## Instruments

- **ROSINA**: Mass spectrometer suite measuring gas composition of 67P's coma
- **OSIRIS**: Camera system (NAC + WAC) imaging the comet nucleus and coma
- **RPC**: Plasma consortium with 5 sensors studying the plasma environment
- **ALICE**: Ultraviolet spectrograph characterizing coma and surface composition
- **COSIMA**: Dust mass spectrometer analyzing collected dust grains
- **RSI**: Radio science instrument probing the comet interior and coma
- **MIDAS**: Atomic force microscope imaging dust grain morphology at nm scale
- **VIRTIS**: Visible/infrared imaging spectrometer mapping surface and coma
- **NAVCAM**: Navigation camera providing wide-field context images
- **GIADA**: Dust impact detector measuring grain flux and momentum
- **SREM**: Standard radiation environment monitor
- **CONSERT**: Radar tomography of the comet nucleus (with Philae)
- **MIRO**: Microwave spectrometer measuring subsurface temperature and volatile outgassing
- **SESAME**: Philae surface science package (acoustic, permittivity, dust)
- **ROMAP**: Philae magnetometer and plasma monitor

## Instrument breakdown

- **ROSINA**: 4,654,393 observations
- **OSIRIS**: 2,352,819 observations
- **RPC**: 322,145 observations
- **ALICE**: 258,704 observations
- **COSIMA**: 185,045 observations
- **RSI**: 257,600 observations
- **MIDAS**: 64,561 observations
- **VIRTIS**: 47,850 observations
- **NAVCAM**: 46,568 observations
- **GIADA**: 7,968 observations
- **SREM**: 6,563 observations
- **CONSERT**: 3,248 observations
- **MIRO**: 2,544 observations
- **SESAME**: 2,328 observations
- **ROMAP**: 1,697 observations

## Schema (key columns)

| Column | Type | Description |
|--------|------|-------------|
| `granule_uid` | string | Unique observation identifier |
| `granule_gid` | string | Group identifier |
| `obs_id` | string | Observation ID |
| `dataproduct_type` | string | Data product type (e.g. spectrum, image, profile) |
| `target_name` | string | Target body (67P, etc.) |
| `target_class` | string | Target class (comet, etc.) |
| `instrument_name` | string | Instrument name |
| `time_min` | float64 | Observation start time (JD) |
| `time_max` | float64 | Observation end time (JD) |
| `c1min`/`c1max` | float64 | Spatial coordinate 1 range |
| `c2min`/`c2max` | float64 | Spatial coordinate 2 range |
| `spectral_range_min`/`max` | float64 | Spectral range bounds |
| `processing_level` | string | Data processing level |
| `creation_date` | string | Data product creation date |
| `access_url` | string | URL to access the data product |
| `access_format` | string | Data format (e.g. application/x-pds) |

The full schema contains up to ~50 columns following the EPN-TAP standard.

## Quick stats

- **8,214,033** total observations
- **15** instruments
- **12** distinct targets
- Time span: JD 2451910.5 – 2457662.5

## Usage

```python
from datasets import load_dataset

# Load a single instrument (fast, low memory)
rosina = load_dataset("juliensimon/esa-rosetta-observations", "rosina", split="train")
print(f"{len(rosina):,} ROSINA observations")

# Load all instruments at once (14M+ rows, needs significant RAM)
ds = load_dataset("juliensimon/esa-rosetta-observations", split="train")

# Available configs: rosina, osiris, rpc, alice, cosima, virtis, navcam, midas
```

## Data source

[ESA Planetary Science Archive](https://psa.esa.int/) — EPN-TAP service at
`https://psa.esa.int/psa-tap/tap/sync`.

## Related datasets

- [esa-exomars-tgo-observations](https://huggingface.co/datasets/juliensimon/esa-exomars-tgo-observations) — ESA ExoMars TGO observation catalog
- [esa-mars-express-observations](https://huggingface.co/datasets/juliensimon/esa-mars-express-observations) — ESA Mars Express observation catalog
- [comets](https://huggingface.co/datasets/juliensimon/comets) — JPL comet orbital elements

## Pipeline

Source code: [juliensimon/space-datasets](https://github.com/juliensimon/space-datasets)

## Support

If you find this dataset useful, please give it a ❤️ on the [dataset page](https://huggingface.co/datasets/juliensimon/esa-rosetta-observations) and share feedback in the Community tab! Also consider giving a ⭐️ to the [space-datasets](https://github.com/juliensimon/space-datasets) repo.

## Citation

```bibtex
@dataset{rosetta_observations,
author = {Simon, Julien},
title = {ESA Rosetta Observations},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/juliensimon/esa-rosetta-observations},
note = {Based on ESA Planetary Science Archive (PSA) EPN-TAP service}
}
```

## License

[CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)